Molecular profiling of African American patients with pancreatic cancer.

e16242 Background: Pancreatic cancer (PC) is a highly fatal disease, and multiple studies have shown that African Americans (AA) have a higher incidence of PC compared with Whites and have a higher risk of PC-related mortality. Differences in known risk factors for PC including cigarette smoking, obesity, diabetes and family history have been demonstrated to not fully account for the racial disparity in incidence. Furthermore, differing mortality rates by race are not entirely attributable to variables such as access to care and disparate management. Whether there are molecular and biologic differences in PC that could explain these disparities is a question that has important scientific and clinical consequences. Unfortunately, most large studies of genomic alterations in PC include very few AA patients (pts). This study tries to characterize molecular alterations seen in AA pts with PC. Methods: The electronic medical record was queried for all self-identified Black or AA pts who were diagnosed with PC and had molecular profiling from November 2015 to November 2021. Patient demographics, clinical characteristics and molecular genomics were analyzed. The primary outcome was percent of patients with specific KRAS mutations. The secondary outcome was differences in KRAS mutations by gender. Results: We identified 50 AA pts with PC who underwent molecular profiling, 19 of whom (38%) were female. As expected, most (66%) were advanced stage at diagnosis. 32 pts (64%) had tissue-based molecular profiling with 18 (36%) having blood-based testing. A KRAS mutation was identified in 38 pts (76%). 15 (30%), 12 (24%), and 11 (22%) had G12V, G12D, and G12R mutations in KRAS, respectively. Men were more likely to have a G12D mutation (35.5% vs 5.3%, P = 0.018), and women were more likely to have a G12V mutation (47.4% vs 19.4%, P = 0.036). There was no significant difference in overall survival between men and women. 30 pts (60%) had additional mutations other than KRAS. 6 pts (12%) did not have a KRAS mutation detected despite the presence of other alterations. 4 pts had no mutations detected on molecular next generation sequencing, 3 of which had blood-based testing. Conclusions: In our review of genomic mutations of AA pts with PC, KRAS G12V mutations were most common. Only 24% of our studied population had KRAS G12D mutations. The frequency of specific KRAS mutations differs from what has been described in the literature among predominantly White pts with PC, where almost half of pts have a G12D mutation and 28% and 17% harbor G12V and G12R mutations, respectively. Two major limitations of this study are the small sample size which limits definitive comparisons between races, and the imperfect sensitivity of molecular profiling. KRAS is thought to be present in > 90% of PC, but the mutation was only identified in 76% of our pts. Further studies should be conducted to determine whether different molecular characteristics could account for racial disparities in PC mortality.