Romiplostim for PARP inhibitor-induced thrombocytopenia in solid tumor malignancies.

TO THE EDITOR: PARP (poly-adenosine diphosphate-ribose polymerase) inhibitors are a series of drugs designed to inhibit DNA repair mechanisms, preventing DNA repair in cancer cells and inducing cell death [1]. Currently approved for the treatment of breast, ovarian, pancreatic, and prostate cancers [2], indications for PARP inhibitor therapy continue to expand. However, patients undergoing PARP inhibitor treatment may encounter multiple hematologic toxicities, including anemia, neutropenia, and thrombocytopenia [3,4]. Based on recent results describing success in the use of romiplostim to treat chemotherapy-induced thrombocytopenia (CIT) in patients receiving traditional myelosuppressive chemotherapy [5–8] we sought to understand whether a similar approach could be used for PARP inhibitor induced thrombocytopenia, for which there are little published data. A case series of 5 patients with PARP inhibitor-induced thrombocytopenia treated with avatrombopag suggests potential effectiveness of thrombopoietin receptor agonists in these patients; however, no data to date has been published evaluating romiplostim for PARP inhibitorinduced thrombocytopenia. Given that none of these agents are FDAapproved for CIT but that the majority of safety and efficacy data for use of TPO-RAs to treat CIT generally is for romiplostim (thereby resulting in its specific inclusion as a treatment option for CIT in recent NCCN guidelines) [9,10], we sought to examine romiplostim use in patients with PARP inhibitor-induced thrombocytopenia. This study was approved by the Institutional Review Board of Mass General Brigham (protocol #2015P000152). An electronic patient data registry at Mass General Brigham was used to identify patients who had received both romiplostim and a PARP inhibitor (niraparib, olaparib, rucaparib, or talazoparib) at any time. This resource was used specifically to identify the study population; patients identified by the query then underwent manual chart review by the authors to determine which patients met inclusion criteria (defined as a patient diagnosed with cancer, prescribed a PARP inhibitor, who developed thrombocytopenia requiring support and was treated with romiplostim). Clinical data regarding patients’ oncologic treatment course, platelet counts, and any adverse events were collected manually from charts by investigators and analyzed. The database query returned 32 patients who met initial criteria (had received romiplostim and a PARP inhibitor). Manual review of each chart found 5 patients who met inclusion criteria, all of whom were prescribed niraparib. Baseline characteristics and parameters for these patients can be found in Table I. Platelet counts for each patient were collected prior to niraparib prescription (baseline platelet count), prior to romiplostim initiation, and during romiplostim therapy (on-romiplostim counts defined as platelet counts drawn one week after each romiplostim administration until one week after the last dose of romiplostim). All five patients had advanced cancer that had progressed beyond initial cytotoxic-based treatment regimens and were subsequently prescribed niraparib. One patient had prior radiation treatment in the setting of bone metastasis. Prior to initiation of romiplostim, all 5 patients had experienced dose-reduction, treatment delay, and/or planned discontinuation of niraparib treatment due to persistent thrombocytopenia. Faced with this, they were treated on romiplostim on our institutional weekly romiplostim CIT treatment pathway, described previously [8]. Throughout the course of treatment, 4 out of 5 patients (80%) showed improvement in platelet counts, with 3 showing marked improvements (Table I). Platelet counts just before the first dose of romiplostim ranged from 14 × 10/L to 97 × 10/L, with a per-patient median of 41 × 10/L. By comparison, on romiplostim therapy, the range of the mean platelet count was 65 × 10/L to 190 × 10/L, with a per-patient median of 118 × 10/L. One patient experienced a single niraparib dose administration delay due to thrombocytopenia, but otherwise platelet levels were maintained such that PARP inhibitor therapy could be continued. No patients experienced bleeding, thrombosis, niraparib dose reduction or treatment discontinuation due to thrombocytopenia while on romiplostim therapy. No patients in our cohort were on any other platelet support agents. In this case series, we describe five patients with PARP inhibitorinduced thrombocytopenia who responded to therapy with romiplostim, allowing for continuation of their treatment regimen without dose-reduction. The objective platelet response rate was similar to Correspondence: Hanny Al‐Samkari, Division of Hematology, Massachusetts General Hospital, Suite 118, Room 112, Zero Emerson Place, Boston, MA 02114, USA. Phone: 617-643-6214. Fax: 617-6438840. Email: hal-samkari@mgh.harvard.edu http://www.tandfonline.com/iplt ISSN: 0953-7104 (print), 1369-1635 (electronic)