Search

Competing Interests: Disclosure: Kazunari Sasaki, Jane Wang, Carsten Kamphues, Stefan Buettner, Johan Gagniere, Victoria Ardilles, Katsunori Imai, Doris Wagner, Ioannis Pozios, Dimitris Papakonstantinou, Emmanouil Pikoulis, Efstathios Antoniou, Daisuke Morioka, Inger Marie Løes, Per Eystein Lønning, Peter Kornprat, Federico N. Aucejo, Hideo Baba, Eduardo de Santibañes, Klaus Kaczirek, Richard Burkhart, Itaru Endo, Katharina Beyer, Martin E Kreis, Timothy M. Pawlik, and Georgios Antonios Margonis have no conflicts of interest/competing interests to declare.

Background: Recent studies have suggested that certain combinations of KRAS or BRAF biomarkers with clinical factors are associated with poor outcomes and may indicate that surgery could be "biologically" futile in otherwise technically resectable colorectal liver metastasis (CRLM). However, these combinations have yet to be validated through external studies., Patients and Methods: We conducted a systematic search to identify these studies. The overall survival (OS) of patients with these combinations was evaluated in a cohort of patients treated at 11 tertiary centers. Additionally, the study investigated whether using high-risk KRAS point mutations in these combinations could be associated with particularly poor outcomes., Results: The recommendations of four studies were validated in 1661 patients. The first three studies utilized KRAS, and their validation showed the following median and 5-year OS: (1) 30 months and 16.9%, (2) 24.3 months and 21.6%, and (3) 46.8 months and 44.4%, respectively. When analyzing only patients with high-risk KRAS mutations, median and 5-year OS decreased to: (1) 26.2 months and 0%, (2) 22.3 months and 15.1%, and (3) not reached and 44.9%, respectively. The fourth study utilized BRAF, and its validation showed a median OS of 10.4 months, with no survivors beyond 21 months., Conclusion: The combinations of biomarkers and clinical factors proposed to render surgery for CRLM futile, as presented in studies 1 (KRAS high-risk mutations) and 4, appear justified. In these studies, there were no long-term survivors, and survival was similar to that of historic cohorts with similar mutational profiles that received systemic therapies alone for unresectable disease., Competing Interests: Disclosure. None to report., (© 2024. Society of Surgical Oncology.)

Introduction: Cachexia is strongly associated with digestive cancers, particularly oesogastric cancer. Mitochondria in adipose tissue are involved in the regulation of metabolism and physiopathology of cancer cachexia in animal studies. Chemotherapeutic regimens used to control tumour development could also alter mitochondrial function in adipose tissue. We hypothesise that cachexia induces an increase in adipose tissue mitochondrial energy metabolism and that chemotherapy can mitigate this. The purpose of the ChiFMeOE study is to identify adipocyte factors involved in the energy imbalance associated with the cachectic process and their response to chemotherapeutic treatments in patients with oesogastric cancer., Methods and Analysis: ChiFMeOE is a single-centre observational study that will prospectively include 60 patients referred to chemotherapy and surgery for oesophageal and gastro-oesophageal junction adenocarcinomas at the University Hospital of Clermont-Ferrand, France. Visceral and subcutaneous adipose tissue biopsies will be collected during surgery scheduled before and after neoadjuvant chemotherapy administration, as well as cachexia and nutritional assessment. The primary outcome is the maximum mitochondrial respiration rate (Vmax) measured by high-resolution respirometry. Secondary outcomes are other mitochondrial parameters (ie, enzymatic activities, proteins content and gene expression), tumour characteristics, nutritional status and body composition., Ethics and Dissemination: The study was approved by an independent institutional review board on June 2023 (Comité de protection des personnes Sud-Méditerranée V; 2023-A00582-43) and declared to the French regulatory authority for research. Written informed consent will be obtained prior to patient inclusion. The principal investigator will be notified of any changes in patient's health status requiring a modification of his management and/or treatment during the course of the protocol. Results will be published in peer-reviewed journals., Trial Registration Number: ClinicalTrials.gov, NCT05954117., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2024. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.)

Background: Minimally invasive surgery has gained momentum for left pancreatic resections. However, debate remains about whether it has any advantage over open surgery for distal pancreatectomy for pancreatic neuroendocrine tumors., Methods: This retrospective review examined pancreatectomies performed for resectable pancreatic neuroendocrine tumors at 21 centers in France between January 2014 and December 2018. Short and long-term outcomes were compared before and after propensity score matching based on tumor size, sex, age, body mass index, center, and method of pancreatic transection., Results: During the period study, 274 patients underwent left pancreatic resection for pancreatic neuroendocrine tumors [109 underwent distal splenopancreatectomy, and 165 underwent spleen-preserving distal pancreatectomy [(splenic vessel preservation (n = 97; 58.7%)/splenic vessel resection (n = 68; 41.3%)]. Before propensity score matching, minimally invasive surgery was associated with a lower rate of major morbidity (P = .004), lower rate of postoperative delayed gastric emptying (P = .04), and higher rate of "textbook" outcomes (P = .04). After propensity score matching, there were 2 groups of 54 patients (n = 30 distal splenopancreatectomy; n = 78 spleen-preserving distal pancreatectomy). Minimally invasive surgery was associated with less blood loss (P = .05), decreased rate of major morbidity (6% vs. 24%; P = .02), less delayed gastric emptying (P = .05) despite similar rates of postoperative fistula, hemorrhage, and reoperation (P > .05). The 5-year overall survival (79% vs. 75%; P = .74) and recurrence-free survival (10% vs 17%; P = .39) were similar., Conclusion: Minimally invasive surgery for left pancreatic resection can be safely proposed for patients with resectable left pancreatic neuroendocrine tumors. Minimally invasive surgery decreases the rate of major complications while providing comparable long-term oncologic outcomes., (Copyright © 2024 Elsevier Inc. All rights reserved.)

Background: The association between the intestinal microbiota and psychiatric disorders is becoming increasingly apparent. The gut microbiota contributes to colorectal carcinogenesis (CRC), as demonstrated with colibactin-producing Escherichia coli (CoPEC)., Aim: To evaluate the association between CoPEC prevalence and anxiety- and depressive-like behaviors with both preclinical and clinical approaches., Methods: Patients followed after a CRC surgery and for whom the prevalence of CoPEC has been investigated underwent a psychiatric interview. Results were compared according to the CoPEC colonization. In parallel C57BL6/J wild type mice and mice with a CRC susceptibility were chronically infected with a CoPEC strain. Their behavior was assessed using the Elevated Plus Maze test, the Forced Swimming Test and the Behavior recognition system PhenoTyper ® ., Results: In a limited cohort, all patients with CoPEC colonization presented with psychiatric disorders several years before cancer diagnosis, whereas only one patient (17%) without CoPEC did. This result was confirmed in C57BL6/J wild-type mice and in a CRC susceptibility mouse model (adenomatous polyposis coli multiple intestinal neoplasia/+ ). Mice exhibited a significant increase in anxiety- and depressive-like behaviors after chronic infection with a CoPEC strain., Conclusion: This finding provides the first evidence that CoPEC infection can induce microbiota-gut-brain axis disturbances in addition to its procarcinogenic properties., Competing Interests: Conflict-of-interest statement: Pr. Bonnet has nothing to disclose., (©The Author(s) 2024. Published by Baishideng Publishing Group Inc. All rights reserved.)

Objective: To investigate the clinical implications of BRAF -mutated (mut BRAF ) colorectal liver metastases (CRLMs)., Background: The clinical implications of mut BRAF status in CRLMs are largely unknown., Methods: Patients undergoing resection for mut BRAF CRLM were identified from prospectively maintained registries of the collaborating institutions. Overall survival (OS) and recurrence-free survival (RFS) were compared among patients with V600E versus non-V600E mutations, KRAS/BRAF comutation versus mut BRAF alone, microsatellite stability status (Microsatellite Stable (MSS) vs instable (MSI-high)), upfront resectable versus converted tumors, extrahepatic versus liver-limited disease, and intrahepatic recurrence treated with repeat hepatectomy versus nonoperative management., Results: A total of 240 patients harboring BRAF -mutated tumors were included. BRAF V600E mutation was associated with shorter OS (30.6 vs 144 mo, P =0.004), but not RFS compared with non-V600E mutations. KRAS/BRAF comutation did not affect outcomes. MSS tumors were associated with shorter RFS (9.1 vs 26 mo, P <0.001) but not OS (33.5 vs 41 mo, P =0.3) compared with MSI-high tumors, whereas patients with resected converted disease had slightly worse RFS (8 vs 11 mo, P =0.01) and similar OS (30 vs 40 mo, P =0.4) compared with those with upfront resectable disease. Patients with extrahepatic disease had worse OS compared with those with liver-limited disease (8.8 vs 40 mo, P <0.001). Repeat hepatectomy after intrahepatic recurrence was associated with improved OS compared with nonoperative management (41 vs 18.7 mo, P =0.004). All results continued to hold true in the multivariable OS analysis., Conclusions: Although surgery may be futile in patients with BRAF -mutated CRLM and concurrent extrahepatic disease, resection of converted disease resulted in encouraging survival in the absence of extrahepatic spread. Importantly, second hepatectomy in select patients with recurrence was associated with improved outcomes. Finally, MSI-high status identifies a better prognostic group, with regard to RFS while patients with non-V600E mutations have excellent prognosis., Competing Interests: The authors report no conflicts of interest., (Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.)

Background: Small bowel adenocarcinoma is a rare cancer, and the role of adjuvant chemotherapy for localized disease is still debated., Methods: This retrospective multicenter study included all consecutive patients who underwent curative surgical resection for localized small bowel adenocarcinoma between 1996 and 2019 from 3 French cohort studies. Prognostic and predictive factors of adjuvant chemotherapy efficacy were analyzed for disease-free survival and overall survival. The inverse probability of treatment weighting method was applied in the Cox regression model using the propensity score derived from multivariable logistic regression., Results: A total of 354 patients were included: median age, 63.5 years; duodenum location, 53.5%; and tumor stage I, II, and III in 31 (8.7%), 144 (40.7%), and 179 (50.6%) patients, respectively. The adjuvant chemotherapy was administered in 0 (0%), 66 (48.5%), and 143 (80.3%) patients with stage I, II, and III, respectively (P < .0001). In the subgroup analysis by inverse probability of treatment weighting method, a statistically significant disease-free survival and overall survival benefit in favor of adjuvant chemotherapy was observed in high-risk stage II (T4 and/or <8 lymph nodes examined) and III (T4 and/or N2) but not for low-risk stage II (T3 and ≥8 lymph nodes examined) and III (T1-3/N1) tumors (Pinteraction < .05). Furthermore, tumor location in jejunum and ileum was also a statistically significant predictive factor of response to adjuvant chemotherapy in stage II and III tumors (Pinteraction < .05)., Conclusion: In localized small bowel adenocarcinoma, adjuvant chemotherapy seems to provide a statistically significant survival benefit for high-risk stage II and III tumors and for jejunum and ileum tumor locations., (© The Author(s) 2023. Published by Oxford University Press.)

Objective: Defining robust and standardized outcome references for distal pancreatectomy (DP) by using Benchmark analysis., Background: Outcomes after DP are recorded in medium or small-sized studies without standardized analysis. Therefore, the best results remain uncertain., Methods: This multicenter study included all patients undergoing DP for resectable benign or malignant tumors in 21 French expert centers in pancreas surgery from 2014 to 2018. A low-risk cohort defined by no significant comorbidities was analyzed to establish 18 outcome benchmarks for DP. These values were tested in high risk, minimally invasive and benign tumor cohorts., Results: A total of 1188 patients were identified and 749 low-risk patients were screened to establish Benchmark cut-offs. Therefore, Benchmark rate for mini-invasive approach was ≥36.8%. Benchmark cut-offs for postoperative mortality, major morbidity grade ≥3a and clinically significant pancreatic fistula rates were 0%, ≤27%, and ≤28%, respectively. The benchmark rate for readmission was ≤16%. For patients with pancreatic adenocarcinoma, cut-offs were ≥75%, ≥69.5%, and ≥66% for free resection margins (R0), 1-year disease-free survival and 3-year overall survival, respectively. The rate of mini-invasive approach in high-risk cohort was lower than the Benchmark cut-off (34.1% vs ≥36.8%). All Benchmark cut-offs were respected for benign tumor group. The proportion of benchmark cases was correlated to outcomes of DP. Centers with a majority of low-risk patients had worse results than those operating complex cases., Conclusion: This large-scale study is the first benchmark analysis of DP outcomes and provides robust and standardized data. This may allow for comparisons between surgeons, centers, studies, and surgical techniques., Competing Interests: The authors report no conflicts of interest., (Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.)

Introduction: The management of mid and low rectal cancer is based on neoadjuvant chemoradiotherapy (CRT) followed by standardised surgery. There is no biomarker in rectal cancer to aid clinicians in foreseeing treatment response. The determination of factors associated with treatment response might allow the identification of patients who require tailored strategies (eg, therapeutic de-escalation or intensification). Colibactin-producing Escherichia coli (CoPEC) has been associated with aggressive colorectal cancer and could be a poor prognostic factor. Currently, no study has evaluated the potential association between intestinal microbiota composition and tumour response to CRT in mid and low rectal cancer. The aim of this study is to assess the association between response to neoadjuvant CRT and faecal intestinal microbiota composition and/or CoPEC prevalence in patients with mid or low rectal cancer., Methods and Analysis: This is a non-randomised bicentric prospective clinical study with a recruitment capacity of 200 patients. Three stool samples will be collected from participants with histological-proven adenocarcinome of mid or low rectum who meet eligibility criteria of the study protocol: one before neoadjuvant treatment start, one in the period between CRT end and surgery and one the day before surgery. In each sample, CoPEC will be detected by culture in special media and molecular (PCR) approaches. The global microbiota composition will be also assessed by the bacterial 16S rRNA gene sequencing. Neoadjuvant CRT response and tumour regression grade will be described using the Dworak system at pathological examination. Clinical data and survival outcomes will also be collected and investigated., Ethics and Dissemination: MICARE was approved by the local ethics committee (Comité de Protection des Personnes Sud-Est II, 18 December 2019. Reference number 2019-A02493-54 and the institutional review board. Patients will be required to provide written informed consent. Results will be published in a peer reviewed journal., Trial Registration Number: NCT04103567., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2022. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.)

Importance: In patients with resectable colorectal cancer liver metastases (CRLM), the choice of surgical technique and resection margin are the only variables that are under the surgeon's direct control and may influence oncologic outcomes. There is currently no consensus on the optimal margin width., Objective: To determine the optimal margin width in CRLM by using artificial intelligence-based techniques developed by the Massachusetts Institute of Technology and to assess whether optimal margin width should be individualized based on patient characteristics., Design, Setting, and Participants: The internal cohort of the study included patients who underwent curative-intent surgery for KRAS-variant CRLM between January 1, 2000, and December 31, 2017, at Johns Hopkins Hospital, Baltimore, Maryland, Memorial Sloan Kettering Cancer Center, New York, New York, and Charité-University of Berlin, Berlin, Germany. Patients from institutions in France, Norway, the US, Austria, Argentina, and Japan were retrospectively identified from institutional databases and formed the external cohort of the study. Data were analyzed from April 15, 2019, to November 11, 2021., Exposures: Hepatectomy., Main Outcomes and Measures: Patients with KRAS-variant CRLM who underwent surgery between 2000 and 2017 at 3 tertiary centers formed the internal cohort (training and testing). In the training cohort, an artificial intelligence-based technique called optimal policy trees (OPTs) was used by building on random forest (RF) predictive models to infer the margin width associated with the maximal decrease in death probability for a given patient (ie, optimal margin width). The RF component was validated by calculating its area under the curve (AUC) in the testing cohort, whereas the OPT component was validated by a game theory-based approach called Shapley additive explanations (SHAP). Patients from international institutions formed an external validation cohort, and a new RF model was trained to externally validate the OPT-based optimal margin values., Results: This cohort study included a total of 1843 patients (internal cohort, 965; external cohort, 878). The internal cohort included 386 patients (median [IQR] age, 58.3 [49.0-68.7] years; 200 men [51.8%]) with KRAS-variant tumors. The AUC of the RF counterfactual model was 0.76 in both the internal training and testing cohorts, which is the highest ever reported. The recommended optimal margin widths for patient subgroups A, B, C, and D were 6, 7, 12, and 7 mm, respectively. The SHAP analysis largely confirmed this by suggesting 6 to 7 mm for subgroup A, 7 mm for subgroup B, 7 to 8 mm for subgroup C, and 7 mm for subgroup D. The external cohort included 375 patients (median [IQR] age, 61.0 [53.0-70.0] years; 218 men [58.1%]) with KRAS-variant tumors. The new RF model had an AUC of 0.78, which allowed for a reliable external validation of the OPT-based optimal margin. The external validation was successful as it confirmed the association of the optimal margin width of 7 mm with a considerable prolongation of survival in the external cohort., Conclusions and Relevance: This cohort study used artificial intelligence-based methodologies to provide a possible resolution to the long-standing debate on optimal margin width in CRLM.

Background: This retrospective multicenter cohort study compared the feasibility and safety of oxaliplatin-based pressurized intraperitoneal aerosol chemotherapy (PIPAC-Ox) with or without intraoperative intravenous 5-fluorouracil (5-FU) and leucovorin (L)., Methods: Our study included consecutive patients with histologically proven unresectable and isolated colorectal peritoneal metastases (cPM) treated with PIPAC-Ox in seven tertiary referral centers between January 2015 and April 2020. Toxicity events and oncological outcomes (histological response, progression-free survival, and overall survival) were compared between patients who received intraoperative intravenous 5-FU/L (PIPAC-Ox + 5-FU/L group) and patients who did not (PIPAC-Ox group)., Results: In total, 101 patients (263 procedures) were included in the PIPAC-Ox group and 30 patients (80 procedures) were included in the PIPAC-Ox + 5-FU/L group. Common Terminology Criteria for Adverse Events v4.0 grade 2 or higher adverse events occurred in 48 of 101 (47.5%) patients in the PIPAC-Ox group and in 13 of 30 (43.3%) patients in the PIPAC-Ox + 5-FU/L group (p = 0.73). The complete histological response rates according to the peritoneal regression grading score were 27% for the PIPAC-Ox + 5-FU/L group and 18% for the PIPAC-Ox group (p = 0.74). No statistically significant differences were observed in overall or progression-free survival between the two groups., Conclusions: The safety and feasibility of PIPAC-Ox + 5-FU/L appears to be similar to the safety and feasibility of PIPAC-Ox alone in patients with unresectable cPM. Oncological outcomes must be evaluated in larger studies., (© 2022. Society of Surgical Oncology.)

Small bowel adenocarcinoma (SBA) is a rare tumour. Large genomic analyses with prognostic assessments are lacking. The NADEGE cohort has enrolled 347 patients with all stage SBA from 2009 to 2012. Next-generation sequencing investigates the presence of 740 hotspot somatic mutations in a panel of 46 genes involved in carcinogenesis. The mismatch repair (MMR) status was assessed by immunochemistry. We have collected 196 tumour samples and 125 had conclusive results for mutation analysis. The number of mutations was 0 in 9.6% of tumours, only 1 in 32.0%, 2 in 26.4% and ≥3 in 32.0%. Overall, at least one genomic alteration was observed in 90.4% of tumour. The most frequent genomic alteration was in KRAS (44.0%), TP53 (38.4%), PIK3CA (20.0%), APC (18.4%), SMAD4 (14.4%) and ERBB2 (7.2%) genes. KRAS mutations were more frequent in synchronous metastatic tumours than in localised tumours (72.7% vs 38.2%, P = .003). There was no significant difference in the mutation rates according to primary location for the most frequently altered gene. ATM, FGFR3 and FGFR1 gene alterations were associated with Lynch syndrome and IDH1 mutations with Crohn disease. dMMR tumours were associated with younger age, localised tumours, less KRAS but more SMARCB1 mutations. No genomic alteration was associated with overall survival. There is a trend for better survival in patient with dMMR tumours. In conclusion, there is a different genomic alteration profile in SBA according to predisposing diseases. No association between genomic alterations and prognoses was observed except for a trend of better prognoses associated with dMMR., (© 2020 Union for International Cancer Control.)

Introduction: Colorectal cancer (CRC) is still associated with poor prognosis, especially in patients with advanced disease. Development of new prognostic tools replacing or supplementing those routinely used is definitely needed, with the aim to optimise and personalise treatment strategies. Gut microbiota composition and body composition profile (obesity, sarcopenia and metabolic syndrome) have recently been reported separately as new relevant prognostic factors for postoperative surgical and oncologic outcomes following CRC surgery. However interactions that exist between these factors have been poorly studied. The purpose of this translational prospective cohort study (METABIOTE) is to investigate potential interactions between gut microbiota, body composition profile and postoperative outcomes and recurrence in patients undergoing surgery for non-metastatic sporadic CRC., Methods and Analysis: This single-centre project aims to prospectively enrol 300 consecutive patients undergoing surgery for non-metastatic sporadic CRC at the University Hospital of Clermont-Ferrand, France for the identification of specific microbial signatures (from tumour, colonic mucosa and stools samples) associated with particular metabolic profiles that could impact postoperative morbidity and oncologic outcomes, using microbiological, molecular and imaging approaches. The primary outcome is the 5-year overall survival (OS). Other outcomes are 5-year CRC-related OS, 5-year disease-free survival, 30-day postoperative morbidity, 90-day postoperative mortality and length of hospital stay., Ethics and Dissemination: This study protocol was reviewed and approved by an independent French regional review board (n°2018-A00352-53, 'Comité de Protection des Personnes Ile de France VII' on 4 July 2018, declared to the competent French authority ('Agence Nationale de Sécurité du Médicament et des produits de santé', France), and registered on the Clinical Trials web-based platform (NCT03843905). Oral and written informed consent will be obtained from each included patient. Study results will be reported to the scientific community at conferences and in peer-reviewed scientific journals., Trial Registration Number: NCT03843905.., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2020. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.)

Introduction: Type II AEG is now considered as oesophageal cancer in the seventh and eighth edition of TNM classification but optimal surgical approach for these tumors remains debated. The objective of the study is to assess and compare surgical and oncological outcomes of two surgical approaches: superior polar oesogastrectomy (SPO) or total gastrectomy (TG) in patients with type II adenocarcinoma of the oesophagogastric junction (AEG)., Material and Methods: 183 patients with type II AEG treated from 1997 to 2010 in 21 French centers by SPO or TG were included in a multicenter retrospective study. The surgical and oncological outcomes were compared between these two surgical approaches., Results: A TG was performed in 64 (35%) patients whereas 119 (65%) patients were treated by SPO with transthoracic approach in 100 of them (83.2%) and transhiatal approach with cervicotomy in 19 (16.8%). Surgical outcomes were comparable between the two approaches with a postoperative mortality rate of 4.9% and a severe operative morbidity rate within 30 days of 15.3%. Median survival in patients operated on by TG was of 46 months compared to 27 months in patients treated by SPO (p = 0.118). At multivariate analysis, TG appears to be an independent good prognostic factor compared to SPO (HR = 1.847; p = 0.008). However, TG was also associated with a higher rate of incomplete resection, (12.5% vs 5.9%; p = 0.120)., Conclusion: When TG allows obtaining tumor-free resection margins, this approach should be preferred to SPO., (Copyright © 2019 Elsevier Ltd, BASO ~ The Association for Cancer Surgery, and the European Society of Surgical Oncology. All rights reserved.)

Purpose: To predict microsatellite instability (MSI) status of colon cancer on preoperative CT imaging using radiomic analysis., Methods: This retrospective study involved radiomic analysis of preoperative CT imaging of patients who underwent resection of stage II-III colon cancer from 2004 to 2012. A radiologist blinded to MSI status manually segmented the tumor region on CT images. 254 Intensity-based radiomic features were extracted from the tumor region. Three prediction models were developed with (1) only clinical features, (2) only radiomic features, and (3) "combined" clinical and radiomic features. Patients were randomly separated into training (n = 139) and test (n = 59) sets. The model was constructed from training data only; the test set was reserved for validation only. Model performance was evaluated using AUC, sensitivity, specificity, PPV, and NPV., Results: Of the total 198 patients, 134 (68%) patients had microsatellite stable tumors and 64 (32%) patients had MSI tumors. The combined model performed slightly better than the other models, predicting MSI with an AUC of 0.80 for the training set and 0.79 for the test set (specificity = 96.8% and 92.5%, respectively), whereas the model with only clinical features achieved an AUC of 0.74 and the model with only radiomic features achieved an AUC of 0.76. The model with clinical features alone had the lowest specificity (70%) compared with the model with radiomic features alone (95%) and the combined model (92.5%)., Conclusions: Preoperative prediction of MSI status via radiomic analysis of preoperative CT adds specificity to clinical assessment and could contribute to personalized treatment selection.