Intitulé du sujet: Multimodal Transcriptomic and Liquid Biopsy Approaches to Predict Prognosis and Chemotherapy Response in Stage II and III Colon Cancer

Sujet

Codirection:

Nombre de mois: 48 mois

Ecole Doctorale: ED 563 - Médicament,Toxicologie, Chimie, Imageries

Unité de recherche et équipe:

Coordonnées de l’équipe:

Secteur: Sciences de la vie / Life Sciences

Langue attendue: Anglais

Niveau de langue attendu: B2

Description

Description du sujet:

Multimodal Transcriptomic and Liquid Biopsy Approaches to Predict Prognosis and Chemotherapy Response in Stage II and III Colon Cancer:

Insights from PRODIGE-13, CIRCULATE-PRODIGE 70, PETACC-8, and IDEA France Trials

This project aims to leverage multiple high-quality colon cancer (CC) datasets to discover and validate transcriptomic signatures that can predict recurrence and clinical outcomes in stage II and III CC. The primary cohort for this project is the PRODIGE-13 clinical trial, where we will conduct RNA sequencing (RNAseq) on stage III CC samples, building on the existing stage II RNAseq data. By integrating insights from tumor biology and the tumor microenvironment (TME), we aim to develop models that inform patient stratification and personalize treatment approaches.

1. Background

Colon cancer (CC) is a frequent and severe disease with more than 1.36 million new cases worldwide (1) making it the second most common cancer. Approximately 70% to 80% of patients with colon cancer are diagnosed without metastatic spread and may be cured by surgery. However, 10% to 60% of them will experience disease recurrence depending on the disease stage. To improve long-term outcomes, adjuvant chemotherapy is used in patients with a substantial risk of disease recurrence. Although all patients receiving adjuvant treatment are exposed to short-term and long-lasting toxicities, only 25% of these patients experience therapeutic benefit. The heterogeneity of clinical responses points to a biological heterogeneity of these cancers, which needs to be understood to improve patient clinical management.

Recently, our team contributed to a consensus stratification of colon cancer (CC) based on transcriptional profiles, resulting in a molecular classification system that categorizes most tumors into four robust consensus molecular subtypes (CMS): CMS1, CMS2, CMS3, and CMS4 (2).

• CMS1: Poorly or undifferentiated proximal CC, strongly associated with MSI, CIMP phenotype, and frequent BRAF mutations.
• CMS2: Epithelial type with high somatic copy number alteration (SCNA), MSS, frequent TP53 mutations, and activation of the WNT/MYC pathway.
• CMS3: Epithelial type with MSS (90%) or MSI (10%), low SCNA, frequent KRAS mutations, and slight prevalence of CIMP-low phenotypes, involving metabolic pathway dysregulation.
• CMS4: High SCNA, MSS status, mesenchymal phenotype, and significant stromal infiltration.

The CMS classification has strong prognostic significance. CMS1 tumors exhibit good progression-free survival (PFS) but poor survival after relapse (SAR). CMS2 tumors have favorable outcomes in terms of OS, PFS, and SAR. CMS3 is linked to intermediate OS, while CMS4 has the worst OS and PFS among the subtypes (2). Our team validated the prognostic relevance of CMS classification in stage III CC patients (n=2045) from the PETACC-8 trial (3), which compared FOLFOX chemotherapy with and without cetuximab (4).

Furthermore, we demonstrated that CMS correlates with tumor immune contexture (5). CMS1 tumors have strong immune and inflammatory infiltration, marked by activated CD8+ and NK cells. CMS2 and CMS3 tumors exhibit low lymphocytic and myeloid expression, while CMS4 shows markers of lymphoid and monocytic cells alongside angiogenic, inflammatory, and immunosuppressive signatures (5).

• Intra-tumor CMS heterogeneity of colon cancers

Our team has leveraged the PETACC-8 cohort to refine the CMS classification by applying the WISP (Weighted In Silico Pathology) deconvolution algorithm, (6) that estimates the proportions of each CMS components within a given sample. Using this pipeline, we discovered that over 50% of tumors are heterogeneous, i.e. that there are at least two weights above 20%, and that the prognosis of the tumors is significantly different according to their heterogeneity (7).

Recently, a study conducted by our team established prognostic models for stage III CC based on transcriptomic signatures of the tumor microenvironment (TME) and cell cycle, utilizing data from the PETACC-8 (training set) and IDEA-France (validation set) trials (3,8). Using 3’RNA sequencing data from these two large, randomized phase III trials, we developed an “Immune Proliferative Stromal” (IPS) score. This score, derived from four transcriptomic signatures reflecting the TME and cell cycle, enabled effective stratification of patients based on their risk of recurrence, complementing well-established prognostic factors. Overexpressed signatures linked to T-cell and B-cell/TLS infiltration and cell cycle activity were associated with favorable prognosis, whereas signatures indicating stroma and M2 macrophage infiltration were associated with poorer outcomes in both independent patient cohorts (Gallois et al. in revision).

• Spatial transcriptomic and intratumor heterogeneity.

The recent advancements in spatial transcriptomic technologies have marked the beginning of a new era in the characterization of tissue architecture (9). Our team has performed Visium analyses on more than 60 samples and Visium HD on 4 samples from the PETACC-8 trial. The samples were selected to include the most representative tumors of both pure and heterogeneous CMS subtypes. Spatial analyses are being utilized to dissect the diversity of cell types and states in relation to their spatial distribution within the tumor. This approach provides insights into the proximity of tumor cells in specific states to particular cell types within the TME, such as endothelial cells, defined cancer-associated fibroblast (CAF) subtypes, or immune cells, and allowed to identify spatial transcriptomic-derived signatures that capture spatial heterogeneity within tumor cells and the TME.

2. Patient cohorts

This research leverages data from four large phase III trials: PRODIGE-13, CIRCULATE-PRODIGE 70, PETACC-8, and IDEA France, which provide unique opportunities for transcriptomic and molecular analyses in localized CC.

• PRODIGE-13: A recent randomized multicenter trial that compared 5-year overall survival (OS) following intensive radiological monitoring versus a lower-intensity program in more than 2,000 patients with stage II and III CC (10).
• CIRCULATE-PRODIGE 70 trial: A multicenter phase III study involving approximately 2,640 stage II CC patients, exploring the efficacy of circulating tumor DNA (ctDNA) screening combined with modified FOLFOX6 chemotherapy (FOLFOX6m) (11).
• PETACC-8: A large pan-European adjuvant colon cancer trial including 2,045 stage III CC patients and comparing 2 different regimens of chemotherapy in adjuvant setting (FOLFOX versus FOLFOX ± cetuximab) (3).
• IDEA France: A randomized trial including 2000 stage III CC patients, comparing 6 months versus 3 months of FOLFOX / CAPOX (investigators’ choice for chemotherapy regimen) (8).

For these trials, extensive clinical, biological, and molecular data are available, including tumor mutational status (RAS, BRAF), microsatellite instability (MMR status), CpG island methylator phenotype (CIMP) status, and survival data (disease-free survival [DFS], OS, and specific adverse events [SAR]).

To date, 3’RNA sequencing has been performed on formalin-fixed paraffin-embedded (FFPE) samples for PETACC-8 (n=1,733), IDEA-France (n=1,263), and PRODIGE-13 (subset of stage II patients n=471).

In addition, for CIRCULATE-PRODIGE 70 we will have access to whole exome sequencing (WES), whole genome sequencing (WGS), and RNAseq data for at least 1,000 tumor samples, along with corresponding ctDNA information.

3. Objectives

The overarching goal of this project is to leverage these rich datasets to redefine molecular classifications, characterize tumor heterogeneity, and identify novel prognostic markers in localized colon cancer.

WISP deconvolution analyses revealed the presence of distinct cellular populations corresponding to different CMS subgroups or homogeneous populations exhibiting overlapping molecular characteristics. By performing tissue- and spatial-level analyses of tumors with mixed CMS phenotypes, we aim to unravel the mechanisms underlying intra-tumor heterogeneity. We will focus on identifying transcriptomic signatures that predict recurrence risk and clinical outcomes, and explore the roles of immune infiltration, stromal components, and cell cycle-related signatures in predicting time to recurrence (TTR).

• Transcriptomic Profiling for PRODIGE-13

The project will initially focus on the PRODIGE-13 trial (27), with the aim of performing 3’RNAseq on 500 stage III CC cases, complementing the existing transcriptomic data generated from 471 stage II cases.

• CMS and WISP Heterogeneity characterization in PRODIGE-13 and CIRCULATE-PRODIGE 70 trials

The molecular CMS classification will be determined using bulk 3’RNAseq in the samples from PRODIGE-13 and CIRCULATE-PRODIGE 70 trials. Correlations between each CMS subtype and clinical and molecular factors will be determined (with gender, age, tumor sidedness, RAS/BRAF mutational status, MMR status, CIMP status). CMS group will be determined using the R packages CMS classifier and CMS caller (2,12). Characterization of the intra tumoral heterogeneity (ITH) of CMS subtypes will be performed by WISP method using the R package « WISP » on the expressions determined by bulk 3’RNAseq.

• Validation of Visium-Derived Signatures

We will integrate findings from spatial transcriptomics conducted on PETACC-8 samples: more than 60 Small Visium samples and 4 Visium HD samples. Spatial transcriptomics-derived signatures, which capture spatial heterogeneity within the tumor cells and TME will be validated against large RNAseq datasets to assess their applicability to bulk transcriptomics and their relevance for predicting clinical outcomes.

• Integration of Circulating DNA Data

To complement tissue-based analyses, we will incorporate ctDNA data from CIRCULATE-PRODIGE 70. By integrating WES, WGS, and RNAseq data with ctDNA dynamics, we aim to explore correlations between transcriptomic profiles, ctDNA levels, recurrence, and minimal residual disease, following the example of our recent study on the IDEA France cohort (Gallois et al. in revision). This liquid biopsy approach will provide a broader view of tumor biology and support non-invasive prognostic and predictive biomarker development.

• Development of a new prognostic molecular classification

Leveraging the PRODIGE-13 dataset, we aim to develop a novel transcriptomic classification centered on survival outcomes. Building upon the IPS (Immune-Proliferative-Stromal) score established in PETACC-8, which stratifies patients by recurrence risk using four key transcriptomic signatures (Gallois et al. in revision), we will adapt and expand this approach to identify new prognostic markers within the PRODIGE-13 cohort. These markers will serve as critical tools to inform and optimize adjuvant chemotherapy strategies. The robustness and clinical utility of this novel classification will be validated across the other colon cancer cohorts, encompassing approximately 5,000 patients, providing a comprehensive and evidence-based framework to refine prognostication and personalize treatment for localized colon cancer.

4. Future Steps: Personalized Care Based on Tumor Biology

This project ultimately aims to pave the way for personalized treatment strategies in colon cancer. By integrating bulk and spatial transcriptomic data with ctDNA analyses, we will generate predictive models that go beyond traditional prognostic factors, such as TNM staging. These models will incorporate detailed transcriptomic profiles to improve risk stratification and optimize treatment decisions, potentially enabling de-escalation or intensification of therapy based on individual tumor biology.

Through the comprehensive characterization of molecular subtypes, tumor heterogeneity, and spatial dynamics, this work will contribute to advancing precision oncology and improving outcomes for colon cancer patients.

Compétences requises:

Academic Background

• A Master’s degree in Medicine with a specialization in Gastrointestinal Surgery or Oncology.
• Strong foundational knowledge in colorectal cancer (CRC) biology and clinical management, particularly in stages II and III.

Technical and Research Skills

1. Clinical Expertise:

• Familiarity with diagnostic and prognostic approaches for colorectal cancer.

2. Molecular Biology and Bioinformatics:

• Basic understanding of transcriptomic data analysis, including RNA sequencing (RNAseq).
• Familiarity with bioinformatics tools and pipelines (e.g., R Studio) for data analysis and visualization.
• Eagerness to acquire and apply computational skills for bulk and spatial transcriptomic analyses.

3. Data Integration and Interpretation:

• Understanding of tumor biology, particularly in colorectal cancer (CRC), including molecular subtypes (e.g., CMS classification), tumor microenvironment (TME), and immune infiltration.
• Familiarity with integrating clinical parameters (e.g., TNM staging, survival outcomes) with molecular data.
• Ability to interpret clinical and molecular data to draw meaningful conclusions about tumor biology and treatment responses.

Language and Communication Skills

• Proficiency in English, both spoken and written, necessary to contribute to academic discussions, write scientific articles, and present findings at international conferences.
• Knowledge of French is advantageous but not required.
• Clear and effective communication with multidisciplinary teams, including biologists, clinicians, and computational scientists.

Research and Organizational Skills

• Ability to manage large-scale projects and collaborate in multidisciplinary teams.
• Experience with writing research papers or presenting findings at conferences.
• Strong organizational and time-management skills to meet the demands of a complex, multi-cohort research project.

Références bibliographiques:

1. Allemani C, Matsuda T, Di Carlo V, Harewood R, Matz M, Nikšić M, et al. Global surveillance of trends in cancer survival 2000-14 (CONCORD-3): analysis of individual records for 37 513 025 patients diagnosed with one of 18 cancers from 322 population-based registries in 71 countries. Lancet Lond Engl. 17 mars 2018;391(10125):1023‑75.
2. Guinney J, Dienstmann R, Wang X, de Reyniès A, Schlicker A, Soneson C, et al. The consensus molecular subtypes of colorectal cancer. Nat Med. nov 2015;21(11):1350‑6.
3. Taieb J, Tabernero J, Mini E, Subtil F, Folprecht G, Van Laethem JL, et al. Oxaliplatin, fluorouracil, and leucovorin with or without cetuximab in patients with resected stage III colon cancer (PETACC-8): an open-label, randomised phase 3 trial. Lancet Oncol. 1 juill 2014;15(8):862‑73.
4. Marisa L, Ayadi M, Balogoun R, Pilati C, Le Malicot K, Lepage C, et al. Clinical utility of colon cancer molecular subtypes: Validation of two main colorectal molecular classifications on the PETACC-8 phase III trial cohort. J Clin Oncol. 20 mai 2017;35(15_suppl):3509‑3509.
5. Becht E, de Reyniès A, Giraldo NA, Pilati C, Buttard B, Lacroix L, et al. Immune and Stromal Classification of Colorectal Cancer Is Associated with Molecular Subtypes and Relevant for Precision Immunotherapy. Clin Cancer Res Off J Am Assoc Cancer Res. 15 août 2016;22(16):4057‑66.
6. Blum Y, Meiller C, Quetel L, Elarouci N, Ayadi M, Tashtanbaeva D, et al. Dissecting heterogeneity in malignant pleural mesothelioma through histo-molecular gradients for clinical applications. Nat Commun. 22 mars 2019;10(1):1333.
7. Marisa L, Blum Y, Taieb J, Ayadi M, Pilati C, Le Malicot K, et al. Intratumor CMS Heterogeneity Impacts Patient Prognosis in Localized Colon Cancer. Clin Cancer Res Off J Am Assoc Cancer Res. 1 sept 2021;27(17):4768‑80.
8. André T, Vernerey D, Mineur L, Bennouna J, Desrame J, Faroux R, et al. Three Versus 6 Months of Oxaliplatin-Based Adjuvant Chemotherapy for Patients With Stage III Colon Cancer: Disease-Free Survival Results From a Randomized, Open-Label, International Duration Evaluation of Adjuvant (IDEA) France, Phase III Trial. J Clin Oncol Off J Am Soc Clin Oncol. 20 2018;36(15):1469‑77.
9. Rao A, Barkley D, França GS, Yanai I. Exploring tissue architecture using spatial transcriptomics. Nature. août 2021;596(7871):211‑20.
10. Lepage C, Phelip JM, Cany L, Faroux R, Manfredi S, Ain JF, et al. Effect of 5 years of imaging and CEA follow-up to detect recurrence of colorectal cancer: The FFCD PRODIGE 13 randomised phase III trial. Dig Liver Dis. 1 juill 2015;47(7):529‑31.
11. Taïeb J, Benhaim L, Laurent Puig P, Le Malicot K, Emile JF, Geillon F, et al. Decision for adjuvant treatment in stage II colon cancer based on circulating tumor DNA:The CIRCULATE-PRODIGE 70 trial. Dig Liver Dis Off J Ital Soc Gastroenterol Ital Assoc Study Liver. juill 2020;52(7):730‑3.
12. Eide PW, Bruun J, Lothe RA, Sveen A. CMScaller: an R package for consensus molecular subtyping of colorectal cancer pre-clinical models. Sci Rep. 30 nov 2017;7:16618.