The course of platinum-based chemotherapy is associated with the risk of hematological complications during poly(ADP-ribose) polymerase (PARP) inhibitor therapy in patients with ovarian cancer.

BACKGROUND

Poly(ADP-ribose) polymerase (PARP) inhibitor maintenance therapy is a standard treatment in selected ovarian cancer patients. However, hematological toxicity remains a clinical problem. The study aimed to identify predictors of early hematologic toxicity during PARP inhibitor maintenance therapy in high-grade serous ovarian cancer.

METHODS

We retrospectively analyzed 130 patients receiving PARP inhibitors (niraparib, 49; olaparib, 81). Blood counts from first-line platinum chemotherapy and from the first 3 months of PARP inhibitor therapy were assessed. Adverse events were graded using the Common Terminology Criteria for Adverse Events v5.0. Weekly changes and minimal values for neutrophils, hemoglobin, and platelets were calculated (excluding post-intervention values).

RESULTS

Thrombocytopenia was mainly related to niraparib treatment, and anemia was mainly related to olaparib treatment. During niraparib therapy, higher human epididymis protein 4 (He4) at diagnosis correlated with greater platelet decline and lower platelet nadir. In multivariable linear regression, higher body weight (β = 1.86, 95% CI 0.05 to 3.67, p = 0.044) and higher minimal hemoglobin during chemotherapy (β = 67.41, 95% CI 16.39 to 118.43, p = 0.011) were independently associated with higher minimal platelet count during niraparib treatment. In multivariable logistic regression, higher minimal neutrophil count during chemotherapy (OR = 4.51, 95% CI 1.47 to 17.96, p = 0.015) and lower minimal platelet count during chemotherapy (OR = 0.975, 95% CI 0.958 to 0.989, p = 0.001) were independently associated with grade 3 or 4 thrombocytopenia during niraparib treatment. In the olaparib cohort, lower minimal neutrophil count during chemotherapy (β = -0.414, 95% CI -0.755 to -0.074, p = 0.018) and higher hemoglobin concentration at qualification for PARP inhibitor therapy (β = 0.533, 95% CI 0.042 to 1.024, p = 0.034) were independently associated with higher minimal hemoglobin during treatment.

CONCLUSIONS

Early, severe cytopenias during PARP inhibitor therapy can be predicted using clinical data. Platinum-related side effects and parameters for disease diagnosis and eligibility for PARP inhibitor therapy identify high-risk patients, supporting risk-adapted monitoring and proactive dose optimization. These findings should be considered hypothesis-generating and require external validation before implementation in clinical decision-making.

CLINICAL TRIAL NUMBER: Not applicable.