Factors associated with platinum sensitivity (PST) and related outcomes in patients with ovarian cancer (OC) in a US community oncology setting: Looking beyond.

Abstract

e17590 Background: Along with residual disease after surgery, response to platinum therapy is a key prognostic indicator in patients with advanced OC. It is a prerequisite for receiving PARP inhibitor maintenance therapy and directs future treatment choices. Our objective was to describe differences in characteristics and outcomes between patients with platinum resistance (PR) or PST in a community oncology setting. Methods: This was a retrospective study of adult female patients with advanced OC (Stage III/IV) who received 4+ cycles of a 1L platinum-based regimen during 01/01/2017 to 06/30/2021 (followed until 12/31/2021) and had ≥ 2 visits within The US Oncology Network. Patients who were platinum resistant (PR, relapse 30 days to < 6 months after platinum discontinuation) or PST (relapse at 6+ months after platinum discontinuation) were included; refractory patients (relapse during or within 30 days of platinum discontinuation) were excluded. Structured and chart review data were used. Multivariable logistic regression assessed the association of factors with PST vs. PR. Results: In total 142 patient charts (108 PST, 34 PR) were reviewed. In addition to platinum-based chemotherapy, 19% of patients also initiated bevacizumab. Most patients were White (63%), diagnosed at Stage IIIC (55%) or IV (33%), and had epithelial OC (82%). Overall, 52%, 23% and 25% of patients had ECOG performance status scores of 0/1, 2+ and not documented, respectively. BRCA status was positive for 7.7% but not documented for 38%, and Charlson comorbidity score (CCS) was 0 for 63% of patients. PST vs. PR patients were younger, 69 (27-90+) years vs. 71 (51-89) [median (range), p = 0.045]. More PST vs. PR patients were normal weight (40% vs. 29%) or overweight (30% vs. 15%) and fewer were obese (24% vs. 41%) (p = 0.011). In multivariable adjusted analysis (excluding underweight and unreported BMI categories for low counts), Stage IV vs. III (p = 0.003) was associated with significantly higher odds of PR after 1L, while obesity vs. healthy weight (p = 0.067) and ECOG 2+ vs. 0/1 (p = 0.057) were associated with numerically higher odds of PR after 1L; age, CCS and 1L bevacizumab use were not associated with PST. Kaplan-Meier median (95% confidence interval [CI]) overall survival in PST vs. PR was not reached [NR] (44.6, NR) vs. 16.7 (13.9, 23.6) months while median (95% CI) progression-free survival was 19.6 (17.6,24.9) months vs.7.9 (7.1,8.5) months, respectively. Conclusions: When exploring factors associated with response to 1L platinum-based treatment of OC adjusting for other factors, higher stage was significantly associated with PR, while patients with obesity and poor ECOG status had nonsignificantly greater odds of PR. PR patients have poorer survival and fewer treatment options, and standard of care treatment of 1L OC in these subgroups should be further explored to address this unmet need.