Early-Onset Colorectal Cancers Exhibit Distinctive Placental-Like Features

Early-Onset Colorectal Cancers Exhibit Distinctive Placental-Like Features | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Early-Onset Colorectal Cancers Exhibit Distinctive Placental-Like Features Gianluca Mauri, Lucia Santorelli, Federica Marasca, Valeria Ranzani, and 34 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7193450/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract The incidence of early-onset colorectal cancer (EO-CRC, diagnosed earlier than age 50) is rising worldwide. Despite distinctive clinicopathological features, whether EO-CRC represents a biologically distinct entity from standard-onset CRC (SO-CRC) remains unclear. To investigate molecular underpinnings of EO-CRC, we applied high-resolution label-free mass spectrometry coupled with transcriptomic approaches on primary tumours, healthy mucosae, and metastases of EO-CRC and SO-CRC patients. Most EO-CRC displayed reactivation of placental-like programs and HERVH reactivation, a family of retrotransposons maintaining pluripotency. These features were retained in patient-derived organoids (PDOs) showing sensitivity to pharmacological ATR (Ataxia Telangiectasia and Rad3-related) inhibition. While these findings point to specific EO-CRC vulnerabilities, they require further validation in larger geographically distinct series. These findings distinguish most EO-CRC from SO-CRC as they possess specific placental mimicry and HERVH reactivation. The placental mimicry and HERVH reactivation observed may provide a molecular rationale for EO-CRC aggressive behaviour and suggest potential avenues for therapeutic targeting. Biological sciences/Cancer/Gastrointestinal cancer/Colorectal cancer Health sciences/Medical research/Translational research Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction The incidence of early-onset colorectal cancers (EO-CRC), defined as colorectal cancer (CRC) occurring in adults under 50 years of age, is rising globally 1 – 5 . EO-CRC often presents with clinicopathological features indicative of intrinsic aggressiveness, including mucinous histology, poor differentiation, and peritoneal spread associated with poorer response to therapy and survival 6 – 8 . Whether these characteristics reflect a distinct underlying biology remains unknown 9 – 11 . Elucidating EO-CRC molecular features is expected to inform the development of tailored screening and therapeutic approaches 12 , 13 . Multiple studies have shown that EO-CRC and standard onset colorectal cancer (SO-CRC) genomic landscapes are broadly similar, with no significant differences in the prevalence of genomic alterations occurring in RAS , APC , and TP53 genes 14 , 15 . Recent studies have reported an enrichment of single base substitution (SBS) signature 88 and small insertion and deletion (ID) signature 18 in EO-CRC, suggesting a potential causative role for the bacterial colibactin genotoxin in 30% of cases 16 . However, given their geographic variability 16 , additional causative factors are likely to contribute to EO-CRC rising incidence. Overall, whether EO-CRC are biologically and molecularly distinct from SO-CRC remain unresolved 12 . We previously hypothesized that cancer development involve the reactivation of molecular pathways active in para-physiological states such as pregnancy 17 . Placentation is a highly orchestrated developmental process enabling rapid, regulated growth in mammals. To establish a functional placenta, trophoblast cells acquire properties strikingly similar to those observed in cancers as they invade healthy tissue, induce neovascularization, and create an immune-tolerant microenvironment 17 . This physiological mimicry highlights a shared biology, where key malignancy hallmarks, including tissue invasion, immune evasion, and sustained proliferative signalling are naturally recapitulated during placentation 17 . As in cancer, placental proliferation is driven by elevated IGF/MAPK signalling, anti-apoptotic pathways activation, genome duplication events leading to polyploidy 18 and occurrence of extensive mutations with the intestinal epithelium associated SBS18 signature 19 being the most prevalent 20 . Strikingly, the normal placenta exhibits high levels of copy number alterations (CNAs), a feature rarely seen in healthy tissues but frequently in CRC 21 . However, the extent to which placental mimicry contributes to CRC pathogenesis remains unclear. Several genomic alterations shared by the placenta and CRCs may arise from DNA replication stress, a condition characterized by slowed or stalled replication forks leading to genomic instability through DNA breaks, fragile site activation, and error-prone repair, resulting in copy number gains or losses 22 . In previous work leading to this study, we demonstrated that embryonic stem cells subjected to replication stress reactivate endogenous retrotransposons (ERVs), driving a cell fate transition toward trophoblast and placental lineages 23 . Among ERVs, Human Endogenous Retroviruses (HERVs), representing nearly 8% of the human genome 24 , have emerged as key transcriptional regulators of placental development 25 . The HERVH subfamily, in particular, plays a central role in maintaining pluripotency and guiding trophoblast differentiation by contributing to core self-renewal transcriptional networks 25 – 28 . Aberrant reactivation of HERVs has been consistently reported in multiple cancers, reflecting a pattern consistent with onco-exaptation , the repurposing of retroelements as cancer regulatory elements 29 – 31 . In this study, we employed a multi-omics approach combining high-resolution label-free mass spectrometry (MS)-based proteomics with transcriptomic analyses. This strategy uncovered a distinctive hallmark of EO-CRC: the reactivation of placental-like transcriptional programs and the aberrant expression of HERVH retroelements. These findings suggest that the reactivation of placental-like transcriptional programs may underlie the aggressiveness of EO-CRC, providing a novel perspective on its pathogenesis and opening avenues towards tailored preventive, diagnostic and therapeutic strategies. Results EO-CRC tumour samples were collected from individuals aged 45 years old or younger, and from a control cohort of SO-CRC patients, diagnosed at age 50 or older ( Supplementary Fig. 1 ). EO-CRC samples were collected from a cohort of 33 EO-CRC patients, of whom 29/33 (88%) diagnosed earlier than age 40. EO-CRC patients diagnosed with hereditary cancer predisposition syndromes (HCPSs) were not studied, and we elected to prioritize study of EO-CRC diagnosed under 40 years of age to ensure a more biologically informative cohort. The clinicopathological features of both EO-CRC and SO-CRC (N = 27) patients are aligned with previous reports 9 , 14 . As expected, most of EO-CRCs were in the rectum or left colon, frequently poorly differentiated, and with mucinous features ( Supplementary Table 1A and B ). Proteomics unveils specific placental features of early-onset colorectal cancer . To gain insight into EO-CRC biological and molecular processes, we leveraged the power of high-resolution label-free MS-based proteomics, allowing proteins comprehensive profiling and disease mechanisms elucidation 32 . By capturing dynamic proteomic changes, MS facilitates the identification of novel biomarkers and therapeutic targets. We applied this approach to compare the protein profiles of primary EO-CRC and SO-CRC tumors, as well as their matched adjacent mucosae (N-EOCRC and N-SOCRC), as controls (Fig. 1A). Firstly, we applied label-free quantitation to assess the protein content in EO-CRC (N = 12) and SO-CRC (N = 7). Results revealed significant proteomic changes upon cancerous conditions in accordance with the clinical classification, as evidenced by distinct separation along principal component 2 (Fig. 1B). Motivated by these findings, we proceeded to compare the EO-CRC and SO-CRC proteomes, conducting statistical analyses to explore differential proteins expression. We uncovered 2675 dysregulated proteins, mostly upregulated in EO-CRC (Fig. 1C, Supplementary Table 2 ). Gene Ontology classification biological process (GOBP), cellular component (GOCC) showed a notable upregulation of proteins associated with the mitochondria, endoplasmic reticulum, lysosome, and vesicles in EO-CRC compared to SO-CRC ( Supplementary Fig. 2A-B, Supplementary Table 2 ). Of note, several of the significantly deregulated proteins in comparing EO-CRC and SO-CRC were associated with cellular component biogenesis, protein modification, response to replication stress, and system development (Fig. 1D, Supplementary Table 2 ), processes often linked with tumour growth and progression. The developmental category emerged with multiple GOBP terms related to embryonic development significantly enriched in the deregulated proteome of the EO-CRC cohort (Fig. 1E, Supplementary Table 2 ). This proteomic profile led us to hypothesize that developmental pathways play a pivotal role in the molecular mechanisms driving EO-CRC, potentially reflecting a reactivation or dysregulation of biological programs typically confined to early developmental stages. To further assess this hypothesis, we investigated whether the differential protein abundances detected between EO-CRC and SO-CRC showed tissue-specific expression patterns, to confirm the developmental pathways relevance. As expected, a large proportion of the differential proteins were distinctive colon-resident. Intriguingly, we identified a significant number of deregulated proteins as typically expressed in highly undifferentiated, embryonic and placental tissues/cell lines (Fig. 1C and F, Supplementary Table 2 ), known for their role in development and regeneration processes. Although gastrointestinal tissues like colon and rectum appeared to dominate the sample proteome landscape, EO-CRC appeared to involve developmentally associated tissues, markedly the placenta. To validate this findings, we repeated the tissue specificity analysis by comparing the EO-CRC deregulated proteome with additional available tissue databases 33 – 35 . These independent investigations further corroborated the over-representation of a placental and trophoblast protein core predominantly expressed in EO-CRC ( Supplementary Fig. 2C, Fig. 1F ), providing a potential mechanistic explanation for the differences between EO-CRC and SO-CRC previously highlighted by the Principal Component Analysis (PCA) (Fig. 1B). Among these proteins, we found proteins involved in trophoblast fusion (e.g., GPC4, LAMA3, ITGB4), immune modulators/defence response (e.g., HLA-DRB5, CD14, BST2), cell adhesion/invasion (e.g., ITGA1, ITGA5, ITGB1, F11R, VASP), proteostasis/stress response (e.g., HSPA5, NUCB1, NUCB2, OTUB1), vesicle trafficking (VAMP7, RAB3A, VPS37C, VPS4B), and several others ( Supplementary Fig. 2F, Supplementary Table 2 ). Proteins involved in DNA metabolism were also predominantly expressed in EO-CRC. In particular, proteins regulating DNA replication under stress such as SMARCAL1, CDK1, CDK2, and RRM2B, and several DNA repair proteins, including ERCC2, ERCC3, ERCC4, UNG, APOBEC3C, PARP10, NBN and many others were significantly more expressed in EO-CRC (Fig. 1G, Supplementary Table 2 ). To ensure that deregulation of these placenta-related features in EO-CRC was not solely attributable to patient age, we included in the study samples from tumour-adjacent mucosae of both EO-CRC and SO-CRC patients. PCA of the full proteomes of the total 35 samples disclosed a distinguishable separation between younger and older individuals ( Supplementary Fig. 2D-E ). The proteomic profile of adjacent mucosae tended to cluster closely with their cancerous counterparts, pointing out the importance of the age factor as a relevant variable. However, the distinct clustering of the EO-CRC subgroup initially observed in the PCA (Fig. 1B) was confirmed. To investigate this further, we conducted additional analyses. Normal colonic samples from young and old individuals were first compared to each other, and each cancer group (EO-CRC and SO-CRC) was independently compared to its age-matched non-cancerous counterpart ( Supplementary Table 2 ). The significantly deregulated proteins identified in all comparisons were then cross-referenced with the previously defined placenta-related signature associated with EO-CRC. To construct this signature, we integrated placental-associated features identified through the tissue specificity analysis (Fig. 1F and Supplementary Fig. 2C ), thereby generating a consolidated placental protein signature relevant to EO-CRC ( Supplementary Table 2 ). Distribution analysis of placental-associated proteins demonstrated a significant enrichment in EO-CRC tumours relative to both SO-CRC tumours and matched normal mucosae from EO-CRC patients (N-EO-CRC), indicating a distinct placental-like protein expression profile unique to early-onset disease (Fig. 1H). This evidence suggests that EO-CRC is characterized by a proteomic profile indicative of a less differentiated cellular state. Some placental-associated proteins were also deregulated when we compared normal tissues from younger and older patients (N-EOCRC vs N-SOCRC). We attribute these differences to age-related changes or to field carcinogenesis in histologically normal mucosa adjacent to tumours 36 . To isolate true cancer-specific signals, we refined the placental protein list by removing any proteins that overlapped with other comparisons or that were detected in normal tissues. This allowed us to define a robust and EO-CRC-specific placental signature, restricted to features uniquely associated with EO-CRC (Fig. 1H-I, Supplementary Table 2 ). Analysis of the control-adjusted placental signature, following the exclusion of shared control features, confirmed the upregulation of these proteins in the EO-CRC. Importantly, it also revealed notable intra-group heterogeneity within the EO-CRC cohort. In contrast to SO-CRC, which displayed a relatively homogeneous distribution of placental core protein expression, EO-CRC exhibited clear stratification. Specifically, a distinct subgroup of patients showed markedly higher expression levels of the placental signature compared to the rest of the EO-CRC cases, indicating a possible molecularly distinct subset within the EO-CRC population (Fig. 1J). To test this hypothesis, we performed hierarchical clustering analysis (HCA) based exclusively on the placental protein core. The resulting dendrogram (Fig. 1K) illustrated distinct clustering patterns, with several clear macro-separations. Notably, SO-CRC patients formed a well-defined cluster, reinforcing their molecular distinction from EO-CRC cases. All but one EO-CRC exhibit the characteristic overexpression of the placental signature compared to SO-CRC. Of notice, HCA revealed two distinct branches also within EO-CRC. This separation highlights a gradient of expression of the placental signature among EO-CRC (Fig. 1L), further supporting our hypothesis that the placental protein profile can discriminate between subpopulations of EO-CRC. To validate these results, we performed a de novo proteomic analysis on an independent patient cohort of EO-CRC (N = 9) and SO-CRC (N = 10), using the same pipeline as in the discovery phase. Consistent with the findings from the discovery cohort, PCA analysis of the validation data set revealed that EO-CRC clustered differently from SO-CRC ( Supplementary Fig. 3A ). The comparison between EO-CRC and SO-CRC again highlighted differential regulation of key biological processes, notably those associated with organism development ( Supplementary Figs. 3B-D, Supplementary Table 3 ). Furthermore, tissue specificity analysis once again demonstrated a distinct enrichment of placental-related proteins in EO-CRC ( Supplementary Figs. 3B and E, Supplementary Table 3 ), reinforcing the robustness and reproducibility of our findings. As in the discovery phase, we normalized the EO-CRC placental-related features by adjusting expression levels in the corresponding control samples ( Supplementary Figs. 3F-G, Supplementary Table 3 ). We then used these control-corrected protein profiles to perform HCA. Also, for this independent cohort, we were able to discriminate a sub-cluster of EO-CRC samples with marked over-expression of the placental signature. While SO-CRC patients remained distinctly separated, one EO-CRC grouped with the SO-CRC patients ( Supplementary Figs. 3H, J, and K ). In conclusion, our proteomic approach highlighted the distinctiveness of protein expression of EO-CRC compared to SO-CRC. Particularly, EO-CRC samples were characterized by a higher expression of proteins usually involved in placental processes, which were not observed in samples collected from older patients. This finding was confirmed even after adjusting our results on the proteomic profiling of adjacent non-cancer mucosa, thus ruling out a potentially merely age-related finding. Given the limited sample size, further validations of our findings are warranted in larger prospective cohorts. Placental features are retained in patient-derived organoids . One approach to overcome this limitation is the use of cultured models such as organoids, which provide greater reproducibility while retaining proteomic profiles that closely mirror those of primary tumors 37 . Hence, whenever possible, we established patient-derived organoids (PDOs) from both EO-CRC and SO-CRC primary tumors. PDOs were successfully generated from two EO-CRC (EOCRC8 and EOCRC18) and two SO-CRC (SOCRC26 and SOCRC27) patients. We performed high-resolution proteomic analysis of the PDOs, applying the same pipeline used for the solid tumour samples (Fig. 2A). PCA of the full proteomes revealed that the organoids retain the proteomic hallmarks of their parental tumours, including the distinction between early- and standard-onset origins (Fig. 2B, Supplementary Table 4 ). A correlation matrix confirmed high intra-sample reproducibility among PDOs, underscoring their robustness as ex vivo models (Fig. 2C). To assess the conservation of the placental-like features in these models, we examined the expression of core placental proteins previously identified in tumour samples. Most of these proteins (85%) remained detectable in the EO-CRC PDOs (Fig. 2D), with cluster analysis revealing distinct patterns of expression across the cohort, with divergency for the EO-CRC8 sample similarly to parental tissues (Fig. 1L and 2E, Supplementary Table 4 ). Specifically, the stratification of proteins observed in PDOs mirrored trends observed in primary tissues, including the enrichment of placenta-related proteins associated with key EO-CRC processes, such as DNA metabolism, tissue invasion, and immunomodulation in EO-CRC (Fig. 2F-G, Supplementary Table 4 ). HCA further confirmed the segregation of EO-CRC and SO-CRC PDOs according to the placental signature, reinforcing the notion that this molecular program is maintained in PDOs (Fig. 2H). Ataxia Telangiectasia and Rad3-related (ATR) signalling is a key mediator of replication stress responses, a pathway which is frequently dysregulated in aggressive, developmentally placental mimicking tumours 23 . On these premises, aiming at functionally validating our findings and exploring potential hints of drug sensitivity in PDOs generated from EO-CRC, we next focused on assessing how ATR inhibitors (ATRis) can impact on PDOs viability. This focus was further supported by our proteomic data, which revealed that EO-CRC samples exhibited elevated expression of proteins involved in DNA metabolism. Notably, proteins implicated in replication stress management along with a range of DNA repair factors were significantly more abundant in EO-CRC (Fig. 1G). These findings suggest that EO-CRC may be particularly vulnerable to therapies targeting replication stress response pathways, such as ATR inhibition. Thus, PDOs were exposed to increasing doses of ATRi Ceralasertib and Berzosertib, over a five-days in vitro viability assay. Dose-response analyses demonstrated differential sensitivity among the PDOs, with EOCRC8 showing the highest expression of placental-like proteomic features and exhibiting the greatest sensitivity to both ATRis when compared to other PDOs including EOCRC18 harbouring fainter expression of placental features (Fig. 2I-J). This observation was confirmed at clinically relevant concentrations of both compounds, with significantly reduced viability of EOCRC8 PDO (Fig. 2K-L ) . Of notice, none of the previously suggested molecular alterations identified in CRC sensitizing to ATRis was retrieved in EOCRC8 PDO ( Supplementary Fig. 4 ) 38 . These findings demonstrate that the placental-like proteomic signature is retained in PDOs and it can potentially predict sensitivity to replication stress-targeting agents such as ATRis. While intriguing, this observation is limited to a single sample and requires broader validation. The retrotransposable element HERVH is overexpressed in early-onset colorectal cancer . In our previous findings, we demonstrated that exposure of mammalian stem cells to DNA replication inhibitors, activating an ATR-dependent replication stress response, led to the induction of trophoblast-specific genes and endogenous retroviruses (ERVs), thereby linking replication stress to retrotransposon reactivation and the emergence of placental-like transcriptional programs 23 . HERV elements have been recognized as key transcriptional regulators of pluripotency in both embryonic stem cells and placental development 26 – 28 . Given EO-CRC placental-like traits, we investigated whether these features were also associated with altered HERV expression. To this end, we performed RNA-sequencing on samples from EO-CRC patients, also including samples isolated from adjacent normal tissue (N = 8), primary tumour tissue (N = 9), and liver metastases (N = 5) ( Supplementary Table 5 and Supplementary Fig. 5A ). We first assessed whether the placental-like signature identified by proteomics was also reflected at the transcriptomic level. Consistently, this signature, as well as additional placental gene sets from public datasets, were enriched in primary tumour samples compared to matched adjacent mucosae (Fig. 3A and Supplementary Fig. 5B ). Notably, these signatures exhibited further enrichment in metastatic lesions relative to primary EO-CRC lesions (Fig. 3B and Supplementary Fig. 5C ), suggesting a progressive acquisition of a placental-like trait during tumour progression and evolution. We next quantified retrotransposable element (RE) expression and performed differential expression analysis. We identified 16 upregulated and 8 downregulated RE subfamilies in EO-CRC versus normal mucosa (Fig. 3C), and 2 upregulated with 13 downregulated REs in metastases versus EO-CRC primary lesions (Fig. 3D). Notably, the differentially expressed REs were predominantly HERVs, mainly from the ERV1, ERVL, and ERVK families. Of particular interest, two ERV1 subfamilies, LTR7Y and HERVH, representing the regulatory and internal components of full-length HERVH elements, showed a specific stepwise increase in expression from normal mucosa to EO-CRC primary tumour and then to metastases (Fig. 3E –F ). These findings suggest a potential role for specific HERVH elements not only as markers of cellular dedifferentiation, consistent with their established expression in pluripotent stem cells, but also as indicators of tumour progression and metastatic driver. Based on these observations, we hypothesized that HERVH may represent a novel molecular feature of the previously described placental features distinguishing EO-CRC from SO-CRC. Therefore, we investigated the expression of HERVH using RNA In Situ Hybridization (RNA-ISH) on FFPE sections from EO-CRC (N = 16) and SO-CRC (N = 14) ( Supplementary Table 5 ). In 8/10 (80%) EO-CRC cases with available matched mucosae, tumours exhibited stronger HERVH expression, with only one case showing no difference and one case showing the reverse (Fig. 4A-B). Of notice, the EOCRC8 PDO (established from an HERVH high expressing sample) was sensitive to ATRis, while EOCRC18 PDO (established from an HERVH low expressing sample) was much less sensitive (Fig. 2I-L and Fig. 4B ). In contrast, 4/12 (33%) SO-CRC expressed higher HERVH levels compared to their matched mucosa consistently with prior reports 23 – 25 . However, they were significantly lower compared to EO-CRC tumour samples (Fig. 4C –D ). Although higher in EO-CRC, HERVH expression was not exclusive, suggesting limited diagnostic specificity. In conclusion, we found that HERVH expression levels were significantly higher in EO-CRC tumours compared to SO-CRC (Fig. 4E) and showed a positive association with tumour progression. This pattern may reflect the acquisition of placental-like features in most EO-CRC, highlighting HERVH as a potential novel diagnostic and therapeutic target (Fig. 5). Discussion EO-CRC incidence is rising globally posing urgent clinical challenges 1 , 39 , 40 . Compared to SO-CRC, EO-CRC present specific clinical features of poor response and survival after available therapies 6 , 9 . These observations point to a distinct biological entity, whose molecular underpinnings remain insufficiently explored 12 . To address this gap, we employed a multi-omics approach combining proteomics and transcriptomics on non-hereditary EO-CRC samples from patients mostly diagnosed under age 40. Additionally, we functionally test our findings in PDOs established from EO-CRC patients. Our analyses revealed that in the majority of EO-CRC placental developmental programs are activated and HERVH, a known regulator of pluripotency in embryonic stem cells, is overexpressed. In detail, proteomic profiling showed that most EO-CRC, but not SO-CRC, are enriched in placentation molecules, angiogenic mediators, and immune checkpoint regulators factors that are physiologically restricted to the maternal-foetal interface. These findings suggest that EO-CRC mimic placental biology, acquiring invasive, immune-evasive, and remodelling capacities characteristic of placental development (Fig. 5). While these data suggest a potential link, functional validation will be needed to establish causality. If confirmed in future studies, this placental mimicry, in conjunction with HERVH reactivation, could define a dual molecular signature that differentiate most EO-CRC from SO-CRC, offering a potential explanation for its aggressive behaviour. At the molecular level, we speculate that chronic replication stress, potentially driven by exposome factors, such as but not limited to microbiome-derived genotoxins 16 , 41 and high-fat diets 9 , 42 , induces epigenetic alterations that massively unlock the expression of a placentals signature, comprising HERVH in the young adult colon (Fig. 5). If so, and contextualized in the available literature, our findings might represent a novel biological hint to rationalize the search for the EO-CRC increase incidence causes by focusing on exposomal agents retaining hypomethylating potentials 43 . Once reactivated, it is reasonable that HERVH could act on chromatin contributing to dedifferentiation and immune evasion 26 , 44 , 45 . This pattern mirrors our prior findings in stem cells undergoing replication stress, where ERV activation drives trophoblast lineage transitions 23 . Supporting this model and accordingly to another recent publication 46 , RNA-seq of paired EO-CRC mucosa, primaries, and metastases shows a step-wise rise in HERVH from normal tissue to metastasis, pointing to a role in cancer progression. Full-length HERVH remains transcriptionally active, mirroring its behaviour in pluripotent and trophoblast cells and exemplifying retrotransposon onco-exaptation 26 , 47 – 49 . The reactivation likely reflects a global loss of repressive chromatin, warranting future epigenomic study. Importantly, placental-like overexpression peaks in EO-CRC and declines with age, affecting younger patients the most. Previous reports have documented that HERV positivity in non-metastatic CRC is associated with worse prognosis and higher risk of disease relapse after surgery 47 . Accordingly, in the same clinical setting from the IDEA collaboration dataset, young age emerged as a negative prognostic factor in stage III CRC and was associated with significantly higher chance of relapse, despite a more aggressive therapeutic strategy 8 . On these premises, the present findings corroborate the understanding that EO-CRC, because of their biological characteristics including placental features and HERVH reactivation, represents an aggressive subset of CRC also potentially resistant to current therapies including the most intensive chemo-regimens 6 , 50 . Future studies are warranted to assess the prevalence of placental features and HERVH reactivation in a CRC patients population stratified for minimal residual disease (MRD) using circulating tumour DNA (ctDNA) after primary tumour resection. These efforts may help determine whether the prevalence of intrinsically aggressive ‘ born-to-be-bad ’ diseases is more common in EO-CRC patients. Previous findings showed that a relevant fraction of CRC cells display pronounced sensitivity to pharmacologic inhibition of ATR 38 , a master kinase that buffers replication stress 22 , able to induce ERVs and trophoblast cell fate transition (Fig. 5) 23 . Our PDOs replicate this vulnerability, supporting ATR inhibitors now in trials for DNA-repair-deficient tumours. Placental-like signatures may predict response to ATR-inhibitors and, if validated, would be the first EO-CRC actionable biomarker. It is tempting to speculate that coupling replication-stress targeting with immune-checkpoint blockade, or with vaccines directed against HERVH peptide, might deserve further investigation. Finally, detecting HERVH RNA in stool or blood could help stratify CRC risk in young adults, for whom universal colonoscopy is unfeasible (Fig. 5) 12 , 51 , 52 . While our findings consistently suggest that most EO-CRC engages a placental-like program and HERVH reactivation, several caveats remain. Functional analyses rely on a limited PDO set, and the observed association with ATR inhibitor sensitivity should be viewed as hypothesis-generating. The specificity of the placental signature to EO-CRC is highly suggestive but requires further confirmation in larger geographically distinct series. Disentangling specific EO-CRC reprogramming from features shared with other aggressive tumour states will benefit from broader pan-cancer comparisons. Finally, although direct modelling of environmental influences is challenging, indirect strategies, such as epigenetic modulation in organoids or analysis of clinical metadata, may help to clarify upstream regulators of these features. In summary, we present an original perspective that might redefine a subset of non-hereditary EO-CRC as a biologically distinct malignancy characterized by reawakening placental programs and HERVH expression. These insights provide a potential rationale for EO-CRC aggressive clinical behaviour and a possible actionable vulnerability, opening a path from molecular discovery to targeted interception. Material and methods Patient identification and enrolment All patients were enrolled within the IANG-CRC ( https://www.frrb.it/it/progetti-finanziati-iang-crc ) or the IANG-CRC2 studies. IANG-CRC was a prospective multi-institutional Italian study funded by Fondazione Regionale Ricerca Biomedica (FRRB), approved by local Ethical Committees at Grande Ospedale Metropolitano Niguarda, Milan, Italy, and Istituto Ricovero Cura Carattere Scientifico (IRCCS) San Raffaele, Milan, Italy (please refer to Supplementary Data 1 for the full protocol). Later, IANG-CRC2 study was designed embedded within the AlfaOmega Master Observational Trial (MOT) (NCT04120935) 53 which was already active at Grande Ospedale Metropolitano Niguarda, Milan, Italy, Vall d'Hebron Institute of Oncology (VHIO), Barcelona, Spain, with the translational partnership of Fondazione Istituto Nazionale Genetica Molecolare (INGM), Padiglione "Romeo ed Enrica Invernizzi", Milan, Italy, and IFOM ETS - The AIRC Institute of Molecular Oncology, Milan Italy. The IANG-CRC2 study followed the IANG-CRC study towards broadening the latter sample size. All EO-CRC patients included in this publication were enrolled within the IANG-CRC or IANG-CRC2 studies, while SO-CRC patients were enrolled within the AlfaOmega MOT (NCT04120935). All patients provided written informed consent for participation in the study and associated procedures. These studies were conducted in accordance with the principles of the Declaration of Helsinki and the International Conference on Harmonisation and Good Clinical Practice guidelines. Whenever available, both fresh and formalin-fixed paraffin-embedded (FFPE) samples were sent for translational analysis from recruiting clinical centers to both IFOM-ETS – The AIRC Institute of Molecular Oncology, Milan, Italy and to Fondazione Istituto Nazionale Genetica Molecolare (INGM), Padiglione "Romeo ed Enrica Invernizzi", Milan, Italy. EO-CRC patients diagnosed with disease at age younger than 45 and with availability of fresh frozen or archival FFPE samples were considered for eligibility for the present study. Forty-five years as age cut-off was arbitrarily chosen based on the updated recent CRC screening recommendations by the American Cancer Society 52 . EO-CRC patients known to be affected with hereditary cancer predisposition syndromes (HCPS) were not eligible for the present study. EO-CRC patients with a first-degree relative (FDR) diagnosed with CRC were considered eligible only if they had undergone extended germline next generation sequencing (NGC) panelling to rule out HCPSs, according to the most recent consensus publication on EO-CRC patients management 54 . Patients diagnosed with somatic MSI EO-CRC were considered eligible only if Lynch syndrome was ruled out by germline profiling. As controls, all SO-CRC patients diagnosed with the disease at age 50 or later and with availability of fresh frozen or archival FFPE samples were eligible. Collection and analysis of clinicopathological features The following data were extrapolated from each enrolled patient: age, sex, cancer familiarity, prior history of inflammatory bowel disease (IBD), primary tumour location, tumour histology, stage at initial diagnosis, tumour grading, RAS / BRAF status, HER2 expression, and MMR status, disease burden with regard to the presence of liver and peritoneal metastases, and last follow-up or death. Continuous variables were summarized as median and interquartile range (IQR). Categorical variables were summarized as frequency and percentage. Baseline characteristics were compared according to age of onset using Mann-Whitney test for continuous variables, and Chi-square test or Fisher’s exact test for categorical variables. Survival was evaluated from the date of initial diagnosis to last follow-up or death occurring for any cause. All tests were two-sided and p-values < 0.05 were considered significant. Statistical analyses were performed using GraphPad Prism 10.2.3 Software. Proteomic workflow Protein extraction and digestion Frozen human colon samples and PDOs were homogenized in 8M Urea, 100 mM Tris-HCl pH = 8 using Dounce tissue homogenizer. To complete the solubilization, 5 on/off cycles with tip sonicator were performed. The samples were then centrifuged (14’000 xg , 30 min, 4°C) and protein concentration were measured in the resulting supernatant via BCA assay (Microplate BCA™ protein Assay Kit, Thermo Scientific), using BSA as standard. 70 µg of proteins for each specimen were digested following in-solution digestion. Briefly, proteins were reduced by 10 mM tris-carboxy-ethyl phosphine (Thermo scientific) and alkylated with 40 mM 2-Chloroacetamide (Sigma-Aldrich) in 8 M Urea 100 mM Tris pH = 8 at for 30 min (r.t., dark shaker). Double protein digestion was carried out. Firstly, the endoproteinase Lys-C (Thermo scientific) was added in 1:50 ratio to initial protein concentration (1h, r.t.), then the hydrolysis was boosted by supplementing trypsin (Roche) (overnight, 37°C), using the same ratio. The resulting peptide solution was desalted and concentrated using µ-C18 Ziptip pipette tips (Millipore) following manufacturer's instruction. Purified samples were resuspended in 5% formic acid (FA) solution and stored at -20°C until the MS analysis. LC-MS/MS and Data Processing The LC/MS analysis was performed using an UHPLC Easy-nLC 1200 (Thermo Scientific) coupled to an Orbitrap Exploris 480 Mass Spectrometer (Thermo Scientific). The hydrolysed peptides (1.5 µl) were separated by applying a linear gradient from 95% solvent A (2% acetonitrile (ACN), 0.1% FA) to 55% solvent B (80% ACN, 0.1% FA). Data was acquired in both Data Dependent Acquisition (DDA) and Data Independent Acquisition (DIA) modes. Raw MS files were processed with Spectronaut software. MS/MS peak lists are searched against the UniProtKB complete proteome 2021 Human Database. Raw data files were uploaded in Spectronaut 15, run in library-assisted search for label-free protein quantification. Peptides were identified using the Pulsar search engine specifying the UniProt Homo Sapiens database (UP000005640) as reference. Methionine Oxidation (M) and Acetyl N-Terminal modification (Protein N-term) were specified as dynamic modifications while carbamidomethyl at cysteine residues (C) was used as static modification. The peptides were required to be at least 7 amino acids long and counting a maximum of 2 missed cleavages. Data were filtered at 1% FDR on the protein level and protein quantities were exported, followed by analysis with different software packages (Excel, R-studio, Perseus). Specifically, a first level of data filtering was applied to exclude contaminant proteins/peptides. The signal/noise values were normalized using log2 transformation, while the protein abundances were grouped and filtered to achieve a minimum valid number equal to 70% in at least one group. Missing values have been replaced by random numbers that are drawn from a normal distribution. Statistical analysis for quantitative evaluation was performed. For comparisons among the different sample cohorts, a non-pairwise t-test was applied, with the level of significance set at p ≤ 0.05. Proteins exhibiting at least a log2 fold change of ± 0.5 were considered differentially regulated. To further explore patterns of protein expression across samples, unsupervised hierarchical clustering analysis was performed using the Ward method and Euclidean distance as the metric. Functional analysis was performed using multiple diverse tools 33 , 34 , 55 , with a focus on GO terms and tissue- or cell type–specific expression patterns. An FDR threshold of < 0.1 was applied to determine significance. Declarations Acknowledgements We are grateful to all members of the A.B. and the V.C. laboratories for the insightful and critical scientific discussion. We acknowledge the scientific and technical assistance of the INGM Imaging Facility, in particular, Chiara Cordiglieri and Alessandra Fasciani (Istituto Nazionale Genetica Molecolare ‘Romeo ed Enrica Invernizzi’ (INGM), Milan, Italy). We acknowledge the scientific and technical contributions of the Cogentech Histopathology Unit and the IFOM Cellular and Preclinical Models Unit. Proteomics data have been generated by the IFOM ETS Cogentech Proteomics and Metabolomics Core Facility (RRID:SCR_026937). G.Patelli is a PhD student within the European School of Molecular Medicine (SEMM). We gratefully acknowledge all young individuals in care and their families for their generous participation in the IANG-CRC study. We extend our deepest appreciation to the families of Andrea Fernandez and Gae Federica Elli, in whose loving memory this work is dedicated. Their extraordinary emotional support and generosity were instrumental in inspiring and sustaining the IANG-CRC program from its inception. Funding Study funded by: Fondazione Regionale Ricerca Biomedica, project IANG-CRC (grant CP2_12/2018 to S.Siena); Fondazione Oncologia Niguarda ETS (grant ‘Giovani CRC’ to G.M.); Grande Ospedale Metropolitano Niguarda (Fondo Divisionale Oncologia Falck); the Italian Ministry of University and Research (MUR), Progetti di Ricerca di Rilevante Interesse Nazionale (PRIN) 2022 under the project ‘A Comprehensive Analysis of Dietary Risk Factors, Colibactin and Tumour Molecular Features in Early-Onset Colorectal Cancer’ (grant 2022LSFKWE to A.S-B., GM.C., and S.Arena); AIRC under 5 per Mille 2018 - ID. 21091 program – PI Alberto Bardelli (A.Bardelli), Group Leader (GL) Salvatore Siena (S.Siena) and Silvia Marsoni (S.Marsoni); Regione Lombardia (Delibera XII/2455/2024 Young Adult Colorectal Cancer Surveillance Program (Ages 20e49) to Grande Ospedale Metropolitano Niguarda; AIRC under IG 2023—ID. 28725 project to V.C. Ricerca Finalizzata, (grant nr GR-2018-12365280). AIRC under IG 2023 - ID. 29286 project to S.Arena; FPRC 5 9 1000 Ministero della Salute 2022 CARESS to S. Arena; Italian Ministry of Health, Ricerca Corrente 2025 to S.Arena; Prin 2022 PNRR finanziato dall’Unione Europea NextGenerationEU M4 C2 I.1.1.- P2022E3BTH to S.Arena. Fondazione AIRC (grant nr 27066 and nr 21073); Fondazione Cariplo (grant nr 2019-3416); Piano Nazionale Ripresa e Resilienza (PNRR) (grant nr G43C22002620007); and Progetti Rilevante Interesse Nazionale (PRIN) (grant nr 2022PKF9S) to B.B.; European Research Council (ERC) under the European Union‘s Horizon 2020 research and innovation programme (TARGET, grant agreement n. 101020342) (A.Bardelli); AIRC under IG 2023 - ID. 28922 project – P.I. Bardelli Alberto (A.Bardelli). Authors Contribution G.M., L.Santorelli, and F.Marasca designed and conducted experiments, analysed and interpreted data and wrote the manuscript. L.Santorelli designed and executed the proteomic analyses. V.R. designed and supervised bioinformatic analyses and revised the manuscript. E.G. designed and performed experiments, analysed and interpreted data. L.Salviati designed and performed bioinformatic experiments and analysed data. B.B. conceived and conceptualized the study, designed experiments, analysed and interpreted data and wrote and finalized the manuscript. S.Abrignani interpreted data, wrote and finalized the manuscript. A.Bachi curated with L.Santorelli the acquisition and interpretation of the proteomic data. V.C. conceived and conceptualized the study, designed experiments, coordinated the experimental work, analysed and interpreted data, wrote the manuscript, and contributed to its editing and critical revision. A.C. and G.Parodi contributed with samples management and performed Mass Spectrometry acquisition. E.B., F.T., K.B., A.A., A.S-B., S.Siena, E.E., N.S, I.B., G.Patelli, M.P., L.L., A.M., GM.C. contributed with patients’ studies and tumour procurement. G.C., A.S., and S.Scardellato performed experiments and bioinformatic analysis. A.S-B., S.Arena, S.Marsoni also contributed with financial support. S.G., S.Mariano, L.M., contributed with patients data and tumour samples management. E.B., M.D-C., MC.A. contributed with pathology diagnosis and tumour samples preparation. A.Bardelli designed experiments, coordinated the experimental work, analysed and interpreted data, wrote the manuscript, and contributed to its editing and critical revision. S.Siena conceived, conceptualized, and coordinated the study including clinical and financial responsibilities, and co-wrote the manuscript. All authors read and approved the manuscript ahead of submission. Conflict of interest S.S. is advisory board member for Agenus, AstraZeneca, Bayer, Bristol Myers Squibb, CheckmAb, Daiichi-Sankyo, GlaxoSmithKline, MSD, Merck, Novartis, Ospedale San Raffaele, Pierre-Fabre, Pfizer, Seagen, and T-One Therapeutics. A.S-B. reported consulting or advisory role for Bayer, Novartis, Pierre Fabre, Servier, and Takeda; and personal honoraria as an invited speaker from Amgen, Guardant Health, Pierre Fabre, and Seagen. A.Bardelli reports receipt of grants/research supports from Neophore, AstraZeneca, Boehringer Ingelheim and honoraria/consultation fees from Guardant Health. A.B. is stock shareholder of Neophore and Kither Biotech. A.B. is an advisory boards member for Neophore. S.Abrignani is a co-founder of the startup CheckMab s.r.l and T-One Therapeutics s.r.l.; F.Marasca and B.B. are co-founders of the startup T-One Therapeutics s.r.l.. S.Arena reports personal fees from MSD Italia and a patent (Italian patent application No. 102022000007535) outside the submitted work. K.B. is advisory board member for AstraZeneca. N.S. reports personal honoraria as an invited speaker for AMGEN, Medistream (OncoBites2025), and travel supports from AMGEN, MERCK, and BAYER. N.S. is an ESMO Fellow (starting from April 2024). F.T. reports travel supports from ROCHE. A.A. is advisory board member for AMGEN e Italfarmaco. I.B. has received accommodation and travel expenses from Amgen, Merck, Sanofi, and Servier; and personal speaker honoraria from Astra Zeneca. E.E. reports the following: Honoraria, Consulting or Advisory Role, and Speakers's Bureau: Agenus, Amgen, Bayer, Bristol Myers Squibb, Boehringer Ingelheim, Cure Teq AG, GlaxoSmithKline, Hoffman La-Roche, Janssen, Johnson&Johnson, Lilly, Medscape, Merck Serono, MSD, Nordic Group BV, Novartis, Organon, Pfizer, Pierre Fabre, Repare Therapeutics Inc., RIN Institute Inc., Rottapharm Biotech, Sanofi, Seagen International GmbH, Servier, and Takeda. References Siegel, R. L., Miller, K. D., Wagle, N. S. & Jemal, A. Cancer statistics, 2023. CA Cancer J Clin 73, 17–48 (2023). Daca-Alvarez, M. et al. Regional patterns of early-onset colorectal cancer from the GEOCODE (Global Early-Onset COlorectal Cancer DatabasE)-European consortium: retrospective cohort study. BJS Open 9, zraf024 (2025). Russo, A. G. et al. Increased incidence of colon cancer among individuals younger than 50 years: A 17 years analysis from the cancer registry of the municipality of Milan, Italy. Cancer Epidemiol 60, 134–140 (2019). Vuik, F. E. et al. Increasing incidence of colorectal cancer in young adults in Europe over the last 25 years. Gut 68, 1820–1826 (2019). Eng, C. et al. Colorectal cancer. Lancet 404, 294–310 (2024). Kanter, K. et al. Care Patterns and Overall Survival in Patients With Early-Onset Metastatic Colorectal Cancer. JCO Oncol Pract 17, e1846–e1855 (2021). Lieu, C. H. et al. Association of age with survival in patients with metastatic colorectal cancer: analysis from the ARCAD Clinical Trials Program. J Clin Oncol 32, 2975–2984 (2014). Fontana, E. et al. Early-Onset Colorectal Adenocarcinoma in the IDEA Database: Treatment Adherence, Toxicities, and Outcomes With 3 and 6 Months of Adjuvant Fluoropyrimidine and Oxaliplatin. J Clin Oncol 39, 4009–4019 (2021). Akimoto, N. et al. Rising incidence of early-onset colorectal cancer - a call to action. Nat Rev Clin Oncol 18, 230–243 (2021). Hofseth, L. J. et al. Early-onset colorectal cancer: initial clues and current views. Nat Rev Gastroenterol Hepatol 17, 352–364 (2020). Eng, C. & Hochster, H. Early-Onset Colorectal Cancer: The Mystery Remains. J Natl Cancer Inst 113, 1608–1610 (2021). Mauri, G. et al. Early-onset cancers: Biological bases and clinical implications. Cell Rep Med 5, 101737 (2024). Ogino, S. & Ugai, T. The global epidemic of early-onset cancer: nature, nurture, or both? Ann Oncol 35, 1071–1073 (2024). Cercek, A. et al. A Comprehensive Comparison of Early-Onset and Average-Onset Colorectal Cancers. J Natl Cancer Inst 113, 1683–1692 (2021). Willauer, A. N. et al. Clinical and molecular characterization of early-onset colorectal cancer. Cancer 125, 2002–2010 (2019). Díaz-Gay, M. et al. Geographic and age variations in mutational processes in colorectal cancer. Nature (2025) doi: 10.1038/s41586-025-09025-8 . Costanzo, V., Bardelli, A., Siena, S. & Abrignani, S. Exploring the links between cancer and placenta development. Open Biol 8, 180081 (2018). Holtan, S. G., Creedon, D. J., Haluska, P. & Markovic, S. N. Cancer and pregnancy: parallels in growth, invasion, and immune modulation and implications for cancer therapeutic agents. Mayo Clin Proc 84, 985–1000 (2009). Lee-Six, H. et al. The landscape of somatic mutation in normal colorectal epithelial cells. Nature 574, 532–537 (2019). Coorens, T. H. H. et al. Inherent mosaicism and extensive mutation of human placentas. Nature 592, 80–85 (2021). Mamlouk, S. et al. DNA copy number changes define spatial patterns of heterogeneity in colorectal cancer. Nat Commun 8, 14093 (2017). Zeman, M. K. & Cimprich, K. A. Causes and consequences of replication stress. Nat Cell Biol 16, 2–9 (2014). Atashpaz, S. et al. ATR expands embryonic stem cell fate potential in response to replication stress. Elife 9, e54756 (2020). Nurk, S. et al. The complete sequence of a human genome. Science 376, 44–53 (2022). Chuong, E. B., Rumi, M. A. K., Soares, M. J. & Baker, J. C. Endogenous retroviruses function as species-specific enhancer elements in the placenta. Nat Genet 45, 325–329 (2013). Lu, X. et al. The retrovirus HERVH is a long noncoding RNA required for human embryonic stem cell identity. Nat Struct Mol Biol 21, 423–425 (2014). Frost, J. M. et al. Regulation of human trophoblast gene expression by endogenous retroviruses. Nat Struct Mol Biol 30, 527–538 (2023). Wang, J. et al. Isolation and cultivation of naive-like human pluripotent stem cells based on HERVH expression. Nat Protoc 11, 327–346 (2016). Attig, J. et al. Human endogenous retrovirus onco-exaptation counters cancer cell senescence through calbindin. J Clin Invest 133, e164397 (2023). Babaian, A. et al. Onco-exaptation of an endogenous retroviral LTR drives IRF5 expression in Hodgkin lymphoma. Oncogene 35, 2542–2546 (2016). Jang, H. S. et al. Transposable elements drive widespread expression of oncogenes in human cancers. Nat Genet 51, 611–617 (2019). Teclemariam, E. T., Pergande, M. R. & Cologna, S. M. Considerations for mass spectrometry-based multi-omic analysis of clinical samples. Expert Rev Proteomics 17, 99–107 (2020). Pan, J.-B. et al. PaGenBase: a pattern gene database for the global and dynamic understanding of gene function. PLoS One 8, e80747 (2013). Palasca, O., Santos, A., Stolte, C., Gorodkin, J. & Jensen, L. J. TISSUES 2.0: an integrative web resource on mammalian tissue expression. Database (Oxford) 2018, bay028 (2018). Rudolph, J. D. & Cox, J. A Network Module for the Perseus Software for Computational Proteomics Facilitates Proteome Interaction Graph Analysis. J Proteome Res 18, 2052–2064 (2019). Curtius, K., Wright, N. A. & Graham, T. A. An evolutionary perspective on field cancerization. Nat Rev Cancer 18, 19–32 (2018). Lindhorst, P. H. & Hummon, A. B. Proteomics of Colorectal Cancer: Tumors, Organoids, and Cell Cultures-A Minireview. Front Mol Biosci 7, 604492 (2020). Durinikova, E. et al. Targeting the DNA Damage Response Pathways and Replication Stress in Colorectal Cancer. Clin Cancer Res 28, 3874–3889 (2022). Ben-Aharon, I. et al. Early-Onset Cancer in the Gastrointestinal Tract Is on the Rise-Evidence and Implications. Cancer Discov 13, 538–551 (2023). Jayakrishnan, T. & Ng, K. Early-Onset Gastrointestinal Cancers: A Review. JAMA (2025) doi: 10.1001/jama.2025.10218 . Sogari, A. et al. Tolerance to colibactin correlates with homologous recombination proficiency and resistance to irinotecan in colorectal cancer cells. Cell Rep Med 5, 101376 (2024). Arima, K. et al. Western-Style Diet, pks Island-Carrying Escherichia coli, and Colorectal Cancer: Analyses From Two Large Prospective Cohort Studies. Gastroenterology 163, 862–874 (2022). Young, A. S. et al. The need for a cancer exposome atlas: a scoping review. JNCI Cancer Spectr 9, pkae122 (2025). Schmidleithner, L., Stüve, P. & Feuerer, M. Transposable elements as instructors of the immune system. Nat Rev Immunol (2025) doi: 10.1038/s41577-025-01172-3 . Ugai, T. et al. Immune cell profiles in the tumor microenvironment of early-onset, intermediate-onset, and later-onset colorectal cancer. Cancer Immunol Immunother 71, 933–942 (2022). Viot, J. et al. Deciphering human endogenous retrovirus expression in colorectal cancers: exploratory analysis regarding prognostic value in liver metastases. EBioMedicine 116, 105727 (2025). Golkaram, M. et al. HERVs establish a distinct molecular subtype in stage II/III colorectal cancer with poor outcome. NPJ Genom Med 6, 13 (2021). Mazumder, H. et al. Human endogenous retroviruses (HERVs) associated with glioblastoma risk and prognosis. Cancer Gene Ther 32, 622–632 (2025). Matteucci, C. et al. Expression of HERV-K and embryonic genes in chronic lymphocytic leukemia and their association with therapy regimens. Blood Adv bloodadvances.2024014181 (2025) doi: 10.1182/bloodadvances.2024014181 . Kneuertz, P. J. et al. Overtreatment of young adults with colon cancer: more intense treatments with unmatched survival gains. JAMA Surg 150, 402–409 (2015). Mauri, G., Patelli, G., Crisafulli, G., Siena, S. & Bardelli, A. Tumor ‘age’ in early-onset colorectal cancer. Cell 188, 589–593 (2025). Mehta, S. J., Morris, A. M. & Kupfer, S. S. Colorectal Cancer Screening Starting at Age 45 Years-Ensuring Benefits Are Realized by All. JAMA Netw Open 4, e2112593 (2021). Lazzari, L. et al. Abstract CT214: AlfaOmega- a master protocol empowering precision research in colorectal cancer. Cancer Research 79, CT214 (2019). Cavestro, G. M. et al. Delphi Initiative for Early-Onset Colorectal Cancer (DIRECt) International Management Guidelines. Clin Gastroenterol Hepatol 21, 581–603.e33 (2023). Ge, S. X., Jung, D. & Yao, R. ShinyGO: a graphical gene-set enrichment tool for animals and plants. Bioinformatics 36, 2628–2629 (2020). Perez-Riverol, Y. et al. The PRIDE database at 20 years: 2025 update. Nucleic Acids Res 53, D543–D553 (2025). Sato, T. et al. Long-term expansion of epithelial organoids from human colon, adenoma, adenocarcinoma, and Barrett’s epithelium. Gastroenterology 141, 1762–1772 (2011). Dijkstra, K. K. et al. Generation of Tumor-Reactive T Cells by Co-culture of Peripheral Blood Lymphocytes and Tumor Organoids. Cell 174, 1586–1598.e12 (2018). Marasca, F. et al. LINE1 are spliced in non-canonical transcript variants to regulate T cell quiescence and exhaustion. Nat Genet 54, 180–193 (2022). Desai, N. et al. Diverse repetitive element RNA expression defines epigenetic and immunologic features of colon cancer. JCI Insight 2, e91078 (2017). Additional Declarations Yes there is potential Competing Interest. S.S. is advisory board member for Agenus, AstraZeneca, Bayer, Bristol Myers Squibb, CheckmAb, Daiichi-Sankyo, GlaxoSmithKline, MSD, Merck, Novartis, Ospedale San Raffaele, Pierre-Fabre, Pfizer, Seagen, and T-One Therapeutics. A.S-B. reported consulting or advisory role for Bayer, Novartis, Pierre Fabre, Servier, and Takeda; and personal honoraria as an invited speaker from Amgen, Guardant Health, Pierre Fabre, and Seagen. A.Bardelli reports receipt of grants/research supports from Neophore, AstraZeneca, Boehringer Ingelheim and honoraria/consultation fees from Guardant Health. A.B. is stock shareholder of Neophore and Kither Biotech. A.B. is an advisory boards member for Neophore. S.Abrignani is a co-founder of the startup CheckMab s.r.l and T-One Therapeutics s.r.l.; F.Marasca and B.B. are co-founders of the startup T-One Therapeutics s.r.l.. S.Arena reports personal fees from MSD Italia and a patent (Italian patent application No. 102022000007535) outside the submitted work. K.B. is advisory board member for AstraZeneca. N.S. reports personal honoraria as an invited speaker for AMGEN, Medistream (OncoBites2025), and travel supports from AMGEN, MERCK, and BAYER. N.S. is an ESMO Fellow (starting from April 2024). F.T. reports travel supports from ROCHE. A.A. is advisory board member for AMGEN e Italfarmaco. I.B. has received accommodation and travel expenses from Amgen, Merck, Sanofi, and Servier; and personal speaker honoraria from Astra Zeneca. E.E. reports the following: Honoraria, Consulting or Advisory Role, and Speakers's Bureau: Agenus, Amgen, Bayer, Bristol Myers Squibb, Boehringer Ingelheim, Cure Teq AG, GlaxoSmithKline, Hoffman La-Roche, Janssen, Johnson&Johnson, Lilly, Medscape, Merck Serono, MSD, Nordic Group BV, Novartis, Organon, Pfizer, Pierre Fabre, Repare Therapeutics Inc., RIN Institute Inc., Rottapharm Biotech, Sanofi, Seagen International GmbH, Servier, and Takeda. Supplementary Files SupplementaryFigure1.pdf Supplementary Figure 1. Samples disposition from early-onset and standard-onset colorectal cancer patients. Identification and collection of both fresh and archival samples from both early-onset (EO-CRC) and standard-onset (SO-CRC) patients at three Institutions across Italy and Spain: Grande Ospedale Metropolitano Niguarda, and San Raffaele Hospital, Milan, Italy, and Vall d’Hebron Institute of Oncology (VHIO), Barcelona, Spain. Patients diagnosed with hereditary cancer predisposition syndromes (HSCPs) were excluded from our analysis (N=3, EO-CRC patients). Keys : EO-CRC = early-onset colorectal cancer; SO-CRC = standard-onset colorectal cancer; FFPE = formalin-fixed paraffin-embedded samples; PDO = patient-derived organoid; HSCPs = hereditary cancer predisposition syndromes; * = Primary tumour tissue was not available for analysis; however, to ensure statistical balance and an adequate number of healthy mucosae samples for a robust comparative analysis, adjacent healthy mucosae from both matched and unmatched SO-CRC patients in the cohort. SupplementaryFigure2.pdf Supplementary Figure 2. Proteomic profiling of early-onset colorectal cancer (EO-CRC) primary tumour samples reveal a specific placenta signature over-expression. A, B) List and frequency of GOCC terms related to the up-regulated (A) and down-regulated (B) proteins after the comparison between young and old cancerous proteomes. C) Principal Component Analysis (PCA) of the Mass Spectrometry (MS) full proteome from EO-CRC and SO-CRC affected and non-affected adjacent tissues, N-EO-CRC and N-SO-CRC. D) Hierarchical clustering of CRC and corresponding normal mucosa samples based on full proteome expression profiles. Clustering was performed using Euclidean distance and Ward linkage, highlighting molecular differences between EO-CRC and SO-CRC groups. E) Distribution of tissue- and cell-specificity terms associated with deregulated proteins in EO-CRC compared to SO-CRC, based on annotations from three reference databases. F) Differential expression of proteins involved in tissue invasion, immunomodulation, and defence response in EO-CRC) versus SO-CRC. Boxplots display the log2 LFQ intensity (group mean) for key proteins associated with these processes. Statistical significance was determined using an unpaired t-test (****p < 0.0001, *p < 0.05). SupplementaryFigure3.pdf Supplementary Figure 3. Proteomic Profiling of EO-CRC samples reveal a specific placenta signature over-expression (Validation cohort). A) PCA of the MS full proteomes from EO-CRC and SO-CRC solid samples of validation cohort. B) Volcano plot of the identified and quantified proteins in EO-CRC vs SO-CRC samples. In orange the placental related terms. C, D) List and frequency of GOBP terms related to the deregulated proteins after comparison between EO-CRC and SO-CRC proteomes. Top enriched terms (C) and specific items related to system development (D). E) Distribution of tissue- and cell-specificity terms associated with deregulated proteins in EOCRC compared to SO-CRC, based on annotations from three reference databases. F) Chart represents the frequency of proteins identified as placental- and not placental-related in the differently expressed proteomes of cancerous and non-affected samples. G) Plot describes the total number (%) of proteins belonging to the placenta core presents in all the samples and the remaining number (%) of placenta proteins specific to the EO-CRC group, after control adjustment. H) Violin plots of placental core proteins showing the relative change in abundance levels between EO-CRC and SO-CRC classes, after control adjustment. J) Unsupervised hierarchical clustering of CRC samples based on expression profiles of placenta-associated proteins, following control adjustment. Clustering (Euclidean distance and ward linkage method) reveal distinct molecular signatures between EO-CRC and SO-CRC cases. K) Heatmap illustrating the expression of placenta-related proteins differentially expressed between EOCRC and SOCRC. Samples were clustered using ward method after control adjustment. Each column represents a sample and each row a protein, with colour intensity indicating normalized expression levels. The resulting clustering pattern underscores differences in placenta-associated molecular features between EO-CRC and SO-CRC groups. SupplementaryFigure4.pdf Supplementary Figure 4. DNA repair proteins expression as assessed by immunohistochemistry in the EOCRC8 patient derived organoid. As suggested by previous publication ( Durinikova et al. Clin Can Res, 2022 ), we ruled out the presence of key molecular alterations explaining the sensitivity to ATR inhibitors (ceralasertib and berzosertib) observed in the EOCRC8 tumoroid. Particularly, RAD51, RAD51C and ATM proteins expression was retained in the EOCRC8 tumoroid, which was instead retaining placental and replication stress features observed in the match fresh tumour tissue. SupplementaryFigure5.pdf Supplementary Figure 5. Transcriptomic profiling highlights placenta-like signature enrichment in EO-CRC tumour and metastasis. A) Principal Component Analysis (PCA) on DESeq2 variance-stabilized RNA-seq data from EO-CRC, adjacent normal tissue, and metastases. B) Gene set–enrichment analysis (GSEA), presented as enrichment score profiles for Placenta development geneset from MSigDB in the comparison between EOCRC and normal adjacent tissue (Nominal p-value < 0.001, NES=1.62, FDR=0.02). C) Gene set–enrichment analysis (GSEA), presented as enrichment score profiles for Placenta development geneset from MSigDB in the comparison between EOCRC metastasis and EOCRC. (Nominal p-value <0.001, NES=1.76, FDR=0.006). SupplementaryTable1.docx Supplementary table 1. Clinicopathological features of early-onset colorectal cancers (n=33) compared to standard-onset colorectal cancers (n=27) included in the present study. Keys : * Fisher’s Exact Test was performed. ** Located in caecum, ascending colon, liver flexure and transverse colon. ^ Located in splenic flexure, descending colon and sigmoid colon. NOS Not otherwise specified. ^^ In this patients Lynch syndrome was ruled out by performing extended germline panelling, even if the tumour somatic mismatch repair status was not assessed. SupplementaryTable2.xlsx Supplementary Table 2. Proteomic profiling of EO-CRC samples in the discovery cohort reveals a distinct overexpression of a placenta-associated signature. List of the table summarizing all analyses performed on the discovery cohort (statistical and functional). SupplementaryTable3.xlsx Supplementary Table 3. Proteomic profiling of EO-CRC samples in the validation cohort reveals a distinct overexpression of a placenta-associated signature. List of the table summarizing all analyses performed on the discovery cohort (statistical and functional). SupplementaryTable4.xlsx Supplementary Table 4. Patient-derived organoids retain placental features. List of the table summarizing all the proteomic analyses performed on the discovery cohort (statistical and functional). SupplementaryTable5.xlsx Supplementary Table 5. Sample and experimental details of RNA-ISH and RNA-seq data. Comprehensive metadata and differential expression summary of EO-CRC and SO-CRC patients in RNA-seq and ISH analyses, including lists of samples and RNA-seq sequencing details. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7193450","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":493219818,"identity":"1e3d8521-e8df-4c5f-8628-afd66cabadc2","order_by":0,"name":"Gianluca Mauri","email":"","orcid":"","institution":"Department of Hematology Oncology, and Molecular Medicine, Grande Ospedale Metropolitano Niguarda; Università degli Studi di Milano; and IFOM-ETS The AIRC Institute of Molecular Oncology","correspondingAuthor":false,"prefix":"","firstName":"Gianluca","middleName":"","lastName":"Mauri","suffix":""},{"id":493219819,"identity":"9fdf5d8b-0e23-4523-b454-1b5f0922ed65","order_by":1,"name":"Lucia Santorelli","email":"","orcid":"","institution":"Università degli Studi di Milano, and IFOM-ETS The AIRC Institute of Molecular Oncology, Milan, Italy","correspondingAuthor":false,"prefix":"","firstName":"Lucia","middleName":"","lastName":"Santorelli","suffix":""},{"id":493219820,"identity":"66d13a56-4195-49dd-80df-549b7f7e3ab4","order_by":2,"name":"Federica Marasca","email":"","orcid":"","institution":"INGM, Istituto Nazionale Genetica Molecolare \"Romeo ed Enrica Invernizzi\"","correspondingAuthor":false,"prefix":"","firstName":"Federica","middleName":"","lastName":"Marasca","suffix":""},{"id":493219821,"identity":"0a2b21eb-e0c8-4d25-9450-b485331d342a","order_by":3,"name":"Valeria Ranzani","email":"","orcid":"","institution":"INGM","correspondingAuthor":false,"prefix":"","firstName":"Valeria","middleName":"","lastName":"Ranzani","suffix":""},{"id":493219822,"identity":"f31daeb0-a120-404f-a913-72e3f034be96","order_by":4,"name":"Giovanni Crisafulli","email":"","orcid":"https://orcid.org/0000-0002-5511-1555","institution":"IFOM-ETS The AIRC Institute of Molecular Oncology","correspondingAuthor":false,"prefix":"","firstName":"Giovanni","middleName":"","lastName":"Crisafulli","suffix":""},{"id":493219823,"identity":"c5f9679a-57f0-4e78-98aa-b31e6f161ad4","order_by":5,"name":"Erica Gasparotto","email":"","orcid":"","institution":"INGM, Istituto Nazionale Genetica Molecolare \"Romeo ed Enrica Invernizzi\"","correspondingAuthor":false,"prefix":"","firstName":"Erica","middleName":"","lastName":"Gasparotto","suffix":""},{"id":493219824,"identity":"51b47744-9b9c-4ec8-b385-b938da5ac8e4","order_by":6,"name":"Marta Puzzono","email":"","orcid":"","institution":"Gastroenterology and Gastrointestinal Endoscopy Unit, Vita-Salute San Raffaele University, IRCCS San Raffaele Scientific Institute","correspondingAuthor":false,"prefix":"","firstName":"Marta","middleName":"","lastName":"Puzzono","suffix":""},{"id":493219825,"identity":"e542f094-07ad-485e-9501-83b6660d9e2a","order_by":7,"name":"Iosune Baraibar","email":"","orcid":"","institution":"Vall Hebron University Hospital and Vall Hebron Institute of Oncology (VHIO)","correspondingAuthor":false,"prefix":"","firstName":"Iosune","middleName":"","lastName":"Baraibar","suffix":""},{"id":493219826,"identity":"1669d02d-acad-4ba4-86e2-5962c8d5b079","order_by":8,"name":"Lorenzo Salviati","email":"","orcid":"","institution":"INGM, Istituto Nazionale Genetica Molecolare \"Romeo ed Enrica Invernizzi\"","correspondingAuthor":false,"prefix":"","firstName":"Lorenzo","middleName":"","lastName":"Salviati","suffix":""},{"id":493219827,"identity":"8e4e2687-3e8a-4327-bbb8-56108aaf6f50","order_by":9,"name":"Alberto Sogari","email":"","orcid":"","institution":"IFOM-ETS The AIRC Institute of Molecular Oncology, and Department of Oncology, Molecular Biotechnology Center, University of Torino","correspondingAuthor":false,"prefix":"","firstName":"Alberto","middleName":"","lastName":"Sogari","suffix":""},{"id":493219828,"identity":"7050bf00-64b7-4fbf-8303-e9b1047fe4a9","order_by":10,"name":"Sharon Scardellato","email":"","orcid":"","institution":"IFOM-ETS The AIRC Institute of Molecular Oncology","correspondingAuthor":false,"prefix":"","firstName":"Sharon","middleName":"","lastName":"Scardellato","suffix":""},{"id":493219829,"identity":"6ed99f7a-76da-4d9b-bad7-cea095c9c38a","order_by":11,"name":"Silvia Ghezzi","email":"","orcid":"","institution":"Department of Hematology Oncology, and Molecular Medicine, Grande Ospedale Metropolitano Niguarda, Milan, Italy","correspondingAuthor":false,"prefix":"","firstName":"Silvia","middleName":"","lastName":"Ghezzi","suffix":""},{"id":493219830,"identity":"3dd08fdd-2d67-43dc-836b-919735d1ff06","order_by":12,"name":"Sara Mariano","email":"","orcid":"","institution":"Department of Hematology Oncology, and Molecular Medicine, Grande Ospedale Metropolitano Niguarda, Milan, Italy","correspondingAuthor":false,"prefix":"","firstName":"Sara","middleName":"","lastName":"Mariano","suffix":""},{"id":493219831,"identity":"0d00f6b1-1394-4be5-ae9c-8c0ee8088b30","order_by":13,"name":"Nadia Saoudi-González","email":"","orcid":"","institution":"Vall Hebron University Hospital and Vall Hebron Institute of Oncology (VHIO)","correspondingAuthor":false,"prefix":"","firstName":"Nadia","middleName":"","lastName":"Saoudi-González","suffix":""},{"id":493219832,"identity":"1ce8c0cc-64d0-4fb2-8c56-a6a94b3b19d4","order_by":14,"name":"Letizia Monti","email":"","orcid":"","institution":"Department of Hematology Oncology, and Molecular Medicine, Grande Ospedale Metropolitano Niguarda","correspondingAuthor":false,"prefix":"","firstName":"Letizia","middleName":"","lastName":"Monti","suffix":""},{"id":493219833,"identity":"384a638c-9741-4c77-9f0d-881aa872cdd2","order_by":15,"name":"Alessandro Mannucci","email":"","orcid":"https://orcid.org/0000-0002-1655-6762","institution":"IRCCS San Raffaele Scientific Institute","correspondingAuthor":false,"prefix":"","firstName":"Alessandro","middleName":"","lastName":"Mannucci","suffix":""},{"id":493219834,"identity":"78e7bb6e-ee6a-4330-b9e9-988ee244ff76","order_by":16,"name":"Martina Di Como","email":"","orcid":"","institution":"Department of Pathology, Grande Ospedale Metropolitano Niguarda","correspondingAuthor":false,"prefix":"","firstName":"Martina","middleName":"Di","lastName":"Como","suffix":""},{"id":493219835,"identity":"7368c1c3-54dc-4310-886c-6d53b1fc01dd","order_by":17,"name":"Federica Tosi","email":"","orcid":"","institution":"Department of Hematology, Oncology, and Molecular Medicine, Grande Ospedale Metropolitano Niguarda, Milan, Italy","correspondingAuthor":false,"prefix":"","firstName":"Federica","middleName":"","lastName":"Tosi","suffix":""},{"id":493219836,"identity":"86e7f914-58b0-4789-8887-16b96996c5f5","order_by":18,"name":"Erica Bonazzina","email":"","orcid":"","institution":"Ospedale Niguarda Ca' Granda","correspondingAuthor":false,"prefix":"","firstName":"Erica","middleName":"","lastName":"Bonazzina","suffix":""},{"id":493219837,"identity":"29e737d8-8b8b-4b03-bcc4-d08684c078eb","order_by":19,"name":"Giorgia Parodi","email":"","orcid":"","institution":"Cogentech S.C.a.R.L.","correspondingAuthor":false,"prefix":"","firstName":"Giorgia","middleName":"","lastName":"Parodi","suffix":""},{"id":493219838,"identity":"9405465d-42e8-4d0a-bced-081d528c8be7","order_by":20,"name":"Maria Costanza Aquilano","email":"","orcid":"https://orcid.org/0000-0002-2955-0845","institution":"Department of Pathology, Grande Ospedale Metropolitano Niguarda, Milan, Italy","correspondingAuthor":false,"prefix":"","firstName":"Maria","middleName":"Costanza","lastName":"Aquilano","suffix":""},{"id":493219839,"identity":"343c2c8f-03cf-4f75-b450-b5b12c26570c","order_by":21,"name":"Angela Cattaneo","email":"","orcid":"","institution":"Cogentech S.C.a.R.L.","correspondingAuthor":false,"prefix":"","firstName":"Angela","middleName":"","lastName":"Cattaneo","suffix":""},{"id":493219840,"identity":"03f1a6f9-500d-4d06-a907-40ef3c2ab0dd","order_by":22,"name":"Giorgio Patelli","email":"","orcid":"https://orcid.org/0000-0001-8697-2158","institution":"Department of Hematology Oncology, and Molecular Medicine, Grande Ospedale Metropolitano Niguarda; Università degli Studi di Milano; and IFOM-ETS The AIRC Institute of Molecular Oncology","correspondingAuthor":false,"prefix":"","firstName":"Giorgio","middleName":"","lastName":"Patelli","suffix":""},{"id":493219841,"identity":"32c22bbf-2fc5-46f9-a210-dd3134b84159","order_by":23,"name":"Alessio Amatu","email":"","orcid":"https://orcid.org/0000-0001-5396-3378","institution":"Niguarda Cancer Center, Grande Ospedale Metropolitano Niguarda","correspondingAuthor":false,"prefix":"","firstName":"Alessio","middleName":"","lastName":"Amatu","suffix":""},{"id":493219842,"identity":"793a3ef0-8da0-4223-962a-b968542a3c94","order_by":24,"name":"Emanuela Bonoldi","email":"","orcid":"","institution":"Department of Pathology, Grande Ospedale Metropolitano Niguarda, Milan, Italy","correspondingAuthor":false,"prefix":"","firstName":"Emanuela","middleName":"","lastName":"Bonoldi","suffix":""},{"id":493219843,"identity":"de5cbd19-1970-4bcf-80ff-0f4ec02664a7","order_by":25,"name":"Luca Lazzari","email":"","orcid":"https://orcid.org/0000-0001-6841-4240","institution":"IFOM ETS - The AIRC Institute of Molecular Oncology","correspondingAuthor":false,"prefix":"","firstName":"Luca","middleName":"","lastName":"Lazzari","suffix":""},{"id":493219844,"identity":"9ee2e416-3c37-4a13-98a3-ceca969649fe","order_by":26,"name":"Elena Elez","email":"","orcid":"https://orcid.org/0000-0002-4653-6324","institution":"Vall d´Hebron Institute of Oncology (VHIO)","correspondingAuthor":false,"prefix":"","firstName":"Elena","middleName":"","lastName":"Elez","suffix":""},{"id":493219845,"identity":"42ec1515-8ad2-4aaa-b624-bfcb35003b88","order_by":27,"name":"Sabrina Arena","email":"","orcid":"https://orcid.org/0000-0002-1318-2494","institution":"University of Torino and IRCCS-FPO","correspondingAuthor":false,"prefix":"","firstName":"Sabrina","middleName":"","lastName":"Arena","suffix":""},{"id":493219846,"identity":"5be0b783-4837-46b4-b731-38786818bde5","order_by":28,"name":"Katia Bencardino","email":"","orcid":"","institution":"Department of Hematology Oncology, and Molecular Medicine, Grande Ospedale Metropolitano Niguarda","correspondingAuthor":false,"prefix":"","firstName":"Katia","middleName":"","lastName":"Bencardino","suffix":""},{"id":493219847,"identity":"e820919a-0468-471d-a9e7-1dfde37ff680","order_by":29,"name":"Silvia Marsoni","email":"","orcid":"","institution":"IFOM, FIRC Institute of Molecular Oncology, Milan, Italy","correspondingAuthor":false,"prefix":"","firstName":"Silvia","middleName":"","lastName":"Marsoni","suffix":""},{"id":493219848,"identity":"257593ff-55e9-42ea-876f-4a219d69cf03","order_by":30,"name":"Giulia Martina Cavestro","email":"","orcid":"","institution":"Gastroenterology and Gastrointestinal Endoscopy Unit, Vita-Salute San Raffaele University, IRCCS San Raffaele Scientific Institute","correspondingAuthor":false,"prefix":"","firstName":"Giulia","middleName":"Martina","lastName":"Cavestro","suffix":""},{"id":493219849,"identity":"14f5d0bb-505e-4475-a21d-478f664c0312","order_by":31,"name":"Andrea Sartore-Bianchi","email":"","orcid":"https://orcid.org/0000-0003-0780-0409","institution":"Grande Ospedale Metropolitano Niguarda; Università degli Studi di Milano","correspondingAuthor":false,"prefix":"","firstName":"Andrea","middleName":"","lastName":"Sartore-Bianchi","suffix":""},{"id":493219850,"identity":"ef838736-556c-4186-b6f3-adef5a668778","order_by":32,"name":"Angela Bachi","email":"","orcid":"https://orcid.org/0000-0003-4842-6556","institution":"IFOM","correspondingAuthor":false,"prefix":"","firstName":"Angela","middleName":"","lastName":"Bachi","suffix":""},{"id":493219851,"identity":"c39838e1-ef1f-407c-98ca-abb537321e8a","order_by":33,"name":"Sergio Abrignani","email":"","orcid":"https://orcid.org/0000-0002-0794-3285","institution":"Istituto Nazionale Genetica Molecolare","correspondingAuthor":false,"prefix":"","firstName":"Sergio","middleName":"","lastName":"Abrignani","suffix":""},{"id":493219852,"identity":"0501e13e-46e6-4499-a1f4-e5b465042ac7","order_by":34,"name":"Vincenzo Costanzo","email":"","orcid":"","institution":"Department of Oncology and Hemato-Oncology, Università degli Studi di Milano, and IFOM-ETS The AIRC Institute of Molecular Oncology","correspondingAuthor":false,"prefix":"","firstName":"Vincenzo","middleName":"","lastName":"Costanzo","suffix":""},{"id":493219853,"identity":"0a960072-3436-4002-a426-6849c3ae7fcd","order_by":35,"name":"Beatrice Bodega","email":"","orcid":"https://orcid.org/0000-0003-0527-9234","institution":"Istituto Nazionale di Genetica Molecolare","correspondingAuthor":false,"prefix":"","firstName":"Beatrice","middleName":"","lastName":"Bodega","suffix":""},{"id":493219854,"identity":"78e6c5d6-f615-44fb-8e34-5c2360cbe70b","order_by":36,"name":"Alberto Bardelli","email":"","orcid":"https://orcid.org/0000-0003-1647-5070","institution":"University of Torino","correspondingAuthor":false,"prefix":"","firstName":"Alberto","middleName":"","lastName":"Bardelli","suffix":""},{"id":493219817,"identity":"516534f0-b11e-4c02-a1c5-af01a2d09c48","order_by":37,"name":"Salvatore Siena","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA9UlEQVRIiWNgGAWjYFAC5gY48wAQyzEwMDYwJDBI4NECVABUAddiDNOCRw+SFhBIhFmKU4t8+8E2iZ8/bPL5GdgfHvi4wy59w+3m1g0PGCzqcGkxOJPYJtmTkGY5s4HH4ODMM8m5G+4cbLuBz2EGDInNBjwJhw0MDvAwHOZtY87dcCMRvxb5/ofNhn+AWuwPsD84/LetPt2AkBaGG4mNj8G2AG08zNh2OIGgFoMbDxsfy6SlGUgcBvqlt+244UywFgMJyQacDks+cPCNjY0Bf3v74w8/26rl+W6kP7v5o6KOH6fD4IAZ1XbCGkbBKBgFo2AU4AYA1xpcISo+XUQAAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0002-2681-2846","institution":"Università degli Studi di Milano and Grande Ospedale Metropolitano Niguarda","correspondingAuthor":true,"prefix":"","firstName":"Salvatore","middleName":"","lastName":"Siena","suffix":""}],"badges":[],"createdAt":"2025-07-23 07:40:14","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7193450/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7193450/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":88424570,"identity":"222d976d-7548-4887-b606-ed30eb7834fd","added_by":"auto","created_at":"2025-08-06 09:47:54","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":275726,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eProteomic profiling of EO-CRC samples reveals a specific placenta signature over-expression (Discovery cohort). \u003c/strong\u003eA) Schematic of the proteomic workflow applied to fresh frozen samples collected. B) Principal Component Analysis (PCA) of the Mass Spectrometry (MS) full proteomes from EO-CRC and SO-CRC solid samples of discovery cohort. C) Volcano plot of the identified and quantified proteins in EO-CRC vs SO-CRC samples. In orange the placental related terms. D, E) List and frequency of Gene Ontology classification biological process (GOBP) terms related to the deregulated proteins after comparison between EO-CRC and SO-CRC proteomes. Top enriched terms (D) and specific items related to system development (E). F) Distribution of residence specificity terms obtained by cells line and tissue type pattern genes enrichment analysis. G) Differential expression of DNA metabolism-related proteins in EO-CRC versus SO-CRC. Boxplots show the log2 Label Free Quantity (LFQ) intensity (group mean) for key proteins involved in DNA metabolism (***p \u0026lt; 0.0001, unpaired t-test). Protein regulating DNA replication stress (SMARCAL1, CDK1, CDK2, RRM2B) and DNA repair (ERCC2, ERCC3, NBN, UNG, PARP10, APOBEC3C) are listed. H) Chart represents the frequency of proteins identified as placental- and not placental-related in the differently expressed proteomes of cancerous and matched-normal samples. I) Plot describes the total number (%) of proteins belonging to the placenta core presents in all the samples and the remaining number (%) of placenta proteins specific for the EO-CRC group, after control adjustment. J) Violin plots of stem core proteins showing the relative change in abundance levels between EO-CRC and SO-CRC classes, after control adjustment. K) Hierarchical clustering of CRC samples based on placental protein expression profiles, after control adjustment. Clustering was performed using Euclidean distance and Ward linkage, highlighting molecular differences between EOCRC and SO-CRC subtypes. L) Heatmap of differentially expressed proteins in EO-CRC and SO-CRC. Samples were hierarchically clustered using ward’s method after control adjustment. Rows represent proteins, and columns represent individual samples. Colour intensity indicates relative protein expression levels (centered and scaled). Distinct clustering patterns reflect differences in protein expression profiles of placenta-related features between EOCRC and SOCRC groups.\u003c/p\u003e","description":"","filename":"Figure141.png","url":"https://assets-eu.researchsquare.com/files/rs-7193450/v1/ce4d6228b7dc17dcba232cbf.png"},{"id":88423534,"identity":"c71de772-ca23-43fd-820d-f45ba0ff1ad5","added_by":"auto","created_at":"2025-08-06 09:39:54","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":193368,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePatient-derived organoids retain placental features. \u003c/strong\u003eFour patient-derived organoids (PDOs) or tumoroids were established from two early-onset colorectal cancer (EO-CRC) patients [EOCRC8 (PDO4) and EOCRC18 (PDO1)] and from two standard-onset CRC (SO-CRC) patients [SOCRC26 (PDO2) and EOCRC27 (PDO3)]. A) Schematic of the proteomic workflow applied to CRC-PDOs B) PCA of the MS full proteomes from CRC-PDOs and EO-CRC and SO-CRC solid samples from both validation and discovery cohorts. C) Correlation matrix of PDOs samples. Heatmap showing pairwise Pearson correlation coefficients among four PDO samples. D) The plot illustrates the overlap and conservation of placental protein signatures between tissue and PDO models. E) Expression profiles of placenta core proteins grouped into three clusters based on their abundance (log₂ LFQ intensity) across PDOs samples. F) Bar plot displaying the percentage of placenta signature proteins detected in each cluster, relative to the total placenta-related proteins identified in PDOs. G) Donut plot showing the distribution of principal molecular processes associated with placental proteins highlighted in PDOs. Each donut corresponds to a specific protein cluster, with segments representing the relative contribution of key biological processes. H) Hierarchical clustering analysis of placenta signature proteins across the PDOs cohort (Euclidean distance, ward linkage method). I-L) PDOs represent clinically relevant models to assess sensitivity to two different ATRi under clinical development in vitro in a 5-day-long viability assay. The results at the endpoint were normalized to control wells containing DMSO vehicle and were plotted into two different modalities. I, J) Dose-response curves to Ceralasertib and Berzosertib respectively. K, L) Histograms (a percentage of viability) at clinically relevant concentrations of individual inhibitors (Ceralasertib 1 μM and Berzosertib 0.3 μM) respectively. EOCRC8 was more sensitive to ceralasertib compared to EOCRC18 (p=0.0098), SOCRC26 (p=0.0167), and SOCRC27 (p=00019). EOCRC8 was more sensitive to berzosertib compared to EOCRC18 (p=0.0001), SOCRC26 (p=0.0015), and SOCRC27 (p=0.0014). MG-132 was used as a positive control for organoid death. Statistical analyses were performed using GraphPad Prism 10.5.0.\u003c/p\u003e","description":"","filename":"Figure142.png","url":"https://assets-eu.researchsquare.com/files/rs-7193450/v1/fc55da3e092da448853a03eb.png"},{"id":88423537,"identity":"b6f301dd-a3ca-4e6e-944c-1d5fb04f1be5","added_by":"auto","created_at":"2025-08-06 09:39:54","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":76977,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAnalysis of retrotransposable elements expression revealed that HERVH is expressed in EO-CRC tumour progression\u003c/strong\u003e. A-B) Gene set–enrichment analysis (GSEA), presented as enrichment score profiles for the placenta signature derived from proteomic data in the comparison between (A) EOCRC and normal adjacent tissue (nominal p-value \u0026lt; 0.001, NES=1.39, FDR\u0026lt;0.001) and (B) EOCRC metastasis and EOCRC (nominal p-value \u0026lt;0.001, NES=1.38, FDR \u0026lt; 0.001). Volcano plots showing differentially expressed transposable element subfamilies in (C) EOCRC versus adjacent normal tissue and in (D) metastasis versus EOCRC. Selected upregulated TE subfamilies are highlighted. E-F) Boxplot of normalized TE counts of (E) HERVH-int and (F) LTR7Y in normal, EOCRC and metastasis conditions. Wilcoxon rank sum exact test of normal versus tumor and normal versus metastasis (HERVH-int: *P-value \u0026lt; 0.05, from left to the right P-value 0.0296, 0.0225. LTR7Y: **P-value \u0026lt; 0.01, *** P-value \u0026lt; 0.001, from left to the right P-value 0.00761, 0.00077).\u003c/p\u003e","description":"","filename":"Figure143.png","url":"https://assets-eu.researchsquare.com/files/rs-7193450/v1/2216f739bc9cf9f6caed3c50.png"},{"id":88423541,"identity":"87e71cf3-494f-47a4-99e1-4878a4b974e9","added_by":"auto","created_at":"2025-08-06 09:39:54","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":530078,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eHERVH expression is upregulated in EO-CRC patients. \u003c/strong\u003eA) Representative fluorescence microscopy images of HERVH-ISH (magenta) acquired with a widefield microscope equipped with a spinning-disk confocal unit performed in adjacent normal mucosa and tumor tissue in EO-CRC patients. Images were processed with Richardson–Lucy two-dimensional deconvolution using NIS-Elements analysis AR imaging software. Nuclei were stained with 4,6-diamidino-2-phenylindole (DAPI). Original magnification, 40×. B) Relative quantification of HERVH dots number of per cell in EO-CRC patients (N=16). Statistical significance was determined using two-tailed Mann–Whitney test (from left to the right p\u0026lt; 0.001, \u0026lt; 0.001, = 0.006, = 0.02, \u0026lt; 0.001, = 0.001, \u0026lt; 0.001, = 0.004). C) Representative fluorescence microscopy images of HERVH-ISH (magenta) acquired with a widefield microscope equipped with a spinning-disk confocal unit performed in adjacent normal mucosa and tumour tissue in SO-CRC patients. Images were processed with Richardson–Lucy two-dimensional deconvolution using NIS-Elements analysis AR imaging software. Nuclei were stained with DAPI. Original magnification, 40×. D) Relative quantification of HERVH dots number of per cell in SO-CRC patients (N=14). Statistical significance was determined using two-tailed Mann–Whitney test (from left to the right p\u0026lt; 0.001, \u0026lt; 0.001, \u0026lt; 0.001, \u0026lt; 0.001, = 0.01, = 0.03, = 0.0027, \u0026lt; 0.001, = 0.04, = 0.002). E) Box plots showing the average number of HERVH-positive dots per cell for each patient, quantified in both adjacent normal mucosa and tumor tissue from EO-CRC and SO-CRC patients. Each dot represents the average value for an individual patient. Statistical significance was determined using two-tailed Mann–Whitney test (from left to the right p= \u0026lt; 0.001, 0.002). All statistical analysis was performed with GraphPad Prism 10.4.1.\u003c/p\u003e","description":"","filename":"Figure144.png","url":"https://assets-eu.researchsquare.com/files/rs-7193450/v1/9f35a007a88851833aa55ea3.png"},{"id":88424571,"identity":"25473936-4b40-474a-8f82-e0e7c27fef7d","added_by":"auto","created_at":"2025-08-06 09:47:54","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":2355181,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSchematic summary of the early-onset colorectal cancer placental-like features and their potential impact of both the tumour and its microenvironment\u003c/strong\u003e. The tumour is schematically represented in a fashion resembling placental capability of infiltrating adjacent tissues. Here, the tumour microenvironment is hypothesized and represented to be characterized by exhausted immune cells, and vessels early infiltrated by spreading tumour cells. The distinctive placental-like features identified in early-onset colorectal cancer (EO-CRC) can serve as: i) rationale to explain the aggressive clinical features of EO-CRC; ii) a novel biological hint to refine the search for EO-CRC incidence increase causes focusing on environmental/dietary hypomethylating agents; iii) a potential target for treatment specific for EO-CRC; iv) HERVH as a novel potential biomarker to screen individuals at higher risk to develop EO-CRC.\u003c/p\u003e","description":"","filename":"Figure145.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7193450/v1/54c5acfe3f0df5a583d3de23.jpg"},{"id":90011272,"identity":"84f97ed1-61bf-4562-8afa-62d6ac9b4b99","added_by":"auto","created_at":"2025-08-27 10:49:24","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4712144,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7193450/v1/080927ff-f6fa-4745-966e-1f61aae28a31.pdf"},{"id":88423539,"identity":"61d92015-f92f-4308-b44f-f1bd93b7bc33","added_by":"auto","created_at":"2025-08-06 09:39:54","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":226799,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSupplementary Figure 1. Samples disposition from early-onset and standard-onset colorectal cancer patients.\u003c/strong\u003e Identification and collection of both fresh and archival samples from both early-onset (EO-CRC) and standard-onset (SO-CRC) patients at three Institutions across Italy and Spain: Grande Ospedale Metropolitano Niguarda, and San Raffaele Hospital, Milan, Italy, and Vall d’Hebron Institute of Oncology (VHIO), Barcelona, Spain. Patients diagnosed with hereditary cancer predisposition syndromes (HSCPs) were excluded from our analysis (N=3, EO-CRC patients). \u003cu\u003eKeys\u003c/u\u003e: EO-CRC = early-onset colorectal cancer; SO-CRC = standard-onset colorectal cancer; FFPE = formalin-fixed paraffin-embedded samples; PDO = patient-derived organoid; HSCPs = hereditary cancer predisposition syndromes; * = Primary tumour tissue was not available for analysis; however, to ensure statistical balance and an adequate number of healthy mucosae samples for a robust comparative analysis, adjacent healthy mucosae from both matched and unmatched SO-CRC patients in the cohort.\u003c/p\u003e","description":"","filename":"SupplementaryFigure1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7193450/v1/de9484252fe9a4bea7a13d89.pdf"},{"id":88423538,"identity":"6e9eb2c4-6c0e-4789-af11-d1d4fed05989","added_by":"auto","created_at":"2025-08-06 09:39:54","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":821850,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSupplementary Figure 2. Proteomic profiling of early-onset colorectal cancer (EO-CRC) primary tumour samples reveal a specific placenta signature over-expression. \u003c/strong\u003eA, B) List and frequency of GOCC terms related to the up-regulated (A) and down-regulated (B) proteins after the comparison between young and old cancerous proteomes. C) Principal Component Analysis (PCA) of the Mass Spectrometry (MS) full proteome from EO-CRC and SO-CRC affected and non-affected adjacent tissues, N-EO-CRC and N-SO-CRC. D) Hierarchical clustering of CRC and corresponding normal mucosa samples based on full proteome expression profiles. Clustering was performed using Euclidean distance and Ward linkage, highlighting molecular differences between EO-CRC and SO-CRC groups. E) Distribution of tissue- and cell-specificity terms associated with deregulated proteins in EO-CRC compared to SO-CRC, based on annotations from three reference databases. F) Differential expression of proteins involved in tissue invasion, immunomodulation, and defence response in EO-CRC) versus SO-CRC. Boxplots display the log2 LFQ intensity (group mean) for key proteins associated with these processes. Statistical significance was determined using an unpaired t-test (****p \u0026lt; 0.0001, *p \u0026lt; 0.05).\u003c/p\u003e","description":"","filename":"SupplementaryFigure2.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7193450/v1/e718e664006b09a0abac3f10.pdf"},{"id":88423542,"identity":"9d8ff78f-1eb5-4720-aec8-478b312c531f","added_by":"auto","created_at":"2025-08-06 09:39:54","extension":"pdf","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":7615525,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSupplementary Figure 3. Proteomic Profiling of EO-CRC samples reveal a specific placenta signature over-expression (Validation cohort).\u003c/strong\u003e\u003cem\u003e\u003cstrong\u003e \u003c/strong\u003e\u003c/em\u003eA) PCA of the MS full proteomes from EO-CRC and SO-CRC solid samples of validation cohort. B) Volcano plot of the identified and quantified proteins in EO-CRC vs SO-CRC samples. In orange the placental related terms. C, D) List and frequency of GOBP terms related to the deregulated proteins after comparison between EO-CRC and SO-CRC proteomes. Top enriched terms (C) and specific items related to system development (D). E) Distribution of tissue- and cell-specificity terms associated with deregulated proteins in EOCRC compared to SO-CRC, based on annotations from three reference databases. F) Chart represents the frequency of proteins identified as placental- and not placental-related in the differently expressed proteomes of cancerous and non-affected samples. G) Plot describes the total number (%) of proteins belonging to the placenta core presents in all the samples and the remaining number (%) of placenta proteins specific to the EO-CRC group, after control adjustment. H) Violin plots of placental core proteins showing the relative change in abundance levels between EO-CRC and SO-CRC classes, after control adjustment. J) Unsupervised hierarchical clustering of CRC samples based on expression profiles of placenta-associated proteins, following control adjustment. Clustering (Euclidean distance and ward linkage method) reveal distinct molecular signatures between EO-CRC and SO-CRC cases. K) Heatmap illustrating the expression of placenta-related proteins differentially expressed between EOCRC and SOCRC. Samples were clustered using ward method after control adjustment. Each column represents a sample and each row a protein, with colour intensity indicating normalized expression levels. The resulting clustering pattern underscores differences in placenta-associated molecular features between EO-CRC and SO-CRC groups.\u003c/p\u003e","description":"","filename":"SupplementaryFigure3.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7193450/v1/2f3457a10638b852ba815cca.pdf"},{"id":88423536,"identity":"ba0f3be5-7e0d-4ae2-a026-27dad32d55de","added_by":"auto","created_at":"2025-08-06 09:39:54","extension":"pdf","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":59712,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSupplementary Figure 4. DNA repair proteins expression as assessed by immunohistochemistry in the EOCRC8 patient derived organoid\u003c/strong\u003e. As suggested by previous publication (\u003cem\u003eDurinikova et al. Clin Can Res, 2022\u003c/em\u003e), we ruled out the presence of key molecular alterations explaining the sensitivity to ATR inhibitors (ceralasertib and berzosertib) observed in the EOCRC8 tumoroid. Particularly, RAD51, RAD51C and ATM proteins expression was retained in the EOCRC8 tumoroid, which was instead retaining placental and replication stress features observed in the match fresh tumour tissue.\u003c/p\u003e","description":"","filename":"SupplementaryFigure4.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7193450/v1/64758d8cf41bd39cf6e80a0e.pdf"},{"id":88424572,"identity":"3bc965a2-6489-451e-8118-9cbf42120d96","added_by":"auto","created_at":"2025-08-06 09:47:54","extension":"pdf","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":609226,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSupplementary Figure 5. Transcriptomic profiling highlights placenta-like signature enrichment in EO-CRC tumour and metastasis. \u003c/strong\u003eA) Principal Component Analysis (PCA) on DESeq2 variance-stabilized RNA-seq data from EO-CRC, adjacent normal tissue, and metastases. B) Gene set–enrichment analysis (GSEA), presented as enrichment score profiles for Placenta development geneset from MSigDB in the comparison between EOCRC and normal adjacent tissue (Nominal p-value \u0026lt; 0.001, NES=1.62, FDR=0.02). C) Gene set–enrichment analysis (GSEA), presented as enrichment score profiles for Placenta development geneset from MSigDB in the comparison between EOCRC metastasis and EOCRC. (Nominal p-value \u0026lt;0.001, NES=1.76, FDR=0.006).\u003c/p\u003e","description":"","filename":"SupplementaryFigure5.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7193450/v1/7498590f07afa4dd660e84a8.pdf"},{"id":88423544,"identity":"e52138d3-591c-49bc-8295-d054cbdd9dce","added_by":"auto","created_at":"2025-08-06 09:39:54","extension":"docx","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":755925,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSupplementary table 1. Clinicopathological features of early-onset colorectal cancers (n=33) compared to standard-onset colorectal cancers (n=27) included in the present study\u003c/strong\u003e. \u003cu\u003eKeys\u003c/u\u003e: * Fisher’s Exact Test was performed. ** Located in caecum, ascending colon, liver flexure and transverse colon. ^ Located in splenic flexure, descending colon and sigmoid colon. NOS Not otherwise specified. ^^ In this patients Lynch syndrome was ruled out by performing extended germline panelling, even if the tumour somatic mismatch repair status was not assessed.\u003c/p\u003e","description":"","filename":"SupplementaryTable1.docx","url":"https://assets-eu.researchsquare.com/files/rs-7193450/v1/9f5e931bcf145530bb0bf4b8.docx"},{"id":88423548,"identity":"bb8296a8-1bb0-4465-bce6-00c09d4905e2","added_by":"auto","created_at":"2025-08-06 09:39:54","extension":"xlsx","order_by":7,"title":"","display":"","copyAsset":false,"role":"supplement","size":7711095,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSupplementary Table 2. Proteomic profiling of EO-CRC samples in the discovery cohort reveals a distinct overexpression of a placenta-associated signature. \u003c/strong\u003eList of the table summarizing all analyses performed on the discovery cohort (statistical and functional).\u003c/p\u003e","description":"","filename":"SupplementaryTable2.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7193450/v1/087a50c2d3766bb5bf898df5.xlsx"},{"id":88423549,"identity":"ec6f3c95-d744-4651-8fc6-10da9abdaac2","added_by":"auto","created_at":"2025-08-06 09:39:54","extension":"xlsx","order_by":8,"title":"","display":"","copyAsset":false,"role":"supplement","size":8238276,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSupplementary Table 3. Proteomic profiling of EO-CRC samples in the validation cohort reveals a distinct overexpression of a placenta-associated signature. \u003c/strong\u003eList of the table summarizing all analyses performed on the discovery cohort (statistical and functional).\u003c/p\u003e","description":"","filename":"SupplementaryTable3.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7193450/v1/f86e7502aa3b9d27659cc4ae.xlsx"},{"id":88423547,"identity":"144737eb-850b-4663-821c-4442b151bfc0","added_by":"auto","created_at":"2025-08-06 09:39:54","extension":"xlsx","order_by":9,"title":"","display":"","copyAsset":false,"role":"supplement","size":2764711,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSupplementary Table 4. Patient-derived organoids retain placental features. \u003c/strong\u003eList of the table summarizing all the proteomic analyses performed on the discovery cohort (statistical and functional).\u003c/p\u003e","description":"","filename":"SupplementaryTable4.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7193450/v1/388f56aee3be69ef00a405bd.xlsx"},{"id":88423546,"identity":"654b5eed-e3d4-47b6-b8d1-c261d3846541","added_by":"auto","created_at":"2025-08-06 09:39:54","extension":"xlsx","order_by":10,"title":"","display":"","copyAsset":false,"role":"supplement","size":107905,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSupplementary Table 5\u003c/strong\u003e. \u003cstrong\u003eSample and experimental details of RNA-ISH and RNA-seq data.\u003c/strong\u003e Comprehensive metadata and differential expression summary of EO-CRC and SO-CRC patients in RNA-seq and ISH analyses, including lists of samples and RNA-seq sequencing details.\u003c/p\u003e","description":"","filename":"SupplementaryTable5.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7193450/v1/77edda0f74afde99fa021323.xlsx"}],"financialInterests":"\u003cb\u003eYes\u003c/b\u003e there is potential Competing Interest.\nS.S. is advisory board member for Agenus, AstraZeneca, Bayer, Bristol Myers Squibb, CheckmAb, Daiichi-Sankyo, GlaxoSmithKline, MSD, Merck, Novartis, Ospedale San Raffaele, Pierre-Fabre, Pfizer, Seagen, and T-One Therapeutics. A.S-B. reported consulting or advisory role for Bayer, Novartis, Pierre Fabre, Servier, and Takeda; and personal honoraria as an invited speaker from Amgen, Guardant Health, Pierre Fabre, and Seagen. A.Bardelli reports receipt of grants/research supports from Neophore, AstraZeneca, Boehringer Ingelheim and honoraria/consultation fees from Guardant Health. A.B. is stock shareholder of Neophore and Kither Biotech. A.B. is an advisory boards member for Neophore. S.Abrignani is a co-founder of the startup CheckMab s.r.l and T-One Therapeutics s.r.l.; F.Marasca and B.B. are co-founders of the startup T-One Therapeutics s.r.l.. S.Arena reports personal fees from MSD Italia and a patent (Italian patent application No. 102022000007535) outside the submitted work. K.B. is advisory board member for AstraZeneca. N.S. reports personal honoraria as an invited speaker for AMGEN, Medistream (OncoBites2025), and travel supports from AMGEN, MERCK, and BAYER. N.S. is an ESMO Fellow (starting from April 2024). F.T. reports travel supports from ROCHE. A.A. is advisory board member for AMGEN e Italfarmaco. I.B. has received accommodation and travel expenses from Amgen, Merck, Sanofi, and Servier; and personal speaker honoraria from Astra Zeneca. E.E. reports the following: Honoraria, Consulting or Advisory Role, and Speakers's Bureau: Agenus, Amgen, Bayer, Bristol Myers Squibb, Boehringer Ingelheim, Cure Teq AG, GlaxoSmithKline, Hoffman La-Roche, Janssen, Johnson\u0026Johnson, Lilly, Medscape, Merck Serono, MSD, Nordic Group BV, Novartis, Organon, Pfizer, Pierre Fabre, Repare Therapeutics Inc., RIN Institute Inc., Rottapharm Biotech, Sanofi, Seagen International GmbH, Servier, and Takeda.","formattedTitle":"Early-Onset Colorectal Cancers Exhibit Distinctive Placental-Like Features","fulltext":[{"header":"Introduction","content":"\u003cp\u003eThe incidence of early-onset colorectal cancers (EO-CRC), defined as colorectal cancer (CRC) occurring in adults under 50 years of age, is rising globally \u003csup\u003e\u003cspan additionalcitationids=\"CR2 CR3 CR4\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e. EO-CRC often presents with clinicopathological features indicative of intrinsic aggressiveness, including mucinous histology, poor differentiation, and peritoneal spread associated with poorer response to therapy and survival \u003csup\u003e\u003cspan additionalcitationids=\"CR7\" citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. Whether these characteristics reflect a distinct underlying biology remains unknown \u003csup\u003e\u003cspan additionalcitationids=\"CR10\" citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eElucidating EO-CRC molecular features is expected to inform the development of tailored screening and therapeutic approaches \u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e. Multiple studies have shown that EO-CRC and standard onset colorectal cancer (SO-CRC) genomic landscapes are broadly similar, with no significant differences in the prevalence of genomic alterations occurring in \u003cem\u003eRAS\u003c/em\u003e, \u003cem\u003eAPC\u003c/em\u003e, and \u003cem\u003eTP53\u003c/em\u003e genes \u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e,\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e. Recent studies have reported an enrichment of single base substitution (SBS) signature 88 and small insertion and deletion (ID) signature 18 in EO-CRC, suggesting a potential causative role for the bacterial colibactin genotoxin in 30% of cases \u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e. However, given their geographic variability \u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e, additional causative factors are likely to contribute to EO-CRC rising incidence. Overall, whether EO-CRC are biologically and molecularly distinct from SO-CRC remain unresolved \u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eWe previously hypothesized that cancer development involve the reactivation of molecular pathways active in para-physiological states such as pregnancy \u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e. Placentation is a highly orchestrated developmental process enabling rapid, regulated growth in mammals. To establish a functional placenta, trophoblast cells acquire properties strikingly similar to those observed in cancers as they invade healthy tissue, induce neovascularization, and create an immune-tolerant microenvironment \u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e. This physiological mimicry highlights a shared biology, where key malignancy hallmarks, including tissue invasion, immune evasion, and sustained proliferative signalling are naturally recapitulated during placentation \u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e. As in cancer, placental proliferation is driven by elevated IGF/MAPK signalling, anti-apoptotic pathways activation, genome duplication events leading to polyploidy \u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e and occurrence of extensive mutations with the intestinal epithelium associated SBS18 signature \u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e being the most prevalent \u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e. Strikingly, the normal placenta exhibits high levels of copy number alterations (CNAs), a feature rarely seen in healthy tissues but frequently in CRC \u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e. However, the extent to which placental mimicry contributes to CRC pathogenesis remains unclear. Several genomic alterations shared by the placenta and CRCs may arise from DNA replication stress, a condition characterized by slowed or stalled replication forks leading to genomic instability through DNA breaks, fragile site activation, and error-prone repair, resulting in copy number gains or losses \u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eIn previous work leading to this study, we demonstrated that embryonic stem cells subjected to replication stress reactivate endogenous retrotransposons (ERVs), driving a cell fate transition toward trophoblast and placental lineages \u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e. Among ERVs, Human Endogenous Retroviruses (HERVs), representing nearly 8% of the human genome \u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e, have emerged as key transcriptional regulators of placental development \u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. The HERVH subfamily, in particular, plays a central role in maintaining pluripotency and guiding trophoblast differentiation by contributing to core self-renewal transcriptional networks \u003csup\u003e\u003cspan additionalcitationids=\"CR26 CR27\" citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e. Aberrant reactivation of HERVs has been consistently reported in multiple cancers, reflecting a pattern consistent with \u003cem\u003eonco-exaptation\u003c/em\u003e, the repurposing of retroelements as cancer regulatory elements \u003csup\u003e\u003cspan additionalcitationids=\"CR30\" citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eIn this study, we employed a multi-omics approach combining high-resolution label-free mass spectrometry (MS)-based proteomics with transcriptomic analyses. This strategy uncovered a distinctive hallmark of EO-CRC: the reactivation of placental-like transcriptional programs and the aberrant expression of HERVH retroelements. These findings suggest that the reactivation of placental-like transcriptional programs may underlie the aggressiveness of EO-CRC, providing a novel perspective on its pathogenesis and opening avenues towards tailored preventive, diagnostic and therapeutic strategies.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eEO-CRC tumour samples were collected from individuals aged 45 years old or younger, and from a control cohort of SO-CRC patients, diagnosed at age 50 or older (\u003cb\u003eSupplementary Fig.\u0026nbsp;1\u003c/b\u003e). EO-CRC samples were collected from a cohort of 33 EO-CRC patients, of whom 29/33 (88%) diagnosed earlier than age 40. EO-CRC patients diagnosed with hereditary cancer predisposition syndromes (HCPSs) were not studied, and we elected to prioritize study of EO-CRC diagnosed under 40 years of age to ensure a more biologically informative cohort. The clinicopathological features of both EO-CRC and SO-CRC (N\u0026thinsp;=\u0026thinsp;27) patients are aligned with previous reports \u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e,\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e. As expected, most of EO-CRCs were in the rectum or left colon, frequently poorly differentiated, and with mucinous features (\u003cb\u003eSupplementary Table\u0026nbsp;1A\u003c/b\u003e and \u003cb\u003eB\u003c/b\u003e).\u003c/p\u003e\u003cp\u003e\u003cspan type=\"ItalicUnderline\" class=\"ItalicUnderline\" name=\"Emphasis\"\u003eProteomics unveils specific placental features of early-onset colorectal cancer\u003c/span\u003e.\u003c/p\u003e\u003cp\u003eTo gain insight into EO-CRC biological and molecular processes, we leveraged the power of high-resolution label-free MS-based proteomics, allowing proteins comprehensive profiling and disease mechanisms elucidation \u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e. By capturing dynamic proteomic changes, MS facilitates the identification of novel biomarkers and therapeutic targets. We applied this approach to compare the protein profiles of primary EO-CRC and SO-CRC tumors, as well as their matched adjacent mucosae (N-EOCRC and N-SOCRC), as controls (Fig.\u0026nbsp;1A). Firstly, we applied label-free quantitation to assess the protein content in EO-CRC (N\u0026thinsp;=\u0026thinsp;12) and SO-CRC (N\u0026thinsp;=\u0026thinsp;7). Results revealed significant proteomic changes upon cancerous conditions in accordance with the clinical classification, as evidenced by distinct separation along principal component 2 (Fig.\u0026nbsp;1B). Motivated by these findings, we proceeded to compare the EO-CRC and SO-CRC proteomes, conducting statistical analyses to explore differential proteins expression. We uncovered 2675 dysregulated proteins, mostly upregulated in EO-CRC (Fig.\u0026nbsp;1C, \u003cb\u003eSupplementary Table\u0026nbsp;2\u003c/b\u003e). Gene Ontology classification biological process (GOBP), cellular component (GOCC) showed a notable upregulation of proteins associated with the mitochondria, endoplasmic reticulum, lysosome, and vesicles in EO-CRC compared to SO-CRC (\u003cb\u003eSupplementary Fig.\u0026nbsp;2A-B, Supplementary Table\u0026nbsp;2\u003c/b\u003e). Of note, several of the significantly deregulated proteins in comparing EO-CRC and SO-CRC were associated with cellular component biogenesis, protein modification, response to replication stress, and system development (Fig.\u0026nbsp;1D, \u003cb\u003eSupplementary Table\u0026nbsp;2\u003c/b\u003e), processes often linked with tumour growth and progression. The developmental category emerged with multiple GOBP terms related to embryonic development significantly enriched in the deregulated proteome of the EO-CRC cohort (Fig.\u0026nbsp;1E, \u003cb\u003eSupplementary Table\u0026nbsp;2\u003c/b\u003e).\u003c/p\u003e\u003cp\u003eThis proteomic profile led us to hypothesize that developmental pathways play a pivotal role in the molecular mechanisms driving EO-CRC, potentially reflecting a reactivation or dysregulation of biological programs typically confined to early developmental stages. To further assess this hypothesis, we investigated whether the differential protein abundances detected between EO-CRC and SO-CRC showed tissue-specific expression patterns, to confirm the developmental pathways relevance. As expected, a large proportion of the differential proteins were distinctive colon-resident. Intriguingly, we identified a significant number of deregulated proteins as typically expressed in highly undifferentiated, embryonic and placental tissues/cell lines (Fig.\u0026nbsp;1C and \u003cb\u003eF, Supplementary Table\u0026nbsp;2\u003c/b\u003e), known for their role in development and regeneration processes. Although gastrointestinal tissues like colon and rectum appeared to dominate the sample proteome landscape, EO-CRC appeared to involve developmentally associated tissues, markedly the placenta.\u003c/p\u003e\u003cp\u003eTo validate this findings, we repeated the tissue specificity analysis by comparing the EO-CRC deregulated proteome with additional available tissue databases \u003csup\u003e\u003cspan additionalcitationids=\"CR34\" citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e. These independent investigations further corroborated the over-representation of a placental and trophoblast protein core predominantly expressed in EO-CRC (\u003cb\u003eSupplementary Fig.\u0026nbsp;2C, Fig.\u0026nbsp;1F\u003c/b\u003e), providing a potential mechanistic explanation for the differences between EO-CRC and SO-CRC previously highlighted by the Principal Component Analysis (PCA) (Fig.\u0026nbsp;1B). Among these proteins, we found proteins involved in trophoblast fusion (e.g., GPC4, LAMA3, ITGB4), immune modulators/defence response (e.g., HLA-DRB5, CD14, BST2), cell adhesion/invasion (e.g., ITGA1, ITGA5, ITGB1, F11R, VASP), proteostasis/stress response (e.g., HSPA5, NUCB1, NUCB2, OTUB1), vesicle trafficking (VAMP7, RAB3A, VPS37C, VPS4B), and several others (\u003cb\u003eSupplementary Fig.\u0026nbsp;2F, Supplementary Table\u0026nbsp;2\u003c/b\u003e). Proteins involved in DNA metabolism were also predominantly expressed in EO-CRC. In particular, proteins regulating DNA replication under stress such as SMARCAL1, CDK1, CDK2, and RRM2B, and several DNA repair proteins, including ERCC2, ERCC3, ERCC4, UNG, APOBEC3C, PARP10, NBN and many others were significantly more expressed in EO-CRC (Fig.\u0026nbsp;1G, \u003cb\u003eSupplementary Table\u0026nbsp;2\u003c/b\u003e).\u003c/p\u003e\u003cp\u003eTo ensure that deregulation of these placenta-related features in EO-CRC was not solely attributable to patient age, we included in the study samples from tumour-adjacent mucosae of both EO-CRC and SO-CRC patients. PCA of the full proteomes of the total 35 samples disclosed a distinguishable separation between younger and older individuals (\u003cb\u003eSupplementary Fig.\u0026nbsp;2D-E\u003c/b\u003e). The proteomic profile of adjacent mucosae tended to cluster closely with their cancerous counterparts, pointing out the importance of the age factor as a relevant variable. However, the distinct clustering of the EO-CRC subgroup initially observed in the PCA (Fig.\u0026nbsp;1B) was confirmed. To investigate this further, we conducted additional analyses. Normal colonic samples from young and old individuals were first compared to each other, and each cancer group (EO-CRC and SO-CRC) was independently compared to its age-matched non-cancerous counterpart (\u003cb\u003eSupplementary Table\u0026nbsp;2\u003c/b\u003e). The significantly deregulated proteins identified in all comparisons were then cross-referenced with the previously defined placenta-related signature associated with EO-CRC. To construct this signature, we integrated placental-associated features identified through the tissue specificity analysis (Fig.\u0026nbsp;1F and \u003cb\u003eSupplementary Fig.\u0026nbsp;2C\u003c/b\u003e), thereby generating a consolidated placental protein signature relevant to EO-CRC (\u003cb\u003eSupplementary Table\u0026nbsp;2\u003c/b\u003e). Distribution analysis of placental-associated proteins demonstrated a significant enrichment in EO-CRC tumours relative to both SO-CRC tumours and matched normal mucosae from EO-CRC patients (N-EO-CRC), indicating a distinct placental-like protein expression profile unique to early-onset disease (Fig.\u0026nbsp;1H). This evidence suggests that EO-CRC is characterized by a proteomic profile indicative of a less differentiated cellular state. Some placental-associated proteins were also deregulated when we compared normal tissues from younger and older patients (N-EOCRC vs N-SOCRC). We attribute these differences to age-related changes or to field carcinogenesis in histologically normal mucosa adjacent to tumours \u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e. To isolate true cancer-specific signals, we refined the placental protein list by removing any proteins that overlapped with other comparisons or that were detected in normal tissues. This allowed us to define a robust and EO-CRC-specific placental signature, restricted to features uniquely associated with EO-CRC (Fig.\u0026nbsp;1H-I, \u003cb\u003eSupplementary Table\u0026nbsp;2\u003c/b\u003e). Analysis of the control-adjusted placental signature, following the exclusion of shared control features, confirmed the upregulation of these proteins in the EO-CRC. Importantly, it also revealed notable intra-group heterogeneity within the EO-CRC cohort. In contrast to SO-CRC, which displayed a relatively homogeneous distribution of placental core protein expression, EO-CRC exhibited clear stratification. Specifically, a distinct subgroup of patients showed markedly higher expression levels of the placental signature compared to the rest of the EO-CRC cases, indicating a possible molecularly distinct subset within the EO-CRC population (Fig.\u0026nbsp;1J). To test this hypothesis, we performed hierarchical clustering analysis (HCA) based exclusively on the placental protein core. The resulting dendrogram (Fig.\u0026nbsp;1K) illustrated distinct clustering patterns, with several clear macro-separations. Notably, SO-CRC patients formed a well-defined cluster, reinforcing their molecular distinction from EO-CRC cases. All but one EO-CRC exhibit the characteristic overexpression of the placental signature compared to SO-CRC. Of notice, HCA revealed two distinct branches also within EO-CRC. This separation highlights a gradient of expression of the placental signature among EO-CRC (Fig.\u0026nbsp;1L), further supporting our hypothesis that the placental protein profile can discriminate between subpopulations of EO-CRC.\u003c/p\u003e\u003cp\u003eTo validate these results, we performed a \u003cem\u003ede novo\u003c/em\u003e proteomic analysis on an independent patient cohort of EO-CRC (N\u0026thinsp;=\u0026thinsp;9) and SO-CRC (N\u0026thinsp;=\u0026thinsp;10), using the same pipeline as in the discovery phase. Consistent with the findings from the discovery cohort, PCA analysis of the validation data set revealed that EO-CRC clustered differently from SO-CRC (\u003cb\u003eSupplementary Fig.\u0026nbsp;3A\u003c/b\u003e). The comparison between EO-CRC and SO-CRC again highlighted differential regulation of key biological processes, notably those associated with organism development (\u003cb\u003eSupplementary Figs.\u0026nbsp;3B-D, Supplementary Table\u0026nbsp;3\u003c/b\u003e). Furthermore, tissue specificity analysis once again demonstrated a distinct enrichment of placental-related proteins in EO-CRC (\u003cb\u003eSupplementary Figs.\u0026nbsp;3B and E, Supplementary Table\u0026nbsp;3\u003c/b\u003e), reinforcing the robustness and reproducibility of our findings. As in the discovery phase, we normalized the EO-CRC placental-related features by adjusting expression levels in the corresponding control samples (\u003cb\u003eSupplementary Figs.\u0026nbsp;3F-G, Supplementary Table\u0026nbsp;3\u003c/b\u003e). We then used these control-corrected protein profiles to perform HCA. Also, for this independent cohort, we were able to discriminate a sub-cluster of EO-CRC samples with marked over-expression of the placental signature. While SO-CRC patients remained distinctly separated, one EO-CRC grouped with the SO-CRC patients (\u003cb\u003eSupplementary Figs.\u0026nbsp;3H, J, and K\u003c/b\u003e). In conclusion, our proteomic approach highlighted the distinctiveness of protein expression of EO-CRC compared to SO-CRC. Particularly, EO-CRC samples were characterized by a higher expression of proteins usually involved in placental processes, which were not observed in samples collected from older patients. This finding was confirmed even after adjusting our results on the proteomic profiling of adjacent non-cancer mucosa, thus ruling out a potentially merely age-related finding. Given the limited sample size, further validations of our findings are warranted in larger prospective cohorts.\u003c/p\u003e\u003cp\u003e\u003cspan type=\"ItalicUnderline\" class=\"ItalicUnderline\" name=\"Emphasis\"\u003ePlacental features are retained in patient-derived organoids\u003c/span\u003e.\u003c/p\u003e\u003cp\u003eOne approach to overcome this limitation is the use of cultured models such as organoids, which provide greater reproducibility while retaining proteomic profiles that closely mirror those of primary tumors \u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e. Hence, whenever possible, we established patient-derived organoids (PDOs) from both EO-CRC and SO-CRC primary tumors. PDOs were successfully generated from two EO-CRC (EOCRC8 and EOCRC18) and two SO-CRC (SOCRC26 and SOCRC27) patients. We performed high-resolution proteomic analysis of the PDOs, applying the same pipeline used for the solid tumour samples (Fig.\u0026nbsp;2A). PCA of the full proteomes revealed that the organoids retain the proteomic hallmarks of their parental tumours, including the distinction between early- and standard-onset origins (Fig.\u0026nbsp;2B, \u003cb\u003eSupplementary Table\u0026nbsp;4\u003c/b\u003e). A correlation matrix confirmed high intra-sample reproducibility among PDOs, underscoring their robustness as \u003cem\u003eex vivo\u003c/em\u003e models (Fig.\u0026nbsp;2C). To assess the conservation of the placental-like features in these models, we examined the expression of core placental proteins previously identified in tumour samples. Most of these proteins (85%) remained detectable in the EO-CRC PDOs (Fig.\u0026nbsp;2D), with cluster analysis revealing distinct patterns of expression across the cohort, with divergency for the EO-CRC8 sample similarly to parental tissues (Fig.\u0026nbsp;1L and 2E, \u003cb\u003eSupplementary Table\u0026nbsp;4\u003c/b\u003e). Specifically, the stratification of proteins observed in PDOs mirrored trends observed in primary tissues, including the enrichment of placenta-related proteins associated with key EO-CRC processes, such as DNA metabolism, tissue invasion, and immunomodulation in EO-CRC (Fig.\u0026nbsp;2F-G, \u003cb\u003eSupplementary Table\u0026nbsp;4\u003c/b\u003e). HCA further confirmed the segregation of EO-CRC and SO-CRC PDOs according to the placental signature, reinforcing the notion that this molecular program is maintained in PDOs (Fig.\u0026nbsp;2H).\u003c/p\u003e\u003cp\u003eAtaxia Telangiectasia and Rad3-related (ATR) signalling is a key mediator of replication stress responses, a pathway which is frequently dysregulated in aggressive, developmentally placental mimicking tumours \u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e. On these premises, aiming at functionally validating our findings and exploring potential hints of drug sensitivity in PDOs generated from EO-CRC, we next focused on assessing how ATR inhibitors (ATRis) can impact on PDOs viability. This focus was further supported by our proteomic data, which revealed that EO-CRC samples exhibited elevated expression of proteins involved in DNA metabolism. Notably, proteins implicated in replication stress management along with a range of DNA repair factors were significantly more abundant in EO-CRC (Fig.\u0026nbsp;1G). These findings suggest that EO-CRC may be particularly vulnerable to therapies targeting replication stress response pathways, such as ATR inhibition. Thus, PDOs were exposed to increasing doses of ATRi Ceralasertib and Berzosertib, over a five-days in vitro viability assay. Dose-response analyses demonstrated differential sensitivity among the PDOs, with EOCRC8 showing the highest expression of placental-like proteomic features and exhibiting the greatest sensitivity to both ATRis when compared to other PDOs including EOCRC18 harbouring fainter expression of placental features (Fig.\u0026nbsp;2I-J). This observation was confirmed at clinically relevant concentrations of both compounds, with significantly reduced viability of EOCRC8 PDO (Fig.\u0026nbsp;2K-L\u003cb\u003e)\u003c/b\u003e. Of notice, none of the previously suggested molecular alterations identified in CRC sensitizing to ATRis was retrieved in EOCRC8 PDO (\u003cb\u003eSupplementary Fig.\u0026nbsp;4\u003c/b\u003e) \u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e. These findings demonstrate that the placental-like proteomic signature is retained in PDOs and it can potentially predict sensitivity to replication stress-targeting agents such as ATRis. While intriguing, this observation is limited to a single sample and requires broader validation.\u003c/p\u003e\u003cp\u003e\u003cspan type=\"ItalicUnderline\" class=\"ItalicUnderline\" name=\"Emphasis\"\u003eThe retrotransposable element\u003c/span\u003e \u003cspan type=\"ItalicUnderline\" class=\"ItalicUnderline\" name=\"Emphasis\"\u003eHERVH is overexpressed in early-onset colorectal cancer\u003c/span\u003e.\u003c/p\u003e\u003cp\u003eIn our previous findings, we demonstrated that exposure of mammalian stem cells to DNA replication inhibitors, activating an ATR-dependent replication stress response, led to the induction of trophoblast-specific genes and endogenous retroviruses (ERVs), thereby linking replication stress to retrotransposon reactivation and the emergence of placental-like transcriptional programs \u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e. HERV elements have been recognized as key transcriptional regulators of pluripotency in both embryonic stem cells and placental development \u003csup\u003e\u003cspan additionalcitationids=\"CR27\" citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e. Given EO-CRC placental-like traits, we investigated whether these features were also associated with altered HERV expression. To this end, we performed RNA-sequencing on samples from EO-CRC patients, also including samples isolated from adjacent normal tissue (N\u0026thinsp;=\u0026thinsp;8), primary tumour tissue (N\u0026thinsp;=\u0026thinsp;9), and liver metastases (N\u0026thinsp;=\u0026thinsp;5) (\u003cb\u003eSupplementary Table\u0026nbsp;5\u003c/b\u003e and \u003cb\u003eSupplementary Fig.\u0026nbsp;5A\u003c/b\u003e). We first assessed whether the placental-like signature identified by proteomics was also reflected at the transcriptomic level. Consistently, this signature, as well as additional placental gene sets from public datasets, were enriched in primary tumour samples compared to matched adjacent mucosae (Fig.\u0026nbsp;3A and \u003cb\u003eSupplementary Fig.\u0026nbsp;5B\u003c/b\u003e). Notably, these signatures exhibited further enrichment in metastatic lesions relative to primary EO-CRC lesions (Fig.\u0026nbsp;3B and \u003cb\u003eSupplementary Fig.\u0026nbsp;5C\u003c/b\u003e), suggesting a progressive acquisition of a placental-like trait during tumour progression and evolution. We next quantified retrotransposable element (RE) expression and performed differential expression analysis. We identified 16 upregulated and 8 downregulated RE subfamilies in EO-CRC versus normal mucosa (Fig.\u0026nbsp;3C), and 2 upregulated with 13 downregulated REs in metastases versus EO-CRC primary lesions (Fig.\u0026nbsp;3D). Notably, the differentially expressed REs were predominantly HERVs, mainly from the ERV1, ERVL, and ERVK families. Of particular interest, two ERV1 subfamilies, LTR7Y and HERVH, representing the regulatory and internal components of full-length HERVH elements, showed a specific stepwise increase in expression from normal mucosa to EO-CRC primary tumour and then to metastases (Fig.\u0026nbsp;3E\u003cb\u003e\u0026ndash;F\u003c/b\u003e). These findings suggest a potential role for specific HERVH elements not only as markers of cellular dedifferentiation, consistent with their established expression in pluripotent stem cells, but also as indicators of tumour progression and metastatic driver.\u003c/p\u003e\u003cp\u003eBased on these observations, we hypothesized that HERVH may represent a novel molecular feature of the previously described placental features distinguishing EO-CRC from SO-CRC. Therefore, we investigated the expression of HERVH using RNA In Situ Hybridization (RNA-ISH) on FFPE sections from EO-CRC (N\u0026thinsp;=\u0026thinsp;16) and SO-CRC (N\u0026thinsp;=\u0026thinsp;14) (\u003cb\u003eSupplementary Table\u0026nbsp;5\u003c/b\u003e). In 8/10 (80%) EO-CRC cases with available matched mucosae, tumours exhibited stronger HERVH expression, with only one case showing no difference and one case showing the reverse (Fig.\u0026nbsp;4A-B). Of notice, the EOCRC8 PDO (established from an HERVH high expressing sample) was sensitive to ATRis, while EOCRC18 PDO (established from an HERVH low expressing sample) was much less sensitive (Fig.\u0026nbsp;2I-L and \u003cb\u003eFig.\u0026nbsp;4B\u003c/b\u003e). In contrast, 4/12 (33%) SO-CRC expressed higher HERVH levels compared to their matched mucosa consistently with prior reports \u003csup\u003e\u003cspan additionalcitationids=\"CR24\" citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. However, they were significantly lower compared to EO-CRC tumour samples (Fig.\u0026nbsp;4C\u003cb\u003e\u0026ndash;D\u003c/b\u003e). Although higher in EO-CRC, HERVH expression was not exclusive, suggesting limited diagnostic specificity. In conclusion, we found that HERVH expression levels were significantly higher in EO-CRC tumours compared to SO-CRC (Fig.\u0026nbsp;4E) and showed a positive association with tumour progression. This pattern may reflect the acquisition of placental-like features in most EO-CRC, highlighting HERVH as a potential novel diagnostic and therapeutic target (Fig.\u0026nbsp;5).\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eEO-CRC incidence is rising globally posing urgent clinical challenges \u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e,\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e. Compared to SO-CRC, EO-CRC present specific clinical features of poor response and survival after available therapies \u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e. These observations point to a distinct biological entity, whose molecular underpinnings remain insufficiently explored \u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eTo address this gap, we employed a multi-omics approach combining proteomics and transcriptomics on non-hereditary EO-CRC samples from patients mostly diagnosed under age 40. Additionally, we functionally test our findings in PDOs established from EO-CRC patients. Our analyses revealed that in the majority of EO-CRC placental developmental programs are activated and HERVH, a known regulator of pluripotency in embryonic stem cells, is overexpressed. In detail, proteomic profiling showed that most EO-CRC, but not SO-CRC, are enriched in placentation molecules, angiogenic mediators, and immune checkpoint regulators factors that are physiologically restricted to the maternal-foetal interface. These findings suggest that EO-CRC mimic placental biology, acquiring invasive, immune-evasive, and remodelling capacities characteristic of placental development (Fig.\u0026nbsp;5). While these data suggest a potential link, functional validation will be needed to establish causality. If confirmed in future studies, this placental mimicry, in conjunction with HERVH reactivation, could define a dual molecular signature that differentiate most EO-CRC from SO-CRC, offering a potential explanation for its aggressive behaviour.\u003c/p\u003e\u003cp\u003eAt the molecular level, we speculate that chronic replication stress, potentially driven by exposome factors, such as but not limited to microbiome-derived genotoxins \u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e,\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e and high-fat diets \u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e,\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e, induces epigenetic alterations that massively unlock the expression of a placentals signature, comprising HERVH in the young adult colon (Fig.\u0026nbsp;5). If so, and contextualized in the available literature, our findings might represent a novel biological hint to rationalize the search for the EO-CRC increase incidence causes by focusing on exposomal agents retaining hypomethylating potentials \u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e. Once reactivated, it is reasonable that HERVH could act on chromatin contributing to dedifferentiation and immune evasion \u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e,\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e,\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e. This pattern mirrors our prior findings in stem cells undergoing replication stress, where ERV activation drives trophoblast lineage transitions \u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e. Supporting this model and accordingly to another recent publication \u003csup\u003e\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e, RNA-seq of paired EO-CRC mucosa, primaries, and metastases shows a step-wise rise in HERVH from normal tissue to metastasis, pointing to a role in cancer progression. Full-length HERVH remains transcriptionally active, mirroring its behaviour in pluripotent and trophoblast cells and exemplifying retrotransposon \u003cem\u003eonco-exaptation\u003c/em\u003e \u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e,\u003cspan additionalcitationids=\"CR48\" citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e. The reactivation likely reflects a global loss of repressive chromatin, warranting future epigenomic study. Importantly, placental-like overexpression peaks in EO-CRC and declines with age, affecting younger patients the most.\u003c/p\u003e\u003cp\u003ePrevious reports have documented that HERV positivity in non-metastatic CRC is associated with worse prognosis and higher risk of disease relapse after surgery \u003csup\u003e\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e. Accordingly, in the same clinical setting from the IDEA collaboration dataset, young age emerged as a negative prognostic factor in stage III CRC and was associated with significantly higher chance of relapse, despite a more aggressive therapeutic strategy \u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. On these premises, the present findings corroborate the understanding that EO-CRC, because of their biological characteristics including placental features and HERVH reactivation, represents an aggressive subset of CRC also potentially resistant to current therapies including the most intensive chemo-regimens \u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e. Future studies are warranted to assess the prevalence of placental features and HERVH reactivation in a CRC patients population stratified for minimal residual disease (MRD) using circulating tumour DNA (ctDNA) after primary tumour resection. These efforts may help determine whether the prevalence of intrinsically aggressive \u0026lsquo;\u003cem\u003eborn-to-be-bad\u003c/em\u003e\u0026rsquo; diseases is more common in EO-CRC patients.\u003c/p\u003e\u003cp\u003ePrevious findings showed that a relevant fraction of CRC cells display pronounced sensitivity to pharmacologic inhibition of ATR \u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e, a master kinase that buffers replication stress \u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e, able to induce ERVs and trophoblast cell fate transition (Fig.\u0026nbsp;5) \u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e. Our PDOs replicate this vulnerability, supporting ATR inhibitors now in trials for DNA-repair-deficient tumours. Placental-like signatures may predict response to ATR-inhibitors and, if validated, would be the first EO-CRC actionable biomarker. It is tempting to speculate that coupling replication-stress targeting with immune-checkpoint blockade, or with vaccines directed against HERVH peptide, might deserve further investigation. Finally, detecting HERVH RNA in stool or blood could help stratify CRC risk in young adults, for whom universal colonoscopy is unfeasible (Fig.\u0026nbsp;5) \u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e,\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eWhile our findings consistently suggest that most EO-CRC engages a placental-like program and HERVH reactivation, several caveats remain. Functional analyses rely on a limited PDO set, and the observed association with ATR inhibitor sensitivity should be viewed as hypothesis-generating. The specificity of the placental signature to EO-CRC is highly suggestive but requires further confirmation in larger geographically distinct series. Disentangling specific EO-CRC reprogramming from features shared with other aggressive tumour states will benefit from broader pan-cancer comparisons. Finally, although direct modelling of environmental influences is challenging, indirect strategies, such as epigenetic modulation in organoids or analysis of clinical metadata, may help to clarify upstream regulators of these features.\u003c/p\u003e\u003cp\u003eIn summary, we present an original perspective that might redefine a subset of non-hereditary EO-CRC as a biologically distinct malignancy characterized by reawakening placental programs and HERVH expression. These insights provide a potential rationale for EO-CRC aggressive clinical behaviour and a possible actionable vulnerability, opening a path from molecular discovery to targeted interception.\u003c/p\u003e"},{"header":"Material and methods","content":"\u003cp\u003e\u003cb\u003ePatient identification and enrolment\u003c/b\u003e\u003c/p\u003e\u003cp\u003eAll patients were enrolled within the IANG-CRC (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.frrb.it/it/progetti-finanziati-iang-crc\u003c/span\u003e\u003cspan address=\"https://www.frrb.it/it/progetti-finanziati-iang-crc\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e)\u003c/span\u003e or the IANG-CRC2 studies. IANG-CRC was a prospective multi-institutional Italian study funded by Fondazione Regionale Ricerca Biomedica (FRRB), approved by local Ethical Committees at Grande Ospedale Metropolitano Niguarda, Milan, Italy, and Istituto Ricovero Cura Carattere Scientifico (IRCCS) San Raffaele, Milan, Italy (please refer to \u003cb\u003eSupplementary Data 1\u003c/b\u003e for the full protocol). Later, IANG-CRC2 study was designed embedded within the AlfaOmega Master Observational Trial (MOT) (NCT04120935) \u003csup\u003e\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u003c/sup\u003e which was already active at Grande Ospedale Metropolitano Niguarda, Milan, Italy, Vall d'Hebron Institute of Oncology (VHIO), Barcelona, Spain, with the translational partnership of Fondazione Istituto Nazionale Genetica Molecolare (INGM), Padiglione \"Romeo ed Enrica Invernizzi\", Milan, Italy, and IFOM ETS - The AIRC Institute of Molecular Oncology, Milan Italy. The IANG-CRC2 study followed the IANG-CRC study towards broadening the latter sample size. All EO-CRC patients included in this publication were enrolled within the IANG-CRC or IANG-CRC2 studies, while SO-CRC patients were enrolled within the AlfaOmega MOT (NCT04120935). All patients provided written informed consent for participation in the study and associated procedures. These studies were conducted in accordance with the principles of the Declaration of Helsinki and the International Conference on Harmonisation and Good Clinical Practice guidelines. Whenever available, both fresh and formalin-fixed paraffin-embedded (FFPE) samples were sent for translational analysis from recruiting clinical centers to both IFOM-ETS \u0026ndash; The AIRC Institute of Molecular Oncology, Milan, Italy and to Fondazione Istituto Nazionale Genetica Molecolare (INGM), Padiglione \"Romeo ed Enrica Invernizzi\", Milan, Italy.\u003c/p\u003e\u003cp\u003eEO-CRC patients diagnosed with disease at age younger than 45 and with availability of fresh frozen or archival FFPE samples were considered for eligibility for the present study. Forty-five years as age cut-off was arbitrarily chosen based on the updated recent CRC screening recommendations by the American Cancer Society \u003csup\u003e\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e\u003c/sup\u003e. EO-CRC patients known to be affected with hereditary cancer predisposition syndromes (HCPS) were not eligible for the present study. EO-CRC patients with a first-degree relative (FDR) diagnosed with CRC were considered eligible only if they had undergone extended germline next generation sequencing (NGC) panelling to rule out HCPSs, according to the most recent consensus publication on EO-CRC patients management \u003csup\u003e\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e\u003c/sup\u003e. Patients diagnosed with somatic MSI EO-CRC were considered eligible only if Lynch syndrome was ruled out by germline profiling. As controls, all SO-CRC patients diagnosed with the disease at age 50 or later and with availability of fresh frozen or archival FFPE samples were eligible.\u003c/p\u003e\u003cp\u003e\u003cb\u003eCollection and analysis of clinicopathological features\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThe following data were extrapolated from each enrolled patient: age, sex, cancer familiarity, prior history of inflammatory bowel disease (IBD), primary tumour location, tumour histology, stage at initial diagnosis, tumour grading, \u003cem\u003eRAS\u003c/em\u003e/\u003cem\u003eBRAF\u003c/em\u003e status, HER2 expression, and MMR status, disease burden with regard to the presence of liver and peritoneal metastases, and last follow-up or death.\u003c/p\u003e\u003cp\u003eContinuous variables were summarized as median and interquartile range (IQR). Categorical variables were summarized as frequency and percentage. Baseline characteristics were compared according to age of onset using Mann-Whitney test for continuous variables, and Chi-square test or Fisher\u0026rsquo;s exact test for categorical variables. Survival was evaluated from the date of initial diagnosis to last follow-up or death occurring for any cause. All tests were two-sided and p-values\u0026thinsp;\u0026lt;\u0026thinsp;0.05 were considered significant. Statistical analyses were performed using GraphPad Prism 10.2.3 Software.\u003c/p\u003e\u003cp\u003e\u003cb\u003eProteomic workflow\u003c/b\u003e\u003c/p\u003e\u003cp\u003e\u003cspan type=\"ItalicUnderline\" class=\"ItalicUnderline\" name=\"Emphasis\"\u003eProtein extraction and digestion\u003c/span\u003e\u003c/p\u003e\u003cp\u003eFrozen human colon samples and PDOs were homogenized in 8M Urea, 100 mM Tris-HCl pH\u0026thinsp;=\u0026thinsp;8 using Dounce tissue homogenizer. To complete the solubilization, 5 on/off cycles with tip sonicator were performed. The samples were then centrifuged (14\u0026rsquo;000 \u003cem\u003exg\u003c/em\u003e, 30 min, 4\u0026deg;C) and protein concentration were measured in the resulting supernatant via BCA assay (Microplate BCA\u0026trade; protein Assay Kit, Thermo Scientific), using BSA as standard. 70 \u0026micro;g of proteins for each specimen were digested following in-solution digestion. Briefly, proteins were reduced by 10 mM tris-carboxy-ethyl phosphine (Thermo scientific) and alkylated with 40 mM 2-Chloroacetamide (Sigma-Aldrich) in 8 M Urea 100 mM Tris pH\u0026thinsp;=\u0026thinsp;8 at for 30 min (r.t., dark shaker).\u003c/p\u003e\u003cp\u003eDouble protein digestion was carried out. Firstly, the endoproteinase Lys-C (Thermo scientific) was added in 1:50 ratio to initial protein concentration (1h, r.t.), then the hydrolysis was boosted by supplementing trypsin (Roche) (overnight, 37\u0026deg;C), using the same ratio. The resulting peptide solution was desalted and concentrated using \u0026micro;-C18 Ziptip pipette tips (Millipore) following manufacturer's instruction. Purified samples were resuspended in 5% formic acid (FA) solution and stored at -20\u0026deg;C until the MS analysis.\u003c/p\u003e\u003cp\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003eLC-MS/MS and Data Processing\u003c/span\u003e\u003c/p\u003e\u003cp\u003eThe LC/MS analysis was performed using an UHPLC Easy-nLC 1200 (Thermo Scientific) coupled to an Orbitrap Exploris 480 Mass Spectrometer (Thermo Scientific). The hydrolysed peptides (1.5 \u0026micro;l) were separated by applying a linear gradient from 95% solvent A (2% acetonitrile (ACN), 0.1% FA) to 55% solvent B (80% ACN, 0.1% FA). Data was acquired in both Data Dependent Acquisition (DDA) and Data Independent Acquisition (DIA) modes. Raw MS files were processed with Spectronaut software. MS/MS peak lists are searched against the UniProtKB complete proteome 2021 Human Database. Raw data files were uploaded in Spectronaut 15, run in library-assisted search for label-free protein quantification. Peptides were identified using the Pulsar search engine specifying the UniProt Homo Sapiens database (UP000005640) as reference. Methionine Oxidation (M) and Acetyl N-Terminal modification (Protein N-term) were specified as dynamic modifications while carbamidomethyl at cysteine residues (C) was used as static modification. The peptides were required to be at least 7 amino acids long and counting a maximum of 2 missed cleavages. Data were filtered at 1% FDR on the protein level and protein quantities were exported, followed by analysis with different software packages (Excel, R-studio, Perseus). Specifically, a first level of data filtering was applied to exclude contaminant proteins/peptides. The signal/noise values were normalized using log2 transformation, while the protein abundances were grouped and filtered to achieve a minimum valid number equal to 70% in at least one group. Missing values have been replaced by random numbers that are drawn from a normal distribution.\u003c/p\u003e\u003cp\u003eStatistical analysis for quantitative evaluation was performed. For comparisons among the different sample cohorts, a non-pairwise t-test was applied, with the level of significance set at \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026le;\u0026thinsp;0.05. Proteins exhibiting at least a log2 fold change of \u0026plusmn;\u0026thinsp;0.5 were considered differentially regulated. To further explore patterns of protein expression across samples, unsupervised hierarchical clustering analysis was performed using the Ward method and Euclidean distance as the metric. Functional analysis was performed using multiple diverse tools \u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e,\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e,\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e\u003c/sup\u003e, with a focus on GO terms and tissue- or cell type\u0026ndash;specific expression patterns. An FDR threshold of \u0026lt;\u0026thinsp;0.1 was applied to determine significance.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe are grateful to all members of the A.B. and the V.C. laboratories for the insightful and critical scientific discussion. We acknowledge the scientific and technical assistance of the INGM Imaging Facility, in particular, Chiara Cordiglieri and Alessandra Fasciani (Istituto Nazionale Genetica Molecolare \u0026lsquo;Romeo ed Enrica Invernizzi\u0026rsquo; (INGM), Milan, Italy). We acknowledge the scientific and technical contributions of the Cogentech Histopathology Unit and the IFOM Cellular and Preclinical Models Unit. Proteomics data have been generated by the IFOM ETS Cogentech Proteomics and Metabolomics Core Facility (RRID:SCR_026937). G.Patelli is a PhD student within the European School of Molecular Medicine (SEMM).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWe gratefully acknowledge all young individuals in care and their families for their generous participation in the IANG-CRC study. We extend our deepest appreciation to the families of Andrea Fernandez and Gae Federica Elli, in whose loving memory this work is dedicated. Their extraordinary emotional support and generosity were instrumental in inspiring and sustaining the IANG-CRC program from its inception.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eStudy funded by: Fondazione Regionale Ricerca Biomedica, project IANG-CRC (grant CP2_12/2018 to S.Siena); Fondazione Oncologia Niguarda ETS (grant \u0026lsquo;Giovani CRC\u0026rsquo; to G.M.); Grande Ospedale Metropolitano Niguarda (Fondo Divisionale Oncologia Falck); the Italian Ministry of University and Research (MUR), Progetti di Ricerca di Rilevante Interesse Nazionale (PRIN) 2022 under the project \u0026lsquo;A Comprehensive Analysis of Dietary Risk Factors, Colibactin and Tumour Molecular Features in Early-Onset Colorectal Cancer\u0026rsquo; (grant 2022LSFKWE to A.S-B., GM.C., and S.Arena); AIRC under 5 per Mille 2018 - ID. 21091 program \u0026ndash; PI Alberto Bardelli (A.Bardelli), Group Leader (GL) Salvatore Siena (S.Siena) and Silvia Marsoni (S.Marsoni); Regione Lombardia (Delibera XII/2455/2024 Young Adult Colorectal Cancer Surveillance Program (Ages 20e49) to Grande Ospedale Metropolitano Niguarda; AIRC under IG 2023\u0026mdash;ID. 28725 project to V.C. Ricerca Finalizzata, (grant nr GR-2018-12365280). AIRC under IG 2023 - ID. 29286 project to S.Arena; FPRC 5 9 1000 Ministero della Salute 2022 CARESS to S. Arena; Italian Ministry of Health, Ricerca Corrente 2025 to S.Arena; Prin 2022 PNRR finanziato dall\u0026rsquo;Unione Europea NextGenerationEU M4 C2 I.1.1.- P2022E3BTH to S.Arena. Fondazione AIRC (grant nr 27066 and nr 21073); Fondazione Cariplo (grant nr 2019-3416); Piano Nazionale Ripresa e Resilienza (PNRR) (grant nr G43C22002620007); and Progetti Rilevante Interesse Nazionale (PRIN) (grant nr 2022PKF9S) to B.B.; European Research Council (ERC) under the European Union\u0026lsquo;s Horizon 2020 research and innovation programme (TARGET, grant agreement n. 101020342) (A.Bardelli); AIRC under IG 2023 - ID. 28922 project \u0026ndash; P.I. Bardelli Alberto (A.Bardelli).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors Contribution\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eG.M., L.Santorelli, and F.Marasca designed and conducted experiments, analysed and interpreted data and wrote the manuscript. L.Santorelli designed and executed the proteomic analyses. V.R. designed and supervised bioinformatic analyses and revised the manuscript. E.G. designed and performed experiments, analysed and interpreted data. L.Salviati designed and performed bioinformatic experiments and analysed data. B.B. conceived and conceptualized the study, designed experiments, analysed and interpreted data and wrote and finalized the manuscript. S.Abrignani interpreted data, wrote and finalized the manuscript. A.Bachi \u0026nbsp;curated with L.Santorelli the acquisition and interpretation of the proteomic data. V.C. conceived and conceptualized the study, designed experiments, coordinated the experimental work, analysed and interpreted data, wrote the manuscript, and contributed to its editing and critical revision. A.C. and G.Parodi contributed with samples management and performed Mass Spectrometry acquisition. E.B., F.T., K.B., A.A., A.S-B., S.Siena, E.E., N.S, I.B., G.Patelli, M.P., L.L., A.M., GM.C. contributed with patients\u0026rsquo; studies and tumour procurement. G.C., A.S., and S.Scardellato performed experiments and bioinformatic analysis. A.S-B., S.Arena, S.Marsoni also contributed with financial support. S.G., S.Mariano, L.M., contributed with patients data and tumour samples management. E.B., M.D-C., MC.A. contributed with pathology diagnosis and tumour samples preparation. A.Bardelli designed experiments, coordinated the experimental work, analysed and interpreted data, wrote the manuscript, and contributed to its editing and critical revision. S.Siena conceived, conceptualized, and coordinated the study including clinical and financial responsibilities, and co-wrote the manuscript. All authors read and approved the manuscript ahead of submission.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eS.S. is advisory board member for Agenus, AstraZeneca, Bayer, Bristol Myers Squibb, CheckmAb, Daiichi-Sankyo, GlaxoSmithKline, MSD, Merck, Novartis, Ospedale San Raffaele, Pierre-Fabre, Pfizer, Seagen, and T-One Therapeutics. A.S-B. reported consulting or advisory role for Bayer, Novartis, Pierre Fabre, Servier, and Takeda; and personal honoraria as an invited speaker from Amgen, Guardant Health, Pierre Fabre, and Seagen. A.Bardelli reports receipt of grants/research supports from Neophore, AstraZeneca, Boehringer Ingelheim and honoraria/consultation fees from Guardant Health. A.B. is stock shareholder of Neophore and Kither Biotech. A.B. is an advisory boards member for Neophore. S.Abrignani is a co-founder of the startup CheckMab s.r.l and T-One Therapeutics s.r.l.; F.Marasca and B.B. are co-founders of the startup T-One Therapeutics s.r.l.. S.Arena reports personal fees from MSD Italia and a patent (Italian patent application No. 102022000007535) outside the submitted work. K.B. is advisory board member for AstraZeneca. N.S. reports personal honoraria as an invited speaker for AMGEN, Medistream (OncoBites2025), and travel supports from AMGEN, MERCK, and BAYER. N.S. is an ESMO Fellow (starting from April 2024). F.T. reports travel supports from ROCHE. A.A. is advisory board member for AMGEN e Italfarmaco. I.B. has received accommodation and travel expenses from Amgen, Merck, Sanofi, and Servier; and personal speaker honoraria from Astra Zeneca. E.E. reports the following: Honoraria, Consulting or Advisory Role, and Speakers\u0026apos;s Bureau: Agenus, Amgen, Bayer, Bristol Myers Squibb, Boehringer Ingelheim, Cure Teq AG, GlaxoSmithKline, Hoffman La-Roche, Janssen, Johnson\u0026amp;Johnson, Lilly, Medscape, Merck Serono, MSD, Nordic Group BV, Novartis, Organon, Pfizer, Pierre Fabre, Repare Therapeutics Inc., RIN Institute Inc., Rottapharm Biotech, Sanofi, Seagen International GmbH, Servier, and Takeda.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eSiegel, R. L., Miller, K. D., Wagle, N. S. \u0026amp; Jemal, A. Cancer statistics, 2023. \u003cem\u003eCA Cancer J Clin\u003c/em\u003e 73, 17\u0026ndash;48 (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDaca-Alvarez, M. \u003cem\u003eet al.\u003c/em\u003e Regional patterns of early-onset colorectal cancer from the GEOCODE (Global Early-Onset COlorectal Cancer DatabasE)-European consortium: retrospective cohort study. \u003cem\u003eBJS Open\u003c/em\u003e 9, zraf024 (2025).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRusso, A. G. \u003cem\u003eet al.\u003c/em\u003e Increased incidence of colon cancer among individuals younger than 50 years: A 17 years analysis from the cancer registry of the municipality of Milan, Italy. \u003cem\u003eCancer Epidemiol\u003c/em\u003e 60, 134\u0026ndash;140 (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVuik, F. E. \u003cem\u003eet al.\u003c/em\u003e Increasing incidence of colorectal cancer in young adults in Europe over the last 25 years. \u003cem\u003eGut\u003c/em\u003e 68, 1820\u0026ndash;1826 (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEng, C. \u003cem\u003eet al.\u003c/em\u003e Colorectal cancer. \u003cem\u003eLancet\u003c/em\u003e 404, 294\u0026ndash;310 (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKanter, K. \u003cem\u003eet al.\u003c/em\u003e Care Patterns and Overall Survival in Patients With Early-Onset Metastatic Colorectal Cancer. \u003cem\u003eJCO Oncol Pract\u003c/em\u003e 17, e1846\u0026ndash;e1855 (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLieu, C. H. \u003cem\u003eet al.\u003c/em\u003e Association of age with survival in patients with metastatic colorectal cancer: analysis from the ARCAD Clinical Trials Program. \u003cem\u003eJ Clin Oncol\u003c/em\u003e 32, 2975\u0026ndash;2984 (2014).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFontana, E. \u003cem\u003eet al.\u003c/em\u003e Early-Onset Colorectal Adenocarcinoma in the IDEA Database: Treatment Adherence, Toxicities, and Outcomes With 3 and 6 Months of Adjuvant Fluoropyrimidine and Oxaliplatin. \u003cem\u003eJ Clin Oncol\u003c/em\u003e 39, 4009\u0026ndash;4019 (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAkimoto, N. \u003cem\u003eet al.\u003c/em\u003e Rising incidence of early-onset colorectal cancer - a call to action. \u003cem\u003eNat Rev Clin Oncol\u003c/em\u003e 18, 230\u0026ndash;243 (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHofseth, L. J. \u003cem\u003eet al.\u003c/em\u003e Early-onset colorectal cancer: initial clues and current views. \u003cem\u003eNat Rev Gastroenterol Hepatol\u003c/em\u003e 17, 352\u0026ndash;364 (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEng, C. \u0026amp; Hochster, H. Early-Onset Colorectal Cancer: The Mystery Remains. \u003cem\u003eJ Natl Cancer Inst\u003c/em\u003e 113, 1608\u0026ndash;1610 (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMauri, G. \u003cem\u003eet al.\u003c/em\u003e Early-onset cancers: Biological bases and clinical implications. \u003cem\u003eCell Rep Med\u003c/em\u003e 5, 101737 (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eOgino, S. \u0026amp; Ugai, T. The global epidemic of early-onset cancer: nature, nurture, or both? \u003cem\u003eAnn Oncol\u003c/em\u003e 35, 1071\u0026ndash;1073 (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCercek, A. \u003cem\u003eet al.\u003c/em\u003e A Comprehensive Comparison of Early-Onset and Average-Onset Colorectal Cancers. \u003cem\u003eJ Natl Cancer Inst\u003c/em\u003e 113, 1683\u0026ndash;1692 (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWillauer, A. N. \u003cem\u003eet al.\u003c/em\u003e Clinical and molecular characterization of early-onset colorectal cancer. \u003cem\u003eCancer\u003c/em\u003e 125, 2002\u0026ndash;2010 (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eD\u0026iacute;az-Gay, M. \u003cem\u003eet al.\u003c/em\u003e Geographic and age variations in mutational processes in colorectal cancer. \u003cem\u003eNature\u003c/em\u003e (2025) doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41586-025-09025-8\u003c/span\u003e\u003cspan address=\"10.1038/s41586-025-09025-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCostanzo, V., Bardelli, A., Siena, S. \u0026amp; Abrignani, S. Exploring the links between cancer and placenta development. \u003cem\u003eOpen Biol\u003c/em\u003e 8, 180081 (2018).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHoltan, S. G., Creedon, D. J., Haluska, P. \u0026amp; Markovic, S. N. Cancer and pregnancy: parallels in growth, invasion, and immune modulation and implications for cancer therapeutic agents. \u003cem\u003eMayo Clin Proc\u003c/em\u003e 84, 985\u0026ndash;1000 (2009).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLee-Six, H. \u003cem\u003eet al.\u003c/em\u003e The landscape of somatic mutation in normal colorectal epithelial cells. \u003cem\u003eNature\u003c/em\u003e 574, 532\u0026ndash;537 (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCoorens, T. H. H. \u003cem\u003eet al.\u003c/em\u003e Inherent mosaicism and extensive mutation of human placentas. \u003cem\u003eNature\u003c/em\u003e 592, 80\u0026ndash;85 (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMamlouk, S. \u003cem\u003eet al.\u003c/em\u003e DNA copy number changes define spatial patterns of heterogeneity in colorectal cancer. \u003cem\u003eNat Commun\u003c/em\u003e 8, 14093 (2017).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZeman, M. K. \u0026amp; Cimprich, K. A. Causes and consequences of replication stress. \u003cem\u003eNat Cell Biol\u003c/em\u003e 16, 2\u0026ndash;9 (2014).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAtashpaz, S. \u003cem\u003eet al.\u003c/em\u003e ATR expands embryonic stem cell fate potential in response to replication stress. \u003cem\u003eElife\u003c/em\u003e 9, e54756 (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNurk, S. \u003cem\u003eet al.\u003c/em\u003e The complete sequence of a human genome. \u003cem\u003eScience\u003c/em\u003e 376, 44\u0026ndash;53 (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChuong, E. B., Rumi, M. A. K., Soares, M. J. \u0026amp; Baker, J. C. Endogenous retroviruses function as species-specific enhancer elements in the placenta. \u003cem\u003eNat Genet\u003c/em\u003e 45, 325\u0026ndash;329 (2013).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLu, X. \u003cem\u003eet al.\u003c/em\u003e The retrovirus HERVH is a long noncoding RNA required for human embryonic stem cell identity. \u003cem\u003eNat Struct Mol Biol\u003c/em\u003e 21, 423\u0026ndash;425 (2014).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFrost, J. M. \u003cem\u003eet al.\u003c/em\u003e Regulation of human trophoblast gene expression by endogenous retroviruses. \u003cem\u003eNat Struct Mol Biol\u003c/em\u003e 30, 527\u0026ndash;538 (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang, J. \u003cem\u003eet al.\u003c/em\u003e Isolation and cultivation of naive-like human pluripotent stem cells based on HERVH expression. \u003cem\u003eNat Protoc\u003c/em\u003e 11, 327\u0026ndash;346 (2016).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAttig, J. \u003cem\u003eet al.\u003c/em\u003e Human endogenous retrovirus onco-exaptation counters cancer cell senescence through calbindin. \u003cem\u003eJ Clin Invest\u003c/em\u003e 133, e164397 (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBabaian, A. \u003cem\u003eet al.\u003c/em\u003e Onco-exaptation of an endogenous retroviral LTR drives IRF5 expression in Hodgkin lymphoma. \u003cem\u003eOncogene\u003c/em\u003e 35, 2542\u0026ndash;2546 (2016).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJang, H. S. \u003cem\u003eet al.\u003c/em\u003e Transposable elements drive widespread expression of oncogenes in human cancers. \u003cem\u003eNat Genet\u003c/em\u003e 51, 611\u0026ndash;617 (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTeclemariam, E. T., Pergande, M. R. \u0026amp; Cologna, S. M. Considerations for mass spectrometry-based multi-omic analysis of clinical samples. \u003cem\u003eExpert Rev Proteomics\u003c/em\u003e 17, 99\u0026ndash;107 (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePan, J.-B. \u003cem\u003eet al.\u003c/em\u003e PaGenBase: a pattern gene database for the global and dynamic understanding of gene function. \u003cem\u003ePLoS One\u003c/em\u003e 8, e80747 (2013).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePalasca, O., Santos, A., Stolte, C., Gorodkin, J. \u0026amp; Jensen, L. J. TISSUES 2.0: an integrative web resource on mammalian tissue expression. \u003cem\u003eDatabase (Oxford)\u003c/em\u003e 2018, bay028 (2018).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRudolph, J. D. \u0026amp; Cox, J. A Network Module for the Perseus Software for Computational Proteomics Facilitates Proteome Interaction Graph Analysis. \u003cem\u003eJ Proteome Res\u003c/em\u003e 18, 2052\u0026ndash;2064 (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCurtius, K., Wright, N. A. \u0026amp; Graham, T. A. An evolutionary perspective on field cancerization. \u003cem\u003eNat Rev Cancer\u003c/em\u003e 18, 19\u0026ndash;32 (2018).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLindhorst, P. H. \u0026amp; Hummon, A. B. Proteomics of Colorectal Cancer: Tumors, Organoids, and Cell Cultures-A Minireview. \u003cem\u003eFront Mol Biosci\u003c/em\u003e 7, 604492 (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDurinikova, E. \u003cem\u003eet al.\u003c/em\u003e Targeting the DNA Damage Response Pathways and Replication Stress in Colorectal Cancer. \u003cem\u003eClin Cancer Res\u003c/em\u003e 28, 3874\u0026ndash;3889 (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBen-Aharon, I. \u003cem\u003eet al.\u003c/em\u003e Early-Onset Cancer in the Gastrointestinal Tract Is on the Rise-Evidence and Implications. \u003cem\u003eCancer Discov\u003c/em\u003e 13, 538\u0026ndash;551 (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJayakrishnan, T. \u0026amp; Ng, K. Early-Onset Gastrointestinal Cancers: A Review. \u003cem\u003eJAMA\u003c/em\u003e (2025) doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1001/jama.2025.10218\u003c/span\u003e\u003cspan address=\"10.1001/jama.2025.10218\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSogari, A. \u003cem\u003eet al.\u003c/em\u003e Tolerance to colibactin correlates with homologous recombination proficiency and resistance to irinotecan in colorectal cancer cells. \u003cem\u003eCell Rep Med\u003c/em\u003e 5, 101376 (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eArima, K. \u003cem\u003eet al.\u003c/em\u003e Western-Style Diet, pks Island-Carrying Escherichia coli, and Colorectal Cancer: Analyses From Two Large Prospective Cohort Studies. \u003cem\u003eGastroenterology\u003c/em\u003e 163, 862\u0026ndash;874 (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYoung, A. S. \u003cem\u003eet al.\u003c/em\u003e The need for a cancer exposome atlas: a scoping review. \u003cem\u003eJNCI Cancer Spectr\u003c/em\u003e 9, pkae122 (2025).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSchmidleithner, L., St\u0026uuml;ve, P. \u0026amp; Feuerer, M. Transposable elements as instructors of the immune system. \u003cem\u003eNat Rev Immunol\u003c/em\u003e (2025) doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41577-025-01172-3\u003c/span\u003e\u003cspan address=\"10.1038/s41577-025-01172-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eUgai, T. \u003cem\u003eet al.\u003c/em\u003e Immune cell profiles in the tumor microenvironment of early-onset, intermediate-onset, and later-onset colorectal cancer. \u003cem\u003eCancer Immunol Immunother\u003c/em\u003e 71, 933\u0026ndash;942 (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eViot, J. \u003cem\u003eet al.\u003c/em\u003e Deciphering human endogenous retrovirus expression in colorectal cancers: exploratory analysis regarding prognostic value in liver metastases. \u003cem\u003eEBioMedicine\u003c/em\u003e 116, 105727 (2025).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGolkaram, M. \u003cem\u003eet al.\u003c/em\u003e HERVs establish a distinct molecular subtype in stage II/III colorectal cancer with poor outcome. \u003cem\u003eNPJ Genom Med\u003c/em\u003e 6, 13 (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMazumder, H. \u003cem\u003eet al.\u003c/em\u003e Human endogenous retroviruses (HERVs) associated with glioblastoma risk and prognosis. \u003cem\u003eCancer Gene Ther\u003c/em\u003e 32, 622\u0026ndash;632 (2025).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMatteucci, C. \u003cem\u003eet al.\u003c/em\u003e Expression of HERV-K and embryonic genes in chronic lymphocytic leukemia and their association with therapy regimens. \u003cem\u003eBlood Adv\u003c/em\u003e bloodadvances.2024014181 (2025) doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1182/bloodadvances.2024014181\u003c/span\u003e\u003cspan address=\"10.1182/bloodadvances.2024014181\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKneuertz, P. J. \u003cem\u003eet al.\u003c/em\u003e Overtreatment of young adults with colon cancer: more intense treatments with unmatched survival gains. \u003cem\u003eJAMA Surg\u003c/em\u003e 150, 402\u0026ndash;409 (2015).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMauri, G., Patelli, G., Crisafulli, G., Siena, S. \u0026amp; Bardelli, A. Tumor \u0026lsquo;age\u0026rsquo; in early-onset colorectal cancer. \u003cem\u003eCell\u003c/em\u003e 188, 589\u0026ndash;593 (2025).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMehta, S. J., Morris, A. M. \u0026amp; Kupfer, S. S. Colorectal Cancer Screening Starting at Age 45 Years-Ensuring Benefits Are Realized by All. \u003cem\u003eJAMA Netw Open\u003c/em\u003e 4, e2112593 (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLazzari, L. \u003cem\u003eet al.\u003c/em\u003e Abstract CT214: AlfaOmega- a master protocol empowering precision research in colorectal cancer. \u003cem\u003eCancer Research\u003c/em\u003e 79, CT214 (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCavestro, G. M. \u003cem\u003eet al.\u003c/em\u003e Delphi Initiative for Early-Onset Colorectal Cancer (DIRECt) International Management Guidelines. \u003cem\u003eClin Gastroenterol Hepatol\u003c/em\u003e 21, 581\u0026ndash;603.e33 (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGe, S. X., Jung, D. \u0026amp; Yao, R. ShinyGO: a graphical gene-set enrichment tool for animals and plants. \u003cem\u003eBioinformatics\u003c/em\u003e 36, 2628\u0026ndash;2629 (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePerez-Riverol, Y. \u003cem\u003eet al.\u003c/em\u003e The PRIDE database at 20 years: 2025 update. \u003cem\u003eNucleic Acids Res\u003c/em\u003e 53, D543\u0026ndash;D553 (2025).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSato, T. \u003cem\u003eet al.\u003c/em\u003e Long-term expansion of epithelial organoids from human colon, adenoma, adenocarcinoma, and Barrett\u0026rsquo;s epithelium. \u003cem\u003eGastroenterology\u003c/em\u003e 141, 1762\u0026ndash;1772 (2011).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDijkstra, K. K. \u003cem\u003eet al.\u003c/em\u003e Generation of Tumor-Reactive T Cells by Co-culture of Peripheral Blood Lymphocytes and Tumor Organoids. \u003cem\u003eCell\u003c/em\u003e 174, 1586\u0026ndash;1598.e12 (2018).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMarasca, F. \u003cem\u003eet al.\u003c/em\u003e LINE1 are spliced in non-canonical transcript variants to regulate T cell quiescence and exhaustion. \u003cem\u003eNat Genet\u003c/em\u003e 54, 180\u0026ndash;193 (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDesai, N. \u003cem\u003eet al.\u003c/em\u003e Diverse repetitive element RNA expression defines epigenetic and immunologic features of colon cancer. \u003cem\u003eJCI Insight\u003c/em\u003e 2, e91078 (2017).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-7193450/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7193450/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe incidence of early-onset colorectal cancer (EO-CRC, diagnosed earlier than age 50) is rising worldwide. Despite distinctive clinicopathological features, whether EO-CRC represents a biologically distinct entity from standard-onset CRC (SO-CRC) remains unclear. To investigate molecular underpinnings of EO-CRC, we applied high-resolution label-free mass spectrometry coupled with transcriptomic approaches on primary tumours, healthy mucosae, and metastases of EO-CRC and SO-CRC patients. Most EO-CRC displayed reactivation of placental-like programs and HERVH reactivation, a family of retrotransposons maintaining pluripotency. These features were retained in patient-derived organoids (PDOs) showing sensitivity to pharmacological ATR (Ataxia Telangiectasia and Rad3-related) inhibition. While these findings point to specific EO-CRC vulnerabilities, they require further validation in larger geographically distinct series. These findings distinguish most EO-CRC from SO-CRC as they possess specific placental mimicry and HERVH reactivation. The placental mimicry and HERVH reactivation observed may provide a molecular rationale for EO-CRC aggressive behaviour and suggest potential avenues for therapeutic targeting.\u003c/p\u003e","manuscriptTitle":"Early-Onset Colorectal Cancers Exhibit Distinctive Placental-Like Features","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-06 09:39:49","doi":"10.21203/rs.3.rs-7193450/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"ef8a6b35-3413-4eef-94f7-4f6e9dc230b9","owner":[],"postedDate":"August 6th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":52375775,"name":"Biological sciences/Cancer/Gastrointestinal cancer/Colorectal cancer"},{"id":52375776,"name":"Health sciences/Medical research/Translational research"}],"tags":[],"updatedAt":"2025-08-27T10:41:14+00:00","versionOfRecord":[],"versionCreatedAt":"2025-08-06 09:39:49","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7193450","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7193450","identity":"rs-7193450","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}