The Immunoscore as a Prognostic Marker for Patients With a Colorectal Cancer

The main objective is to assess the impact of the immunoscore in the clinical practice. To this end, the immune parameter (CD3, CD8, CD45RO) of colorectal cancers (stage I-IV) will be evaluated prospectively through a large multicenter study. 400 patients with CRC stage I to IV will be included (6 centers for inclusion; Paris-HEGP, Dijon, Bobigny-Avicenne, Besancon, Poitiers, Rouen). The number of patients to be included has been calculated on the basis of an inclusion period of 24 months, a follow-up of 3 years (a collection of data during the 4th and 5th year is planned with a co-financing), an expected event rate of 20%, a prognostic value of the immune parameter similar to that observed in the retrospective study of 399 patients with CRC (*1) and a power of the test of 90%. During both initial management and follow-up, all therapeutic procedures will be registered (surgery, radiotherapy, chemotherapy…). Data will be collected from clinical examination, radiological exams and biological tests that are usually performed during a hospitalization for colorectal cancer: Full clinical examination and questioning about the presence of personal and family history of CRC, adenomas and other cancers; Abdominal ultrasound, chest X-ray; Computed tomography (CT) and magnetic resonance imaging (MRI) according to clinician's requirement; Standard laboratory tests and blood assay for carcinoembryonic antigen (CEA); The Histopathologic study of the tumor containing the status of the surgical excision margins, the histological type of cancer, the level of parietal invasion, the number of lymph nodes examined and the number of metastatic lymph nodes, according to the TNM classification. Patients will be followed-up every 3 months during the first 2 years, and every 6 months for the rest of the follow-up period, as recommended. Each patient will be followed up to 5 years. Clinical, biological and radiological data will be registered into a specific CRF. Verification of inclusions and monitoring of data will be conducted by the HEGP Clinical Research Unit (URC HEGP) and the principal investigator. A plan for data-management will be built in line with the monitoring. A questionnaire will also be sent to the patients every six month for five years. This questionnaire was built by Pr F Pagès and by psychologists and psychiatrists to search for information concerning (i) the presence of an immune disorder (such as allergy, autoimmunity, inflammatory process) and (ii) the psychological status of the patients (eg. depression). After monitoring, all data will be stored using a double data entry system in a MS-Access database designed by the URC-HEGP and in the TME.dB database created by the team of the principal investigator. The 2 databases will be merged, and the final data base will be frozen for statistical analysis. Sample preparation: The pathologist of each center will choose a tumor block containing the core of the tumor and the invasive margin. This tumor block is part of the tumor samples routinely processed by the department of Pathology for diagnosis. three tissue sections of 4-microns will be made and deposit on a glass slide to perform the immunohistochemistry. Four tissue sections of 4-microns will be made and deposit on an Eppendorf safe lock tubes to perform the genetic analyses. Anonymous patient names on the slides and tubes will be provided by using a code number and initials of the patient. If biopsies for diagnostic purposes are available (i.e. those performed for the patients during their initial hospitalization (50% of cases), A tissue sections of each biopsy will be done. All slides and tubes will be sent by mail to the HEGP platform for analysis of the immune infiltrate and to the Biology department for the genetic analyses. The immunoscore will be determine using the following steps: Immunostainings for CD3, CD8, CD45RO on Benchmark XT automate; Scanning the immunostainings on Hamamatsu scanner: Quantification of the immune populations in specific tumor areas (the core and the invasive margin) with a dedicated image analysis module based on the recognition of histological structures adapted to colorectal cancer. This module has been developed by our group in connection with the Definiens company. A semi-automatic procedure allows the quantification of stained cells in each tumor region. Each region is divided into tiles of 800 microns side (500 to 700 tiles analyzed per tumor). The density of each marker is calculated using the mean density of the three tiles the most infiltrated in each tumor region. A categorization into high density (Hi) or low (Lo) for each marker in each tumor region of interest will be performed. This categorization is done by comparing the observed density with the optimal cut-off density (min p value approach) we previously determined for the DFS and OS in a retrospective study of 250 CRC. Genetic analyses: The tubes containing the sections for the genetic analyses of patients with inform consent signed for genetic investigation will be transfer to the Department of Biology, Hôpital Européen Georges Pompidou. The % of tumor cells of each specimen will be determined by a pathologist (based on a H\&E evaluation; less than 10%, 10-20%, 20-50%, up to 50% tumor cells) and the DNA extraction will be performed. Genetic investigation will comprise the MSI status, the presence of a K-Ras and BRAF mutation