Recurrent hematological toxicity after second-line treatment in patients with glioblastoma: indication of a potential predisposition

Purpose Myelotoxicity is a well-known adverse effect of alkylating chemotherapy for glioblastoma. While risk factors during first-line therapy are established, little is known about myelotoxicity recurrence in second-line treatment. This study investigates whether first-line myelotoxicity therapy predisposes patients to recurrence in the second-line setting. Patients and Methods We conducted a retrospective cohort study of 589 patients with glioblastoma treated at the Brain Tumor Centre Amsterdam (2005-2022). Of these, 178 received second-line lomustine or rechallenge temozolomide. Myelotoxicity severity was predominantly assessed using nadir hematological values and its duration. A log-link generalized linear model evaluated associations between first-line and second-line myelotoxicity severity, adjusting for covariates. Cox proportional hazards models assessed time to myelotoxicity onset. Results We included 151 patients (mean age 57.1±11.8years; 66.9% male). Lomustine was given to 66.9%. Myelotoxicity occurred in 73.5% of patients, with 19.9% developing severe toxicity. First-line myelotoxicity severity was significantly associated with second-line severity (β=1.3, P <.001). Lomustine correlated with higher myelotoxicity severity than temozolomide (β=1.4, P =.002). Higher first-line myelotoxicity scores predicted earlier onset of any-grade (HR=1.4, P <.001) and severe (HR =2.1, P <.001) myelotoxicity in second-line therapy. Conclusion First-line myelotoxicity for glioblastoma predicts its recurrence and earlier onset in second-line therapy. Patients with toxicity in first-line have an increased risk of severe hematological toxicity upon re-exposure. Lomustine carries a higher risk for myelotoxicity than temozolomide. These findings suggest an inherent predisposition to alkylating chemotherapy-induced myelotoxicity for a subgroup of patients. Integrating prior myelotoxicity history into second-line treatment decisions may improve risk stratification and guide monitoring.