Disparities and trends in biomarker testing in metastatic colorectal cancer.

Abstract

29 Background: Several therapies with biomarker-based indications have received approvals for use in metastatic colorectal cancer (mCRC) in the past decade. Consequently, biomarker testing is integral for treatment decisions, and guidelines recommend universal RAS, BRAF, and microsatellite instability (MSI)/ mismatch repair protein (MMR) testing in patients with mCRC. This study describes the frequency of biomarker testing in mCRC and explores variables influencing testing patterns. Methods: This study used the nationwide Flatiron Health electronic health record (EHR)-derived de-identified database, which contains patient-level data originating from approximately 280 cancer clinics in the United States (~800 sites of care).Adult patients diagnosed with mCRC after January 1, 2013 who received at least 1 line of treatment were included. The percentage of patients who received KRAS, NRAS, BRAF, and MSI/MMR testing as reported in the database was determined. Univariate and multivariable analyses were performed to evaluate factors associated with completion of biomarker testing. Results: A total of 25,937 patients were included in this analysis. Among these, 81.6%, 60.5%, 63.7%, and 73.4% had documented test results for KRAS, NRAS, BRAF, and MSI/MMR, respectively, and 51.6% received completed testing for all 4 biomarkers. In a multivariable analysis including 13,894 patients with complete data available for assessed variables, younger age at diagnosis, non-Hispanic or Latino ethnicity, white versus black race, more recent year of diagnosis, community versus academic practice setting, higher socioeconomic status index, and lower Eastern Cooperative Oncology Group performance status (ECOG PS) were associated with increased completion of biomarker testing. Year of diagnosis and ECOG PS had the largest impact. Frequency of testing completion was 73.7% in those diagnosed in or after 2020 compared to 21.5% in those diagnosed from 2013-2015 (odds ratio [OR] 8.56, p<0.001). Complete testing was performed in 59.3% of those with an ECOS PS of 0, compared to 55.1% (OR 0.86, p<0.001), 45.7% (OR 0.62, p<0.001), 38.0% (OR 0.43, p<0.001), and 16.7% (OR 0.16, p=0.019) of those with an ECOG PS of 1, 2, 3, and 4, respectively. Notably, testing was completed in 47.2% versus 50.5% of patients with Hispanic or Latino ethnicity versus non-Hispanic or Latino ethnicity (OR 0.69; p<0.001), and in 48.8% versus 51.6% of black versus white patients (OR 0.82, p=0.001). Conclusions: Based on results from the Flatiron Health database, rates of guideline-recommended biomarker testing in mCRC remain suboptimal and are impacted by age, ethnicity, race, socioeconomic status, practice setting, ECOG PS, and year of diagnosis. These results highlight important disparities in testing that require attention. Encouragingly, rates of testing have significantly increased over the past decade as new biomarker-directed therapies have become available.