Real-world patterns of molecular testing in advanced melanoma (advMel): adherence to guidelines.

Abstract

e22031 Background: National guidelines (eg, NCCN and ESMO) recommend BRAF mutation testing to inform treatment (tx) decisions in high-risk stage III or IV cutaneous melanoma, but few studies have evaluated its implementation in a real-world setting. Here we report the testing patterns for BRAF and other mutations in advMel via an electronic health record (EHR)–derived database, to assess guideline adherence and implications for patient (pt) care. Methods: This retrospective observational study used Flatiron Health’s nationwide longitudinal database comprising de-identified EHR structured and unstructured data, curated via technology-enabled abstraction. Pts with stage III or IV cutaneous melanoma at initial diagnosis, or a locoregional or distant recurrence, with ≥ 2 clinic encounters starting January 1, 2011 were included in the advMel cohort. Data through October 2019 were analyzed. Results: A total of 9333 pts from 160 US cancer clinics were in the advMel cohort. Overall, 21% of pts had no evidence of BRAF testing. In the adjuvant setting, the percentage of pts with known BRAF status before initiation of systemic tx increased from 34% prior to the year 2018 to 56% post approval of adjuvant targeted tx in 2018. The rate of known BRAF status before initiation of first line (1L) metastatic tx declined from 80% in 2012 to 62% in 2018. Notably, within 30 days after tx initiation, BRAF status was reported in an additional 17% of pts in 1L metastatic setting. Next-generation sequencing (NGS) overtook polymerase chain reaction (PCR) as the most common assay (40% NGS vs 22% PCR in 2018, whilst 8% vs 53% in 2012). Of the 35-day (NGS) or 27-day (PCR) median turnaround time from sample collection to results, 21-25 days occurred from specimen collection to receipt in the lab. Despite this turnaround time, 35% of pts with known BRAF status before tx initiation had NGS testing. Of all pts tested for BRAF, 30% were also tested for NRAS and 32% for KIT. Among pts with a BRAF mutation, concomitant NRAS (1.1%) or KIT (0.6%) mutation was rare. Conclusions: Our real-world analysis shows that BRAF mutation status is known in ≈60% of pts with advMel before initiation of adjuvant or 1L metastatic tx. NGS and PCR are the most common testing platforms, with the use of NGS on the rise. Notably, a prolonged time was observed between collection and lab receipt of the specimen. These results indicate an opportunity for physicians to conduct earlier molecular testing at workup per the guidelines, and for improved logistical efficiency to facilitate timely receipt of genomic results to better inform individualized tx plans.