Cancer Medicine (Jan 2023)

Sociodemographic and clinical factors associated with receipt of biomarker testing in patients with metastatic colorectal cancer

  • Sarah C. Markt,
  • Benjamin D. Booker,
  • Wyatt Bensken,
  • Nicholas K. Schiltz,
  • Fredrick R. Schumacher,
  • Johnie Rose,
  • Greg Cooper,
  • J. Eva Selfridge,
  • Siran M. Koroukian

DOI
https://doi.org/10.1002/cam4.4995
Journal volume & issue
Vol. 12, no. 2
pp. 1850 – 1859

Abstract

Read online

Abstract Background Standard clinical practice and national guidelines dictate somatic testing of metastatic colorectal cancer (mCRC) tumors to guide appropriate therapy; however, previous studies suggest that not all patients are tested. The objective of this study was to investigate potential differences in testing for mCRC by demographic and clinical factors. Methods We performed a retrospective review of de‐identified patient data derived from electronic health records (EHRs) of 25,469 patients diagnosed with mCRC between the years 2013 and 2020. Our outcome was a receipt of the following tests: (a) biomarker testing (BRAF, KRAS, NRAS, MMR/MSI) and (b) next‐generation sequencing (NGS). We interrogated our data using the machine‐learning algorithm Classification and Regression Tree (CART), a unique approach to identifying combinations of, rather than individual demographic and clinical characteristics associated with receipt of testing. Results A total of 25,469 patients were identified with mCRC. Of these, 21,133 (83%) received either biomarker testing only (n = 12,485) or any testing (biomarker + NGS) (n = 8648). The proportion of patients who received any testing increased over calendar time for all age, race, and sex categories. Receipt of any testing was highest (90%) among younger and patients with better performance status, and there was no difference in receipt of any testing by race. The highest percentage of NGS testing was among those with better performance status, <70 years old, commercial or other governmental program payers, and low comorbidity burden; however, those who were Black or Hispanic had a lower prevalence of NGS testing than those who were White. Conclusions and Relevance Considerable variations exist in somatic biomarker testing across subgroups of the population. Identification of genomic alterations can aid in determining targeted treatment and improving clinical outcomes; therefore, equitable use of these testing strategies, particularly NGS, is necessary.

Keywords