Introduction: Racial and ethnic differences in incidence are well-reported in lung cancer associated with Tobacco smoking. ALK gene alterations as driver genetic alterations occur in 2-4% of lung cancer patients. Racial differences in ALK gene-altered lung cancer are not well defined. The objective of this study was to assess the frequency and type of ALK alterations observed in lung cancers across races.

Methods: Patients with a primary lung cancer diagnosis (histology and stage agnostic) that underwent testing with the Tempus xT/xR (tissue) or xF (liquid biopsy) assays (n=27,991) and had a confirmed pathogenic ALK gene alteration (SNVs, CNAs, or fusions) were retrospectively identified within the Tempus database (n=377). Race was determined based on abstracted clinical records, and patients were stratified as either white, Black or African American (BAA), Asian Pacific Islander (API), unknown or other. Somatic pathogenic co-mutations were compared across races using Chi-squared/Fisher’s Exact tests with FDR correction.

Results: Among tissue-tested lung cancer patients (n=17,482), ALK alterations occurred in 1.4% (n=135) white, 1% (n=13) BAA, 3.4% (n=15) API, 3.1% (n=16) other, and 2% (n=108) unknown, which differed across race (p<0.001). The most common ALK alterations were fusions, as detected by DNA or RNA-seq, which also differed across race (p<0.001), with the lowest frequency in BAA patients (0.9%, n=12). EML4-ALK fusions were most common, occurring in 86% (n=116) of white patients with an ALK alteration, 62% (n=8) BAA, 87% (n=13) API, 100% (n=16) other, and 93% (n=100) unknown race (p=0.013). Atypical gene fusion partners ACYP2, RMND5A, BABAM2, MYT1L, and TPR were observed only in BAA, and SOS1, KLC, NPM1, PRKAR1A were observed only in white patients. Interestingly, a subset of patients with ALK fusions also harbored ALK amplifications (n=28), SNVs (n=11), or both (n=2). Similar results were observed in the smaller cohort of patients harboring ALK gene alterations as detected by liquid biopsy (n=90). There were no significant differences in co-mutated genes across races in ALK-altered tumors, including TP53, CDKN2A, or SETD, as detected by liquid or tissue testing.

Conclusions: In this large cohort of lung cancer patients, ALK fusions were more commonly observed in API and “other” groups compared to white or BAA patients, with the lowest frequency in BAA. Some atypical fusion partners were observed in white and BAA that were not observed in other races. These data emphasize the importance of testing lung cancer patients with assays that allow for unbiased fusion detection and identification of novel fusion partners across racially diverse populations to better understand racial and ethnic differences in ALK-rearranged lung cancers.

VIEW THE PUBLICATION (P3.06D.12)

VIEW THE POSTER