Hematologic complications in patients exposed to PARP inhibitors

Patients with BRCA1/2-mutated ovarian, breast, prostate, or pancreatic tumors can be treated with poly ADP-ribose polymerase (PARP) inhibitors. PARP inhibitors, however, are known to cause therapy-related myeloid neoplasms (t-MN) in a subset of patients. Predisposing factors to t-MN development in the context of PARP inhibitor exposure are not well described. To determine the frequency of t-MN in these patients, an institutional cohort of 265 patients with exposure to PARP inhibitors was identified. A subset of these patients with PARP inhibitor-related cytopenias underwent bone marrow biopsies. Among 265 patients, 17 (6.4%) underwent a bone marrow biopsy, which yielded a therapy-related hematologic diagnosis in 47% (8/17). Breast cancer metastasis to the marrow was found in one patient, and hemophagocytic lymphohistiocytosis was found in another. We analyzed the molecular characteristics of t-MNs in 13 PARP inhibitor-exposed patients, including five additional PARP inhibitor-exposed patients diagnosed with t-MNs in community practices. Among patients with t-MNs, five had acute therapy-related myeloid leukemia (t-AML), six were diagnosed with therapy-related myelodysplastic syndrome (t-MDS), and two had therapy-related clonal cytopenias of uncertain significance (t-CCUS). Complex karyotypes were found in four of seven patients who underwent karyotyping (57%). Next-generation sequencing identified TP53 mutations in 7 of 9 patients analyzed (78%). Among patients with germline testing, four (40%) did not have a germline mutation identified, four (40%) had a BRCA1 pathogenic/likely pathogenic (P/LP) variant, and two (20%) had a BRCA2 P/LP variant. Four patients received supportive care and/or observation after blood cancer diagnosis, and six received t-MN-directed therapy. The median survival for patients who received t-MN-directed treatment was 148 days. While cytopenias, particularly anemia, are known to occur with PARP inhibitor therapy, a subset of patients develop chronic cytopenias requiring bone marrow biopsy to evaluate for t-MN. Our study informs the expected findings of such biopsies.