Quantifying intratumoral heterogeneity within sub-regions to predict high-grade patterns in clinical stage I solid lung adenocarcinoma

Quantifying intratumoral heterogeneity within sub-regions to predict high-grade patterns in clinical stage I solid lung adenocarcinoma | 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 Research Article Quantifying intratumoral heterogeneity within sub-regions to predict high-grade patterns in clinical stage I solid lung adenocarcinoma Zhichao Zuo, Jinqiu Deng, Wu Ge, Yinjun Zhou, Haibo Liu, Wei Zhang, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3910257/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 09 Jan, 2025 Read the published version in BMC Cancer → Version 1 posted 4 You are reading this latest preprint version Abstract Background: This study aims to quantify intratumoral heterogeneity(ITH) using preoperative CT scans and evaluate its ability to predict pathological high-grade patterns, specifically micropapillary and/or solid components (MP/S), in patients diagnosed with clinical stage I solid lung adenocarcinoma (LADC). Methods: In this retrospective study, 457 patients postoperatively diagnosed with clinical stage I solid LADC were included from two medical centers, comprising a training set (center 1, n=304) and a test set (center 2, n=153). Sub-regions within the tumor were identified using the K-means method. Both intratumoral ecological diversity features (hereafter referred to as ITH) and conventional radiomics (hereafter referred to as C-radiomics) were extracted to generate ITH scores and C-radiomics scores. Next, univariate and multivariate logistic regression analyses were employed to identify clinical-radiological (Clin-Rad) features associated with the MP/S (+) group for constructing the Clin-Rad classification. Subsequently, a hybrid model which presented as a nomogram was developed, integrating the Clin-Rad classification and ITH score. The performance of models was assessed using the receiver operating characteristic (ROC) curves, and the area under the curve (AUC), accuracy, sensitivity, and specificity were determined. Results: The ITH score outperformed both C-radiomics scores and Clin-Rad classification, as indicated by higher AUC values in the training (0.820 versus 0.810 and 0.700) and test sets (0.805 versus 0.771 and 0.732), respectively. Notably, the hybrid model consistently demonstrated robust predictive capabilities in identifying MP/S (+), achieving AUCs of 0.830 in the training set and 0.849 in the test sets. Conclusion: The ITH of sub-regions within the intratumor has been shown to be a reliable predictor for MP/S (+) in clinical stage I solid LADC. This finding holds the potential to make a significant contribution to clinical decision-making processes. Intratumoral heterogeneity Sub-regions High-grade patterns Clinical stage I Solid lung adenocarcinoma Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Background Lung cancer, the leading cause of global cancer mortality, is primarily defined by lung adenocarcinoma (LADC), which is the predominant histological subtype [ 1 ]. The recently revised 2021 WHO Classification of thoracic neoplasms provides a meticulous delineation of the various histological components associated with invasive non-mucinous LADC, including lepidic, papillary, acinar, micropapillary, solid, and complex glandular elements [ 2 ]. LADC often presents as a heterogeneous amalgamation of various subtypes, with over 94% of LACD cases featuring more than one pathological component [ 3 ]. Numerous investigations have emphasized the association between the micropapillary/solid component (MP/S) and a poor prognosis for patients with LADC [ 4 – 9 ]. In-depth analyses have revealed that the proportion of the MP/S component is intricately linked to an increased risk of adverse outcomes in stage IA LADC cases [ 7 – 9 ]. Currently, the preferred approach for treating LADC is radical resection [ 10 ]. Some studies have suggested that individuals with MP/S components in LADC may necessitate a more extensive surgical removal and intensified adjuvant chemotherapy [ 4 – 6 ]. Given the significant diversity within LADC, identifying MP/S components remains a formidable task. This challenge stems from the constraints imposed by preoperative puncture and intraoperative frozen sections, as well as the limited availability of harvested tissues [ 11 – 12 ]. Consequently, there is an imperative need for a precise and non-invasive method to help predict the presence of MP/S components in individuals diagnosed with invasive LADC before surgery. Numerous studies have highlighted clinical-radiological (Clin-Rad) features that predict the presence of MP/S components through the analysis of CT images and clinical baseline data [ 13 – 14 ]. However, these morphological characteristics lack representativeness, displaying a significant overlap in subtypes, and their performance remains inadequately evaluated, exhibiting considerable interobserver variability. The exploration of radiomic features extracted from CT images provides supplementary information intricately linked to MP/S components [ 15 – 16 ]. Despite this, current radiomics primarily extract characteristics of the entire tumor, disregarding intratumoral heterogeneity (ITH). Unlike previous radiomic methods, the emerging approach to ITH evaluation explicitly divides tumors into subregions containing clusters of voxels with similar attributes, also known as habitats. In the realm of medical imaging, the essence of ITH revolves around the hypothesis that identified subregions, characterized by voxels displaying similar imaging attributes, may inherently reveal a shared tumor biology [ 17 ]. Li et al.'s [ 17 ] study demonstrated an association between the ITH score and tumor phenotypes, such as lymphovascular invasion and pleural invasion, as well as the prognosis of patients with non-small cell lung cancer. In contrast to ground glass nodules (GGN), solid pulmonary nodules (SPN) demonstrate greater invasiveness and heterogeneity. Accordingly, LADC patients with MP/S components more frequently present as SPN in CT images [ 13 – 14 ]. Within our investigation framework, we hypothesized that ITH within subregions of intratumor derived from CT images would exhibit a stronger correlation with MP/S components than that derived from the conventional whole tumor. The main objective was to determine whether such ITH could accuratly predict the MP/S components in clinical stage I solid LADC prior to surgical intervention. Material and Methods Patient selection Adhering to the principles outlined in the Helsinki Declaration, this study obtained ethical approval from the Ethics Committee (No. 2021-07-009) of the participating hospital. The retrospective nature of the study led to the waiver of consent. From January 2020 to January 2023, we identified 868 cases of singularly resected LADCs at Liuzhou People’s Hospital (LZPH), and 732 similar cases at Xiangtan Central Hospital (XTCH). The patient inclusion criteria were as follows: (1) presentation as a SPN in CT scans, without calcification or vacuoles and lacking ground glass opacity; (2) maximum nodule diameter ranging from 0.5 cm to 3.0 cm; (3) postoperative pathology confirmation of non-mucinous LADC, further categorized into the MP/S (-) and MP/S (+) groups; (4) availability of complete thin-slice CT image data (0.625–1.25 mm) within 2 weeks before pathological diagnosis. Exclusion criteria included: (1) postoperative pathological diagnosis of multiple LADCs; (2) prior application of radiotherapy or chemotherapy before CT examination; (3) pathological diagnosis denoting minimally invasive LADCs or LADCs in situ. Figure 1 illustrates the flowchart detailing the process of patient inclusion. CT Image acquisition All non-contrast chest CT images were acquired using helical CT scanners from different manufacturers (SOMATOM go.Up or Brilliance iCT in LZPH; uCT550 or uCT760 in XTCH) with detectors ranging from 64 to 128 rows. The CT protocols used were as follows: 120 kVp, 100–670 mAs, with a pitch of 0.5–1.5. Subsequently, all imaging data were reconstructedusing a medium sharp algorithm with a thickness ranging from 0.60 to 1.25 mm. Clinical and Radiological features analysis and Clin-Rad classification construction The non-enhanced chest CT scans in DICOM format were imported into the RadiAnt DICOM Viewer 2023.1 ( https://www.radiantviewer.com ) for subsequent post-processing. Following this, multiplanar reconstruction was implemented with both lung and mediastinal windows, facilitating further analysis. The lung window exhibited a window width ranging from 1500 to 2000, with a corresponding window level spanning from − 450 to -700 HU. Concurrently, the mediastinal window displayed a window width ranging from 250 to 350 HU, and a window level oscillating between 35 to 50 HU. The comprehensive analysis of CT images involved recording and evaluating various morphological features, including spiculation, shape, boundary, lobulation, vascular convergence signs, and vacuole signs. Two board-certified thoracic radiologists with 8 and 15 years of experience in chest CT imaging, respectively, independently analyzed CT morphological features. These two radiologists were unaware of the clinical and histological findings. Any discrepancies in qualitative indicators were resolved through consensus reached during the discussion. Additionally, consideration was given to clinical baseline information, including clinical stage, age, and sex. Univariable analysis and multivariable backward stepwise logistic regression analysis were performed to identify the most optimal combinations of clinical and radiological variables, and ultimately determine the Clin-Rad classification. Volume of Interest (VOI) Delineation and Sub-Region Clustering The delineation of the volume of interest (VOI) was carried out by a radiologist with over 5 years of experience in chest imaging. Following this, another radiologist with a decade of experience in the same field, made necessary adjustments to the delineation. The segmentation of the VOI was performed using open-source software, specifically ITK-SNAP 4.0 ( http://www.itksnap.org/pmwiki/pmwiki.php ). Subsequently, the VOIs were subdivided into distinct regions based on the clustering of CT values and local entropy values derived from CT images. Local entropy calculations were conducted on each CT image slice using a 9 × 9 moving window. The sub-regions in this study were clustered using the K-means method. The Calinski–Harabasz value served as the criterion to determine the optimal number of clusters at the patient population level [ 18 ]. The study investigated the testing of the number of clusters within the range of 2 to 10. Feature Extraction To address the variations in radiomics features caused by different reconstruction slice thicknesses and pixel sizes, all CT images in this study were reconstructed to a consistent voxel size of 1 × 1 × 5 mm^3. The voxel dimensions of VOIs were adjusted to 64 gray levels to compensate for variations in the CT scanners. Radiomics features, including region volume, shape, intensity, and texture, were quantified for each sub-region, resulting in a total of 1239 features. For comparative analysis, an identical set of 1239 radiomics features was also extracted from the entire tumor region for each patient. The sub-regional partitioning and radiomics feature extraction were meticulously executed using pyradiomics ( https://pyradiomics.readthedocs.io ). Feature selection and prediction model construction To mitigate the potential impact of volume changes arising from sub-regional analysis on radiomics features, we applied the max-relevance and Min-Redundancy (mRMR) approach to assess redundancy among the values of these radiomics features. Subsequently, the least absolute shrinkage and selection operator (LASSO) was employed to identify features highly correlated with the desired outcome. The 10-fold cross-validation strategy was used to determine the optimal λ in the LASSO algorithm, leading to the selection of features with non-zero coefficients. Expanding upon the aforementioned methodology, an ITH score was calculated for sub-regional radiomics features screened by LASSO. This calculation was based on the selected sub-regional radiomics features and their corresponding coefficients. Similarly, for the entire tumor region radiomics features identified via LASSO, a C-radiomics score was derived by utilizing the selected tumor region radiomics features and their associated coefficients. The ITH score and C-radiomics score were employed to predict the presence of MP/S (+) in clinical stage I solid LADC. Hybrid model construction The Clin-Rad classification, in conjunction with the ITH score derived from the sub-regions, was strategically utilized to develop the pioneering hybrid prediction model, which was subsequently transformed into its corresponding nomogram. Rigorous assessment of Predictive performance in both the training and test sets was rigorously assessed by calibration curves. Figure 2 offers a visual representation of the intricate radiomics process, illustrating the successive steps of VOI delineation, meticulous feature extraction, principled dimensionality reduction, judicious feature selection, and the systematic construction of the model. Statistical analysis Statistical analysis was performed using R 4.3.2 ( https://www.r-project.org/ ). Continuous variables with normally distributed were expressed as mean ± standard deviation, and categorical variables were presented as frequencies and percentages. Differences in characteristics between training and test sets were compared via the Chi-square test or Fisher’s exact test. Receiver operating characteristic (ROC) curves were used to evaluate the predictive performance of the models, and the area under the curve (AUC) was used for quantization. The calibration curve was used to test the degree of model calibration. A p < 0.05 was considered as statistical significance. Results Baseline characteristics A total of 457 patients with pathological confirmed clinical stage I solid LADC were enrolled in the study. Among them, 95 cases (20.8%) were diagnosed with MP/S (+) pattern, while another 362 cases (79.2%) were MP/S (-). All the patients were divided into a training set with 304 patients from LZPH, and a test set with 153 patients from XTCH, respectively. Notably, no significant differences (p > 0.05) were observed between these two sets (refer to Table 1 ). Table 1 Comparative Evaluation of Clinico-radiological features between the Training and Test Sets Variables Total (n = 457) Training set (n = 304) Test set (n = 153) p-value Pathological components, N (%) 0.419 MP/S (-) group 362 (79.2) 237 (78) 125 (81.7) MP/S (+) group 95 (20.8) 67 (22) 28 (18.3) Location, N (%) 0.947 RUL 137 (30) 93 (30.6) 44 (28.8) RLL 105 (23) 69 (22.7) 36 (23.5) RML 39 (8.5) 25 (8.2) 14 (9.2) LUL 107 (23.4) 69 (22.7) 38 (24.8) LLL 69 (15.1) 48 (15.8) 21 (13.7) Boundary, N (%) 0.047 Ill-Defined 101 (22.1) 76 (25) 25 (16.3) Well-Defined 356 (77.9) 228 (75) 128 (83.7) Shape, N (%) 1 Irregular 259 (56.7) 172 (56.6) 87 (56.9) Others 198 (43.3) 132 (43.4) 66 (43.1) Lobulation, N (%) 0.102 Absence 93 (20.4) 69 (22.7) 24 (15.7) Presence 364 (79.6) 235 (77.3) 129 (84.3) Spiculation, N (%) 0.645 Absence 112 (24.5) 77 (25.3) 35 (22.9) Presence 345 (75.5) 227 (74.7) 118 (77.1) Vascular Convergence Sign, N (%) 0.09 Absence 129 (28.2) 94 (30.9) 35 (22.9) Presence 328 (71.8) 210 (69.1) 118 (77.1) Vacuole Sign, N (%) 0.054 Absence 370 (81) 238 (78.3) 132 (86.3) Presence 87 (19) 66 (21.7) 21 (13.7) Pleural Indentation, N (%) 0.649 Absence 77 (16.8) 49 (16.1) 28 (18.3) Presence 380 (83.2) 255 (83.9) 125 (81.7) Sex, N (%) 0.771 Male 197 (43.1) 133 (43.8) 64 (41.8) Female 260 (56.9) 171 (56.2) 89 (58.2) Age, Median (Q1, Q3) 61 (54, 67) 61 (54, 67) 62 (53, 68) 0.47 Clinical Stage, N (%) 0.562 cT1a 37 (8.1) 26 (8.6) 11 (7.2) cT1b 188 (41.1) 129 (42.4) 59 (38.6) cT1c 232 (50.8) 149 (49) 83 (54.2) Abbreviation : LUL Left Upper Lobe, LLL Left Lower Lobe, RUL Right Upper Lobe, RML Right Middle Lobe, RLL Right Lower Lobe, MP/S Micropapillary and/or Solid Components, LADC Lung Adenocarcinoma In the training set, clinical baseline characteristics, including predominantly male sex and a higher T stage, were associated with the MP/S (+) pattern in patients with clinical stage I solid LADC. However, CT radiological features such as spiculation, shape, boundary, lobulation, vascular convergence signs, and vacuole signs showed no significant associations in differentiating MP/S (+) from MP/S (-) in clinical stage I solid LADC. More comprehensive information between the MP/S (+) and MP/S (-) groups please refer to Tabel S1. Clin-Rad classification Univariate and multivariate logistic analyses were conducted to discern various factors. Notably, CT morphological features showed no significant associations in differentiating MP/S (+) from MP/S (-) in clinical stage I solid LADC (p > 0.05). However, clinical baseline information, including variables such as sex (OR: 0.891 (95% confidence interval [CI] 0.814–0.976), p = 0.013) and clinical stage (OR: 1.014 (95%CI 1.008–1.021), p < 0.001), presented as independent risk factors for MP/S (+) (refer to Table 2 ). These identified risk factors were subsequently selected as predictive variables for the construction of the Clin-Rad classification. The AUC of the Clin-Rad classification was calculated to be 0.700 for the training set and 0.732 for the test set, respectively. Table 2 Logistic Analysis for MP/S (+) in Clinical Stage I Solid Lung Adenocarcinoma Variables Univariate analyses Multivariate analyses Odd Ratio (95%CI) p-value Odd Ratio (95%CI) p-value Location 1.4 (0.7–2.8) 0.33 Boundary 0.97 (0.52–1.8) 0.94 Shape 0.72 (0.41–1.3) 0.25 Lobulation 1.3 (0.65–2.5) 0.47 Spiculation 1.4 (0.71–2.6) 0.35 Vascular Convergence Sign 1.2 (0.64–2.1) 0.61 Vacuole Sign 0.84 (0.42–1.6) 0.6 Pleural Indentation 1.8 (0.79–4.3) 0.16 Sex 0.47 (0.27–0.82) 0.0075 0.891(0.814–0.976) 0.013 Age 1 (0.98-1) 0.44 Clinical Stage 1.1 (1.1–1.1) p < 0.001 1.014(1.008–1.021) p < 0.001 ITH score and C-radiomics score prediction model After conducting dimensionality reduction and feature selection using Lasso regression (refer to Fig. S1 and Fig. 3 ), we incorporated 20 signatures (as shown in Fig. 4 ) derived from the intratumoral sub-region and an additional 12 signatures (refer to Fig. S2 ) derived from the entire tumor region in the construction of both the ITH score and C-radiomics score. The ITH score was significantly higher in the MP/S (+) group compared to the MP/S (-) group (p < 0.001), as illustrated in Fig. 5 . Notably, the C-radiomics score also demonstrated a higher value in the MP/S (+) group than in the MP/S (-) group (p < 0.001), as shown in Fig. S3 . The ITH score exhibited superior performance than C-radiomics score in differentiating MP/S (+) from MP/S (-) in clinical stage I solid LADC, as indicated by the AUC of ITH score (0.820 in the training set and 0.805 in the test set) and the AUC of C-radiomics score (0.810 in the training set and 0.771 in the test set). The correlation heatmaps visually illustrate the intricate relationships among the ITH score, pathological components, and Clin-Rad features (refer to Fig. 6 ). These heatmaps reveal that a higher ITH score is associated with the MP/S (+) pattern, a male-dominated composition, a higher clinical stage, and more frequent invasive morphological features. Hybrid model and nomogram The ITH score, in conjunction with the Clin-Rad classification, was strategically leveraged to create the innovative hybrid model. Significantly, this hybrid model consistently exhibited robust predictive capabilities for identifying MP/S (+), achieving AUCs of 0.830 in the training set and 0.849 in the test sets (refer to Table 3 , Fig. 7 ). Furthermore, the hybrid model was transformed into a nomogram to quantify the risk of occurrence (Fig. 8 ). The calibration curve indicated good consistency between the predicted risk and the actual risk in both the training and test sets (Fig. 9 ). Table 3 Assessing Diagnostic Performance for Predicting the MP/S (+) Pattern across Diverse Models Models Cohorts Area under curve Accuracy Sensitivity Specificity Clin-Rad classification Training set 0.700 0.664 0.701 0.654 Test set 0.732 0.725 0.643 0.744 ITH score Training set 0.820 0.734 0.821 0.709 Test set 0.805 0.810 0.607 0.856 Hybrid model Training set 0.830 0.730 0.851 0.696 Test set 0.849 0.739 0.929 0.696 Discussion The ITH score, derived from sub-regions, provides a more nuanced reflection of intratumoral ecological diversity compared to the C-radiomics score derived from the entire tumor. The hybrid model, comprising the ITH score and the Clin-Rad classification, demonstrated more comprehensive diagnostic efficiency in distinguishing MP/S (+) from MP/S (-) in clinical stage I solid LADC. To quantify the risk of occurrence, the hybrid model was translated into a nomogram, demonstrating commendable differentiation and calibration capabilities. Several previous studies have utilized Clin-Rad features to construct models for predicting MP/S (+) from MP/S (-) in clinical stage I solid LADC. In a study conducted by Dong et al. [ 14 ], tumor size, density, and lobulation were identified as risk factors for predicting MP/S (+). However, in this current study, CT radiological features such as spiculation, shape, boundary, lobulation, vascular convergence signs, and vacuole signs did not show significant associations in differentiating MP/S (+) from MP/S (-) in clinical stage I solid LADC. We hypothesize that these morphological characteristics are not adequately representative, and their performance has not been sufficiently evaluated, exhibiting significant variability among observers and leading to poor repeatability. Radiomics enables the extraction of high-throughput features from medical images, thereby assisting in the identification of histological heterogeneity. This process enables the identification of additional factors that may not be readily discernible through visual inspection. Numerous previously published radiomics studies have investigated the significance of preoperative categorization based on the histological subtypes of LADC [ 15 – 16 , 19 – 22 ]. However, these studies included LADC patients, covering SPN and GGN types of lung nodules observed in CT scans. Subgroup analyses for SPN or GGN types were not conducted. Due to the high heterogeneity and invasiveness, the MP/S (+) pattern more commonly presents as SPN in CT images [ 13 – 14 ]. This observation leads us to infer that a higher frequency of SPN in the enrolled population could significantly improve the diagnostic efficiency of radiomics, potentially resulting in model overfitting. Furthermore, these aforementioned radiomics studies mainly capture characteristics of the entire tumor, neglecting intratumoral ecological diversity. Within our investigative framework, we analyzed ITH within subregions derived from CT images to predict histological subtype categorization. To the best of our knowledge, this marks the first report on the ITH score for predicting MP/S (+) from MP/S (-) in clinical stage I solid LADC. In the current investigation, the ITH score has demonstrated superior efficacy in discerning between MP/S (+) and MP/S (-) in clinical stage I solid LADC, outperforming the C-radiomics score and Clin-Rad classification. This discovery aligns seamlessly with the conclusions drawn by Li et al. [ 17 ], indicating that the ITH score provides a more intricate representation of intratumoral ecological diversity compared to the C-radiomics score. In Li et al.'s scholarly work, the ITH score manifested correlations with various tumor phenotypes, including lymphovascular invasion and pleural invasion, significantly influencing prognosis of individuals with non-small cell lung cancer. To further explore the relationship between the ITH score and tumor heterogeneity and biological behavior, this study employed correlation heatmaps, revealing that a higher ITH score is concomitant with the MP/S (+) pattern, a predominance of male-dominant characteristics, an advanced clinical stage, and increased occurrences of invasive morphological features. Finally, acknowledging the diagnostic significance of Clin-Rad in histological subtypes, we combined it with the ITH score to construct a hybrid model. Notably, this hybrid model consistently demonstrated robust predictive capabilities in identifying MP/S (+), achieving AUCs of 0.830 in the training set and 0.849 in the test sets. The hybrid model was subsequently transformed into a nomogram to quantify the risk of occurrence. The calibration curve attested to the commendable consistency between the predicted risk and the actual risk in both the training and test sets. This may be attributed to the incorporation of the hybrid model, which integrates multi-domain and abundant information, thereby enhancing diagnostic efficiency. Our investigation has encountered certain limitations. Firstly, it is crucial to acknowledge the retrospective nature of our research, in which potential selection bias arises from exclusively including patients with post-surgery pathological results. Secondly, the bicentric design introduces variability in acquisition parameters, image quality, and the potential for co-registration errors. This presents sources of variability that could confound the analysis. Additionally, the presence of artifacts and technical constraints may impact the reliability and reproducibility of radiomic features. Thirdly, the relatively short follow-up period after surgery prevented the development of a predictive model for disease outcomes. Lastly, it is essential to highlight that our study specifically focused on clinical stage I solid LADC, which may limit the generalizability of our findings to other stages. Conclusion In conclusion, the ITH within sub-regions proved to be a reliable predictor for MP/S (+) in clinical stage I solid LADC. We integrated the Clin-Rad classification with the ITH score to formulate a hybrid model, which was then transformed into a nomogram. This nomogram accurately quantifies the risk of MP/S (+). The nomogram demonstrated exceptional discrimination and calibration, indicating the potential to make a significant contribution to clinical decision-making processes. Abbreviations ITH Intratumoral heterogeneity quantification MP/S Micropapillary and/or solid components LADC Lung adenocarcinoma Clin-Rad Clinical-radiological ROC Receiver operating characteristic AUC Area under the curve GGN Ground glass nodules SPN Solid pulmonary nodules LZPH Liuzhou People’s Hospital XTCH Xiangtan Central Hospital VOI Volume of interest mRMR Max-relevance and Min-Redundancy LASSO Least absolute shrinkage and selection operator AUC The area under the curve CI Confidence interval Declarations Ethics approval and consent to participate Ethical approval was obtained from Medical Ethics Committee of Xiangtan Central Hospital (reference number: No. 2021-07-009), which ensured that the rights and interests of the research participants were not compromised. All methods were performed following the relevant guidelines and regulations. Medical Ethics Committee of Xiangtan Central Hospital approved the retrospective study and waived the requirement for informed consent. Availability of data and materials Not applicable Consent for publication Not applicable Competing interests The authors declare that they have no competing interests. Funding Not applicable Authors' contributions Z.Z.: Conceptualization, Methodology, Original Draft Writing. J.D.: Review & Editing of Writing, Supervision. W.G.: Software Development, Formal Analysis, Data Curation. Y.Z.: Validation, Provision of Resources. H.L.: Validation, Resource Allocation, Statistical Analysis, and Software Development. W.Z. and Y.Z.: Theoretical Guidance, Suggestions for Article Revision, Supportive Contributions. All authors contributed to the article and approved the submitted version. Acknowledgements Not applicable References Sung H, Ferlay J, Siegel RL, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin. 2021;71(3):209–49. http://doi.org/10.3322/caac.21660 . Nicholson AG, Tsao MS, Beasley MB, et al. The 2021 WHO Classification of Lung Tumors: Impact of Advances Since 2015. J Thorac Oncol. 2022;17(3):362–87. http://doi.org/10.1016/j.jtho.2021.11.003 . Hung JJ, Jeng WJ, Chou TY, et al. Prognostic value of the new International Association for the Study of Lung Cancer/American Thoracic Society/European Respiratory Society lung adenocarcinoma classification on death and recurrence in completely resected stage I lung adenocarcinoma. Ann Surg. 2013;258(6):1079–86. http://doi.org/10.1097/SLA.0b013e31828920c0 . Tsubokawa N, Mimae T, Sasada S, et al. Negative prognostic influence of micropapillary pattern in stage IA lung adenocarcinoma. Eur J Cardiothorac Surg. 2016;49(1):293–9. http://doi.org/10.1093/ejcts/ezv058 . Yanagawa N, Shiono S, Abiko M, et al. The Clinical Impact of Solid and Micropapillary Patterns in Resected Lung Adenocarcinoma. J Thorac Oncol. 2016;11(11):1976–83. http://doi.org/10.1016/j.jtho.2016.06.014 . Mikubo M, Tamagawa S, Kondo Y, et al. Micropapillary and solid components as high-grade patterns in IASLC grading system of lung adenocarcinoma: Clinical implications and management. Lung Cancer. 2024;187:107445. http://doi.org/10.1016/j.lungcan.2023.107445 . Liu W, Zhang Q, Zhang T, Li L, Xu C. Minor histological components predict the recurrence of patients with resected stage I acinar- or papillary-predominant lung adenocarcinoma. Front Oncol. 2022;12:1090544. http://doi.org/10.3389/fonc.2022.1090544 . Chen C, Chen ZJ, Li WJ, et al. Impact of minimal solid and micropapillary components on invasive lung adenocarcinoma recurrence. Ann Diagn Pathol. 2022;59:151945. http://doi.org/10.1016/j.anndiagpath.2022.151945 . Huang W, Zhang H, Zhang Z, et al. A prognostic nomogram based on a new classification of combined micropapillary and solid components for stage IA invasive lung adenocarcinoma. J Surg Oncol. 2022;125(4):796–808. http://doi.org/10.1002/jso.26760 . Van Schil PE, Asamura H, Rusch VW, et al. Surgical implications of the new IASLC/ATS/ERS adenocarcinoma classification. Eur Respir J. 2012;39(2):478–86. http://doi.org/10.1183/09031936.00027511 . Yeh YC, Nitadori J, Kadota K, et al. Using frozen section to identify histological patterns in stage I lung adenocarcinoma of ≤ 3 cm: accuracy and interobserver agreement. Histopathology. 2015;66(7):922–38. http://doi.org/10.1111/his.12468 . Trejo Bittar HE, Incharoen P, Althouse AD, Dacic S. Accuracy of the IASLC/ATS/ERS histological subtyping of stage I lung adenocarcinoma on intraoperative frozen sections. Mod Pathol. 2015;28(8):1058–63. http://doi.org/10.1038/modpathol.2015.71 . Lederlin M, Puderbach M, Muley T, et al. Correlation of radio- and histomorphological pattern of pulmonary adenocarcinoma. Eur Respir J. 2013;41(4):943–51. http://doi.org/10.1183/09031936.00056612 . Dong H, Yin LK, Qiu YG, et al. Prediction of high-grade patterns of stage IA lung invasive adenocarcinoma based on high-resolution CT features: a bicentric study. Eur Radiol. 2023;33(6):3931–40. http://doi.org/10.1007/s00330-022-09379-x . Liu Y, Chang Y, Zha X, et al. A Combination of Radiomic Features, Imaging Characteristics, and Serum Tumor Biomarkers to Predict the Possibility of the High-Grade Subtypes of Lung Adenocarcinoma. Acad Radiol. 2022;29(12):1792–801. http://doi.org/10.1016/j.acra.2022.02.024 . Dong H, Wang X, Qiu Y, et al. Establishment and visualization of a model based on high-resolution CT qualitative and quantitative features for prediction of micropapillary or solid components in invasive lung adenocarcinoma. J Cancer Res Clin Oncol. 2023;149(12):10519–30. http://doi.org/10.1007/s00432-023-04854-4 . Li J, Qiu Z, Zhang C, et al. ITHscore: comprehensive quantification of intra-tumor heterogeneity in NSCLC by multi-scale radiomic features. Eur Radiol. 2023;33(2):893–903. http://doi.org/10.1007/s00330-022-09055-0 . Liu Y, Li Z, Xiong H, et al. Understanding and enhancement of internal clustering validation measures. IEEE Trans Cybern. 2013;43(3):982–94. http://doi.org/10.1109/tsmcb.2012.2220543 . Song SH, Park H, Lee G, et al. Imaging Phenotyping Using Radiomics to Predict Micropapillary Pattern within Lung Adenocarcinoma. J Thorac Oncol. 2017;12(4):624–32. http://doi.org/10.1016/j.jtho.2016.11.2230 . Park S, Lee SM, Noh HN, et al. Differentiation of predominant subtypes of lung adenocarcinoma using a quantitative radiomics approach on CT. Eur Radiol. 2020;30(9):4883–92. http://doi.org/10.1007/s00330-020-06805-w . Chen LW, Yang SM, Wang HJ, et al. Prediction of micropapillary and solid pattern in lung adenocarcinoma using radiomic values extracted from near-pure histopathological subtypes. Eur Radiol. 2021;31(7):5127–38. http://doi.org/10.1007/s00330-020-07570-6 . Xu Y, Ji W, Hou L, et al. Enhanced CT-Based Radiomics to Predict Micropapillary Pattern Within Lung Invasive Adenocarcinoma. Front Oncol. 2021;11:704994. http://doi.org/10.3389/Fonc.2021.704994 . Legend. Additional Declarations No competing interests reported. Supplementary Files Fig.S1.tif Fig. S1 showcases the art of feature selection for signatures derived from the entire tumor region using Lasso regression and a 10-fold cross-validation. Fig.S2.tif Fig. S2 presents carefully chosen signatures derived from the entire tumor region using Lasso regression. Fig.S3.tif Fig. S3 conducts a nuanced comparison of C-radiomics scores between the MP/S (+) and MP/S (-) groups, revealing notably elevated C-radiomics scores in the MP/S (+) group (p<0.001). TableS1.docx Cite Share Download PDF Status: Published Journal Publication published 09 Jan, 2025 Read the published version in BMC Cancer → Version 1 posted Editorial decision: Revision requested 08 Feb, 2024 Submission checks completed at journal 08 Feb, 2024 Editor assigned by journal 08 Feb, 2024 First submitted to journal 30 Jan, 2024 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-3910257","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":271912225,"identity":"3c5763ef-1678-42f7-8bb3-9704c704697b","order_by":0,"name":"Zhichao Zuo","email":"","orcid":"","institution":"Xiangtan Central Hospital","correspondingAuthor":false,"prefix":"","firstName":"Zhichao","middleName":"","lastName":"Zuo","suffix":""},{"id":271912226,"identity":"888dab8d-511c-44bd-91b5-af7cf145f83f","order_by":1,"name":"Jinqiu Deng","email":"","orcid":"","institution":"Xiangtan University","correspondingAuthor":false,"prefix":"","firstName":"Jinqiu","middleName":"","lastName":"Deng","suffix":""},{"id":271912227,"identity":"ded02679-600a-416d-b46c-a9e0c0fabcf2","order_by":2,"name":"Wu Ge","email":"","orcid":"","institution":"Xiangtan Central Hospital","correspondingAuthor":false,"prefix":"","firstName":"Wu","middleName":"","lastName":"Ge","suffix":""},{"id":271912228,"identity":"d2f84f08-d47d-4b6d-8ad2-09eb07581f66","order_by":3,"name":"Yinjun Zhou","email":"","orcid":"","institution":"Xiangtan Central Hospital","correspondingAuthor":false,"prefix":"","firstName":"Yinjun","middleName":"","lastName":"Zhou","suffix":""},{"id":271912229,"identity":"b7633323-92d9-496c-8361-47570761977a","order_by":4,"name":"Haibo Liu","email":"","orcid":"","institution":"Xiangtan Central Hospital","correspondingAuthor":false,"prefix":"","firstName":"Haibo","middleName":"","lastName":"Liu","suffix":""},{"id":271912230,"identity":"09e51977-28ef-485d-b43a-1fdc2e903e20","order_by":5,"name":"Wei Zhang","email":"","orcid":"","institution":"Liuzhou People's Hospital","correspondingAuthor":false,"prefix":"","firstName":"Wei","middleName":"","lastName":"Zhang","suffix":""},{"id":271912231,"identity":"c959d2c7-78c6-40ea-80dc-f53aaece298b","order_by":6,"name":"Ying Zeng","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA30lEQVRIiWNgGAWjYFAC5oMPJP/YyNm3NzY+/ECcFrZkA8uGNGMDnsPNxhLEaeFRE6hsOJy4QSK9TYCHGA3yM3LYGG7uYE7cLvmwjUGCwU5Ot4GAFsYZuccezjzDZrxzdmLbgwKGZGOzAwS0MEvkpRtLsPHINtxObDeQYDiQuI2QFjaJHDPpP2wSjA03D7ZJ8BCjhQeoRUKyzUBxww1GIrVI8DxLNpA4k2As2ZMIDGQDIvwi35588IFExX85fvbjDx9+qLCTI6iFQSABmWdASDkI8BM0dBSMglEwCkY8AADI5ESlz3iw/AAAAABJRU5ErkJggg==","orcid":"","institution":"Xiangtan Central Hospital","correspondingAuthor":true,"prefix":"","firstName":"Ying","middleName":"","lastName":"Zeng","suffix":""}],"badges":[],"createdAt":"2024-01-30 10:17:40","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3910257/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3910257/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s12885-025-13445-0","type":"published","date":"2025-01-09T15:57:05+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":51024916,"identity":"8d62093c-a585-417a-8c09-24518cc59b90","added_by":"auto","created_at":"2024-02-12 21:41:19","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":660094,"visible":true,"origin":"","legend":"\u003cp\u003eillustrates the process of patient inclusion with intricate detail through a comprehensive flowchart.\u003c/p\u003e","description":"","filename":"Fig.1.png","url":"https://assets-eu.researchsquare.com/files/rs-3910257/v1/666a1372508ea6ba1bced771.png"},{"id":51024917,"identity":"9d1b29f3-fc63-455c-bc91-9293ad4f84de","added_by":"auto","created_at":"2024-02-12 21:41:19","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":427545,"visible":true,"origin":"","legend":"\u003cp\u003eprovides a captivating visual representation of the radiomics process, delineating the successive steps of VOI delineation, meticulous feature extraction, principled dimensionality reduction, judicious feature selection, and systematic model construction.\u003c/p\u003e","description":"","filename":"Fig.2.png","url":"https://assets-eu.researchsquare.com/files/rs-3910257/v1/f45e830b4e186f7113af99d3.png"},{"id":51024918,"identity":"6e79e518-9e94-43a1-9ea2-f5f4f06e647d","added_by":"auto","created_at":"2024-02-12 21:41:19","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":140938,"visible":true,"origin":"","legend":"\u003cp\u003eshowcases the art of feature selection for signatures derived from the intratumoral sub-region using Lasso regression and a 10-fold cross-validation.\u003c/p\u003e","description":"","filename":"Fig.3.png","url":"https://assets-eu.researchsquare.com/files/rs-3910257/v1/5628536e31d35900bce27999.png"},{"id":51024920,"identity":"4bc443e1-356d-4cd7-844c-c037395af50b","added_by":"auto","created_at":"2024-02-12 21:41:19","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":225187,"visible":true,"origin":"","legend":"\u003cp\u003epresents carefully chosen signatures derived from the intratumoral sub-region using Lasso regression.\u003c/p\u003e","description":"","filename":"Fig.4.png","url":"https://assets-eu.researchsquare.com/files/rs-3910257/v1/fbff7f912c03ee81aa95d681.png"},{"id":51024926,"identity":"5011130d-18f7-44ac-9ca9-409d879d05fd","added_by":"auto","created_at":"2024-02-12 21:41:19","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":225748,"visible":true,"origin":"","legend":"\u003cp\u003econducts a nuanced comparison of ITH scores between the MP/S (+) and MP/S (-) groups, revealing notably elevated ITH scores in the MP/S (+) group (p\u0026lt;0.001).\u003c/p\u003e","description":"","filename":"Fig.5.png","url":"https://assets-eu.researchsquare.com/files/rs-3910257/v1/327bfca22bcf1b0d495733d6.png"},{"id":51024919,"identity":"af67bd0c-6875-4587-be36-c5e629114c7c","added_by":"auto","created_at":"2024-02-12 21:41:19","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":162176,"visible":true,"origin":"","legend":"\u003cp\u003eportrays correlation heatmaps that vividly illustrate the intricate relationships among the ITH score, pathological components, and Clin-Rad features. These heatmaps underscore that a higher ITH score corresponds to the MP/S (+) pattern, a prevalence of male composition, an advanced clinical stage, and more frequent instances of invasive morphological features.\u003c/p\u003e","description":"","filename":"Fig.6.png","url":"https://assets-eu.researchsquare.com/files/rs-3910257/v1/94ccea789dbf5a87c3e7d080.png"},{"id":51024922,"identity":"336413b5-3c90-49f8-8c74-011b0915fb03","added_by":"auto","created_at":"2024-02-12 21:41:19","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":174974,"visible":true,"origin":"","legend":"\u003cp\u003eoffers a comprehensive comparison of the diagnostic performance of Clin-Rad classification, ITH score, and the hybrid model. The hybrid model consistently manifests superior predictive capabilities in both the training and test sets compared to Clin-Rad classification and ITH score.\u003c/p\u003e","description":"","filename":"Fig.7.png","url":"https://assets-eu.researchsquare.com/files/rs-3910257/v1/9cf15354def2447a3f29d8cd.png"},{"id":51024925,"identity":"65e4f527-b333-44f8-b2eb-b9774e02b9b6","added_by":"auto","created_at":"2024-02-12 21:41:19","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":95224,"visible":true,"origin":"","legend":"\u003cp\u003etransforms the hybrid model into a sophisticated nomogram, enabling the quantification of the risk of occurrence.\u003c/p\u003e","description":"","filename":"Fig.8.png","url":"https://assets-eu.researchsquare.com/files/rs-3910257/v1/4622401f5a9c4ac30a4a0be2.png"},{"id":51024928,"identity":"d5c87502-9815-4064-bc23-3abe93537a2a","added_by":"auto","created_at":"2024-02-12 21:41:20","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":475751,"visible":true,"origin":"","legend":"\u003cp\u003eexhibits the calibration curve of the nomogram, demonstrating commendable consistency between the predicted and actual risks in both the training and test sets.\u003c/p\u003e","description":"","filename":"Fig.9.png","url":"https://assets-eu.researchsquare.com/files/rs-3910257/v1/0dce77268c3f94c3d909072a.png"},{"id":73693819,"identity":"e40d5a61-e814-4aa6-b761-94a05660699e","added_by":"auto","created_at":"2025-01-13 16:07:38","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3086535,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3910257/v1/cfa81f11-66ab-4610-87e9-1598722816a3.pdf"},{"id":51024923,"identity":"8f9fb4d7-2b44-424a-bbdd-8d10e03f4df2","added_by":"auto","created_at":"2024-02-12 21:41:19","extension":"tif","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":10239072,"visible":true,"origin":"","legend":"\u003cp\u003eFig. S1 showcases the art of feature selection for signatures derived from the entire tumor region using Lasso regression and a 10-fold cross-validation.\u003c/p\u003e","description":"","filename":"Fig.S1.tif","url":"https://assets-eu.researchsquare.com/files/rs-3910257/v1/f8d6955d3209b7183ebae874.tif"},{"id":51024927,"identity":"8eef822b-f85a-4c87-a18c-9febfeb69a60","added_by":"auto","created_at":"2024-02-12 21:41:20","extension":"tif","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":7907050,"visible":true,"origin":"","legend":"\u003cp\u003eFig. S2 presents carefully chosen signatures derived from the entire tumor region using Lasso regression.\u003c/p\u003e","description":"","filename":"Fig.S2.tif","url":"https://assets-eu.researchsquare.com/files/rs-3910257/v1/5e44f9a396ef2dd8ceec493a.tif"},{"id":51025900,"identity":"4b428dc2-25c7-4965-8db2-fee1cdc6555e","added_by":"auto","created_at":"2024-02-12 21:49:20","extension":"tif","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":10539745,"visible":true,"origin":"","legend":"\u003cp\u003eFig. S3 conducts a nuanced comparison of C-radiomics scores between the MP/S (+) and MP/S (-) groups, revealing notably elevated C-radiomics scores in the MP/S (+) group (p\u0026lt;0.001).\u003c/p\u003e","description":"","filename":"Fig.S3.tif","url":"https://assets-eu.researchsquare.com/files/rs-3910257/v1/59491fa8e5ef1427172e0e27.tif"},{"id":51024921,"identity":"a24ae96a-34e5-4cfe-ad46-961215bd3f9b","added_by":"auto","created_at":"2024-02-12 21:41:19","extension":"docx","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":22380,"visible":true,"origin":"","legend":"","description":"","filename":"TableS1.docx","url":"https://assets-eu.researchsquare.com/files/rs-3910257/v1/1ce1e3be94005f374a33bb52.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Quantifying intratumoral heterogeneity within sub-regions to predict high-grade patterns in clinical stage I solid lung adenocarcinoma","fulltext":[{"header":"Background","content":"\u003cp\u003eLung cancer, the leading cause of global cancer mortality, is primarily defined by lung adenocarcinoma (LADC), which is the predominant histological subtype [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. The recently revised 2021 WHO Classification of thoracic neoplasms provides a meticulous delineation of the various histological components associated with invasive non-mucinous LADC, including lepidic, papillary, acinar, micropapillary, solid, and complex glandular elements [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. LADC often presents as a heterogeneous amalgamation of various subtypes, with over 94% of LACD cases featuring more than one pathological component [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. Numerous investigations have emphasized the association between the micropapillary/solid component (MP/S) and a poor prognosis for patients with LADC [\u003cspan additionalcitationids=\"CR5 CR6 CR7 CR8\" citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. In-depth analyses have revealed that the proportion of the MP/S component is intricately linked to an increased risk of adverse outcomes in stage IA LADC cases [\u003cspan additionalcitationids=\"CR8\" citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eCurrently, the preferred approach for treating LADC is radical resection [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. Some studies have suggested that individuals with MP/S components in LADC may necessitate a more extensive surgical removal and intensified adjuvant chemotherapy [\u003cspan additionalcitationids=\"CR5\" citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. Given the significant diversity within LADC, identifying MP/S components remains a formidable task. This challenge stems from the constraints imposed by preoperative puncture and intraoperative frozen sections, as well as the limited availability of harvested tissues [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. Consequently, there is an imperative need for a precise and non-invasive method to help predict the presence of MP/S components in individuals diagnosed with invasive LADC before surgery.\u003c/p\u003e \u003cp\u003eNumerous studies have highlighted clinical-radiological (Clin-Rad) features that predict the presence of MP/S components through the analysis of CT images and clinical baseline data [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. However, these morphological characteristics lack representativeness, displaying a significant overlap in subtypes, and their performance remains inadequately evaluated, exhibiting considerable interobserver variability. The exploration of radiomic features extracted from CT images provides supplementary information intricately linked to MP/S components [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. Despite this, current radiomics primarily extract characteristics of the entire tumor, disregarding intratumoral heterogeneity (ITH). Unlike previous radiomic methods, the emerging approach to ITH evaluation explicitly divides tumors into subregions containing clusters of voxels with similar attributes, also known as habitats. In the realm of medical imaging, the essence of ITH revolves around the hypothesis that identified subregions, characterized by voxels displaying similar imaging attributes, may inherently reveal a shared tumor biology [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. Li et al.'s [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e] study demonstrated an association between the ITH score and tumor phenotypes, such as lymphovascular invasion and pleural invasion, as well as the prognosis of patients with non-small cell lung cancer.\u003c/p\u003e \u003cp\u003eIn contrast to ground glass nodules (GGN), solid pulmonary nodules (SPN) demonstrate greater invasiveness and heterogeneity. Accordingly, LADC patients with MP/S components more frequently present as SPN in CT images [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. Within our investigation framework, we hypothesized that ITH within subregions of intratumor derived from CT images would exhibit a stronger correlation with MP/S components than that derived from the conventional whole tumor. The main objective was to determine whether such ITH could accuratly predict the MP/S components in clinical stage I solid LADC prior to surgical intervention.\u003c/p\u003e"},{"header":"Material and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003ePatient selection\u003c/h2\u003e \u003cp\u003e Adhering to the principles outlined in the Helsinki Declaration, this study obtained ethical approval from the Ethics Committee (No. 2021-07-009) of the participating hospital. The retrospective nature of the study led to the waiver of consent.\u003c/p\u003e \u003cp\u003eFrom January 2020 to January 2023, we identified 868 cases of singularly resected LADCs at Liuzhou People\u0026rsquo;s Hospital (LZPH), and 732 similar cases at Xiangtan Central Hospital (XTCH). The patient inclusion criteria were as follows: (1) presentation as a SPN in CT scans, without calcification or vacuoles and lacking ground glass opacity; (2) maximum nodule diameter ranging from 0.5 cm to 3.0 cm; (3) postoperative pathology confirmation of non-mucinous LADC, further categorized into the MP/S (-) and MP/S (+) groups; (4) availability of complete thin-slice CT image data (0.625\u0026ndash;1.25 mm) within 2 weeks before pathological diagnosis. Exclusion criteria included: (1) postoperative pathological diagnosis of multiple LADCs; (2) prior application of radiotherapy or chemotherapy before CT examination; (3) pathological diagnosis denoting minimally invasive LADCs or LADCs in situ. Figure\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e illustrates the flowchart detailing the process of patient inclusion.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003eCT Image acquisition\u003c/h2\u003e \u003cp\u003eAll non-contrast chest CT images were acquired using helical CT scanners from different manufacturers (SOMATOM go.Up or Brilliance iCT in LZPH; uCT550 or uCT760 in XTCH) with detectors ranging from 64 to 128 rows. The CT protocols used were as follows: 120 kVp, 100\u0026ndash;670 mAs, with a pitch of 0.5\u0026ndash;1.5. Subsequently, all imaging data were reconstructedusing a medium sharp algorithm with a thickness ranging from 0.60 to 1.25 mm.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003eClinical and Radiological features analysis and Clin-Rad classification construction\u003c/h2\u003e \u003cp\u003eThe non-enhanced chest CT scans in DICOM format were imported into the RadiAnt DICOM Viewer 2023.1 (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.radiantviewer.com\u003c/span\u003e\u003cspan address=\"https://www.radiantviewer.com\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) for subsequent post-processing. Following this, multiplanar reconstruction was implemented with both lung and mediastinal windows, facilitating further analysis. The lung window exhibited a window width ranging from 1500 to 2000, with a corresponding window level spanning from \u0026minus;\u0026thinsp;450 to -700 HU. Concurrently, the mediastinal window displayed a window width ranging from 250 to 350 HU, and a window level oscillating between 35 to 50 HU. The comprehensive analysis of CT images involved recording and evaluating various morphological features, including spiculation, shape, boundary, lobulation, vascular convergence signs, and vacuole signs.\u003c/p\u003e \u003cp\u003eTwo board-certified thoracic radiologists with 8 and 15 years of experience in chest CT imaging, respectively, independently analyzed CT morphological features. These two radiologists were unaware of the clinical and histological findings. Any discrepancies in qualitative indicators were resolved through consensus reached during the discussion. Additionally, consideration was given to clinical baseline information, including clinical stage, age, and sex.\u003c/p\u003e \u003cp\u003eUnivariable analysis and multivariable backward stepwise logistic regression analysis were performed to identify the most optimal combinations of clinical and radiological variables, and ultimately determine the Clin-Rad classification.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003eVolume of Interest (VOI) Delineation and Sub-Region Clustering\u003c/h2\u003e \u003cp\u003eThe delineation of the volume of interest (VOI) was carried out by a radiologist with over 5 years of experience in chest imaging. Following this, another radiologist with a decade of experience in the same field, made necessary adjustments to the delineation. The segmentation of the VOI was performed using open-source software, specifically ITK-SNAP 4.0 (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.itksnap.org/pmwiki/pmwiki.php\u003c/span\u003e\u003cspan address=\"http://www.itksnap.org/pmwiki/pmwiki.php\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Subsequently, the VOIs were subdivided into distinct regions based on the clustering of CT values and local entropy values derived from CT images. Local entropy calculations were conducted on each CT image slice using a 9 \u0026times; 9 moving window. The sub-regions in this study were clustered using the K-means method. The Calinski\u0026ndash;Harabasz value served as the criterion to determine the optimal number of clusters at the patient population level [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. The study investigated the testing of the number of clusters within the range of 2 to 10.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eFeature Extraction\u003c/h2\u003e \u003cp\u003eTo address the variations in radiomics features caused by different reconstruction slice thicknesses and pixel sizes, all CT images in this study were reconstructed to a consistent voxel size of 1 \u0026times; 1 \u0026times; 5 mm^3. The voxel dimensions of VOIs were adjusted to 64 gray levels to compensate for variations in the CT scanners. Radiomics features, including region volume, shape, intensity, and texture, were quantified for each sub-region, resulting in a total of 1239 features. For comparative analysis, an identical set of 1239 radiomics features was also extracted from the entire tumor region for each patient. The sub-regional partitioning and radiomics feature extraction were meticulously executed using pyradiomics (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://pyradiomics.readthedocs.io\u003c/span\u003e\u003cspan address=\"https://pyradiomics.readthedocs.io\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eFeature selection and prediction model construction\u003c/h2\u003e \u003cp\u003eTo mitigate the potential impact of volume changes arising from sub-regional analysis on radiomics features, we applied the max-relevance and Min-Redundancy (mRMR) approach to assess redundancy among the values of these radiomics features. Subsequently, the least absolute shrinkage and selection operator (LASSO) was employed to identify features highly correlated with the desired outcome. The 10-fold cross-validation strategy was used to determine the optimal λ in the LASSO algorithm, leading to the selection of features with non-zero coefficients.\u003c/p\u003e \u003cp\u003eExpanding upon the aforementioned methodology, an ITH score was calculated for sub-regional radiomics features screened by LASSO. This calculation was based on the selected sub-regional radiomics features and their corresponding coefficients. Similarly, for the entire tumor region radiomics features identified via LASSO, a C-radiomics score was derived by utilizing the selected tumor region radiomics features and their associated coefficients. The ITH score and C-radiomics score were employed to predict the presence of MP/S (+) in clinical stage I solid LADC.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003eHybrid model construction\u003c/h2\u003e \u003cp\u003eThe Clin-Rad classification, in conjunction with the ITH score derived from the sub-regions, was strategically utilized to develop the pioneering hybrid prediction model, which was subsequently transformed into its corresponding nomogram. Rigorous assessment of Predictive performance in both the training and test sets was rigorously assessed by calibration curves. Figure\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e offers a visual representation of the intricate radiomics process, illustrating the successive steps of VOI delineation, meticulous feature extraction, principled dimensionality reduction, judicious feature selection, and the systematic construction of the model.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003eStatistical analysis\u003c/h2\u003e \u003cp\u003eStatistical analysis was performed using R 4.3.2 (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.r-project.org/\u003c/span\u003e\u003cspan address=\"https://www.r-project.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Continuous variables with normally distributed were expressed as mean\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviation, and categorical variables were presented as frequencies and percentages. Differences in characteristics between training and test sets were compared via the Chi-square test or Fisher\u0026rsquo;s exact test. Receiver operating characteristic (ROC) curves were used to evaluate the predictive performance of the models, and the area under the curve (AUC) was used for quantization. The calibration curve was used to test the degree of model calibration. A p\u0026thinsp;\u0026lt;\u0026thinsp;0.05 was considered as statistical significance.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eBaseline characteristics\u003c/h2\u003e \u003cp\u003eA total of 457 patients with pathological confirmed clinical stage I solid LADC were enrolled in the study. Among them, 95 cases (20.8%) were diagnosed with MP/S (+) pattern, while another 362 cases (79.2%) were MP/S (-). All the patients were divided into a training set with 304 patients from LZPH, and a test set with 153 patients from XTCH, respectively. Notably, no significant differences (p\u0026thinsp;\u0026gt;\u0026thinsp;0.05) were observed between these two sets (refer to Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\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\u003eComparative Evaluation of Clinico-radiological features between the Training and Test Sets\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\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 \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariables\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTotal (n\u0026thinsp;=\u0026thinsp;457)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTraining set (n\u0026thinsp;=\u0026thinsp;304)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTest set (n\u0026thinsp;=\u0026thinsp;153)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ep-value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePathological components, N (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.419\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMP/S (-) group\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e362 (79.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e237 (78)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e125 (81.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMP/S (+) group\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e95 (20.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e67 (22)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e28 (18.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLocation, N (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.947\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRUL\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e137 (30)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e93 (30.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e44 (28.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRLL\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e105 (23)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e69 (22.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e36 (23.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRML\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e39 (8.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e25 (8.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e14 (9.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLUL\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e107 (23.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e69 (22.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e38 (24.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLLL\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e69 (15.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e48 (15.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e21 (13.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBoundary, N (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.047\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIll-Defined\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e101 (22.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e76 (25)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e25 (16.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWell-Defined\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e356 (77.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e228 (75)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e128 (83.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eShape, N (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIrregular\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e259 (56.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e172 (56.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e87 (56.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOthers\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e198 (43.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e132 (43.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e66 (43.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLobulation, N (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.102\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAbsence\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e93 (20.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e69 (22.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e24 (15.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePresence\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e364 (79.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e235 (77.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e129 (84.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSpiculation, N (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.645\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAbsence\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e112 (24.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e77 (25.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e35 (22.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePresence\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e345 (75.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e227 (74.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e118 (77.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVascular Convergence Sign, N (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.09\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAbsence\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e129 (28.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e94 (30.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e35 (22.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePresence\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e328 (71.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e210 (69.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e118 (77.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVacuole Sign, N (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.054\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAbsence\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e370 (81)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e238 (78.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e132 (86.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePresence\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e87 (19)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e66 (21.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e21 (13.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePleural Indentation, N (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.649\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAbsence\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e77 (16.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e49 (16.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e28 (18.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePresence\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e380 (83.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e255 (83.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e125 (81.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSex, N (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.771\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e197 (43.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e133 (43.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e64 (41.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFemale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e260 (56.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e171 (56.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e89 (58.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge, Median (Q1, Q3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e61 (54, 67)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e61 (54, 67)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e62 (53, 68)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.47\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eClinical Stage, N (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.562\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ecT1a\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e37 (8.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e26 (8.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e11 (7.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ecT1b\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e188 (41.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e129 (42.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e59 (38.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ecT1c\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e232 (50.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e149 (49)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e83 (54.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"5\" nameend=\"c5\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAbbreviation\u003c/b\u003e: LUL Left Upper Lobe, LLL Left Lower Lobe, RUL Right Upper Lobe, RML Right Middle Lobe, RLL Right Lower Lobe, MP/S Micropapillary and/or Solid Components, LADC Lung Adenocarcinoma\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\u003eIn the training set, clinical baseline characteristics, including predominantly male sex and a higher T stage, were associated with the MP/S (+) pattern in patients with clinical stage I solid LADC. However, CT radiological features such as spiculation, shape, boundary, lobulation, vascular convergence signs, and vacuole signs showed no significant associations in differentiating MP/S (+) from MP/S (-) in clinical stage I solid LADC. More comprehensive information between the MP/S (+) and MP/S (-) groups please refer to Tabel S1.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eClin-Rad classification\u003c/h2\u003e \u003cp\u003eUnivariate and multivariate logistic analyses were conducted to discern various factors. Notably, CT morphological features showed no significant associations in differentiating MP/S (+) from MP/S (-) in clinical stage I solid LADC (p\u0026thinsp;\u0026gt;\u0026thinsp;0.05). However, clinical baseline information, including variables such as sex (OR: 0.891 (95% confidence interval [CI] 0.814\u0026ndash;0.976), p\u0026thinsp;=\u0026thinsp;0.013) and clinical stage (OR: 1.014 (95%CI 1.008\u0026ndash;1.021), p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), presented as independent risk factors for MP/S (+) (refer to Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). These identified risk factors were subsequently selected as predictive variables for the construction of the Clin-Rad classification. The AUC of the Clin-Rad classification was calculated to be 0.700 for the training set and 0.732 for the test set, respectively.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eLogistic Analysis for MP/S (+) in Clinical Stage I Solid Lung Adenocarcinoma\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" 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=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eVariables\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e \u003cp\u003eUnivariate analyses\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e \u003cp\u003eMultivariate analyses\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOdd Ratio (95%CI)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ep-value\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eOdd Ratio (95%CI)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003ep-value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLocation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.4 (0.7\u0026ndash;2.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.33\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBoundary\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.97 (0.52\u0026ndash;1.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.94\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eShape\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.72 (0.41\u0026ndash;1.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLobulation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.3 (0.65\u0026ndash;2.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.47\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSpiculation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.4 (0.71\u0026ndash;2.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.35\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVascular Convergence Sign\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.2 (0.64\u0026ndash;2.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.61\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVacuole Sign\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.84 (0.42\u0026ndash;1.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePleural Indentation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.8 (0.79\u0026ndash;4.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSex\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.47 (0.27\u0026ndash;0.82)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0075\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.891(0.814\u0026ndash;0.976)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.013\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1 (0.98-1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.44\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eClinical Stage\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.1 (1.1\u0026ndash;1.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.014(1.008\u0026ndash;1.021)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eITH score and C-radiomics score prediction model\u003c/h2\u003e \u003cp\u003eAfter conducting dimensionality reduction and feature selection using Lasso regression (refer to Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e and Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e), we incorporated 20 signatures (as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e) derived from the intratumoral sub-region and an additional 12 signatures (refer to Fig. \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e) derived from the entire tumor region in the construction of both the ITH score and C-radiomics score. The ITH score was significantly higher in the MP/S (+) group compared to the MP/S (-) group (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), as illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e. Notably, the C-radiomics score also demonstrated a higher value in the MP/S (+) group than in the MP/S (-) group (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), as shown in Fig. \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003e. The ITH score exhibited superior performance than C-radiomics score in differentiating MP/S (+) from MP/S (-) in clinical stage I solid LADC, as indicated by the AUC of ITH score (0.820 in the training set and 0.805 in the test set) and the AUC of C-radiomics score (0.810 in the training set and 0.771 in the test set).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe correlation heatmaps visually illustrate the intricate relationships among the ITH score, pathological components, and Clin-Rad features (refer to Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e6\u003c/span\u003e). These heatmaps reveal that a higher ITH score is associated with the MP/S (+) pattern, a male-dominated composition, a higher clinical stage, and more frequent invasive morphological features.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eHybrid model and nomogram\u003c/h2\u003e \u003cp\u003eThe ITH score, in conjunction with the Clin-Rad classification, was strategically leveraged to create the innovative hybrid model. Significantly, this hybrid model consistently exhibited robust predictive capabilities for identifying MP/S (+), achieving AUCs of 0.830 in the training set and 0.849 in the test sets (refer to Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e7\u003c/span\u003e). Furthermore, the hybrid model was transformed into a nomogram to quantify the risk of occurrence (Fig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e8\u003c/span\u003e). The calibration curve indicated good consistency between the predicted risk and the actual risk in both the training and test sets (Fig.\u0026nbsp;\u003cspan refid=\"Fig12\" class=\"InternalRef\"\u003e9\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eAssessing Diagnostic Performance for Predicting the MP/S (+) Pattern across Diverse Models\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\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=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eModels\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCohorts\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eArea under curve\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAccuracy\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSensitivity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eClin-Rad classification\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTraining set\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.700\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.664\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.701\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.654\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTest set\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.732\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.725\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.643\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.744\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eITH score\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTraining set\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.820\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.734\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.821\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.709\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTest set\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.805\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.810\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.607\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.856\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHybrid model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTraining set\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.830\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.730\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.851\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.696\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTest set\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.849\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.739\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.929\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.696\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\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe ITH score, derived from sub-regions, provides a more nuanced reflection of intratumoral ecological diversity compared to the C-radiomics score derived from the entire tumor. The hybrid model, comprising the ITH score and the Clin-Rad classification, demonstrated more comprehensive diagnostic efficiency in distinguishing MP/S (+) from MP/S (-) in clinical stage I solid LADC. To quantify the risk of occurrence, the hybrid model was translated into a nomogram, demonstrating commendable differentiation and calibration capabilities.\u003c/p\u003e \u003cp\u003eSeveral previous studies have utilized Clin-Rad features to construct models for predicting MP/S (+) from MP/S (-) in clinical stage I solid LADC. In a study conducted by Dong et al. [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e], tumor size, density, and lobulation were identified as risk factors for predicting MP/S (+). However, in this current study, CT radiological features such as spiculation, shape, boundary, lobulation, vascular convergence signs, and vacuole signs did not show significant associations in differentiating MP/S (+) from MP/S (-) in clinical stage I solid LADC. We hypothesize that these morphological characteristics are not adequately representative, and their performance has not been sufficiently evaluated, exhibiting significant variability among observers and leading to poor repeatability.\u003c/p\u003e \u003cp\u003eRadiomics enables the extraction of high-throughput features from medical images, thereby assisting in the identification of histological heterogeneity. This process enables the identification of additional factors that may not be readily discernible through visual inspection. Numerous previously published radiomics studies have investigated the significance of preoperative categorization based on the histological subtypes of LADC [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan additionalcitationids=\"CR20 CR21\" citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. However, these studies included LADC patients, covering SPN and GGN types of lung nodules observed in CT scans. Subgroup analyses for SPN or GGN types were not conducted. Due to the high heterogeneity and invasiveness, the MP/S (+) pattern more commonly presents as SPN in CT images [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. This observation leads us to infer that a higher frequency of SPN in the enrolled population could significantly improve the diagnostic efficiency of radiomics, potentially resulting in model overfitting. Furthermore, these aforementioned radiomics studies mainly capture characteristics of the entire tumor, neglecting intratumoral ecological diversity. Within our investigative framework, we analyzed ITH within subregions derived from CT images to predict histological subtype categorization. To the best of our knowledge, this marks the first report on the ITH score for predicting MP/S (+) from MP/S (-) in clinical stage I solid LADC.\u003c/p\u003e \u003cp\u003eIn the current investigation, the ITH score has demonstrated superior efficacy in discerning between MP/S (+) and MP/S (-) in clinical stage I solid LADC, outperforming the C-radiomics score and Clin-Rad classification. This discovery aligns seamlessly with the conclusions drawn by Li et al. [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e], indicating that the ITH score provides a more intricate representation of intratumoral ecological diversity compared to the C-radiomics score. In Li et al.'s scholarly work, the ITH score manifested correlations with various tumor phenotypes, including lymphovascular invasion and pleural invasion, significantly influencing prognosis of individuals with non-small cell lung cancer. To further explore the relationship between the ITH score and tumor heterogeneity and biological behavior, this study employed correlation heatmaps, revealing that a higher ITH score is concomitant with the MP/S (+) pattern, a predominance of male-dominant characteristics, an advanced clinical stage, and increased occurrences of invasive morphological features.\u003c/p\u003e \u003cp\u003eFinally, acknowledging the diagnostic significance of Clin-Rad in histological subtypes, we combined it with the ITH score to construct a hybrid model. Notably, this hybrid model consistently demonstrated robust predictive capabilities in identifying MP/S (+), achieving AUCs of 0.830 in the training set and 0.849 in the test sets. The hybrid model was subsequently transformed into a nomogram to quantify the risk of occurrence. The calibration curve attested to the commendable consistency between the predicted risk and the actual risk in both the training and test sets. This may be attributed to the incorporation of the hybrid model, which integrates multi-domain and abundant information, thereby enhancing diagnostic efficiency.\u003c/p\u003e \u003cp\u003eOur investigation has encountered certain limitations. Firstly, it is crucial to acknowledge the retrospective nature of our research, in which potential selection bias arises from exclusively including patients with post-surgery pathological results. Secondly, the bicentric design introduces variability in acquisition parameters, image quality, and the potential for co-registration errors. This presents sources of variability that could confound the analysis. Additionally, the presence of artifacts and technical constraints may impact the reliability and reproducibility of radiomic features. Thirdly, the relatively short follow-up period after surgery prevented the development of a predictive model for disease outcomes. Lastly, it is essential to highlight that our study specifically focused on clinical stage I solid LADC, which may limit the generalizability of our findings to other stages.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn conclusion, the ITH within sub-regions proved to be a reliable predictor for MP/S (+) in clinical stage I solid LADC. We integrated the Clin-Rad classification with the ITH score to formulate a hybrid model, which was then transformed into a nomogram. This nomogram accurately quantifies the risk of MP/S (+). The nomogram demonstrated exceptional discrimination and calibration, indicating the potential to make a significant contribution to clinical decision-making processes.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cdiv class=\"DefinitionList\"\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eITH\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eIntratumoral heterogeneity quantification\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eMP/S\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eMicropapillary and/or solid components\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eLADC\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eLung adenocarcinoma\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eClin-Rad\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eClinical-radiological\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eROC\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eReceiver operating characteristic\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eAUC\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eArea under the curve\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eGGN\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eGround glass nodules\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eSPN\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eSolid pulmonary nodules\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eLZPH\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eLiuzhou People\u0026rsquo;s Hospital\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eXTCH\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eXiangtan Central Hospital\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eVOI\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eVolume of interest\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003emRMR\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eMax-relevance and Min-Redundancy\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eLASSO\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eLeast absolute shrinkage and selection operator\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eAUC\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eThe area under the curve\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eCI\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eConfidence interval\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eEthical approval was obtained from Medical Ethics Committee of Xiangtan Central Hospital (reference number: No. 2021-07-009), which ensured that the rights and interests of the research participants were not compromised. All methods were performed following the relevant guidelines and regulations. Medical Ethics Committee of Xiangtan Central Hospital approved the retrospective study and waived the requirement for informed consent.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors' contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eZ.Z.: Conceptualization, Methodology, Original Draft Writing.\u003c/p\u003e\n\u003cp\u003eJ.D.: Review \u0026amp; Editing of Writing, Supervision.\u003c/p\u003e\n\u003cp\u003eW.G.: Software Development, Formal Analysis, Data Curation.\u003c/p\u003e\n\u003cp\u003eY.Z.: Validation, Provision of Resources.\u003c/p\u003e\n\u003cp\u003eH.L.: Validation, Resource Allocation, Statistical Analysis, and Software Development.\u003c/p\u003e\n\u003cp\u003eW.Z. and Y.Z.: Theoretical Guidance, Suggestions for Article Revision, Supportive Contributions.\u003c/p\u003e\n\u003cp\u003eAll authors contributed to the article and approved the submitted version.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eSung H, Ferlay J, Siegel RL, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin. 2021;71(3):209\u0026ndash;49. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://doi.org/10.3322/caac.21660\u003c/span\u003e\u003cspan address=\"10.3322/caac.21660\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNicholson AG, Tsao MS, Beasley MB, et al. The 2021 WHO Classification of Lung Tumors: Impact of Advances Since 2015. J Thorac Oncol. 2022;17(3):362\u0026ndash;87. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://doi.org/10.1016/j.jtho.2021.11.003\u003c/span\u003e\u003cspan address=\"10.1016/j.jtho.2021.11.003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHung JJ, Jeng WJ, Chou TY, et al. Prognostic value of the new International Association for the Study of Lung Cancer/American Thoracic Society/European Respiratory Society lung adenocarcinoma classification on death and recurrence in completely resected stage I lung adenocarcinoma. Ann Surg. 2013;258(6):1079\u0026ndash;86. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://doi.org/10.1097/SLA.0b013e31828920c0\u003c/span\u003e\u003cspan address=\"10.1097/SLA.0b013e31828920c0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTsubokawa N, Mimae T, Sasada S, et al. Negative prognostic influence of micropapillary pattern in stage IA lung adenocarcinoma. Eur J Cardiothorac Surg. 2016;49(1):293\u0026ndash;9. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://doi.org/10.1093/ejcts/ezv058\u003c/span\u003e\u003cspan address=\"10.1093/ejcts/ezv058\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYanagawa N, Shiono S, Abiko M, et al. The Clinical Impact of Solid and Micropapillary Patterns in Resected Lung Adenocarcinoma. J Thorac Oncol. 2016;11(11):1976\u0026ndash;83. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://doi.org/10.1016/j.jtho.2016.06.014\u003c/span\u003e\u003cspan address=\"10.1016/j.jtho.2016.06.014\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMikubo M, Tamagawa S, Kondo Y, et al. Micropapillary and solid components as high-grade patterns in IASLC grading system of lung adenocarcinoma: Clinical implications and management. Lung Cancer. 2024;187:107445. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://doi.org/10.1016/j.lungcan.2023.107445\u003c/span\u003e\u003cspan address=\"10.1016/j.lungcan.2023.107445\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu W, Zhang Q, Zhang T, Li L, Xu C. Minor histological components predict the recurrence of patients with resected stage I acinar- or papillary-predominant lung adenocarcinoma. Front Oncol. 2022;12:1090544. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://doi.org/10.3389/fonc.2022.1090544\u003c/span\u003e\u003cspan address=\"10.3389/fonc.2022.1090544\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen C, Chen ZJ, Li WJ, et al. Impact of minimal solid and micropapillary components on invasive lung adenocarcinoma recurrence. Ann Diagn Pathol. 2022;59:151945. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://doi.org/10.1016/j.anndiagpath.2022.151945\u003c/span\u003e\u003cspan address=\"10.1016/j.anndiagpath.2022.151945\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuang W, Zhang H, Zhang Z, et al. A prognostic nomogram based on a new classification of combined micropapillary and solid components for stage IA invasive lung adenocarcinoma. J Surg Oncol. 2022;125(4):796\u0026ndash;808. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://doi.org/10.1002/jso.26760\u003c/span\u003e\u003cspan address=\"10.1002/jso.26760\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVan Schil PE, Asamura H, Rusch VW, et al. Surgical implications of the new IASLC/ATS/ERS adenocarcinoma classification. Eur Respir J. 2012;39(2):478\u0026ndash;86. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://doi.org/10.1183/09031936.00027511\u003c/span\u003e\u003cspan address=\"10.1183/09031936.00027511\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYeh YC, Nitadori J, Kadota K, et al. Using frozen section to identify histological patterns in stage I lung adenocarcinoma of \u0026le;\u0026thinsp;3 cm: accuracy and interobserver agreement. Histopathology. 2015;66(7):922\u0026ndash;38. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://doi.org/10.1111/his.12468\u003c/span\u003e\u003cspan address=\"10.1111/his.12468\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTrejo Bittar HE, Incharoen P, Althouse AD, Dacic S. Accuracy of the IASLC/ATS/ERS histological subtyping of stage I lung adenocarcinoma on intraoperative frozen sections. Mod Pathol. 2015;28(8):1058\u0026ndash;63. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://doi.org/10.1038/modpathol.2015.71\u003c/span\u003e\u003cspan address=\"10.1038/modpathol.2015.71\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLederlin M, Puderbach M, Muley T, et al. Correlation of radio- and histomorphological pattern of pulmonary adenocarcinoma. Eur Respir J. 2013;41(4):943\u0026ndash;51. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://doi.org/10.1183/09031936.00056612\u003c/span\u003e\u003cspan address=\"10.1183/09031936.00056612\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDong H, Yin LK, Qiu YG, et al. Prediction of high-grade patterns of stage IA lung invasive adenocarcinoma based on high-resolution CT features: a bicentric study. Eur Radiol. 2023;33(6):3931\u0026ndash;40. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://doi.org/10.1007/s00330-022-09379-x\u003c/span\u003e\u003cspan address=\"10.1007/s00330-022-09379-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu Y, Chang Y, Zha X, et al. A Combination of Radiomic Features, Imaging Characteristics, and Serum Tumor Biomarkers to Predict the Possibility of the High-Grade Subtypes of Lung Adenocarcinoma. Acad Radiol. 2022;29(12):1792\u0026ndash;801. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://doi.org/10.1016/j.acra.2022.02.024\u003c/span\u003e\u003cspan address=\"10.1016/j.acra.2022.02.024\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDong H, Wang X, Qiu Y, et al. Establishment and visualization of a model based on high-resolution CT qualitative and quantitative features for prediction of micropapillary or solid components in invasive lung adenocarcinoma. J Cancer Res Clin Oncol. 2023;149(12):10519\u0026ndash;30. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://doi.org/10.1007/s00432-023-04854-4\u003c/span\u003e\u003cspan address=\"10.1007/s00432-023-04854-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi J, Qiu Z, Zhang C, et al. ITHscore: comprehensive quantification of intra-tumor heterogeneity in NSCLC by multi-scale radiomic features. Eur Radiol. 2023;33(2):893\u0026ndash;903. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://doi.org/10.1007/s00330-022-09055-0\u003c/span\u003e\u003cspan address=\"10.1007/s00330-022-09055-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu Y, Li Z, Xiong H, et al. Understanding and enhancement of internal clustering validation measures. IEEE Trans Cybern. 2013;43(3):982\u0026ndash;94. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://doi.org/10.1109/tsmcb.2012.2220543\u003c/span\u003e\u003cspan address=\"10.1109/tsmcb.2012.2220543\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSong SH, Park H, Lee G, et al. Imaging Phenotyping Using Radiomics to Predict Micropapillary Pattern within Lung Adenocarcinoma. J Thorac Oncol. 2017;12(4):624\u0026ndash;32. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://doi.org/10.1016/j.jtho.2016.11.2230\u003c/span\u003e\u003cspan address=\"10.1016/j.jtho.2016.11.2230\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePark S, Lee SM, Noh HN, et al. Differentiation of predominant subtypes of lung adenocarcinoma using a quantitative radiomics approach on CT. Eur Radiol. 2020;30(9):4883\u0026ndash;92. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://doi.org/10.1007/s00330-020-06805-w\u003c/span\u003e\u003cspan address=\"10.1007/s00330-020-06805-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen LW, Yang SM, Wang HJ, et al. Prediction of micropapillary and solid pattern in lung adenocarcinoma using radiomic values extracted from near-pure histopathological subtypes. Eur Radiol. 2021;31(7):5127\u0026ndash;38. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://doi.org/10.1007/s00330-020-07570-6\u003c/span\u003e\u003cspan address=\"10.1007/s00330-020-07570-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXu Y, Ji W, Hou L, et al. Enhanced CT-Based Radiomics to Predict Micropapillary Pattern Within Lung Invasive Adenocarcinoma. Front Oncol. 2021;11:704994. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://doi.org/10.3389/Fonc.2021.704994\u003c/span\u003e\u003cspan address=\"10.3389/Fonc.2021.704994\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLegend.\u003c/span\u003e\u003c/li\u003e\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":"bmc-cancer","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bcan","sideBox":"Learn more about [BMC Cancer](http://bmccancer.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bcan/default.aspx","title":"BMC Cancer","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Intratumoral heterogeneity, Sub-regions, High-grade patterns, Clinical stage I, Solid lung adenocarcinoma","lastPublishedDoi":"10.21203/rs.3.rs-3910257/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3910257/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground: \u003c/strong\u003eThis study aims to quantify intratumoral heterogeneity(ITH) using preoperative CT scans and evaluate its ability to predict pathological high-grade patterns, specifically micropapillary and/or solid components (MP/S), in patients diagnosed with clinical stage I solid lung adenocarcinoma (LADC).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods:\u003c/strong\u003e In this retrospective study, 457 patients postoperatively diagnosed with clinical stage I solid LADC were included from two medical centers, comprising a training set (center 1, n=304) and a test set (center 2, n=153). Sub-regions within the tumor were identified using the K-means method. Both intratumoral ecological diversity features (hereafter referred to as ITH) and conventional radiomics (hereafter referred to as C-radiomics) were extracted to generate ITH scores and C-radiomics scores. Next, univariate and multivariate logistic regression analyses were employed to identify clinical-radiological (Clin-Rad) features associated with the MP/S (+) group for constructing the Clin-Rad classification. Subsequently, a hybrid model which presented as a nomogram was developed, integrating the Clin-Rad classification and ITH score. The performance of models was assessed using the receiver operating characteristic (ROC) curves, and the area under the curve (AUC), accuracy, sensitivity, and specificity were determined.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults: \u003c/strong\u003eThe ITH score outperformed both C-radiomics scores and Clin-Rad classification, as indicated by higher AUC values in the training (0.820 versus 0.810 and 0.700) and test sets (0.805 versus 0.771 and 0.732), respectively. Notably, the hybrid model consistently demonstrated robust predictive capabilities in identifying MP/S (+), achieving AUCs of 0.830 in the training set and 0.849 in the test sets.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion:\u003c/strong\u003e The ITH of sub-regions within the intratumor has been shown to be a reliable predictor for MP/S (+) in clinical stage I solid LADC. This finding holds the potential to make a significant contribution to clinical decision-making processes.\u003c/p\u003e","manuscriptTitle":"Quantifying intratumoral heterogeneity within sub-regions to predict high-grade patterns in clinical stage I solid lung adenocarcinoma","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-02-12 21:41:14","doi":"10.21203/rs.3.rs-3910257/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-02-09T04:50:28+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-02-08T12:15:55+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-02-08T12:15:55+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Cancer","date":"2024-01-30T10:08:35+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-cancer","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bcan","sideBox":"Learn more about [BMC Cancer](http://bmccancer.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bcan/default.aspx","title":"BMC Cancer","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"e43ac359-8452-4790-b992-52fc285657de","owner":[],"postedDate":"February 12th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-01-13T15:59:36+00:00","versionOfRecord":{"articleIdentity":"rs-3910257","link":"https://doi.org/10.1186/s12885-025-13445-0","journal":{"identity":"bmc-cancer","isVorOnly":false,"title":"BMC Cancer"},"publishedOn":"2025-01-09 15:57:05","publishedOnDateReadable":"January 9th, 2025"},"versionCreatedAt":"2024-02-12 21:41:14","video":"","vorDoi":"10.1186/s12885-025-13445-0","vorDoiUrl":"https://doi.org/10.1186/s12885-025-13445-0","workflowStages":[]},"version":"v1","identity":"rs-3910257","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3910257","identity":"rs-3910257","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}