Racial and ethnic disparities in presentation and outcomes of patients with gastrointestinal stromal tumors (GIST) in the era of targeted therapy.

e18573 Background: GIST is the most common malignant mesenchymal tumor of the gastrointestinal tract and is a compelling model for targeted therapies. Racial and ethnic minorities are underrepresented in clinical research, and information regarding disparities in GIST is limited. Our sarcoma center is located within a large and diverse population of Hispanic, LatinX, and African American/African Caribbean patients (pts). We aimed to identify racial and ethnic disparities among pts with GIST. Methods: We retrospectively reviewed all pts with primary or recurrent GIST between January 2003 and October 2022 in our institution. All data were retrieved from electronic medical records. SPSS Statistics was utilized, Fisher exact test and log-rank test were used for analysis. A two-sided p < 0.05 was considered statistically significant. Results: A total of 1150 pts had a diagnosis of GIST during the study period. Information regarding race and ethnicity was available in 711 pts. There were 42.4% White Non-Hispanic/LatinX (WN), 33.2% Hispanic/LatinX (HL), and 24.4% Black Non-Hispanic/LatinX (BN). The stomach was the most common primary site in all three groups, especially in the BN population (80.9%) vs WN (53%) and HL (61.9%), followed by the small intestine (p < 0.001). The WN pts more commonly presented with metastatic disease on diagnosis (16.2%) vs BN (12.7%) and HL (8.9) (p = 0.042) and had higher grade tumors (37.7%) vs BN (24.9%) and HL (23.7%) (p < 0.001). Median disease-free survival (mDFS) after radical resection was significantly lower in BN population vs WN and HL (p = 0.014). Disease-specific survival (DSS) in BN was also significantly reduced compared to the other groups (p = 0.024). A total of 445 pts had some molecular testing done, and BN were significantly less tested (38.5%) (p < 0.001). The most common primary driver mutation was KIT exon 11 throughout all groups, especially in the BN group corresponding to 80% of the cases tested (p = 0.008). The median overall survival (mOS) among pts with metastatic disease was 7.1 years (95% CI = 5.6-8.6). The HL population had a trend toward the longest mOS (8.8 years), followed by the WN (7.7 years), followed by the BN population (5.8 years) (p = 0.073). A total of 75 of these patients were tested for secondary mutations, with no statistical difference among the groups. Conclusions: We found in our study the WN population was significantly more likely to present with metastatic disease and higher-grade tumors than BN or HL pts. Nonetheless, the BN population were substantially less likely to undergo mutation testing and had a trend toward a worse mOS by more than 1 year. All pts had their care within the same healthcare system, implicating access to mutation testing and socioeconomic barriers playing a role in inferior outcomes. Further studies are needed to understand molecular differences and decrease the existing gap among these populations.