Real world study of racial disparities associated with toxicities of sacituzumab govitecan.

person Shista Priyadarshini Penn State Cancer Institute, Hershey, PA info_outline Shista Priyadarshini, Justin Petucci, Avnish Katoch, Vasant Honavar, Monali K. Vasekar

Authors person Shista Priyadarshini Penn State Cancer Institute, Hershey, PA info_outline Shista Priyadarshini, Justin Petucci, Avnish Katoch, Vasant Honavar, Monali K. Vasekar Organizations Penn State Cancer Institute, Hershey, PA, Institute for Computational and Data Sciences, University Park, PA, Clinical and Translational Sciences Institute, University Park, PA, Penn State Milton S. Hershey Medical Center, Hershey, PA Abstract Disclosures Research Funding No funding sources reported Background: Sacituzumab govitecan (SG) is approved in the treatment of metastatic hormone receptor positive and triple negative breast cancer (BC). The efficacy of SG across racial subgroups is known but no real-world data compares the racial differences in toxicities. We aim to expand knowledge about racial disparities regarding SG toxicities and any ineffective treatment or drug failure. Methods: We utilized TriNetX, a multi-health care organization electronic health record database, to identify African American (AA) and Caucasian (C) cohort for outcome comparison. Inclusion criteria required a C50 ICD-10-CM diagnosis code and at least one SG treatment. Age, gender, and race-controlled Cox proportional-hazard (CPH) model was used to analyze the time until the first pause or cessation of SG (defined as first deviation from the approved 21-day SG cycle). Several measures of deviation were noted 0, ±1 and ±3 days. Association between race and toxicity outcomes (controlled for age and gender) for anemia, neutropenia, thrombocytopenia, nausea and inpatient encounters were evaluated using multiple logistic regression (MLR). The odds (OR) and hazard (HR) ratios with corrected p-values are reported as an effect size and significance estimation. Results: 823 patients met the inclusion criteria (545 C, 116 AA & 162 unknown). The mean age (in years) at the initiation of 1st SG treatment was 58.22 ± 12.7 for C and 58.19 ± 12.18 for AA. AA had a 1.26 times higher hazard of SG treatment pause/cessation as compared to C. Blacks experienced an 88% increase in the odds of having a delay longer than 7-days between 2 doses of 1st cycle SG. Males were found to have an 80% increase in the pause/cessation rate as compared to females. The median number of SG treatments was 6 (3 cycles) for both AA and C. SG was noted as 4th chemo line (median) among all races. When compared to C, AA had decreased odds for neutropenia, increased odds for anemia and increased odds for hospitalization. Hazard Ratio and corrected p-values for outcomes comparing the AA and C cohorts. Conclusions: Survival analysis indicates that AA and male patients have an increased SG treatment pause/cessation rate. While the result is sensitive to specific conditions used to calculate the deviation, an increased odds of delay in treatment is also found using logistic regression modeling. Additionally, AA experience increased odds for anemia and late-stage hospitalization. These findings highlight the impact of race on BC outcomes, treatment toxicities and the need for further research to enhance treatment outcomes in diverse populations. Model and Time Window Treatment Cessation Anemia Neutropenia Hospitalization CPH: Strict (± 0 Days) Race - 1.26 (0.025) Gender - 0.55 (0.044) - - - MLR: Last and Missed SG - 1.14 (0.592) 0.51 (0.025) 1.14 (0.592) MLR: Within 10 days of 1st SG 1.88 (0.003) 2.25 (0.032) 0.64 (0.139) 1.23 (0.453) MLR: SG Cycle 10 &11 - 2.03 (0.315) 0.22 (0.046) 2.58 (0.035)