1203P Prognostic implication of clinical decision support system (CDSS) for colorectal cancer

Abstract

Oncologists are challenged to personalize care with rapidly changing scientific evidence treatment guidelines and drug availability. IBM Watson for Oncology (WFO) is a cognitive computing clinical decision support systems (CDSS) that provides oncologists with evidence-based treatment recommendation for a variety of cancer treatment. WFO has been proven to be useful for decision-making in breast and lung cancers, but to date, research on colorectal cancer is limited. In the present study, we compared concordance of WFO with multidisciplinary tumor board from Daegu Catholic University Medical Center (DCUMC) and investigated the impact on patient prognosis We enrolled 200 patients with colorectal cancer, all treated between 2016 and 2018. Cases were processed using WFO, and the output was compared to blinded tumor board recommendations. Concordance was achieved when the DCUMC suggestion was in the “Recommended” or “For consideration” categories given by WFO. A total 200 cases were assessed, 129 (64.5%) were colon cancer and 71 (35.5%) were rectal cancer. The overall concordance rate was 74%; 71.3% for colon cancer, 78.9% for rectal cancer. Among all patients, the progression free time (PFS) was significantly better in concordant patients than in nonconcordant patients. (36m vs 20m, p=0.001) Treatment recommendations made by the DCUMC and WFO were highly concordant in colorectal cancer patients. Survival was better in concordant patients than in nonconcordant patients.