Abstract

3622 Background: According to current guidelines, most patients with high-risk stage II and stage III colon cancer should receive adjuvant chemotherapy (ACT). However, risk determination is controversial and most patients do not benefit from ACT. Histotype Px Colorectal is a novel artificial intelligence-based biomarker validated for R0 stage II-III colorectal adenocarcinoma that analyzes digitized routine H&E-stained FFPE tumor resections. Combined with clinical parameters, patients are stratified into distinct low, intermediate, and high-risk groups. This pilot study seeks to validate the applicability of Histotype Px Colorectal in an independent cohort from the United States. Methods: This was a retrospective analysis of patients diagnosed with pathological stage II-III colon adenocarcinoma at Ohio State University from 2016-2020. Anonymized slides were collected and digitized using the Aperio AT2 scanner. Clinical parameters and outcome data were extracted from patient records. Histotype Px Colorectal biomarker was blindly applied to each scan and subsequently linked to clinical outcomes. Statistical analysis was performed using Cox proportional hazards regression analysis. The pre-specified primary outcome was cancer-specific survival with a secondary outcome of time-to-recurrence. Results: Baseline characteristics of the 159 eligible patients included median age of 63 years (range 22-91), 52% female, 52% stage III, 64% right-sided tumors, and 31% MSI. 18% of stage II and 70% of stage III patients received ACT, respectively, with FOLFOX in 61%. Median follow-up for all patients was 54.2 months. For stage II patients, 21% were classified as intermediate-risk and 79% as low-risk. Only 6 (8%) of the 76 stage II patients had a cancer-related death with 3 of those patients classified as intermediate-risk. For stage III patients, 26% were classified as high-risk, 22% as intermediate-risk, and 52% as low-risk. 18 (22%) of the 83 stage III patients had a cancer-related death with 14 of those patients classified as high or intermediate-risk. Overall, Histotype Px Colorectal was a significant predictor of cancer-specific survival with HR=7.80 (95% CI 2.96-20.56, p<0.001) for high vs. low-risk patients and 2.81 (95% CI 0.98-8.04, p=0.05) for intermediate vs. low-risk patients. In addition, it was found to be a significant predictor for time-to-recurrence with HR=7.59 (3.56-16.20, p<0.001) for high vs. low-risk patients and 2.67 (1.20-5.96, p=0.02) for intermediate vs. low-risk patients. Conclusions: The findings from this study highlight the potential utility of this innovative biomarker in guiding clinical decisions regarding ACT. Further research involving a larger and more diverse patient cohort and subsequent clinical studies are planned to solidify these initial findings and to enable personalized treatment strategies based on individual risk assessments.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call