Romiplostim in a Case Series of Chemotherapy-Induced Thrombocytopenia

Abstract 1170

Thrombocytopenia is a common effect of cytotoxic chemotherapy regimens in cancer patients, often necessitating reduction or delay in subsequent cycles. While the thrombocytopenia typically resolves with time, a subset of patients (pts) with chemotherapy-induced thrombocytopenia (CIT) have prolonged thrombocytopenia, of idiopathic etiology, with interference with subsequent treatment. We describe the use of weekly romiplostim, a second generation thrombopoietin mimetic 'peptibody' in 7 pts who had prolonged thrombocytopenia beyond anticipated recovery from chemotherapy-induced platelet nadir.

Five pts had metastatic solid tumors including soft tissue sarcoma on ifosfamide based therapy (n=2), colon cancer on 5-FU/ oxaliplatin based therapy(n=1), cholangiocarcinoma on gemcitabine therapy(n=1) and glioblastoma multiforme (GBM) on etoposide/CCNU (n=1). One patient had recurrent Non Hodgkin lymphoma (NHL) on rituximab and gemcitabine based therapy, and one pt had locally advanced gastric adenocarcinoma being treated with an irinotecan based regimen.

All had isolated thrombocytopenia (< 60K/mcL) that persisted at least 5 weeks after the last dose of chemotherapy (range 5–48 weeks) causing chemotherapy delay. Five pts had received two or more previous lines of chemotherapy. Pre-treatment bone marrow biopsies were performed in six pts. Five showed normal cellularity with normal megakaryocytes while one showed 10% cellularity. In 2 pts, romiplostim was begun at 1 mcg/kg. All other pts were started at 2 mcg/kg. Doses were escalated by 1 mcg/kg weekly, as is done for chronic ITP, to a maximum of 4 mcg/kg.

In all pts, platelets (plts) rose to greater than 100K/mcL between 2–4 weeks from initiation of romiplostim. Chemotherapy was re-instituted at full doses in six pts; the seventh pt with locally advanced gastric cancer was taken to surgery to attempt resection. Weekly romiplostim was continued throughout subsequent chemotherapy cycles in these six pts, maintaining counts adequate (>100K/mcL) to continue chemotherapy on schedule. Nadir platelet counts were maintained greater than 50K/mcL in 5/6 pts while on chemotherapy. In one pt with NHL, nadir platelet was 22K/mcL but demonstrated adequate recovery to receive next cycle on schedule.

No treatment-related toxicity was observed. One pt, with prior history of thrombosis, developed new deep venous thrombosis while on romiplostim when platelet count was 278K/mcL. He has since been maintained for the next 12 months on both low molecular weight heparin and romiplostim titrated to platelet count 100–200K/mcL without further thrombotic events. This patient had a bone marrow biopsy performed after 14 months of therapy at maintenance dose of 4 mcg/kg. This was normal without reticulin deposition.

No resistance to romiplostim was observed, and no pt required subsequent delay or reduction of chemotherapy due to thrombocytopenia. All pts received at least 4 additional cycles of chemotherapy. If chemotherapy was stopped or altered, it was due to oncologic considerations. One patient with stage IV sarcoma stopped romiplostim and chemotherapy at time of disease progression; no rebound thrombocytopenia was observed. No pts developed appearance of tear drop or nucleated red cells on review of peripheral smear morphology.

Our case series is the first therapeutic interventional experience with a second-generation thrombopoietin agonist for CIT demonstrating benefit. Our series differs from prior studies in that we selected only those pts who had already demonstrated persistent thrombocytopenia, and we administered the romiplostim weekly, throughout the course of chemotherapy. We show promising clinical benefit in this situation, which supports the implementation of a larger study of weekly romiplostim for treatment of CIT.