Single-Center Experience with Romiplostim for Management of Chemotherapy-Induced Thrombocytopenia (CIT)

Background: Chemotherapy-induced thrombocytopenia (CIT) is a frequent complication in cancer patients receiving myelosuppressive chemotherapy. Current management is limited to platelet transfusions, chemotherapy dose reductions, and delays in treatment. Romiplostim is a thrombopoeitin receptor agonist that increases platelet production by stimulating the maturation of megakaryocytes. It is currently approved for the treatment of idiopathic thrombocytopenic purpura (ITP). Literature describing its efficacy and safety in cancer patients undergoing chemotherapy is limited. The optimal dose and schedule of romiplostim in this setting is also unclear. Here we describe our single-center experience with the use of romiplostim for the management of CIT.

Methods: This retrospective case series describes 32 patients who received romiplostim for the management of CIT at DFCI/BWH Cancer Center from August 2008 to April 2016. Patient characteristics collected include age, gender, cancer diagnosis, and chemotherapy. Romiplostim dose adjustments were made per clinican discretion. Four patients received only one dose; these patients were only included in adverse event outcomes analysis. Efficacy outcome measures (N=28 patients) included average dose, average time between doses, achievement of platelet count ≥ 100 x10⁹/L, and completion of at least 2 subsequent chemotherapy cycles without dose-limiting toxicity or delays. Romiplostim-associated adverse events were assessed for safety.

Results: Median age was 54 years and 59% of patients had gastrointestinal malignancies. The most commonly-implicated chemotherapeutic agents were platinum-based. Platelet counts improved in all 28 patients included in the efficacy population allowing all 28 patients to receive additional chemotherapy. Median baseline platelets were 68 x 10⁹/L. Twenty-six patients achieved a platelet count of ≥ 100 x 10⁹/L. Twenty-five patients were able to receive ≥ 2 subsequent cycles of chemotherapy without delays or dose reductions for thrombocytopenia. The median duration of treatment was 131.5 days (IQR 66.5,255.5). Dose and timing of romiplostim administration varied greatly at the discretion of the treating physician. Overall, romiplostim administration was avoided on days of chemotherapy adminitration although 2 patients did have romiplostim administered on the same day. The median average dose was 2.4 mcg/kg (IQR 2, 5.4) and the median average time between doses was 2.1 weeks (IQR 1.0-2.8).

In the safety analysis, 4 patients developed venous thromboembolism. Our rate of thrombotic events 12.5% (0.035-0.29, 95% CI) was high, but may be consistent with the general rate of thrombosis in the oncology population (Timp et al., 2013). All patients who experienced thrombotic adverse events had underlying malignancies that were associated with high risk for thrombosis (3 pancreatic cancer, 1 astrocytoma). Of note, 1 patient was diagnosed with AML in the setting of romiplostim treatment. However, rising leukocyte counts prior to adminstration may have indicated a pre-existing hematologic malignancy. This case highlights safety concerns regarding romiplostim and the risk of aggravating underlying hematologic cancers.

Conclusions: In our retrospective case series, romiplostim showed efficacy in improving platelet counts in patients with CIT allowing for the resumption of chemotherapy. The rate of thrombosis was high, but may be consistent with that reported in the literature. Further studies are required to understand the optimal dose, timing strategies, and safety of romiplostim in the setting of CIT and to develop strategies to mitigate adverse reactions.