Stage-related transcriptional changes in HR-positive/HER2-negative breast cancer

Stage-related transcriptional changes in HR-positive/HER2-negative breast cancer | 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 Stage-related transcriptional changes in HR-positive/HER2-negative breast cancer William S. O. Symmans, Kevin Tran, Chunxiao Fu, Hongxia Sun, Rebekah Gould, and 6 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8207456/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 9 You are reading this latest preprint version Abstract It is important to understand how the biology of breast cancer might differ with the burden of disease. Analyzing the expression of cancer-relevant genes with increasing diagnostic stage can provide insight into the changing biology of breast cancer and may also be clinically useful. Here, we evaluated the transcriptional activity according to stage of breast cancer (American Joint Committee on Cancer, AJCC Stage I to IV) from 1,152 patients with hormone receptor-positive HER2-negative breast cancer. Fresh tumor samples were prospectively collected for clinical-translational research with transcriptional profiling. Most transcriptional signatures remained consistent across stage categories, but the significant changes associated with increasing stage were attributed to endocrine escape, decreasing differentiation, increased complexity of signal transduction, and a few metabolic changes. The clinical relevance and implications of these findings are discussed. The expression data from these high-quality samples are a new publicly available resource for researchers. Health sciences/Biomarkers Biological sciences/Cancer Biological sciences/Computational biology and bioinformatics Health sciences/Oncology Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 INTRODUCTION Staging of breast cancer using the American Joint Committee on Cancer (AJCC) system is an important measure of the extent and risk of the disease. 1 Consequently, staging is an essential tool used for planning the course of treatment and making adjustments based on the estimated level of risk. However, a biological assessment that reflects stage and state of disease is also critical for understanding the full spectrum of a patient’s disease and for developing a personalized therapeutic plan. For example, gene expression profiles have greatly impacted treatment approaches for patients with Stage I-IIA hormone receptor-positive, HER2-negative (HR+/HER2-) cancer. 2 – 4 There is also potential to apply gene expression profiles to more advanced stages of disease to guide treatment. In this context, a comparison of static and dynamic changes in gene expression profiles across the stage categories of HR+/HER2- breast cancer can provide a useful perspective on which clinically or biologically relevant profiles can differ by stage of disease in a population. Our approach was to compare populations defined by stage categories to test for associations between stage of disease and expression level of single genes, multigene prognostic or predictive signatures with established relevance to breast cancer, and more generic cancer biology signatures, many of which relate to the 14 established Hallmarks of Cancer. 5 We assembled a large cohort of fresh clinical samples that were prospectively collected on research protocols and optimally preserved for high-quality RNA. 6 – 9 All had gene expression profiling performed using whole transcriptome microarrays. Adaptation of clinical sampling methods for research sample procurement to different clinical scenarios allowed us to successfully obtain cancer samples for different research protocols. That is because fine needle aspiration (FNA) cytology is generally well accepted for research collection when there is advanced disease, or when diagnosis has already been established, and the patient will proceed with neoadjuvant treatment. 6 In those scenarios, there may be no need for another core biopsy for research. Otherwise, the prospective collection of fresh tissue samples is more acceptable when patients have been diagnosed with invasive breast cancer, and their primary treatment is surgery. Thus, fresh tumor prospective research samples may come from FNA cytology, core biopsy tissue, or fresh breast cancer tissue collected at the time of surgery, and the type of sample is influenced by the stage of the disease and clinical logistics. This also introduces a major preanalytical effect due to differences in the cellular populations within cytology and tissue samples. 10 Therefore, we formally compared matched pairs of cytology and tissue samples that were collected in a subset of patients at the time of surgery, specifically to calibrate gene expression measurements between cytology and tissue samples. 8 This comparison enabled us to account for the type of clinical sample when interpreting differences in gene expression levels as attributable to differences by stage of disease. Overall, this study examines stage-associated changes in the expression of single genes and gene expression signatures of proven relevance to patients with HR+/HER2- breast cancer, along with gene expression profiles that represent important biological signatures of cancer. Notably, we studied this in high-quality samples that were collected prospectively in clinical research protocols and accounted for whether those were cytology or tissue samples. Finally, we present a novel cohort of prospectively collected, high-quality RNA expression profiles from clinical research protocols, which is a new data resource for others to study for stage-related changes in HR + breast cancer. RESULTS Patient and Sample Characteristics The study cohort consisted of 1,152 HR+/HER2- breast cancers (See REMARK diagram in Supplementary Figure S1 ). There was sufficient representation of samples across five stage categories (I, IIA, IIB, III, IV), although a greater number of samples were from Stage IIA (Fig. 1 ). Tissue samples were more commonly collected from patients with lower stage disease (Stages I and IIA) and were usually from surgically resected breast cancer. Cytology samples were more frequently collected from patients with more advanced stage, particularly Stages III and IV (Fig. 1 ). The cohort included matched pairs of cytology and tissue samples collected at the time of surgery from 75 patients. Differences across the Spectrum of Stage I to Stage IV Table 1 shows the results from the selected transcripts and multigene signatures that had significant regression of expression values as Stage increased (p < 0.001). All other results are shown in Supplementary Table S1 . Table 1 Breast Cancer-Specific Genes and Clinical Signatures of Interest Sample Matched Pairs Clinical Samples Cohort Cytology vs Tissue Cytology vs Tissue AJCC Stage Effect Signature Estimate 95% CI p-value Estimate 95% CI p-value Estimate 95% CI p-value Single Genes (Probesets) ESR1 Probeset 0.311 -0.121, 0.743 0.157 -0.233 -0.415, -0.051 0.012 -0.121 -0.188, -0.055 < 0.001 PGR Probeset 0.096 -0.088, 0.280 0.302 0.025 -0.049, 0.098 0.512 -0.058 -0.084, -0.031 < 0.001 ERBB2 Probeset 0.279 0.072, 0.487 0.009 -0.154 -0.257, -0.051 0.003 -0.056 -0.094, -0.019 0.003 AURKA Probeset 0.109 -0.041, 0.259 0.153 0.184 0.109, 0.258 < 0.001 0.063 0.036, 0.090 < 0.001 Breast Cancer Multigene Signatures OncotypeDX -5.068 -14.496, 4.360 0.290 1.690 -0.958, 4.339 0.211 2.198 1.233, 3.162 < 0.001 rorS 5.285 -3.262, 13.831 0.224 11.141 7.747, 14.536 < 0.001 5.102 3.866, 6.338 < 0.001 endoPredict 0.134 -0.651, 0.920 0.736 0.907 0.595, 1.220 < 0.001 0.480 0.366, 0.594 < 0.001 GGI 0.094 0.004, 0.184 0.042 0.132 0.097, 0.167 < 0.001 0.033 0.020, 0.045 < 0.001 SETERPR 0.092 -0.139, 0.322 0.434 -0.040 -0.126, 0.046 0.363 -0.104 -0.135, -0.073 < 0.001 PIK3ges 0.108 -0.130, 0.346 0.371 -0.176 -0.267, -0.085 < 0.001 -0.072 -0.105, -0.039 < 0.001 Cancer Biology Signatures Hallmark_PI3K_AKT_MTOR_signaling 0.014 -0.033, 0.062 0.559 -0.022 -0.044, -0.000 0.045 0.014 0.006, 0.022 0.001 SRC Signature -0.021 -0.040, -0.001 0.041 -0.022 -0.033, -0.012 < 0.001 -0.009 -0.013, -0.005 < 0.001 IMMUNE Signature -1458.583 -1946.473, -970.693 < 0.001 93.247 -149.601, 336.095 0.452 146.009 57.573, 234.444 0.001 CIN70 Signature 0.126 0.036, 0.216 0.006 0.111 0.075, 0.148 < 0.001 0.025 0.012, 0.038 < 0.001 HALLMARK_SPERMATOGENESIS 0.063 0.021, 0.105 0.003 0.182 0.162, 0.202 < 0.001 0.014 0.007, 0.021 < 0.001 HALLMARK_HEME_METABOLISM -0.001 -0.034, 0.032 0.951 0.056 0.035, 0.078 < 0.001 0.018 0.010, 0.026 < 0.001 A cancer biology signature was included from the 50 listed in the Broad Institute Gene Set Enrichment Analysis if its stage effect was highly significant (p < 0.001). For the matched pairs cohort, this meant that cytology and tissue were significantly different. For the clinical samples cohort, this meant that the gradient of the regression line across staging was significantly up or down. The “estimate” column quantifies the difference in means between cytology and tissue for matched pair samples, with a positive value indicating that the cytology was higher than tissue. “Estimate” in the clinical cohort is the slope of the regression line with increasing stage category as an integer. Single Genes That Discriminate Breast Cancer Subsets: There was no significant difference in the expression of ESR1 , PGR , or AURKA in the matched pairs of cytology and tissue samples (Table 1 ). Expression of ERBB2 was significantly higher in cytology samples (Table 1 ). We observed a small, but statistically significant change in the expression levels of ESR1 , PGR , and ERBB2 ; and a significant increase in the expression level of AURKA with increasing stage category (Table 1 , Fig. 2 ). Multigene Signatures: GGI was the only multigene signature that differed by sample type in matched sample pairs, being slightly higher in cytology samples (Table 1 ). The expression levels of multigene prognostic signatures were significantly increased in higher stages of disease (Fig. 3 ). Specifically, Recurrence Score and GGI demonstrated consistent, incremental increases (Fig. 3 A, 3 B), whereas ROR-S and EndoPredict showed a gradual increase from Stage I to IIB, then a greater increase from Stages IIB through IV (Fig. 3 C, 3 D). On the other hand, there was a steady downward trend in the signature that quantifies endocrine transcriptional activity attributable to estrogen receptor and progesterone receptor, i.e., the sensitivity to endocrine therapy \(\:{SET}_{ER/PR}\) index (Fig. 3 E). There was the same downward trend with a signature that represents gene expression related to activating mutation of the PIK3CA gene in HR + breast cancers and represents overactivity of PI3-kinase mediated signal transduction (i.e., PIK3ges). The expression levels of PIK3ges had a relatively flat trend from Stage I to IIB (Fig. 3 F). The type of tumor sample contributed independently to the stage-related trends for every multigene signature, except for Recurrence Score and \(\:{SET}_{ER/PR}\) index (Table 1 ). General Biological Signatures Attributable to Cancer: The expression levels of multigene signatures representing generic cancer biology were mostly consistent across the stages of breast cancer. Those with a highly significant stage-related change (p < 0.001) are shown in Table 1 , the others in Supplementary Table S1 . There was a slight increase in the expression of a cancer biology signature representing signal transduction through the PI3-kinase, AKT, and MTOR pathway (Fig. 4 A), and a steady decrease in the expression of a multigene signature for signal transduction via SRC (Fig. 4 B). The expression of a multigene signature representing immune activity within the tumor increased slightly as the stage increased (Fig. 4 C). A signature for chromosomal instability increased steadily from Stages I to IV (Fig. 4 D), as did a spermatogenesis signature representing the “stem-like” activation of cancer cells to promote proliferation (Table 1 ). Finally, a signature representing deregulated porphyrin-related heme metabolism remained flat from Stages I to III but increased significantly from III to IV (Fig. 4 E, Table 1 ). Stage-related Differences within Stage IV Based on Progression Events Figure 5 presents the most significant changes in single genes or multigene signatures within Stage IV cancer samples categorized by when they were collected: de novo , 1st or 2nd recurrence, or 3rd or later recurrence. The expression of the estrogen receptor gene was decreased in samples from later recurrence (Fig. 5 A), as was the expression of the \(\:{SET}_{ER/PR}\) index of endocrine activity (Fig. 5 B). Samples with a mutation of ESR1 were from recurrent disease appeared to have higher expression levels of ESR1 and \(\:{SET}_{ER/PR}\) index than samples without an ESR1 mutation, although high expression was also seen in some metastases that did not have ESR1 mutation. Average \(\:{SET}_{ER/PR}\) index expression declined on the first recurrence of breast cancer, then remained relatively stable in samples from later recurrences (Fig. 5 B). The decrease in median expression level of PIK3ges had a similar pattern to \(\:{SET}_{ER/PR}\) index. There was a strong upward trend in the expression of genes representing deregulated cellular energetics through heme metabolism with increasing number of recurrence events (Fig. 5 D). DISCUSSION In our study, most transcriptional signatures did not change significantly, including several important biological signatures representing glycolysis, hypoxia, and apoptosis. However, certain biological themes of stage-related change were noteworthy. Generally, we observed a lower state of differentiation (including diminution of endocrine signaling), changes in signal transduction profiles, an increase in proliferation and genomic grade, increasing immune activity and instability of the genome, and some evidence of metabolic dysregulation in more advanced stages of HR+/HER2- breast cancer. A more detailed analysis of Stage IV cancers based on the number of progression events suggested a shifting away from endocrine signaling and PI3-kinase-mediated activity, with a shift towards some aspects of changing energy metabolism. Our findings point towards a modest overall change in the cellular functions within the cancer as it advances in stage. Ultimately, the treatment protocol adapts to the dynamic biological changes acquired through the course of the disease, eventually informing the potential transition to cytotoxic treatment as a late resort in an advanced HR + setting. We observed endocrine escape as decreasing SET ER/PR index and ESR1 with increasing stage. The former was consistent through Stages I to III, and from de novo through progressive Stage IV disease, i.e., through the entire continuum of stage and progression. We are aware that a recent longitudinal study of patient-matched HR+/HER2- breast tumors reported an increase in ESR1 gene expression from early cancer to metastatic. 11 However, this does not necessarily conflict with our findings because that study measured gene expression only in samples from patients who developed metastatic cancer, whereas most of our patients would not have recurred. The cancers most susceptible to hormonal therapy are less likely to metastasize, and vice versa, possibly reflected by a limited emergence of acquired ESR1 mutation. This difference in population may also explain why we observed a decrease in ERBB2 expression, while they found an increase. Our study enhances the perspective of the current evolving paradigm in clinical practice for close monitoring and early detection of recurrent disease in high-risk patients. Recent trials reported that early detection of ESR1 mutation before radiographic progression could prompt a treatment change that improved progression-free survival in patients with early stage disease. 12 There is now a validated benefit to closely monitoring high-risk disease in early stages, guided by the emergence of ESR1 as a targetable genetic alteration. 13 ESR1 mutations, captured in ctDNA, are likely an adaptive consequence of secondary resistance to Aromatase Inhibitors (AI). 14 In this setting, treatment aims to address the dynamic evolution of the biological state of the disease, and longitudinal molecular characterization methods offer an opportunity for precision medicine; Selective Estrogen Receptor Degraders (SERDs) were proven more effective than the standard AI in treating HR+/HER2- breast cancer with mutated ESR1 . Tracking ESR1 mutation status in ctDNA is a sensitive way to monitor progression and diagnose early recurrence of endocrine-resistant disease. After endocrine activity, the second most important driver in hormone receptor-positive breast cancer is signal transduction, particularly through the PI3-kinase pathways. PIK3CA is the most commonly mutated gene in HR+/HER2- breast cancer at a rate of 40%, so characterizing its expression across the stages gives insight into a key driver of the disease. 15 We observed a seemingly discordant decrease in the multigene PI3-kinase gene expression signature with increasing stage and number of recurrences and an increase in the collection of genes representing activating mutations of the PIK3-AKT-MTOR pathway. One explanation is that signaling complexity increases with stage, such that the cancer might not be as reliant on PIK3CA activity, specifically as alternative uses of this signal transduction mechanism evolve. 16 In other words, the early biology of PI3-kinase seems to be replaced by other signal transduction opportunities as the cancer progresses. However, the status of PIK3CA mutation status in primary and metastatic breast cancer has been shown to be largely concordant (69%). 17 Another possible explanation is that an upstream activator of the pathway overpowers the declining PI3-kinase activity, thereby increasing the expression of AKT and MTOR. Or, despite diminished PI3-kinase activity, a repressor of downstream activity is removed. A different measure of signal transduction in cancer changed in the opposite direction to PI3K-AKT-MTOR across the stages. SRC is an upstream activator of PI3K-AKT-MTOR signaling and other signaling pathways. 18 The balance of these two signatures suggested a subtle shift towards PI3K-AKT-MTOR signaling in advancing stages of cancer and away from the upstream SRC pathway that also promotes cell growth and survival. In addition, the expression of the ERBB2 gene for HER2 declined with advancing stage. Overall, within these trends, it seemed that membrane signaling remained consistent across Stages I to IIB, then dropped off slightly in Stages III and IV. The presence of a PIK3CA mutation is addressed in clinical studies that investigate primary resistance to endocrine therapy. Recently, Inavolisib (PI3Kα inhibitor) plus Palbociclib (CDK4/6 inhibitor) and Fulvestrant (SERD) was the first FDA-approved triplet combination treatment used as a first-line treatment for patients with PIK3CA -mutated, locally advanced or metastatic HR + breast cancer. These patients were considered to have endocrine-resistant disease after recurrence on or after completing adjuvant endocrine therapy. 19 Additionally, the CAPItello-291 trial investigated the combination of Capivasertib (AKT inhibitor) and Fulvestrant in a population that progressed after AI treatment with or without CDK4/6 inhibitor. 20 They demonstrated a progression-free survival benefit independent of both the CDK4/6 inhibitor exposure and the detection of AKT pathway alteration. When HR+/HER2- breast cancer does progress, it seemingly becomes less differentiated. Overall, we observed that the multigene expression signatures that represent lesser differentiation were increased with advancing stage. With ROR-S and EndoPredict, but not Recurrence Score, this change was most pronounced at later stages, which fits into the concept that the expression of genes associated with a diminished state of differentiation should reflect more aggressive biology in the cancer. 21 Expression profiles representing proliferation had a more complex association with stage. Aurora kinase A expression increased, matching our expectations that more advanced cancer tends to have a higher level of proliferative signaling because it is innately more aggressive or has progressed to a more aggressive, genetically unstable state. 5 Another marker of proliferation, the “hallmark” of spermatogenesis, representing the “stem-like” activation of cancer cells to promote proliferation, 22 showed an upward trend that was slightly exaggerated by the difference in sample type. One interpretation might be that proliferation was more likely to be driven by alternative growth factor pathways. Our findings are relevant to the current approach to management of metastatic and other locally advanced breast cancers. As blockers of proliferation, CDK4/6 inhibitors, in addition to endocrine therapy with AI or Fulvestrant, represent the current standard of care as a first-line therapy in the metastatic setting. However, the lack of a specific biomarker for treatment selection and the toxic effect of their extended use remain challenges in clinical practice, especially in their integration into adjuvant treatment. Maintenance of CDK4/6 inhibitors throughout progression in order to overcome endocrine treatment resistance represents an emerging practice, as there is a reported benefit for disease control. 23 , 24 Diminishing endocrine sensitivity necessitates the adjustment of the endocrine backbone while continuing with CDK4/6 inhibitors. However, there was conflicting data on the efficacy and predictive value of an ESR1 mutation alone and in combination with other known somatic gene alterations associated with CDK4/6 inhibitor exposure in this therapeutic strategy. 25 Despite new endocrine and targeted treatment options following CDK4/6 inhibitor usage, beyond first-line, chemotherapy remains a valid choice, especially for clinically-aggressive disease without any druggable target. When looking at the broader tumor microenvironment, we observed that an immune signature representing overall immune presence in these immunologically cold HR+/HER2- tumors increased with advancing stage. 26 This conflicts with the finding that metastatic lesions demonstrate low immunogenicity because they include a larger representation of immune-avoidant clones; however, it should be noted that most of our patients were not from Stage IV. 26 A growing, infiltrating, sustained cancer will evade anti-cancer immune signaling and activation. 27 Although statistically significant, the increase in immune cell gene expression with advancing stage was subtle, with little change from Stage III to IV. Therefore, the association that is observed could be due to less-differentiated tumors generally having slightly more cellular immune infiltration. Nevertheless, immune checkpoint inhibitors have not demonstrated efficacy in an HR + setting. At the chromosomal level, genomic instability signaling increased across stages, perhaps because mutations were more prevalent in higher-stage disease. This fits with an understanding of how cancer reflects an ever-destabilizing genome. 28 However, it is important to mention that we do not have DNA sequence or cytogenetic data, so we can only extrapolate about this possible effect. Metabolic dysregulation is another important factor that drives changes. As cancer progresses, its energetic demands also increase. 5 The cancer biology signature heme metabolism, which describes the metabolic shift towards the overproduction of heme intermediates like porphyrin for sustaining the cancer, 29 was significantly higher in Stage IV in our study. It was also the only so-called “hallmark” transcriptional signature to progressively rise within Stage IV cancers based on the number of progression events. In cancer, there is a non-homeostatic upregulation of heme intermediates, called porphyrins, to fuel the cancer in a process called porphyrin overdrive 29 Porphyrin overdrive stands out from other metabolic phenotypes like the Warburg Effect, glutamine addiction, and increased fatty acid anabolism because it represents the cancer-specific accumulation of heme intermediates. 29 Cancer cells rely on these intermediates for essential cellular processes, such as proliferation and the protection against iron-induced ferroptosis. 30 Imbalanced heme metabolism is a targetable metabolic phenotype because it is not present in healthy cells and is essential to the survival of the cancer 29 , so our finding invites further study for treatment options to exploit this phenotype. The site of metastasis did not appear to be associated with the level of heme metabolism gene expression signature in the biopsy samples (Supplementary Figure S2). Other signatures that represented metabolism, but did not change significantly, were fatty acid metabolism, glycolysis, and oxidative phosphorylation (Table S2). However, other important metabolic phenotypes of cancer are not well represented by gene expression signatures, so we cannot speculate on their stage-related trends. The glycolytic state did not vary in mRNA expression, but this is not the most suitable measure anyway; metabolism occurs at the protein and biochemical level, so DNA or RNA expression profiles are only remotely related and cannot definitively measure this. Indeed, our samples could not take into account downstream enzymatic activity, so it is possible that even though we see changes at the transcriptional level, enzymatic activity may not change. Glycolytic state is a very important parameter of treatment resistance, even relevant to immunotherapy. A highly glycolytic state creates an unfavorable tumor microenvironment because it increases acidity, which in turn increases the activity of immune-suppressive T-cells. 31 This concept invites new clinical strategies to select patients who may particularly benefit from immunotherapy. Overall, it is important to note that some signatures in our study contribute to the development of cancer, while others promote the progression of cancer to higher stages. Similarly, for some signatures, the differences in expression between cytology and tissue can be explained by the natural features of those signatures. For example, we would expect that CIN70 and spermatogenesis showed higher expression in the cytology at every stage because the chromosomal instability measure would be best captured within the more concentrated overall cellularity of an FNA, rather than the tissue sample, which would include stroma too. There was a similar confounding effect of sample type with tissue, as it has mesenchymal stroma, whereas cytology has very little, 10 so signatures like angiogenesis and epithelial-mesenchymal transition were much higher in tissue than in cytology. Here, we would like to address the limitations of our study. To start, our study, taking place over many years, used a mature data set from samples that were freshly collected into RNA later and evaluated during that time by Affymetrix gene expression microarray technology, which predates contemporary RNA sequencing. Still, these signatures of interest are published as accurate when using these microarrays, and thus, we can make reasonable inferences. 32 We also acknowledge that these were not longitudinal samples capturing progression within a patient’s cancer timeline. Rather, each one came from a different patient. Studies of paired primary and subsequent metastasis provide more information about the individual’s molecular progression, but do not address the diversity across stages of primary disease. 11 , 33 Another limitation is that when assessing gene expression as a function of stage, as expected, there were fewer cytology samples from early stages and fewer tissue samples from later stages. This sample bias was a result of the context and method in which prospective samples can reasonably be obtained in prospective research protocols. Nevertheless, we feel their exact number and placement at these “boundary stages” is less important than the overall message of the trend (i.e., whether they go up or down or remain flat, when they change, and how quickly). Additionally, there were very few 1st-recurrence patients in our Stage IV-specific trends, because it is a difficult time to discuss with the patient collecting optional research samples when the diagnosis has not been established at the time of biopsy. In that clinical setting, we would need to have relied on archival fixed biopsy samples and those should be prioritized for clinical molecular testing. On the other hand, an advantage of our methodology was that our samples were all from fresh samples and unaffected by preanalytical effects of cold ischemia, fixation or storage (Sura et al., 2025). 34 – 36 Moreover, for our Stage IV analysis, we were not able to take into account how long the interval was from prior treatment before they had the biopsy, for example, whether it was an early or late recurrence. It would also have been helpful to know the PIK3CA mutation status when evaluating the related gene expression signatures in the samples from metastases. The gene expression signatures and the cancer biology signatures are not validated diagnostic signatures, so technical variance may be an issue. Also, they are approximations of biology, not the same as studying biological function. This is true for most of the gene expression signatures that we analyzed. Still, our results provide insights that would guide more specific research. In conclusion, our study summarizes the most significant transcriptional changes from high-quality clinical research biopsies that are related to advancing stage across a large population of patients with HR+/HER2- breast cancer. We evaluated their biological associations with contemporary treatments for this disease and identified a few novel transcriptional changes with potential for novel therapeutic approaches. METHODS The cohort for these analyses were from the compilation of gene expression data from past studies wherein fresh tumor samples of HR+/HER2- breast cancer were profiled on U133A microarrays using uniform methodology. All patients gave informed written consent to take part in the study and for the use of tissue material for research purposes. Definition of HR+/HER2- breast cancer was based on the results of clinical biomarker testing, as reported at the institution where the sample was collected. HR-positive status was defined as > 1% tumor nuclei stained positive for ER or PR by immunohistochemistry. Sample cohorts Samples from primary surgical resection specimens at MDACC were collected fresh at time of intraoperative specimen assessment by a dedicated breast pathologist (WFS) under research IRB protocols with MD Anderson Institutional Review Board (IRB) approval: LAB04-0093, LAB08-0823, LAB08-0824, and 2011-0007. A shave of the tissue from one side of the cut tumor mass was assessed by imprint cytology (to confirm malignancy) and diced in a petri dish and stored frozen at -80°C in RNAlater until use. A subset of fresh tumor samples collected at MDACC were included in the cohort with matched cytology and tissue samples. 8 For those, a cytologic sample was collected before the tissue sample. The freshly cut surface of the tumor was gently scraped using a small scalpel blade, a portion of that liquid sample was placed on a glass slide for assessment of the cytologic smear (to confirm malignancy) and the remainder was mixed into RNAlater solution. This was repeated 2–3 times with the liquid sample entirely added to the RNAlater solution. Immediately thereafter, the subjacent tumor tissue was shaved, diced, and added to a separate vial of RNAlater solution, as described above. Frozen tumor samples from collaborating institutions were stored frozen at -80°C and shipped on dry ice, as previously reported. 7 Samples from needle biopsies, either fine needle aspiration cytology or core biopsy tissue, were prospectively collected and immediately placed in RNAlater solution and held at room temperature at least 30 minutes to allow penetration of the preservative, then stored frozen at -80°C in RNAlater until use, as previously described (approved IRB protocols LAB99-402, LAB04-0093, 2011-0007). 7 , 9 We categorized AJCC Stage into groups of Stage I, IIA, IIB, III and IV. If the sample was obtained from Stage I-III breast cancer prior to neoadjuvant treatment, then we used clinical Stage, otherwise we used pathologic Stage if surgery was the primary treatment. For patients with documented Stage IV (i.e., documented distant metastasis), we further classified into de novo (i.e., Stage IV at first presentation), 1st or 2nd metastatic relapse, or 3rd or later metastatic relapse. Gene expression profiling for target and reference transcripts RNA was extracted, processed and hybridized to Affymetrix human genome U133A microarrays (U133A GeneChip, Affymetrix, Santa Clara, CA, USA) as described previously. In brief, the raw intensity files were processed using the MAS5.0 algorithm to generate probe set-level intensities, normalized to a median array intensity of 600, log2-transformed. 7 , 9 Selection of genes and signatures RNA was purified and hybridized to Affymetrix U133A microarrays (U133A GeneChip; Affymetrix, Santa Clara, CA) as reported previously. 7 , 9 Briefly, raw data was processed using the MAS5 algorithm as implemented in the affy3 package to generate probe—level intensities, scaled to a median intensity of 600 and log2—transformed. The data was then scaled to 1,322 breast cancer reference genes within each sample following MAS5 normalization. 7 , 9 As single genes we selected the four most relevant genes for classification of breast cancer, including the transcripts for estrogen receptor, progesterone receptor, and HER2 that represent the standard molecular therapeutic targets in breast cancer. 37 Research versions of the EndoPredict and Oncotype DX signatures, of the PAM50 and SCMGENE classifiers, the GGI and PIK3CA signatures and gene modules were calculated using the genefu5 package. 32 The SET Index was calculated according to the original publication. 8 We obtained the cancer “Hallmarks” signatures from the Gene Set Enrichment Analysis Human MSigDB Collection. In the matched pairs, the tissue samples were profiled at three different laboratories and the average of those triplicate results was used to compare with the cytology profile that was performed at MDACC, as previously reported. 8 Statistical methods Pearson’s correlation coefficient was used to compare cross-tissue concordance correlation coefficient (CCC) between the matched pairs of tissue and cytology samples for each transcript and signature. To evaluate the effect of disease stage on the genomic signatures, we fit linear regression models fitted to median values (MASS package7) and adjusted for the type of sample (cytology vs. tissue) as a covariate. P-values were obtained by comparing the gradient of the regression line with zero (t-test). The same method was used to evaluate linear regression according to the number of recurrence events in the Stage IV cohort. All statistical analyses and computations were performed in R v. 4.5.1 and Bioconductor. 38 , 39 Declarations Data sharing statement. The microarray and accompanying data are being uploaded to NCBI GEO prior to publication. Clinical trial number: not applicable AUTHOR CONTRIBUTIONS LP, KT, RS, VV, WFS and EA contributed to the conception and design of the work; CF, HS, RG, LP, TK, RS, VV, WFS and EA contributed to the acquisition of samples and data; WSOS, KT, and WFS contributed to the analysis; WSOS, KT, WFS and EA contributed to the interpretation of data; WSOS, KT, WFS and EA drafted the work, and all authors reviewed the manuscript. ACKNOWLEDGEMENTS None. FUNDING This work was supported by the following research grants: Breast Cancer Research Foundation, BCRF-158 (WFS); Susan G. Komen Foundation, SAC110034 (WFS); National Cancer Institute, HHSN261200800001E (WFS); Cancer Prevention and Research Institute of Texas, RP180712 (WFS). CONFLICT OF INTEREST The authors have no competing interests to declare. Non-competing interests that might be perceived to influence the discussion in this paper were: WFS reports stock and other ownership interests in ISIS Pharmaceuticals, Delphi Diagnostics, and Eiger BioPharmaceuticals; consulting or advisory roles for AstraZeneca; research funding from Pfizer (paid to Institution); Co-inventor, US Patent No. 11,459,617 “Targeted Measure of Transcriptional Activity Related to Hormone Receptors,” issued on October 4, 2022 (applicant proprietor: University of Texas MD Anderson Cancer Center) and licensed to Delphi Diagnostics; and an uncompensated scientific advisor relationship with Delphi Diagnostics. CF reports stock in Delphi Diagnostics and co-inventor on US Patent No. 11,459,617. The other authors do not have any relevant non-competing interests. References Giuliano, A. E. et al. Breast Cancer-Major changes in the American Joint Committee on Cancer eighth edition cancer staging manual. CA Cancer J Clin 67 , 290-303 (2017). https://doi.org/10.3322/caac.21393 Sparano, J. A. et al. Adjuvant Chemotherapy Guided by a 21-Gene Expression Assay in Breast Cancer. N Engl J Med 379 , 111-121 (2018). https://doi.org/10.1056/NEJMoa1804710 Kalinsky, K. et al. 21-Gene Assay to Inform Chemotherapy Benefit in Node-Positive Breast Cancer. N Engl J Med 385 , 2336-2347 (2021). https://doi.org/10.1056/NEJMoa2108873 Cardoso, F. et al. 70-Gene Signature as an Aid to Treatment Decisions in Early-Stage Breast Cancer. N Engl J Med 375 , 717-729 (2016). https://doi.org/10.1056/NEJMoa1602253 Hanahan, D. Hallmarks of Cancer: New Dimensions. Cancer Discov 12 , 31-46 (2022). https://doi.org/10.1158/2159-8290.CD-21-1059 Hess, K. R. et al. Pharmacogenomic predictor of sensitivity to preoperative chemotherapy with paclitaxel and fluorouracil, doxorubicin, and cyclophosphamide in breast cancer. J Clin Oncol 24 , 4236-4244 (2006). https://doi.org/10.1200/JCO.2006.05.6861 Symmans, W. F. et al. Genomic index of sensitivity to endocrine therapy for breast cancer. J Clin Oncol 28 , 4111-4119 (2010). https://doi.org/10.1200/JCO.2010.28.4273 Sinn, B. V. et al. SET(ER/PR): a robust 18-gene predictor for sensitivity to endocrine therapy for metastatic breast cancer. NPJ Breast Cancer 5 , 16 (2019). https://doi.org/10.1038/s41523-019-0111-0 Hatzis, C. et al. A genomic predictor of response and survival following taxane-anthracycline chemotherapy for invasive breast cancer. JAMA 305 , 1873-1881 (2011). https://doi.org/10.1001/jama.2011.593 Symmans, W. F. et al. Total RNA yield and microarray gene expression profiles from fine-needle aspiration biopsy and core-needle biopsy samples of breast carcinoma. Cancer 97 , 2960-2971 (2003). https://doi.org/10.1002/cncr.11435 Aftimos, P. et al. Genomic and Transcriptomic Analyses of Breast Cancer Primaries and Matched Metastases in AURORA, the Breast International Group (BIG) Molecular Screening Initiative. Cancer Discov 11 , 2796-2811 (2021). https://doi.org/10.1158/2159-8290.CD-20-1647 Bidard, F. C. et al. First-Line Camizestrant for Emerging ESR1-Mutated Advanced Breast Cancer. N Engl J Med 393 , 569-580 (2025). https://doi.org/10.1056/NEJMoa2502929 Bidard, F.-C. et al. Switch to fulvestrant and palbociclib versus no switch in advanced breast cancer with rising ESR1 mutation during aromatase inhibitor and palbociclib therapy (PADA-1): a randomised, open-label, multicentre, phase 3 trial. The Lancet Oncology 23 , 1367-1377 (2022). https://doi.org/10.1016/S1470-2045(22)00555-1 Bidard, F. C. et al. Elacestrant (oral selective estrogen receptor degrader) Versus Standard Endocrine Therapy for Estrogen Receptor-Positive, Human Epidermal Growth Factor Receptor 2-Negative Advanced Breast Cancer: Results From the Randomized Phase III EMERALD Trial. J Clin Oncol 40 , 3246-3256 (2022). https://doi.org/10.1200/JCO.22.00338 Koboldt, D. C. et al. Comprehensive molecular portraits of human breast tumours. Nature 490 , 61-70 (2012). https://doi.org/10.1038/nature11412 Sever, R. & Brugge, J. S. Signal transduction in cancer. Cold Spring Harb Perspect Med 5 (2015). https://doi.org/10.1101/cshperspect.a006098 van Geelen, C. T. et al. Clinical implications of prospective genomic profiling of metastatic breast cancer patients. Breast Cancer Res 22 , 91 (2020). https://doi.org/10.1186/s13058-020-01328-0 Beadnell, T. C. et al. Src-mediated regulation of the PI3K pathway in advanced papillary and anaplastic thyroid cancer. Oncogenesis 7 , 23 (2018). https://doi.org/10.1038/s41389-017-0015-5 Wedam, S. et al. US Food and Drug Administration Approval Summary: Inavolisib With Palbociclib and Fulvestrant for Endocrine-Resistant, PIK3CA-Mutated, Hormone Receptor-Positive, Human Epidermal Growth Factor Receptor 2-Negative, Locally Advanced or Metastatic Breast Cancer. J Clin Oncol 43 , 3123-3131 (2025). https://doi.org/10.1200/JCO-25-00663 Turner, N. C. et al. Capivasertib in Hormone Receptor-Positive Advanced Breast Cancer. N Engl J Med 388 , 2058-2070 (2023). https://doi.org/10.1056/NEJMoa2214131 Jögi, A., Vaapil, M., Johansson, M. & Påhlman, S. Cancer cell differentiation heterogeneity and aggressive behavior in solid tumors. Ups J Med Sci 117 , 217-224 (2012). https://doi.org/10.3109/03009734.2012.659294 Liberzon, A. et al. The Molecular Signatures Database (MSigDB) hallmark gene set collection. Cell Syst 1 , 417-425 (2015). https://doi.org/10.1016/j.cels.2015.12.004 Gerratana, L. et al. Circulating Tumor Cell Dynamics after CDK4/6 Inhibitor for Hormone Receptor-Positive Metastatic Breast Cancer: A Biomarker Analysis from the PACE Phase II Study. Clin Cancer Res 31 , 4510-4517 (2025). https://doi.org/10.1158/1078-0432.CCR-25-0327 Kalinsky, K. et al. Randomized Phase II Trial of Endocrine Therapy With or Without Ribociclib After Progression on Cyclin-Dependent Kinase 4/6 Inhibition in Hormone Receptor-Positive, Human Epidermal Growth Factor Receptor 2-Negative Metastatic Breast Cancer: MAINTAIN Trial. J Clin Oncol 41 , 4004-4013 (2023). https://doi.org/10.1200/JCO.22.02392 Mayer, E. et al. Abstract GS3-06: GS3-06 Palbociclib After CDK4/6i and Endocrine Therapy (PACE): A Randomized Phase II Study of Fulvestrant, Palbociclib, and Avelumab for Endocrine Pre-treated ER+/HER2- Metastatic Breast Cancer. Cancer Research (2023). https://doi.org/10.1158/1538-7445.sabcs22-gs3-06 Esteva, F. J., Hubbard-Lucey, V. M., Tang, J. & Pusztai, L. Immunotherapy and targeted therapy combinations in metastatic breast cancer. Lancet Oncol 20 , e175-e186 (2019). https://doi.org/10.1016/S1470-2045(19)30026-9 Kim, S. K. & Cho, S. W. The Evasion Mechanisms of Cancer Immunity and Drug Intervention in the Tumor Microenvironment. Front Pharmacol 13 , 868695 (2022). https://doi.org/10.3389/fphar.2022.868695 Yates, L. R. et al. Genomic Evolution of Breast Cancer Metastasis and Relapse. Cancer Cell 32 , 169-184 e167 (2017). https://doi.org/10.1016/j.ccell.2017.07.005 Adapa, S. R. et al. Porphyrin overdrive rewires cancer cell metabolism. Life Sci Alliance 7 (2024). https://doi.org/10.26508/lsa.202302547 Lei, G., Zhuang, L. & Gan, B. Targeting ferroptosis as a vulnerability in cancer. Nat Rev Cancer 22 , 381-396 (2022). https://doi.org/10.1038/s41568-022-00459-0 Zappasodi, R. et al. CTLA-4 blockade drives loss of Treg stability in glycolysis-low tumours. Nature 591 , 652-658 (2021). https://doi.org/10.1038/s41586-021-03326-4 Gendoo, D. M. et al. Genefu: an R/Bioconductor package for computation of gene expression-based signatures in breast cancer. Bioinformatics 32 , 1097-1099 (2016). https://doi.org/10.1093/bioinformatics/btv693 Garcia-Recio, S. et al. Multiomics in primary and metastatic breast tumors from the AURORA US network finds microenvironment and epigenetic drivers of metastasis. Nat Cancer 4 , 128-147 (2023). https://doi.org/10.1038/s43018-022-00491-x Hatzis, C. et al. Effects of tissue handling on RNA integrity and microarray measurements from resected breast cancers. J Natl Cancer Inst 103 , 1871-1883 (2011). https://doi.org/10.1093/jnci/djr438 Li, J., Fu, C., Speed, T. P., Wang, W. & Symmans, W. F. Accurate RNA Sequencing From Formalin-Fixed Cancer Tissue To Represent High-Quality Transcriptome From Frozen Tissue. JCO Precis Oncol 2018 (2018). https://doi.org/10.1200/PO.17.00091 Marczyk, M. et al. Assessment of stained direct cytology smears of breast cancer for whole transcriptome and targeted messenger RNA sequencing. Cancer Cytopathol 131 , 289-299 (2023). https://doi.org/10.1002/cncy.22679 Haibe-Kains, B. et al. A three-gene model to robustly identify breast cancer molecular subtypes. J Natl Cancer Inst 104 , 311-325 (2012). https://doi.org/10.1093/jnci/djr545 Team, R. C. R: A language and environment for statistical computing , (2015). Huber, W. et al. Orchestrating high-throughput genomic analysis with Bioconductor. Nat Methods 12 , 115-121 (2015). https://doi.org/10.1038/nmeth.3252 Additional Declarations Competing interest reported. The authors have no competing interests to declare. Non-competing interests that might be perceived to influence the discussion in this paper were: WFS reports stock and other ownership interests in ISIS Pharmaceuticals, Delphi Diagnostics, and Eiger BioPharmaceuticals; consulting or advisory roles for AstraZeneca; research funding from Pfizer (paid to Institution); Co-inventor, US Patent No. 11,459,617 “Targeted Measure of Transcriptional Activity Related to Hormone Receptors,” issued on October 4, 2022 (applicant proprietor: University of Texas MD Anderson Cancer Center) and licensed to Delphi Diagnostics; and an uncompensated scientific advisor relationship with Delphi Diagnostics. CF reports stock in Delphi Diagnostics and co-inventor on US Patent No. 11,459,617. The other authors do not have any relevant non-competing interests. Supplementary Files SupplementStagerelatedtranscriptionalchanges20251125submit.pdf Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 13 May, 2026 Reviews received at journal 25 Jan, 2026 Reviewers agreed at journal 11 Dec, 2025 Reviewers agreed at journal 11 Dec, 2025 Reviewers agreed at journal 10 Dec, 2025 Reviewers invited by journal 09 Dec, 2025 Editor assigned by journal 30 Nov, 2025 Submission checks completed at journal 29 Nov, 2025 First submitted to journal 25 Nov, 2025 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-8207456","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":558997720,"identity":"6e695747-de67-49dc-8296-89c293617cc0","order_by":0,"name":"William S. O. Symmans","email":"","orcid":"","institution":"The University of Texas MD Anderson Cancer Center","correspondingAuthor":false,"prefix":"","firstName":"William","middleName":"S. O.","lastName":"Symmans","suffix":""},{"id":558997723,"identity":"4dbc2fa0-2395-4295-a459-ea9d4014bf17","order_by":1,"name":"Kevin Tran","email":"","orcid":"","institution":"The University of Texas MD Anderson Cancer Center","correspondingAuthor":false,"prefix":"","firstName":"Kevin","middleName":"","lastName":"Tran","suffix":""},{"id":558997726,"identity":"940c825f-60a0-407e-a091-c239129df4aa","order_by":2,"name":"Chunxiao Fu","email":"","orcid":"","institution":"The University of Texas MD Anderson Cancer Center","correspondingAuthor":false,"prefix":"","firstName":"Chunxiao","middleName":"","lastName":"Fu","suffix":""},{"id":558997727,"identity":"54803a2f-3674-4c6a-836f-fa4ab3680b72","order_by":3,"name":"Hongxia Sun","email":"","orcid":"","institution":"The University of Texas MD Anderson Cancer Center","correspondingAuthor":false,"prefix":"","firstName":"Hongxia","middleName":"","lastName":"Sun","suffix":""},{"id":558997730,"identity":"325262f6-2def-46b3-8329-84d25fd407e9","order_by":4,"name":"Rebekah Gould","email":"","orcid":"","institution":"The University of Texas MD Anderson Cancer Center","correspondingAuthor":false,"prefix":"","firstName":"Rebekah","middleName":"","lastName":"Gould","suffix":""},{"id":558997732,"identity":"b6734f7e-7c3e-40a2-851c-351a89e17a02","order_by":5,"name":"Lajos Pusztai","email":"","orcid":"","institution":"Yale University","correspondingAuthor":false,"prefix":"","firstName":"Lajos","middleName":"","lastName":"Pusztai","suffix":""},{"id":558997737,"identity":"086326b5-354a-4cea-86f1-7e33ac81efd2","order_by":6,"name":"Tari King","email":"","orcid":"","institution":"Emory University","correspondingAuthor":false,"prefix":"","firstName":"Tari","middleName":"","lastName":"King","suffix":""},{"id":558997738,"identity":"a28ad546-eead-42ba-88f6-872a2a08f0e5","order_by":7,"name":"Roberto Salgado","email":"","orcid":"","institution":"ZAS Hospitals","correspondingAuthor":false,"prefix":"","firstName":"Roberto","middleName":"","lastName":"Salgado","suffix":""},{"id":558997741,"identity":"789baa3c-69e8-4621-8c1a-513117d9af23","order_by":8,"name":"Vicente Valero","email":"","orcid":"","institution":"The University of Texas MD Anderson Cancer Center","correspondingAuthor":false,"prefix":"","firstName":"Vicente","middleName":"","lastName":"Valero","suffix":""},{"id":558997742,"identity":"3f5f56b6-ce00-414a-9e94-6f472cee7513","order_by":9,"name":"William Fraser Symmans","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA+klEQVRIiWNgGAWjYBAC9gYog5+BwYCBgY0ILTwHoAzJBpK1GBwgWgv/GdNNN2ru2BtfO7zxc0GZTb58e+8Bho97anFrkcgxu51z7FnitttpxdIzzqVZbjhzLoFxxrPjOLXYS/AAtbAdTgCSBtK8bYcNDCRyDJh5DhzD5zCg4n+H7Y1n5xj/5m37byA/g5AWBqDDctsOM26QzjED2nLAgOEGWEsNHr+kld3O7TucOON2Wpk1z7lkA4MzZwwOzjhwAI/DDm+7nfPtsD3/7OTNt3nK7Azk23sMH3w4UIdTCwMDhwGmGNCKw3i0sD/AKozPllEwCkbBKBhhAABZTVlK5XuR/gAAAABJRU5ErkJggg==","orcid":"","institution":"The University of Texas MD Anderson Cancer Center","correspondingAuthor":true,"prefix":"","firstName":"William","middleName":"Fraser","lastName":"Symmans","suffix":""},{"id":558997743,"identity":"f0b4c384-1aa3-4d15-b7b3-7960cc2036f7","order_by":10,"name":"Eleni Andreopoulou","email":"","orcid":"","institution":"Weill Cornell Medicine","correspondingAuthor":false,"prefix":"","firstName":"Eleni","middleName":"","lastName":"Andreopoulou","suffix":""}],"badges":[],"createdAt":"2025-11-26 01:38:18","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8207456/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8207456/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":98429017,"identity":"84789551-05ba-4b65-8e37-dc9b9576a100","added_by":"auto","created_at":"2025-12-17 16:42:42","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":7167406,"visible":true,"origin":"","legend":"","description":"","filename":"Stagerelatedtranscriptionalchanges20251128submit.docx","url":"https://assets-eu.researchsquare.com/files/rs-8207456/v1/1a2e7a995868763fcb88c35d.docx"},{"id":98075063,"identity":"5eb0e4f6-98cd-4e64-8316-524db2ce5f8d","added_by":"auto","created_at":"2025-12-12 13:31:06","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":11719,"visible":true,"origin":"","legend":"","description":"","filename":"f6006a4e4ca642ac8a6ac29b8370c2be.json","url":"https://assets-eu.researchsquare.com/files/rs-8207456/v1/219bf4c10cdfbec43ad7f76d.json"},{"id":98429013,"identity":"dab78a62-e28a-4be4-83b3-9d44762f2c28","added_by":"auto","created_at":"2025-12-17 16:42:42","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":228837,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementStagerelatedtranscriptionalchanges20251125submit.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8207456/v1/6b98fc65c0525c11b6c09870.pdf"},{"id":98075080,"identity":"85e168f8-0939-421e-9c3f-1945a80168fb","added_by":"auto","created_at":"2025-12-12 13:31:06","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":124502,"visible":true,"origin":"","legend":"","description":"","filename":"f6006a4e4ca642ac8a6ac29b8370c2be1enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8207456/v1/224a91aa8512d1903b9469ff.xml"},{"id":98075082,"identity":"fe075ccb-8508-454e-bb0e-69bfb7f9c648","added_by":"auto","created_at":"2025-12-12 13:31:06","extension":"emf","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2847012,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.emf","url":"https://assets-eu.researchsquare.com/files/rs-8207456/v1/43edd8be8c25b55744b73457.emf"},{"id":98075071,"identity":"f052ea7d-8e9f-47ae-930e-50c7e7afe39e","added_by":"auto","created_at":"2025-12-12 13:31:06","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":319592,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8207456/v1/d4eb9915bdb01f9772ef4ded.png"},{"id":98428969,"identity":"adf20d16-9549-437e-bd7c-6d99e9b3e437","added_by":"auto","created_at":"2025-12-17 16:42:38","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":558581,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8207456/v1/07282dcc4b7759f4b966b645.png"},{"id":98429943,"identity":"31389db5-a7e5-4e08-bc83-25eea44deee1","added_by":"auto","created_at":"2025-12-17 16:44:27","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":500571,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8207456/v1/1ee4cace8fcd46fed91baf80.png"},{"id":98429774,"identity":"64515373-5686-4570-ba15-6943deff0c2c","added_by":"auto","created_at":"2025-12-17 16:44:07","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":223928,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8207456/v1/61aa12b6cb123db8565dd87b.png"},{"id":98075077,"identity":"a417bd1f-4527-4beb-86e8-0e5293eea7d6","added_by":"auto","created_at":"2025-12-12 13:31:06","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":9719,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8207456/v1/d6bfa1ea35f5a5087935fafe.png"},{"id":98075076,"identity":"f11e94e8-5524-4adc-bdd3-bb6460204f83","added_by":"auto","created_at":"2025-12-12 13:31:06","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":69830,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8207456/v1/ccafb5a35d4e65e87489fbad.png"},{"id":98429105,"identity":"2019bc0d-39e2-465b-b89f-bd64cc3d5250","added_by":"auto","created_at":"2025-12-17 16:42:47","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":111331,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8207456/v1/13e45a900d01e0e1ccfc5254.png"},{"id":98075081,"identity":"0059d9c9-8a01-49ba-9960-073a218554a4","added_by":"auto","created_at":"2025-12-12 13:31:06","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":99233,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8207456/v1/219a25c5afb6f064ad8a7f7e.png"},{"id":98429088,"identity":"117fb445-4f05-4e3f-a136-9a0b6b869866","added_by":"auto","created_at":"2025-12-17 16:42:45","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":49702,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8207456/v1/1cbb0685f6744f091ec5d507.png"},{"id":98429160,"identity":"17042d01-8755-4947-973f-a2f518b63157","added_by":"auto","created_at":"2025-12-17 16:42:54","extension":"xml","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":122465,"visible":true,"origin":"","legend":"","description":"","filename":"f6006a4e4ca642ac8a6ac29b8370c2be1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8207456/v1/839d1c16249b85bf3eccb0e2.xml"},{"id":98428984,"identity":"5b02c677-bbd6-4e8e-8a5e-338d98342a9b","added_by":"auto","created_at":"2025-12-17 16:42:38","extension":"html","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":136428,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8207456/v1/e34c683388f985be9459cec4.html"},{"id":98075062,"identity":"589c0edc-8214-4e85-b558-d647ae1fd106","added_by":"auto","created_at":"2025-12-12 13:31:06","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":36458,"visible":true,"origin":"","legend":"\u003cp\u003eThe number and type of specimens collected from patients at each AJCC stage. The red bars represent patients from whom cytology samples were collected, and the blue bars represent the patients from whom tissue samples were collected.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8207456/v1/f5ca4dd8965a3c45f48d8122.png"},{"id":98075066,"identity":"69044990-68be-4385-9b66-87a945918e9c","added_by":"auto","created_at":"2025-12-12 13:31:06","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":319592,"visible":true,"origin":"","legend":"\u003cp\u003eBox plots showing mean expression values and interquartile range for relevant single genes across the stage categories. \u003cstrong\u003ea\u003c/strong\u003e. \u003cem\u003eESR1\u003c/em\u003e, \u003cstrong\u003eb\u003c/strong\u003e. \u003cem\u003ePGR\u003c/em\u003e, \u003cstrong\u003ec\u003c/strong\u003e. \u003cem\u003eERBB2\u003c/em\u003e, \u003cstrong\u003ed\u003c/strong\u003e. \u003cem\u003eAURKA\u003c/em\u003e. Each blue dot represents a patient from whom a cytology sample was collected, and the blue line shows the average expression of that sample group at each stage. The red dots and lines represent patient-specific and mean expression levels per stage for the tissue samples, respectively. The black line connects the mean values from all samples. The red and blue lines connect the mean values for cytology and tissue samples, respectively.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8207456/v1/2333089f7e481a6a6289f9cc.png"},{"id":98075064,"identity":"84738dbf-173e-4cf5-b916-2883ccfdd9fd","added_by":"auto","created_at":"2025-12-12 13:31:06","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":558581,"visible":true,"origin":"","legend":"\u003cp\u003eBox plots showing mean expression values and interquartile range for breast cancer multigene signatures across the stage categories. \u003cstrong\u003ea\u003c/strong\u003e. Recurrence Score, \u003cstrong\u003eb\u003c/strong\u003e. ROR-S, \u003cstrong\u003ec\u003c/strong\u003e. EndoPredict, \u003cstrong\u003ed\u003c/strong\u003e. GGI, \u003cstrong\u003ee.\u003c/strong\u003e \u003cem\u003eSET\u003c/em\u003e\u003csub\u003e\u003cem\u003eER/PR \u003c/em\u003e\u003c/sub\u003eindex of endocrine activity, \u003cstrong\u003ef\u003c/strong\u003e. PI3-kinase gene expression signature. Each blue dot represents a patient from whom a cytology sample was collected, and the blue line shows the average expression of that sample group at each stage. The red dots and lines represent patient-specific and mean expression levels per stage for the tissue samples, respectively. The black line connects the mean values from all samples. The red and blue lines connect the mean values for cytology and tissue samples, respectively.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8207456/v1/90eefc9740a9cc5f1d562fb2.png"},{"id":98075068,"identity":"bc045abf-5aae-4ef6-b22a-4ebef07bc5f9","added_by":"auto","created_at":"2025-12-12 13:31:06","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":500571,"visible":true,"origin":"","legend":"\u003cp\u003eBox plots showing mean expression values and interquartile range for noteworthy signatures related to breast cancer or generic cancer biology across the stage categories. \u003cstrong\u003ea. \u003c/strong\u003ePI3K-AKT-MTOR Signaling, \u003cstrong\u003eb\u003c/strong\u003e. SRC Signature, \u003cstrong\u003ec\u003c/strong\u003e. Immune Signature, \u003cstrong\u003ed\u003c/strong\u003e. CIN70, \u003cstrong\u003ee\u003c/strong\u003e. Heme Metabolism. Each blue dot represents a patient from whom a cytology sample was collected, and the blue line shows the average expression of that sample group at each stage. The red dots and lines represent patient-specific and mean expression levels per stage for the tissue samples, respectively. The black line connects the mean values from all samples. The red and blue lines connect the mean values for cytology and tissue samples, respectively.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8207456/v1/a9adf9c92ede15acae6cf5b8.png"},{"id":98075067,"identity":"56380f8a-b318-41ce-b5fc-773a21e6d9cf","added_by":"auto","created_at":"2025-12-12 13:31:06","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":223928,"visible":true,"origin":"","legend":"\u003cp\u003eBox plots showing mean expression values and interquartile range for signatures related to breast cancer or generic cancer biology as the recurrence of disease increases. \u003cstrong\u003ea. \u003c/strong\u003eESR1 Probeset, \u003cstrong\u003eb\u003c/strong\u003e. \u003cem\u003eSET\u003c/em\u003e\u003csub\u003e\u003cem\u003eER/PR\u003c/em\u003e\u003c/sub\u003e index of endocrine activity, \u003cstrong\u003ec\u003c/strong\u003e. PI3-kinase gene expression signature, \u003cstrong\u003ed\u003c/strong\u003e. Heme Metabolism. Blue dots indicate the presence of a mutation, red dots indicate a lack of mutation, and grey dots indicate an unknown mutation status. The black line connects the mean values from all samples.\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8207456/v1/463ae3f3c7545abe53aaec50.png"},{"id":98444470,"identity":"6d6f200a-b34f-40f2-a5cc-34d28ec736d6","added_by":"auto","created_at":"2025-12-17 17:15:59","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2312256,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8207456/v1/ba968e02-646b-4cca-8ced-08991df0eeee.pdf"},{"id":98427342,"identity":"4647d11b-3039-431b-9d43-c44f3875c1a5","added_by":"auto","created_at":"2025-12-17 16:40:10","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":228837,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementStagerelatedtranscriptionalchanges20251125submit.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8207456/v1/0211a9f5dad3fa168d6ac14d.pdf"}],"financialInterests":"Competing interest reported. The authors have no competing interests to declare. Non-competing interests that might be perceived to influence the discussion in this paper were: WFS reports stock and other ownership interests in ISIS Pharmaceuticals, Delphi Diagnostics, and Eiger BioPharmaceuticals; consulting or advisory roles for AstraZeneca; research funding from Pfizer (paid to Institution); Co-inventor, US Patent No. 11,459,617 “Targeted Measure of Transcriptional Activity Related to Hormone Receptors,” issued on October 4, 2022 (applicant proprietor: University of Texas MD Anderson Cancer Center) and licensed to Delphi Diagnostics; and an uncompensated scientific advisor relationship with Delphi Diagnostics. CF reports stock in Delphi Diagnostics and co-inventor on US Patent No. 11,459,617. The other authors do not have any relevant non-competing interests.","formattedTitle":"Stage-related transcriptional changes in HR-positive/HER2-negative breast cancer","fulltext":[{"header":"INTRODUCTION","content":"\u003cp\u003eStaging of breast cancer using the American Joint Committee on Cancer (AJCC) system is an important measure of the extent and risk of the disease.\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e Consequently, staging is an essential tool used for planning the course of treatment and making adjustments based on the estimated level of risk. However, a biological assessment that reflects stage and state of disease is also critical for understanding the full spectrum of a patient\u0026rsquo;s disease and for developing a personalized therapeutic plan. For example, gene expression profiles have greatly impacted treatment approaches for patients with Stage I-IIA hormone receptor-positive, HER2-negative (HR+/HER2-) cancer.\u003csup\u003e\u003cspan additionalcitationids=\"CR3\" citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e There is also potential to apply gene expression profiles to more advanced stages of disease to guide treatment. In this context, a comparison of static and dynamic changes in gene expression profiles across the stage categories of HR+/HER2- breast cancer can provide a useful perspective on which clinically or biologically relevant profiles can differ by stage of disease in a population.\u003c/p\u003e\u003cp\u003eOur approach was to compare populations defined by stage categories to test for associations between stage of disease and expression level of single genes, multigene prognostic or predictive signatures with established relevance to breast cancer, and more generic cancer biology signatures, many of which relate to the 14 established Hallmarks of Cancer.\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e We assembled a large cohort of fresh clinical samples that were prospectively collected on research protocols and optimally preserved for high-quality RNA.\u003csup\u003e\u003cspan additionalcitationids=\"CR7 CR8\" citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e All had gene expression profiling performed using whole transcriptome microarrays.\u003c/p\u003e\u003cp\u003eAdaptation of clinical sampling methods for research sample procurement to different clinical scenarios allowed us to successfully obtain cancer samples for different research protocols. That is because fine needle aspiration (FNA) cytology is generally well accepted for research collection when there is advanced disease, or when diagnosis has already been established, and the patient will proceed with neoadjuvant treatment.\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e In those scenarios, there may be no need for another core biopsy for research. Otherwise, the prospective collection of fresh tissue samples is more acceptable when patients have been diagnosed with invasive breast cancer, and their primary treatment is surgery. Thus, fresh tumor prospective research samples may come from FNA cytology, core biopsy tissue, or fresh breast cancer tissue collected at the time of surgery, and the type of sample is influenced by the stage of the disease and clinical logistics. This also introduces a major preanalytical effect due to differences in the cellular populations within cytology and tissue samples.\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e Therefore, we formally compared matched pairs of cytology and tissue samples that were collected in a subset of patients at the time of surgery, specifically to calibrate gene expression measurements between cytology and tissue samples.\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e This comparison enabled us to account for the type of clinical sample when interpreting differences in gene expression levels as attributable to differences by stage of disease.\u003c/p\u003e\u003cp\u003eOverall, this study examines stage-associated changes in the expression of single genes and gene expression signatures of proven relevance to patients with HR+/HER2- breast cancer, along with gene expression profiles that represent important biological signatures of cancer. Notably, we studied this in high-quality samples that were collected prospectively in clinical research protocols and accounted for whether those were cytology or tissue samples. Finally, we present a novel cohort of prospectively collected, high-quality RNA expression profiles from clinical research protocols, which is a new data resource for others to study for stage-related changes in HR\u0026thinsp;+\u0026thinsp;breast cancer.\u003c/p\u003e"},{"header":"RESULTS","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003ePatient and Sample Characteristics\u003c/h2\u003e\u003cp\u003eThe study cohort consisted of 1,152 HR+/HER2- breast cancers (See REMARK diagram in Supplementary Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). There was sufficient representation of samples across five stage categories (I, IIA, IIB, III, IV), although a greater number of samples were from Stage IIA (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Tissue samples were more commonly collected from patients with lower stage disease (Stages I and IIA) and were usually from surgically resected breast cancer. Cytology samples were more frequently collected from patients with more advanced stage, particularly Stages III and IV (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The cohort included matched pairs of cytology and tissue samples collected at the time of surgery from 75 patients.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eDifferences across the Spectrum of Stage I to Stage IV\u003c/h3\u003e\n\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e shows the results from the selected transcripts and multigene signatures that had significant regression of expression values as Stage increased (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). All other results are shown in Supplementary Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eBreast Cancer-Specific Genes and Clinical Signatures of Interest\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"10\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e\u003cp\u003eSample Matched Pairs\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"6\" nameend=\"c10\" namest=\"c5\"\u003e\u003cp\u003eClinical Samples Cohort\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003ctr\u003e\u003cth align=\"left\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e\u003cp\u003eCytology vs Tissue\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"3\" nameend=\"c7\" namest=\"c5\"\u003e\u003cp\u003eCytology vs Tissue\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"3\" nameend=\"c10\" namest=\"c8\"\u003e\u003cp\u003eAJCC Stage Effect\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSignature\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eEstimate\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003e95% CI\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003ep-value\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eEstimate\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003e95% CI\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003ep-value\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c8\"\u003e\u003cp\u003eEstimate\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c9\"\u003e\u003cp\u003e95% CI\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c10\"\u003e\u003cp\u003ep-value\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"10\" nameend=\"c10\" namest=\"c1\"\u003e\u003cp\u003eSingle Genes (Probesets)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cem\u003eESR1\u003c/em\u003e Probeset\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.311\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-0.121, 0.743\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.157\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e-0.233\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e-0.415, -0.051\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.012\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e-0.121\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e-0.188, -0.055\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cem\u003ePGR\u003c/em\u003e Probeset\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.096\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-0.088, 0.280\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.302\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.025\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e-0.049, 0.098\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.512\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e-0.058\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e-0.084, -0.031\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cem\u003eERBB2\u003c/em\u003e Probeset\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.279\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.072, 0.487\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e0.009\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e-0.154\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e-0.257, -0.051\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003e0.003\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e-0.056\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e-0.094, -0.019\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e\u003cb\u003e0.003\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cem\u003eAURKA\u003c/em\u003e Probeset\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.109\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-0.041, 0.259\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.153\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.184\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.109, 0.258\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.063\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0.036, 0.090\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"10\" nameend=\"c10\" namest=\"c1\"\u003e\u003cp\u003eBreast Cancer Multigene Signatures\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eOncotypeDX\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-5.068\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-14.496, 4.360\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.290\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1.690\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e-0.958, 4.339\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.211\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e2.198\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e1.233, 3.162\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003erorS\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e5.285\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-3.262, 13.831\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.224\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e11.141\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e7.747, 14.536\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e5.102\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e3.866, 6.338\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eendoPredict\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.134\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-0.651, 0.920\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.736\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.907\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.595, 1.220\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.480\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0.366, 0.594\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGGI\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.094\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.004, 0.184\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.042\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.132\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.097, 0.167\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.033\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0.020, 0.045\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSETERPR\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.092\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-0.139, 0.322\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.434\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e-0.040\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e-0.126, 0.046\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.363\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e-0.104\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e-0.135, -0.073\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePIK3ges\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.108\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-0.130, 0.346\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.371\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e-0.176\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e-0.267, -0.085\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e-0.072\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e-0.105, -0.039\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"10\" nameend=\"c10\" namest=\"c1\"\u003e\u003cp\u003eCancer Biology Signatures\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHallmark_PI3K_AKT_MTOR_signaling\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.014\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-0.033, 0.062\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.559\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e-0.022\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e-0.044, -0.000\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.045\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.014\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0.006, 0.022\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e\u003cb\u003e0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSRC Signature\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-0.021\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-0.040, -0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.041\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e-0.022\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e-0.033, -0.012\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e-0.009\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e-0.013, -0.005\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eIMMUNE Signature\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-1458.583\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-1946.473, -970.693\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e93.247\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e-149.601, 336.095\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.452\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e146.009\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e57.573, 234.444\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e\u003cb\u003e0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCIN70 Signature\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.126\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.036, 0.216\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e0.006\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.111\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.075, 0.148\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.025\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0.012, 0.038\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHALLMARK_SPERMATOGENESIS\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.063\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.021, 0.105\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e0.003\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.182\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.162, 0.202\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.014\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0.007, 0.021\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHALLMARK_HEME_METABOLISM\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-0.034, 0.032\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.951\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.056\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.035, 0.078\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.018\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0.010, 0.026\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eA cancer biology signature was included from the 50 listed in the Broad Institute Gene Set Enrichment Analysis if its stage effect was highly significant (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). For the matched pairs cohort, this meant that cytology and tissue were significantly different. For the clinical samples cohort, this meant that the gradient of the regression line across staging was significantly up or down. The \u0026ldquo;estimate\u0026rdquo; column quantifies the difference in means between cytology and tissue for matched pair samples, with a positive value indicating that the cytology was higher than tissue. \u0026ldquo;Estimate\u0026rdquo; in the clinical cohort is the slope of the regression line with increasing stage category as an integer.\u003c/p\u003e\n\u003ch3\u003eSingle Genes That Discriminate Breast Cancer Subsets:\u003c/h3\u003e\n\u003cp\u003eThere was no significant difference in the expression of \u003cem\u003eESR1\u003c/em\u003e, \u003cem\u003ePGR\u003c/em\u003e, or \u003cem\u003eAURKA\u003c/em\u003e in the matched pairs of cytology and tissue samples (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Expression of \u003cem\u003eERBB2\u003c/em\u003e was significantly higher in cytology samples (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). We observed a small, but statistically significant change in the expression levels of \u003cem\u003eESR1\u003c/em\u003e, \u003cem\u003ePGR\u003c/em\u003e, and \u003cem\u003eERBB2\u003c/em\u003e; and a significant increase in the expression level of \u003cem\u003eAURKA\u003c/em\u003e with increasing stage category (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\n\u003ch3\u003eMultigene Signatures:\u003c/h3\u003e\n\u003cp\u003eGGI was the only multigene signature that differed by sample type in matched sample pairs, being slightly higher in cytology samples (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The expression levels of multigene prognostic signatures were significantly increased in higher stages of disease (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Specifically, Recurrence Score and GGI demonstrated consistent, incremental increases (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA, \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB), whereas ROR-S and EndoPredict showed a gradual increase from Stage I to IIB, then a greater increase from Stages IIB through IV (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC, \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD). On the other hand, there was a steady downward trend in the signature that quantifies endocrine transcriptional activity attributable to estrogen receptor and progesterone receptor, i.e., the sensitivity to endocrine therapy \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{SET}_{ER/PR}\\)\u003c/span\u003e\u003c/span\u003e index (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE). There was the same downward trend with a signature that represents gene expression related to activating mutation of the \u003cem\u003ePIK3CA\u003c/em\u003e gene in HR\u0026thinsp;+\u0026thinsp;breast cancers and represents overactivity of PI3-kinase mediated signal transduction (i.e., PIK3ges). The expression levels of PIK3ges had a relatively flat trend from Stage I to IIB (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF). The type of tumor sample contributed independently to the stage-related trends for every multigene signature, except for Recurrence Score and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{SET}_{ER/PR}\\)\u003c/span\u003e\u003c/span\u003e index (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\n\u003ch3\u003eGeneral Biological Signatures Attributable to Cancer:\u003c/h3\u003e\n\u003cp\u003eThe expression levels of multigene signatures representing generic cancer biology were mostly consistent across the stages of breast cancer. Those with a highly significant stage-related change (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) are shown in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, the others in Supplementary Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e. There was a slight increase in the expression of a cancer biology signature representing signal transduction through the PI3-kinase, AKT, and MTOR pathway (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA), and a steady decrease in the expression of a multigene signature for signal transduction via SRC (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB). The expression of a multigene signature representing immune activity within the tumor increased slightly as the stage increased (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC). A signature for chromosomal instability increased steadily from Stages I to IV (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD), as did a spermatogenesis signature representing the \u0026ldquo;stem-like\u0026rdquo; activation of cancer cells to promote proliferation (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Finally, a signature representing deregulated porphyrin-related heme metabolism remained flat from Stages I to III but increased significantly from III to IV (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eE, Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003eStage-related Differences within Stage IV Based on Progression Events\u003c/h2\u003e\u003cp\u003eFigure \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e presents the most significant changes in single genes or multigene signatures within Stage IV cancer samples categorized by when they were collected: \u003cem\u003ede novo\u003c/em\u003e, 1st or 2nd recurrence, or 3rd or later recurrence. The expression of the estrogen receptor gene was decreased in samples from later recurrence (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA), as was the expression of the \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{SET}_{ER/PR}\\)\u003c/span\u003e\u003c/span\u003e index of endocrine activity (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB). Samples with a mutation of \u003cem\u003eESR1\u003c/em\u003e were from recurrent disease appeared to have higher expression levels of \u003cem\u003eESR1\u003c/em\u003e and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{SET}_{ER/PR}\\)\u003c/span\u003e\u003c/span\u003e index than samples without an \u003cem\u003eESR1\u003c/em\u003e mutation, although high expression was also seen in some metastases that did not have \u003cem\u003eESR1\u003c/em\u003e mutation. Average \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{SET}_{ER/PR}\\)\u003c/span\u003e\u003c/span\u003e index expression declined on the first recurrence of breast cancer, then remained relatively stable in samples from later recurrences (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB). The decrease in median expression level of PIK3ges had a similar pattern to \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{SET}_{ER/PR}\\)\u003c/span\u003e\u003c/span\u003e index. There was a strong upward trend in the expression of genes representing deregulated cellular energetics through heme metabolism with increasing number of recurrence events (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eD).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e"},{"header":"DISCUSSION","content":"\u003cp\u003eIn our study, most transcriptional signatures did not change significantly, including several important biological signatures representing glycolysis, hypoxia, and apoptosis. However, certain biological themes of stage-related change were noteworthy. Generally, we observed a lower state of differentiation (including diminution of endocrine signaling), changes in signal transduction profiles, an increase in proliferation and genomic grade, increasing immune activity and instability of the genome, and some evidence of metabolic dysregulation in more advanced stages of HR+/HER2- breast cancer. A more detailed analysis of Stage IV cancers based on the number of progression events suggested a shifting away from endocrine signaling and PI3-kinase-mediated activity, with a shift towards some aspects of changing energy metabolism. Our findings point towards a modest overall change in the cellular functions within the cancer as it advances in stage. Ultimately, the treatment protocol adapts to the dynamic biological changes acquired through the course of the disease, eventually informing the potential transition to cytotoxic treatment as a late resort in an advanced HR\u0026thinsp;+\u0026thinsp;setting.\u003c/p\u003e\u003cp\u003eWe observed endocrine escape as decreasing SET\u003csub\u003eER/PR\u003c/sub\u003e index and \u003cem\u003eESR1\u003c/em\u003e with increasing stage. The former was consistent through Stages I to III, and from \u003cem\u003ede novo\u003c/em\u003e through progressive Stage IV disease, i.e., through the entire continuum of stage and progression. We are aware that a recent longitudinal study of patient-matched HR+/HER2- breast tumors reported an increase in \u003cem\u003eESR1\u003c/em\u003e gene expression from early cancer to metastatic.\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e However, this does not necessarily conflict with our findings because that study measured gene expression only in samples from patients who developed metastatic cancer, whereas most of our patients would not have recurred. The cancers most susceptible to hormonal therapy are less likely to metastasize, and vice versa, possibly reflected by a limited emergence of acquired \u003cem\u003eESR1\u003c/em\u003e mutation. This difference in population may also explain why we observed a decrease in \u003cem\u003eERBB2\u003c/em\u003e expression, while they found an increase.\u003c/p\u003e\u003cp\u003eOur study enhances the perspective of the current evolving paradigm in clinical practice for close monitoring and early detection of recurrent disease in high-risk patients. Recent trials reported that early detection of \u003cem\u003eESR1\u003c/em\u003e mutation before radiographic progression could prompt a treatment change that improved progression-free survival in patients with early stage disease.\u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e There is now a validated benefit to closely monitoring high-risk disease in early stages, guided by the emergence of \u003cem\u003eESR1\u003c/em\u003e as a targetable genetic alteration.\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e \u003cem\u003eESR1\u003c/em\u003e mutations, captured in ctDNA, are likely an adaptive consequence of secondary resistance to Aromatase Inhibitors (AI).\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e In this setting, treatment aims to address the dynamic evolution of the biological state of the disease, and longitudinal molecular characterization methods offer an opportunity for precision medicine; Selective Estrogen Receptor Degraders (SERDs) were proven more effective than the standard AI in treating HR+/HER2- breast cancer with mutated \u003cem\u003eESR1\u003c/em\u003e. Tracking \u003cem\u003eESR1\u003c/em\u003e mutation status in ctDNA is a sensitive way to monitor progression and diagnose early recurrence of endocrine-resistant disease.\u003c/p\u003e\u003cp\u003eAfter endocrine activity, the second most important driver in hormone receptor-positive breast cancer is signal transduction, particularly through the PI3-kinase pathways. \u003cem\u003ePIK3CA\u003c/em\u003e is the most commonly mutated gene in HR+/HER2- breast cancer at a rate of 40%, so characterizing its expression across the stages gives insight into a key driver of the disease.\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e We observed a seemingly discordant decrease in the multigene PI3-kinase gene expression signature with increasing stage and number of recurrences and an increase in the collection of genes representing activating mutations of the PIK3-AKT-MTOR pathway. One explanation is that signaling complexity increases with stage, such that the cancer might not be as reliant on \u003cem\u003ePIK3CA\u003c/em\u003e activity, specifically as alternative uses of this signal transduction mechanism evolve.\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e In other words, the early biology of PI3-kinase seems to be replaced by other signal transduction opportunities as the cancer progresses. However, the status of \u003cem\u003ePIK3CA\u003c/em\u003e mutation status in primary and metastatic breast cancer has been shown to be largely concordant (69%).\u003csup\u003e17\u003c/sup\u003e Another possible explanation is that an upstream activator of the pathway overpowers the declining PI3-kinase activity, thereby increasing the expression of AKT and MTOR. Or, despite diminished PI3-kinase activity, a repressor of downstream activity is removed.\u003c/p\u003e\u003cp\u003eA different measure of signal transduction in cancer changed in the opposite direction to PI3K-AKT-MTOR across the stages. SRC is an upstream activator of PI3K-AKT-MTOR signaling and other signaling pathways.\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e The balance of these two signatures suggested a subtle shift towards PI3K-AKT-MTOR signaling in advancing stages of cancer and away from the upstream SRC pathway that also promotes cell growth and survival. In addition, the expression of the \u003cem\u003eERBB2\u003c/em\u003e gene for HER2 declined with advancing stage. Overall, within these trends, it seemed that membrane signaling remained consistent across Stages I to IIB, then dropped off slightly in Stages III and IV.\u003c/p\u003e\u003cp\u003eThe presence of a \u003cem\u003ePIK3CA\u003c/em\u003e mutation is addressed in clinical studies that investigate primary resistance to endocrine therapy. Recently, Inavolisib (PI3Kα inhibitor) plus Palbociclib (CDK4/6 inhibitor) and Fulvestrant (SERD) was the first FDA-approved triplet combination treatment used as a first-line treatment for patients with \u003cem\u003ePIK3CA\u003c/em\u003e-mutated, locally advanced or metastatic HR\u0026thinsp;+\u0026thinsp;breast cancer. These patients were considered to have endocrine-resistant disease after recurrence on or after completing adjuvant endocrine therapy.\u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e Additionally, the CAPItello-291 trial investigated the combination of Capivasertib (AKT inhibitor) and Fulvestrant in a population that progressed after AI treatment with or without CDK4/6 inhibitor.\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e They demonstrated a progression-free survival benefit independent of both the CDK4/6 inhibitor exposure and the detection of AKT pathway alteration.\u003c/p\u003e\u003cp\u003eWhen HR+/HER2- breast cancer does progress, it seemingly becomes less differentiated. Overall, we observed that the multigene expression signatures that represent lesser differentiation were increased with advancing stage. With ROR-S and EndoPredict, but not Recurrence Score, this change was most pronounced at later stages, which fits into the concept that the expression of genes associated with a diminished state of differentiation should reflect more aggressive biology in the cancer.\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e Expression profiles representing proliferation had a more complex association with stage. Aurora kinase A expression increased, matching our expectations that more advanced cancer tends to have a higher level of proliferative signaling because it is innately more aggressive or has progressed to a more aggressive, genetically unstable state.\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e Another marker of proliferation, the \u0026ldquo;hallmark\u0026rdquo; of spermatogenesis, representing the \u0026ldquo;stem-like\u0026rdquo; activation of cancer cells to promote proliferation,\u003csup\u003e22\u003c/sup\u003e showed an upward trend that was slightly exaggerated by the difference in sample type. One interpretation might be that proliferation was more likely to be driven by alternative growth factor pathways.\u003c/p\u003e\u003cp\u003eOur findings are relevant to the current approach to management of metastatic and other locally advanced breast cancers. As blockers of proliferation, CDK4/6 inhibitors, in addition to endocrine therapy with AI or Fulvestrant, represent the current standard of care as a first-line therapy in the metastatic setting. However, the lack of a specific biomarker for treatment selection and the toxic effect of their extended use remain challenges in clinical practice, especially in their integration into adjuvant treatment. Maintenance of CDK4/6 inhibitors throughout progression in order to overcome endocrine treatment resistance represents an emerging practice, as there is a reported benefit for disease control.\u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e,\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e Diminishing endocrine sensitivity necessitates the adjustment of the endocrine backbone while continuing with CDK4/6 inhibitors. However, there was conflicting data on the efficacy and predictive value of an \u003cem\u003eESR1\u003c/em\u003e mutation alone and in combination with other known somatic gene alterations associated with CDK4/6 inhibitor exposure in this therapeutic strategy.\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e Despite new endocrine and targeted treatment options following CDK4/6 inhibitor usage, beyond first-line, chemotherapy remains a valid choice, especially for clinically-aggressive disease without any druggable target.\u003c/p\u003e\u003cp\u003eWhen looking at the broader tumor microenvironment, we observed that an immune signature representing overall immune presence in these immunologically cold HR+/HER2- tumors increased with advancing stage.\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e This conflicts with the finding that metastatic lesions demonstrate low immunogenicity because they include a larger representation of immune-avoidant clones; however, it should be noted that most of our patients were not from Stage IV.\u003csup\u003e26\u003c/sup\u003e A growing, infiltrating, sustained cancer will evade anti-cancer immune signaling and activation.\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e Although statistically significant, the increase in immune cell gene expression with advancing stage was subtle, with little change from Stage III to IV. Therefore, the association that is observed could be due to less-differentiated tumors generally having slightly more cellular immune infiltration. Nevertheless, immune checkpoint inhibitors have not demonstrated efficacy in an HR\u0026thinsp;+\u0026thinsp;setting.\u003c/p\u003e\u003cp\u003eAt the chromosomal level, genomic instability signaling increased across stages, perhaps because mutations were more prevalent in higher-stage disease. This fits with an understanding of how cancer reflects an ever-destabilizing genome.\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e However, it is important to mention that we do not have DNA sequence or cytogenetic data, so we can only extrapolate about this possible effect.\u003c/p\u003e\u003cp\u003eMetabolic dysregulation is another important factor that drives changes. As cancer progresses, its energetic demands also increase.\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e The cancer biology signature heme metabolism, which describes the metabolic shift towards the overproduction of heme intermediates like porphyrin for sustaining the cancer,\u003csup\u003e29\u003c/sup\u003e was significantly higher in Stage IV in our study. It was also the only so-called \u0026ldquo;hallmark\u0026rdquo; transcriptional signature to progressively rise within Stage IV cancers based on the number of progression events. In cancer, there is a non-homeostatic upregulation of heme intermediates, called porphyrins, to fuel the cancer in a process called porphyrin overdrive\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e Porphyrin overdrive stands out from other metabolic phenotypes like the Warburg Effect, glutamine addiction, and increased fatty acid anabolism because it represents the cancer-specific accumulation of heme intermediates.\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e Cancer cells rely on these intermediates for essential cellular processes, such as proliferation and the protection against iron-induced ferroptosis.\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e Imbalanced heme metabolism is a targetable metabolic phenotype because it is not present in healthy cells and is essential to the survival of the cancer\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e, so our finding invites further study for treatment options to exploit this phenotype. The site of metastasis did not appear to be associated with the level of heme metabolism gene expression signature in the biopsy samples (Supplementary Figure S2). Other signatures that represented metabolism, but did not change significantly, were fatty acid metabolism, glycolysis, and oxidative phosphorylation (Table S2). However, other important metabolic phenotypes of cancer are not well represented by gene expression signatures, so we cannot speculate on their stage-related trends. The glycolytic state did not vary in mRNA expression, but this is not the most suitable measure anyway; metabolism occurs at the protein and biochemical level, so DNA or RNA expression profiles are only remotely related and cannot definitively measure this. Indeed, our samples could not take into account downstream enzymatic activity, so it is possible that even though we see changes at the transcriptional level, enzymatic activity may not change. Glycolytic state is a very important parameter of treatment resistance, even relevant to immunotherapy. A highly glycolytic state creates an unfavorable tumor microenvironment because it increases acidity, which in turn increases the activity of immune-suppressive T-cells.\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e This concept invites new clinical strategies to select patients who may particularly benefit from immunotherapy.\u003c/p\u003e\u003cp\u003eOverall, it is important to note that some signatures in our study contribute to the development of cancer, while others promote the progression of cancer to higher stages. Similarly, for some signatures, the differences in expression between cytology and tissue can be explained by the natural features of those signatures. For example, we would expect that CIN70 and spermatogenesis showed higher expression in the cytology at every stage because the chromosomal instability measure would be best captured within the more concentrated overall cellularity of an FNA, rather than the tissue sample, which would include stroma too. There was a similar confounding effect of sample type with tissue, as it has mesenchymal stroma, whereas cytology has very little,\u003csup\u003e10\u003c/sup\u003e so signatures like angiogenesis and epithelial-mesenchymal transition were much higher in tissue than in cytology.\u003c/p\u003e\u003cp\u003eHere, we would like to address the limitations of our study. To start, our study, taking place over many years, used a mature data set from samples that were freshly collected into RNA later and evaluated during that time by Affymetrix gene expression microarray technology, which predates contemporary RNA sequencing. Still, these signatures of interest are published as accurate when using these microarrays, and thus, we can make reasonable inferences.\u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e We also acknowledge that these were not longitudinal samples capturing progression within a patient\u0026rsquo;s cancer timeline. Rather, each one came from a different patient. Studies of paired primary and subsequent metastasis provide more information about the individual\u0026rsquo;s molecular progression, but do not address the diversity across stages of primary disease.\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e Another limitation is that when assessing gene expression as a function of stage, as expected, there were fewer cytology samples from early stages and fewer tissue samples from later stages. This sample bias was a result of the context and method in which prospective samples can reasonably be obtained in prospective research protocols. Nevertheless, we feel their exact number and placement at these \u0026ldquo;boundary stages\u0026rdquo; is less important than the overall message of the trend (i.e., whether they go up or down or remain flat, when they change, and how quickly). Additionally, there were very few 1st-recurrence patients in our Stage IV-specific trends, because it is a difficult time to discuss with the patient collecting optional research samples when the diagnosis has not been established at the time of biopsy. In that clinical setting, we would need to have relied on archival fixed biopsy samples and those should be prioritized for clinical molecular testing. On the other hand, an advantage of our methodology was that our samples were all from fresh samples and unaffected by preanalytical effects of cold ischemia, fixation or storage (Sura et al., 2025).\u003csup\u003e\u003cspan additionalcitationids=\"CR35\" citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e Moreover, for our Stage IV analysis, we were not able to take into account how long the interval was from prior treatment before they had the biopsy, for example, whether it was an early or late recurrence. It would also have been helpful to know the \u003cem\u003ePIK3CA\u003c/em\u003e mutation status when evaluating the related gene expression signatures in the samples from metastases. The gene expression signatures and the cancer biology signatures are not validated diagnostic signatures, so technical variance may be an issue. Also, they are approximations of biology, not the same as studying biological function. This is true for most of the gene expression signatures that we analyzed. Still, our results provide insights that would guide more specific research.\u003c/p\u003e\u003cp\u003eIn conclusion, our study summarizes the most significant transcriptional changes from high-quality clinical research biopsies that are related to advancing stage across a large population of patients with HR+/HER2- breast cancer. We evaluated their biological associations with contemporary treatments for this disease and identified a few novel transcriptional changes with potential for novel therapeutic approaches.\u003c/p\u003e"},{"header":"METHODS","content":"\u003cp\u003eThe cohort for these analyses were from the compilation of gene expression data from past studies wherein fresh tumor samples of HR+/HER2- breast cancer were profiled on U133A microarrays using uniform methodology. All patients gave informed written consent to take part in the study and for the use of tissue material for research purposes. Definition of HR+/HER2- breast cancer was based on the results of clinical biomarker testing, as reported at the institution where the sample was collected. HR-positive status was defined as \u0026gt;\u0026thinsp;1% tumor nuclei stained positive for ER or PR by immunohistochemistry.\u003c/p\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003eSample cohorts\u003c/h2\u003e\u003cp\u003eSamples from primary surgical resection specimens at MDACC were collected fresh at time of intraoperative specimen assessment by a dedicated breast pathologist (WFS) under research IRB protocols with MD Anderson Institutional Review Board (IRB) approval: LAB04-0093, LAB08-0823, LAB08-0824, and 2011-0007. A shave of the tissue from one side of the cut tumor mass was assessed by imprint cytology (to confirm malignancy) and diced in a petri dish and stored frozen at -80\u0026deg;C in \u003cem\u003eRNAlater\u003c/em\u003e until use. A subset of fresh tumor samples collected at MDACC were included in the cohort with matched cytology and tissue samples.\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e For those, a cytologic sample was collected before the tissue sample. The freshly cut surface of the tumor was gently scraped using a small scalpel blade, a portion of that liquid sample was placed on a glass slide for assessment of the cytologic smear (to confirm malignancy) and the remainder was mixed into \u003cem\u003eRNAlater\u003c/em\u003e solution. This was repeated 2\u0026ndash;3 times with the liquid sample entirely added to the \u003cem\u003eRNAlater\u003c/em\u003e solution. Immediately thereafter, the subjacent tumor tissue was shaved, diced, and added to a separate vial of \u003cem\u003eRNAlater\u003c/em\u003e solution, as described above. Frozen tumor samples from collaborating institutions were stored frozen at -80\u0026deg;C and shipped on dry ice, as previously reported.\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003cp\u003eSamples from needle biopsies, either fine needle aspiration cytology or core biopsy tissue, were prospectively collected and immediately placed in \u003cem\u003eRNAlater\u003c/em\u003e solution and held at room temperature at least 30 minutes to allow penetration of the preservative, then stored frozen at -80\u0026deg;C in \u003cem\u003eRNAlater\u003c/em\u003e until use, as previously described (approved IRB protocols LAB99-402, LAB04-0093, 2011-0007).\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e,\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e We categorized AJCC Stage into groups of Stage I, IIA, IIB, III and IV. If the sample was obtained from Stage I-III breast cancer prior to neoadjuvant treatment, then we used clinical Stage, otherwise we used pathologic Stage if surgery was the primary treatment. For patients with documented Stage IV (i.e., documented distant metastasis), we further classified into \u003cem\u003ede novo\u003c/em\u003e (i.e., Stage IV at first presentation), 1st or 2nd metastatic relapse, or 3rd or later metastatic relapse.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003eGene expression profiling for target and reference transcripts\u003c/h2\u003e\u003cp\u003eRNA was extracted, processed and hybridized to Affymetrix human genome U133A microarrays (U133A GeneChip, Affymetrix, Santa Clara, CA, USA) as described previously. In brief, the raw intensity files were processed using the MAS5.0 algorithm to generate probe set-level intensities, normalized to a median array intensity of 600, log2-transformed.\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e,\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003eSelection of genes and signatures\u003c/h2\u003e\u003cp\u003eRNA was purified and hybridized to Affymetrix U133A microarrays (U133A GeneChip; Affymetrix, Santa Clara, CA) as reported previously.\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e,\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e Briefly, raw data was processed using the MAS5 algorithm as implemented in the affy3 package to generate probe\u0026mdash;level intensities, scaled to a median intensity of 600 and log2\u0026mdash;transformed. The data was then scaled to 1,322 breast cancer reference genes within each sample following MAS5 normalization.\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e,\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003cp\u003eAs single genes we selected the four most relevant genes for classification of breast cancer, including the transcripts for estrogen receptor, progesterone receptor, and HER2 that represent the standard molecular therapeutic targets in breast cancer.\u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e Research versions of the EndoPredict and Oncotype DX signatures, of the PAM50 and SCMGENE classifiers, the GGI and PIK3CA signatures and gene modules were calculated using the genefu5 package.\u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e The SET Index was calculated according to the original publication.\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e We obtained the cancer \u0026ldquo;Hallmarks\u0026rdquo; signatures from the Gene Set Enrichment Analysis Human MSigDB Collection.\u003c/p\u003e\u003cp\u003eIn the matched pairs, the tissue samples were profiled at three different laboratories and the average of those triplicate results was used to compare with the cytology profile that was performed at MDACC, as previously reported.\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003eStatistical methods\u003c/h2\u003e\u003cp\u003ePearson\u0026rsquo;s correlation coefficient was used to compare cross-tissue concordance correlation coefficient (CCC) between the matched pairs of tissue and cytology samples for each transcript and signature. To evaluate the effect of disease stage on the genomic signatures, we fit linear regression models fitted to median values (MASS package7) and adjusted for the type of sample (cytology vs. tissue) as a covariate. P-values were obtained by comparing the gradient of the regression line with zero (t-test). The same method was used to evaluate linear regression according to the number of recurrence events in the Stage IV cohort. All statistical analyses and computations were performed in R v. 4.5.1 and Bioconductor.\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e,\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eData sharing statement.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe microarray and accompanying data are being uploaded to NCBI GEO prior to publication.\u003c/p\u003e\n\u003cp\u003eClinical trial number: not applicable\u003c/p\u003e\n\u003cp\u003eAUTHOR CONTRIBUTIONS\u003c/p\u003e\n\u003cp\u003eLP, KT, RS, VV, WFS and EA contributed to the conception and design of the work; CF, HS, RG, LP, TK, RS, VV, WFS and EA contributed to the acquisition of samples and data; WSOS, KT, and WFS contributed to the analysis; WSOS, KT, WFS and EA contributed to the \u0026nbsp;interpretation of data; WSOS, KT, WFS and EA drafted the work, and all authors reviewed the manuscript.\u003c/p\u003e\n\u003cp\u003eACKNOWLEDGEMENTS\u003c/p\u003e\n\u003cp\u003eNone.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFUNDING\u003c/p\u003e\n\u003cp\u003eThis work was supported by the following research grants: Breast Cancer Research Foundation, BCRF-158 (WFS); Susan G. Komen Foundation, SAC110034 (WFS); National Cancer Institute, HHSN261200800001E (WFS); Cancer Prevention and Research Institute of Texas, RP180712 (WFS).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eCONFLICT OF INTEREST\u003c/p\u003e\n\u003cp\u003eThe authors have no competing interests to declare. Non-competing interests that might be perceived to influence the discussion in this paper were: WFS reports stock and other ownership interests in ISIS Pharmaceuticals, Delphi Diagnostics, and Eiger BioPharmaceuticals; consulting or advisory roles for AstraZeneca; research funding from Pfizer (paid to Institution); Co-inventor, US Patent No. 11,459,617 “Targeted Measure of Transcriptional Activity Related to Hormone Receptors,” issued on October 4, 2022 (applicant proprietor: University of Texas MD Anderson Cancer Center) and licensed to Delphi Diagnostics; and an uncompensated scientific advisor relationship with Delphi Diagnostics. CF reports stock in Delphi Diagnostics and co-inventor on US Patent No. 11,459,617. The other authors do not have any relevant non-competing interests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eGiuliano, A. E.\u003cem\u003e et al.\u003c/em\u003e Breast Cancer-Major changes in the American Joint Committee on Cancer eighth edition cancer staging manual. \u003cem\u003eCA Cancer J Clin\u003c/em\u003e \u003cstrong\u003e67\u003c/strong\u003e, 290-303 (2017). https://doi.org/10.3322/caac.21393\u003c/li\u003e\n\u003cli\u003eSparano, J. A.\u003cem\u003e et al.\u003c/em\u003e Adjuvant Chemotherapy Guided by a 21-Gene Expression Assay in Breast Cancer. \u003cem\u003eN Engl J Med\u003c/em\u003e \u003cstrong\u003e379\u003c/strong\u003e, 111-121 (2018). https://doi.org/10.1056/NEJMoa1804710\u003c/li\u003e\n\u003cli\u003eKalinsky, K.\u003cem\u003e et al.\u003c/em\u003e 21-Gene Assay to Inform Chemotherapy Benefit in Node-Positive Breast Cancer. \u003cem\u003eN Engl J Med\u003c/em\u003e \u003cstrong\u003e385\u003c/strong\u003e, 2336-2347 (2021). https://doi.org/10.1056/NEJMoa2108873\u003c/li\u003e\n\u003cli\u003eCardoso, F.\u003cem\u003e et al.\u003c/em\u003e 70-Gene Signature as an Aid to Treatment Decisions in Early-Stage Breast Cancer. \u003cem\u003eN Engl J Med\u003c/em\u003e \u003cstrong\u003e375\u003c/strong\u003e, 717-729 (2016). https://doi.org/10.1056/NEJMoa1602253\u003c/li\u003e\n\u003cli\u003eHanahan, D. Hallmarks of Cancer: New Dimensions. \u003cem\u003eCancer Discov\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 31-46 (2022). https://doi.org/10.1158/2159-8290.CD-21-1059\u003c/li\u003e\n\u003cli\u003eHess, K. R.\u003cem\u003e et al.\u003c/em\u003e Pharmacogenomic predictor of sensitivity to preoperative chemotherapy with paclitaxel and fluorouracil, doxorubicin, and cyclophosphamide in breast cancer. \u003cem\u003eJ Clin Oncol\u003c/em\u003e \u003cstrong\u003e24\u003c/strong\u003e, 4236-4244 (2006). https://doi.org/10.1200/JCO.2006.05.6861\u003c/li\u003e\n\u003cli\u003eSymmans, W. F.\u003cem\u003e et al.\u003c/em\u003e Genomic index of sensitivity to endocrine therapy for breast cancer. \u003cem\u003eJ Clin Oncol\u003c/em\u003e \u003cstrong\u003e28\u003c/strong\u003e, 4111-4119 (2010). https://doi.org/10.1200/JCO.2010.28.4273\u003c/li\u003e\n\u003cli\u003eSinn, B. V.\u003cem\u003e et al.\u003c/em\u003e SET(ER/PR): a robust 18-gene predictor for sensitivity to endocrine therapy for metastatic breast cancer. \u003cem\u003eNPJ Breast Cancer\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 16 (2019). https://doi.org/10.1038/s41523-019-0111-0\u003c/li\u003e\n\u003cli\u003eHatzis, C.\u003cem\u003e et al.\u003c/em\u003e A genomic predictor of response and survival following taxane-anthracycline chemotherapy for invasive breast cancer. \u003cem\u003eJAMA\u003c/em\u003e \u003cstrong\u003e305\u003c/strong\u003e, 1873-1881 (2011). https://doi.org/10.1001/jama.2011.593\u003c/li\u003e\n\u003cli\u003eSymmans, W. F.\u003cem\u003e et al.\u003c/em\u003e Total RNA yield and microarray gene expression profiles from fine-needle aspiration biopsy and core-needle biopsy samples of breast carcinoma. \u003cem\u003eCancer\u003c/em\u003e \u003cstrong\u003e97\u003c/strong\u003e, 2960-2971 (2003). https://doi.org/10.1002/cncr.11435\u003c/li\u003e\n\u003cli\u003eAftimos, P.\u003cem\u003e et al.\u003c/em\u003e Genomic and Transcriptomic Analyses of Breast Cancer Primaries and Matched Metastases in AURORA, the Breast International Group (BIG) Molecular Screening Initiative. \u003cem\u003eCancer Discov\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 2796-2811 (2021). https://doi.org/10.1158/2159-8290.CD-20-1647\u003c/li\u003e\n\u003cli\u003eBidard, F. C.\u003cem\u003e et al.\u003c/em\u003e First-Line Camizestrant for Emerging ESR1-Mutated Advanced Breast Cancer. \u003cem\u003eN Engl J Med\u003c/em\u003e \u003cstrong\u003e393\u003c/strong\u003e, 569-580 (2025). https://doi.org/10.1056/NEJMoa2502929\u003c/li\u003e\n\u003cli\u003eBidard, F.-C.\u003cem\u003e et al.\u003c/em\u003e Switch to fulvestrant and palbociclib versus no switch in advanced breast cancer with rising \u0026lt;em\u0026gt;ESR1\u0026lt;/em\u0026gt; mutation during aromatase inhibitor and palbociclib therapy (PADA-1): a randomised, open-label, multicentre, phase 3 trial. \u003cem\u003eThe Lancet Oncology\u003c/em\u003e \u003cstrong\u003e23\u003c/strong\u003e, 1367-1377 (2022). https://doi.org/10.1016/S1470-2045(22)00555-1\u003c/li\u003e\n\u003cli\u003eBidard, F. C.\u003cem\u003e et al.\u003c/em\u003e Elacestrant (oral selective estrogen receptor degrader) Versus Standard Endocrine Therapy for Estrogen Receptor-Positive, Human Epidermal Growth Factor Receptor 2-Negative Advanced Breast Cancer: Results From the Randomized Phase III EMERALD Trial. \u003cem\u003eJ Clin Oncol\u003c/em\u003e \u003cstrong\u003e40\u003c/strong\u003e, 3246-3256 (2022). https://doi.org/10.1200/JCO.22.00338\u003c/li\u003e\n\u003cli\u003eKoboldt, D. C.\u003cem\u003e et al.\u003c/em\u003e Comprehensive molecular portraits of human breast tumours. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e490\u003c/strong\u003e, 61-70 (2012). https://doi.org/10.1038/nature11412\u003c/li\u003e\n\u003cli\u003eSever, R. \u0026amp; Brugge, J. S. Signal transduction in cancer. \u003cem\u003eCold Spring Harb Perspect Med\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e (2015). https://doi.org/10.1101/cshperspect.a006098\u003c/li\u003e\n\u003cli\u003evan Geelen, C. T.\u003cem\u003e et al.\u003c/em\u003e Clinical implications of prospective genomic profiling of metastatic breast cancer patients. \u003cem\u003eBreast Cancer Res\u003c/em\u003e \u003cstrong\u003e22\u003c/strong\u003e, 91 (2020). https://doi.org/10.1186/s13058-020-01328-0\u003c/li\u003e\n\u003cli\u003eBeadnell, T. C.\u003cem\u003e et al.\u003c/em\u003e Src-mediated regulation of the PI3K pathway in advanced papillary and anaplastic thyroid cancer. \u003cem\u003eOncogenesis\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 23 (2018). https://doi.org/10.1038/s41389-017-0015-5\u003c/li\u003e\n\u003cli\u003eWedam, S.\u003cem\u003e et al.\u003c/em\u003e US Food and Drug Administration Approval Summary: Inavolisib With Palbociclib and Fulvestrant for Endocrine-Resistant, PIK3CA-Mutated, Hormone Receptor-Positive, Human Epidermal Growth Factor Receptor 2-Negative, Locally Advanced or Metastatic Breast Cancer. \u003cem\u003eJ Clin Oncol\u003c/em\u003e \u003cstrong\u003e43\u003c/strong\u003e, 3123-3131 (2025). https://doi.org/10.1200/JCO-25-00663\u003c/li\u003e\n\u003cli\u003eTurner, N. C.\u003cem\u003e et al.\u003c/em\u003e Capivasertib in Hormone Receptor-Positive Advanced Breast Cancer. \u003cem\u003eN Engl J Med\u003c/em\u003e \u003cstrong\u003e388\u003c/strong\u003e, 2058-2070 (2023). https://doi.org/10.1056/NEJMoa2214131\u003c/li\u003e\n\u003cli\u003eJ\u0026ouml;gi, A., Vaapil, M., Johansson, M. \u0026amp; P\u0026aring;hlman, S. Cancer cell differentiation heterogeneity and aggressive behavior in solid tumors. \u003cem\u003eUps J Med Sci\u003c/em\u003e \u003cstrong\u003e117\u003c/strong\u003e, 217-224 (2012). https://doi.org/10.3109/03009734.2012.659294\u003c/li\u003e\n\u003cli\u003eLiberzon, A.\u003cem\u003e et al.\u003c/em\u003e The Molecular Signatures Database (MSigDB) hallmark gene set collection. \u003cem\u003eCell Syst\u003c/em\u003e \u003cstrong\u003e1\u003c/strong\u003e, 417-425 (2015). https://doi.org/10.1016/j.cels.2015.12.004\u003c/li\u003e\n\u003cli\u003eGerratana, L.\u003cem\u003e et al.\u003c/em\u003e Circulating Tumor Cell Dynamics after CDK4/6 Inhibitor for Hormone Receptor-Positive Metastatic Breast Cancer: A Biomarker Analysis from the PACE Phase II Study. \u003cem\u003eClin Cancer Res\u003c/em\u003e \u003cstrong\u003e31\u003c/strong\u003e, 4510-4517 (2025). https://doi.org/10.1158/1078-0432.CCR-25-0327\u003c/li\u003e\n\u003cli\u003eKalinsky, K.\u003cem\u003e et al.\u003c/em\u003e Randomized Phase II Trial of Endocrine Therapy With or Without Ribociclib After Progression on Cyclin-Dependent Kinase 4/6 Inhibition in Hormone Receptor-Positive, Human Epidermal Growth Factor Receptor 2-Negative Metastatic Breast Cancer: MAINTAIN Trial. \u003cem\u003eJ Clin Oncol\u003c/em\u003e \u003cstrong\u003e41\u003c/strong\u003e, 4004-4013 (2023). https://doi.org/10.1200/JCO.22.02392\u003c/li\u003e\n\u003cli\u003eMayer, E.\u003cem\u003e et al.\u003c/em\u003e Abstract GS3-06: GS3-06 Palbociclib After CDK4/6i and Endocrine Therapy (PACE): A Randomized Phase II Study of Fulvestrant, Palbociclib, and Avelumab for Endocrine Pre-treated ER+/HER2- Metastatic Breast Cancer. \u003cem\u003eCancer Research\u003c/em\u003e (2023). https://doi.org/10.1158/1538-7445.sabcs22-gs3-06\u003c/li\u003e\n\u003cli\u003eEsteva, F. J., Hubbard-Lucey, V. M., Tang, J. \u0026amp; Pusztai, L. Immunotherapy and targeted therapy combinations in metastatic breast cancer. \u003cem\u003eLancet Oncol\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, e175-e186 (2019). https://doi.org/10.1016/S1470-2045(19)30026-9\u003c/li\u003e\n\u003cli\u003eKim, S. K. \u0026amp; Cho, S. W. The Evasion Mechanisms of Cancer Immunity and Drug Intervention in the Tumor Microenvironment. \u003cem\u003eFront Pharmacol\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 868695 (2022). https://doi.org/10.3389/fphar.2022.868695\u003c/li\u003e\n\u003cli\u003eYates, L. R.\u003cem\u003e et al.\u003c/em\u003e Genomic Evolution of Breast Cancer Metastasis and Relapse. \u003cem\u003eCancer Cell\u003c/em\u003e \u003cstrong\u003e32\u003c/strong\u003e, 169-184 e167 (2017). https://doi.org/10.1016/j.ccell.2017.07.005\u003c/li\u003e\n\u003cli\u003eAdapa, S. R.\u003cem\u003e et al.\u003c/em\u003e Porphyrin overdrive rewires cancer cell metabolism. \u003cem\u003eLife Sci Alliance\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e (2024). https://doi.org/10.26508/lsa.202302547\u003c/li\u003e\n\u003cli\u003eLei, G., Zhuang, L. \u0026amp; Gan, B. Targeting ferroptosis as a vulnerability in cancer. \u003cem\u003eNat Rev Cancer\u003c/em\u003e \u003cstrong\u003e22\u003c/strong\u003e, 381-396 (2022). https://doi.org/10.1038/s41568-022-00459-0\u003c/li\u003e\n\u003cli\u003eZappasodi, R.\u003cem\u003e et al.\u003c/em\u003e CTLA-4 blockade drives loss of Treg stability in glycolysis-low tumours. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e591\u003c/strong\u003e, 652-658 (2021). https://doi.org/10.1038/s41586-021-03326-4\u003c/li\u003e\n\u003cli\u003eGendoo, D. M.\u003cem\u003e et al.\u003c/em\u003e Genefu: an R/Bioconductor package for computation of gene expression-based signatures in breast cancer. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cstrong\u003e32\u003c/strong\u003e, 1097-1099 (2016). https://doi.org/10.1093/bioinformatics/btv693\u003c/li\u003e\n\u003cli\u003eGarcia-Recio, S.\u003cem\u003e et al.\u003c/em\u003e Multiomics in primary and metastatic breast tumors from the AURORA US network finds microenvironment and epigenetic drivers of metastasis. \u003cem\u003eNat Cancer\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 128-147 (2023). https://doi.org/10.1038/s43018-022-00491-x\u003c/li\u003e\n\u003cli\u003eHatzis, C.\u003cem\u003e et al.\u003c/em\u003e Effects of tissue handling on RNA integrity and microarray measurements from resected breast cancers. \u003cem\u003eJ Natl Cancer Inst\u003c/em\u003e \u003cstrong\u003e103\u003c/strong\u003e, 1871-1883 (2011). https://doi.org/10.1093/jnci/djr438\u003c/li\u003e\n\u003cli\u003eLi, J., Fu, C., Speed, T. P., Wang, W. \u0026amp; Symmans, W. F. Accurate RNA Sequencing From Formalin-Fixed Cancer Tissue To Represent High-Quality Transcriptome From Frozen Tissue. \u003cem\u003eJCO Precis Oncol\u003c/em\u003e \u003cstrong\u003e2018\u003c/strong\u003e (2018). https://doi.org/10.1200/PO.17.00091\u003c/li\u003e\n\u003cli\u003eMarczyk, M.\u003cem\u003e et al.\u003c/em\u003e Assessment of stained direct cytology smears of breast cancer for whole transcriptome and targeted messenger RNA sequencing. \u003cem\u003eCancer Cytopathol\u003c/em\u003e \u003cstrong\u003e131\u003c/strong\u003e, 289-299 (2023). https://doi.org/10.1002/cncy.22679\u003c/li\u003e\n\u003cli\u003eHaibe-Kains, B.\u003cem\u003e et al.\u003c/em\u003e A three-gene model to robustly identify breast cancer molecular subtypes. \u003cem\u003eJ Natl Cancer Inst\u003c/em\u003e \u003cstrong\u003e104\u003c/strong\u003e, 311-325 (2012). https://doi.org/10.1093/jnci/djr545\u003c/li\u003e\n\u003cli\u003eTeam, R. C. \u003cem\u003eR: A language and environment for statistical computing\u003c/em\u003e, \u0026lt;http://www.R-project.org/\u0026gt; (2015).\u003c/li\u003e\n\u003cli\u003eHuber, W.\u003cem\u003e et al.\u003c/em\u003e Orchestrating high-throughput genomic analysis with Bioconductor. \u003cem\u003eNat Methods\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 115-121 (2015). https://doi.org/10.1038/nmeth.3252\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"npj-breast-cancer","isNatureJournal":false,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"npjbcancer","sideBox":"Learn more about [npj Breast Cancer](http://www.nature.com/npjbcancer/)","snPcode":"41523","submissionUrl":"https://mts-npjbcancer.nature.com/","title":"npj Breast Cancer","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"NPJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-8207456/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8207456/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eIt is important to understand how the biology of breast cancer might differ with the burden of disease. Analyzing the expression of cancer-relevant genes with increasing diagnostic stage can provide insight into the changing biology of breast cancer and may also be clinically useful. Here, we evaluated the transcriptional activity according to stage of breast cancer (American Joint Committee on Cancer, AJCC Stage I to IV) from 1,152 patients with hormone receptor-positive HER2-negative breast cancer. Fresh tumor samples were prospectively collected for clinical-translational research with transcriptional profiling. Most transcriptional signatures remained consistent across stage categories, but the significant changes associated with increasing stage were attributed to endocrine escape, decreasing differentiation, increased complexity of signal transduction, and a few metabolic changes. The clinical relevance and implications of these findings are discussed. The expression data from these high-quality samples are a new publicly available resource for researchers.\u003c/p\u003e","manuscriptTitle":"Stage-related transcriptional changes in HR-positive/HER2-negative breast cancer","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-12-12 13:31:01","doi":"10.21203/rs.3.rs-8207456/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-05-13T20:16:25+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-01-25T17:35:10+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"116993993276043405415073541367789416338","date":"2025-12-11T16:43:24+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"278837449381608976570388298515940361699","date":"2025-12-11T13:16:28+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"238188617862434012817527371561906280843","date":"2025-12-10T12:11:31+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-12-09T11:16:30+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-11-30T16:57:04+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-11-29T05:07:52+00:00","index":"","fulltext":""},{"type":"submitted","content":"npj Breast Cancer","date":"2025-11-26T01:28:05+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"npj-breast-cancer","isNatureJournal":false,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"npjbcancer","sideBox":"Learn more about [npj Breast Cancer](http://www.nature.com/npjbcancer/)","snPcode":"41523","submissionUrl":"https://mts-npjbcancer.nature.com/","title":"npj Breast Cancer","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"NPJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"bdfc1cb1-c8f4-40d0-a8b4-bfcc3d6e2c3f","owner":[],"postedDate":"December 12th, 2025","published":true,"recentEditorialEvents":[{"type":"decision","content":"Accepted","date":"2026-05-13T20:20:02+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-13T20:16:25+00:00","index":47,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":59510828,"name":"Health sciences/Biomarkers"},{"id":59510829,"name":"Biological sciences/Cancer"},{"id":59510830,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":59510831,"name":"Health sciences/Oncology"}],"tags":[],"updatedAt":"2026-05-14T09:04:35+00:00","versionOfRecord":[],"versionCreatedAt":"2025-12-12 13:31:01","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8207456","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8207456","identity":"rs-8207456","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}