Abstract
Introduction/Background*Although most patients with ovarian cancer (OC) respond to first-line (1L) treatment, ≈70% experience recurrence within 3 years. Limited evidence exists on the interaction of prognostic factors and the risk of progressive disease (PD). This study assessed whether the number of risk factors (RFs) impacted time to PD in patients treated with maintenance therapy (MT) or active surveillance (AS) after completing 1L therapy.MethodologyThis retrospective cohort study included patients diagnosed with OC between January 1, 2011, and February 28, 2021, from the Flatiron Health electronic health record-derived de-identified US database. Patients ≥18 years old, with stage III/IV disease, who received 1L platinum-based therapy, had ECOG performance score of 0/1, and had ≥12 weeks of follow-up after 1L treatment were included. Patients were classified into 2 risk categories: moderate risk (stage III disease, no visual residual disease, primary debulking surgery, and BRCA mutant) or high risk (presence of ≥1 of the following: stage IV disease, visible residual disease, interval debulking surgery/no surgery, or BRCA wild type/unknown status). High-risk patients were further grouped by total number of RFs. Patients were followed from index date (ID), defined as end of 1L treatment. The target trial emulation method was used to account for potential selection and immortal time biases. Patients were classified as having received MT if MT was started within 120 days of ID.Result(s)*Of the 1251 patients with advanced OC evaluated, 26% (n=323) initiated 1L MT and 74% (n=928) did not. Of patients who received MT, 5% (n=16) were moderate risk and 95% (n=307) were high risk. Only 4% (n=34) of AS patients were moderate risk and 96% (n=894) were high risk. Time to next treatment (TTNT) decreased with more RFs (table 1). Notably, among patients in the high-risk category, median TTNT was longer in patients who received 1L MT than in those who received AS (table 1).Abstract 724 Table 1Median time to next treatment by number of risk factors (95% CI), monthsConclusion*The number of RFs impacted the risk of PD, irrespective of the type of treatment the patient received after completion of 1L treatment. For high-risk patients, greater TTNT prolongation was associated with 1L MT but not AS.