Reply to: Extended adjuvant temozolomide in newly diagnosed glioblastoma: the more, the better?

We thank Drs. Zhao and Ma for their letter on our study (GEINO 14-01)¹ investigating whether patients benefit from continuing adjuvant temozolomide further than 6 standard cycles. They mention a potential imbalance between arms in isocitrate dehydrogenase (IDH) mutation status, which is especially important given the recent reclassification of glioblastoma.² As we stated in a reply to a previous letter, at first glance, there may seem to be an imbalance in IDH mutations³ since there were more patients with mutated IDH in the control arm (7 vs 2). We assessed IDH mutations in 146 patients, and all cases younger than 56 years with negative immunohistochemistry results were sequenced. Unfortunately, some tissue samples were insufficient and 15 could not be sequenced. The inclusion of these unassessed patients (control arm: 6; experimental arm: 9) could well have balanced the results. Moreover, when we included IDH mutation status in the multivariate analyses, the treatment arm was still not identified as a prognostic factor. Furthermore, we re-analyzed only the 134 patients with wild-type IDH (control arm: 65; experimental arm: 69) and found similar results: no differences between arms in progression-free survival (PFS) (HR: 1.12, 95% CI: 0.78-1.61; P = .52) or overall survival (OS) (HR: 0.88, 95% CI: 0.59-1.31; P = .54) (Figure 1).

Fig. 1.

(A) Progression-free survival and (B) overall survival for the 134 patients with a wild-type IDH.

Drs. Zhao and Ma also point out the potential impact on our results of neurological symptoms, patient age, and other prognostic factors. The neurological symptoms in our patients (persistent grade 1-2 neurological deficiencies) caused no difference in Karnofsky Performance Status (KPS) between arms (P = .68) and were not associated with PFS or OS. There were no age limitations on inclusion provided the patient had received radiation with concomitant and 6 adjuvant cycles of temozolomide. The age distribution between arms was similar (Student t test P = .334), with 28 patients (35.4%) in the control arm and 27 (33.8%) in the experimental arm younger than 55. In fact, after the publication of Perry’s results,⁴ patients could be included if they had received hypofractionated radiotherapy. Three patients in the control arm (3.7%) and 4 (5% in the experimental arm (P = .68) received hypofractionated radiation due to age or fragility.

When we analyzed all potential prognostic factors in addition to IDH mutations and MGMT (O⁶-methylguanine-DNA methyltransferase) methylation (residual disease, KPS, Mini-Mental State Examination (MMSE), need for dexamethasone, neurological symptoms, age), only IDH mutation status, MGMT methylation, and measurable disease at randomization were associated with PFS, and only MMSE, age and need for dexamethasone were associated to OS. These variables were balanced between the 2 arms. Moreover, the multivariate analyses including all these variables plus treatment arm identified only MGMT methylation and absence of residual disease at randomization as independent factors for both PFS and OS.

Based on these data, together with Drs Zhao and Ma’s findings in their pooled study of our trial and a previous meta-analysis,⁵ which confirmed our findings and supported the recent EANO (European Association of Neuro-Oncology) guidelines,⁶ we are confident that continuing adjuvant temozolomide after 6 cycles would not prove beneficial in glioblastoma patients. Updated results will be presented at the upcoming ASCO (American Society of Clinical Oncology) meeting.

Funding

This trial was supported by a grant from the Spanish Institute Carlos III (ISCIII: PI13/01751).

Conflict of interest statement

C.B. reports a grant from Instituto de Salud Carlos III, during the conduct of the study. All other authors declare no competing interests.