Although temozolomide (TMZ) was known to induce thrombocytopenia with subsequent cycle delay, dose reduction and early treatment discontinuation in glioblastoma multiforme (GBM), no early predictive test of these side effects has been yet clearly established. In this context, our aim was to identify the best threshold of early platelet variation predicting TMZ-induced thrombocytopenia during the TMZ maintenance phase and to validate it in an independent series of GBM patients.
It was a retrospective trial including patients suffering from newly diagnosed GBM and treated with TMZ at 75 mg/m2/day concomitant to radiotherapy (RT) followed by TMZ maintenance, according to the Stupp protocol. In a training set, variations of platelet concentrations occurring from the first week to week 6 (ΔW6) were analyzed to identify the most relevant platelet decrease during RT-TMZ associated with at least one clinically relevant TMZ-induced thrombocytopenia (≤100 G/l) in the maintenance phase. An independent validation cohort was used to validate the performance of the ΔW6 threshold.
A total of 147 patients were included: 85 in the training set and 62 in the validation cohort. Twenty seven patients (18%) experienced at least one TMZ-induced thrombocytopenia in the maintenance phase, respectively 14 (16%) and 13 patients (21%) in each cohort; and was the most frequent cause of TMZ schedule changes (49%, 30/61). A platelet decrease at W6 ≥35% (ΔW6≥35%) was identified as the best predictive variation of clinically induced thrombocytopenia with an AUC of 0.83, a sensitivity (Se) of 65% and a specificity (Sp) of 96%. In the validation set, a presence of a ΔW6≥35% of platelet variation was associated with: Se 77% [95% CI 66%-87%], Sp 73% [62%-84%], positive predictive value 42% [29%-54%] and negative predictive value 92% [86%-99%].
Our results showed that a platelet decrease at W6 ≥35% during the RT-TMZ phase may be an early, widely faisable and costless marker of clinically relevant TMZ-induded thrombocyponia during the TMZ maintenance. Prospective studies are needed to evaluate the usefulness of this test for early TMZ schedule adaptation.
Clinical trial identification
Legal entity responsible for the study
Cancer Center Henri Becquerel
Cancer Center Henri Becquerel and IRIB
All authors have declared no conflicts of interest.