Background:Significant disparities exist across the head and neck cancer (HNC) care continuum in presentation, treatment, and survival. Along with demographic factors such as race or ethnicity, tumor-specific molecular signatures may play a role in these differences. In the United States, the current body of literature mainly highlights molecular contrasts between Black and White patients. Here, we further investigate the molecular profiles of Black and White patients with HNC and add to the literature by comparing additional racial/ethnic groups in a large, real-world cohort.

Methods:De-identified records of patients with a primary diagnosis of HNC who underwent Tempus xT testing (Tempus Labs, Chicago, IL) were selected from the Tempus Database. Those with squamous cell histology and self-reported race were selected for retrospective analysis. Briefly, Tempus xT is a targeted, tumor/normal-matched panel that detects single-nucleotide variants, insertions and/or deletions, and copy number variants in 648 genes, as well as chromosomal rearrangements in 22 genes with high sensitivity and specificity. Tumor mutational burden (TMB) was calculated by dividing the number of nonsynonymous mutations by the megabase size of the panel.

Results:Among the 749 patients included, most identified as Non-Hispanic/Latino (NHL) White (n = 570), while 62 identified as NHL Black, 44 as NHL Asian/Pacific Islander (API), 21 as NHL of other races, and 52 as Hispanic/Latino of any race. Age at diagnosis differed between races (P= 0.009). Oral cavity was the most common subsite in our cohort (N = 306, 41%). TMB also differed among races (P= 0.018), with Black patients demonstrating the highest TMB (median = 4.6 mut/Mb [3.0-6.2]) and API the lowest (median = 2.3 mut/Mb [1.5-4.2]). Black patients had higher proportions of TP53 (76%), KMT2D (24%), and CTTN (15%) alterations (not significant across races). There were differences in the frequencies of OR4G11P (q = 0.031) and OR4F5 (q = 0.014) aberrations between races, with Black patients showing the highest frequencies (16% and 15%, respectively). The API population demonstrated higher frequencies of RASA1 (16%), MTAP (11%), CDKN2B (32%), and ERCC3 (11%) mutations (not significant across races). Additionally, there were differences in TERT mutations between races (q = 0.01), with API harboring the highest proportion of TERT co-mutations (52%).