(064) Incidence and risk of secondary myelodysplastic syndrome and acute myeloid leukaemia in ovarian cancer patients treated with poly-ADP ribose polymerase inhibitors in England

Background: Poly-ADP ribose polymerase inhibitors (PARPi) are now standard-of-care for advanced-stage ovarian cancer. A meta-analysis of randomised controlled trials (RCTs) reported incidence of myelodysplastic syndrome (MDS) and acute myeloid leukaemia (AML) as 0.7% in patients treated with PARPi; however, RCTs may be underpowered to detect true incidence. Observational research can support post-market surveillance by investigating MDS/AML incidence in the routine care ovarian cancer population.

Methods: This retrospective cohort study used National Cancer Registration Dataset (NCRD), Systemic Anti-Cancer Therapy (SACT) and Hospital Episode Statistics (HES) data. MDS/AML incidence was calculated in advanced-stage ovarian cancer patients receiving chemotherapy +/- PARPi in first-line or recurrent settings. Patients were followed up for five years from completion of first-line chemotherapy.

Results: 15,775 patients were included. 1,551 received chemotherapy + PARPi maintenance and 14,224 with chemotherapy alone. Overall MDS/AML incidence was 0.18% (n=29/15,775). 0.52% (n=8/1,551) developed MDS/AML following PARPi maintenance compared to 0.15% (n=21/14,224) treated with chemotherapy alone. Relative risk of MDS/AML was 3.49 (95% CI 1.55, 7.87, p<0.01) for PARPi versus chemotherapy alone.

Discussion: MDS/AML incidence was significantly greater in patients receiving PARPi versus chemotherapy alone. Incidence rates were lower than in RCT follow-up. These findings can reassure clinicians when prescribing PARPi.