Differences in tumor somatic mutations by race in VA patients with colon cancer.

e15607 Background: The Veterans Health Administration (VHA) diagnoses approximately 4,000 people with colon cancer each year and conducts Next Generation Sequencing (NGS) on recurrent and metastatic cases. Characterizing the molecular features of recurrent cancers may illuminate recurrence risk and mechanisms. We sought to characterize genetic profiles of recurrent and de novo metastatic tumors among a heterogenous patient population. Methods: We created a national database of colon cancer patients with recurrent or de novo metastatic disease who underwent NGS from 2015-2021. We examined patient demographics and tumor molecular abnormalities including DNA mismatch repair (MMR), DNA damage repair (DDR) abnormalities, and other clinically relevant mutations (KRAS, BRAF, APC, and TP53). Univariate and multivariable analyses were conducted to examine demographic and clinical associations with tumor molecular abnormalities, and with overall survival (OS). Results: We identified 2,366 patients who underwent NGS for recurrent or de novo metastatic disease. Of these, 1,072 patients had complete staging information (I n = 89, II n = 194, III n = 320, IV n = 453). The average age was 69.4 years; 70.5% were self-identified white race, 21.5% Black race and 7.8% other. Those with stage I-III recurrent disease had more BRAF mutations (15.1% vs 9.3% p = 0.03) and fewer KRAS mutations (43.1% vs 51.4% p = 0.03) than those with de novo metastatic disease. In univariate analyses, Black patients were more likely than white patients to have KRAS mutations (60.0% vs 43.7%; p < 0.01) and less likely to have BRAF mutations (6.9% vs 14.0%; p = 0.03). In multivariable logistic regression, Black patients were more likely to have KRAS mutations (OR 5.93, p = 0.03) but there was no difference in BRAF mutations (Table 1). Of note, race was not associated with OS. Among white patients, a multivariable Cox model showed that BRAF, KRAS and TP53 mutations were associated with worse OS (BRAF HR 1.71, 95%CI 1.30-2.26, KRAS HR 1.33, 95%CI 1.09-1.63, TP53 HR 1.31, 95%CI 1.06-1.63, respectively). For Black patients, BRAF was not associated with OS (BRAF HR 0.94, 95%CI 0.47-1.91), while other associations were similar. Across all races, recurrent tumors have better OS from date of NGS than de novo stage IV disease (HR 0.40, 95%CI 0.34-0.47). Conclusions: Among Veterans with recurrent or metastatic colon cancer, black and white patients differ with respect to clinically relevant rates of KRAS mutations. Prognostic implications of BRAF mutations for Black and white patients differ, with BRAF associated with poor OS only for patients with white race. These findings suggest potentially different pathogenesis for patients of different self-identified races. Outcome Input OR 95% CI pvalue KRAS mut Age 1.00 0.99 – 1.02 0.32 Stage (I-III) 0.70 0.55 – 0.90 0.01 Race-Black 5.93 1.44 – 39.98 0.03 BRAF mut Age 1.03 1.01 – 1.05 <0.01 Stage (I-III) 1.62 1.10 – 2.41 0.02 Race-Black 0.70 0.11 – 13.34 0.74