Reply to Unexpectedly low rates of neuropsychiatric adverse effects associated with mefloquine repurposed for the treatment of glioblastoma.

Abstract

We appreciate Dr. Nevin’s comments in response to our recent publication.1 Dr. Nevin raises a concern for the attribution of neuropsychiatric adverse events falsely to memantine instead of mefloquine. Mefloquine was approved by the US Food and Drug Administration in 1989 and is the preferred treatment and prophylaxis for malaria in the United States. Various dosing schedules have been investigated to date. Single-administration doses as high as 1500 mg have been well tolerated for the treatment of infection.2 In the setting of prophylaxis, weekly dosing of 250 mg, with3 or without4 a loading dose of 3 consecutive days, likewise has been well tolerated. In response to the author’s question regarding the difference in sample size for the doublet treatment arms, we need to emphasize that this phase 1 study followed the 3+3 study design with a minimum of 3 patients enrolled at each dose level and the enrollment of additional patients if any dose-limiting toxicity (DLT) was observed. The starting dose in each arm was our target dose, and we de-escalated the dose if excessive toxicity was observed. In contrast to the doublet arms containing metformin and memantine, a DLT event was not observed for the mefloquine arm, thus explaining the smaller sample size. We need to highlight that DLT events as defined per protocol included any neuropsychiatric events of grade ≥3 that lasted for ≥14 days and that did not resolve/regress to grade ≤1 within 14 days of onset after optimal supportive measures and any intolerable grade 2 toxic effect (optimal medical therapy must have failed for the event to be considered a DLT). Adverse events were graded according to the National Cancer Institute’s Common Toxicity Criteria (version 4.03). Nevertheless, toxicities that were considered to be definitively, possibly, or probably related to treatment and relevant to the authors’ concerns were noted, but these events did not result in DLTs for the mefloquine-containing arms. In patients on the treatment arm with mefloquine, tremor (grade 1) was the neuropsychiatric adverse event noted, whereas for patients on the treatment arm with memantine and mefloquine, tremor (grade 1), dizziness (grade 2), insomnia (grade 2), gait disturbances (grade 2), and decreased libido (grade 1) were the neuropsychiatric adverse events noted. For patients treated on the treatment arm with metformin and mefloquine, dizziness (grade 1), sensory peripheral neuropathy (grade 1), insomnia (grade 1), and headache (grade 3) were the neuropsychiatric adverse events noted. Last, on patients treated on the treatment arm with metformin, memantine, and mefloquine, dizziness (grade 2), tremor (grade 1), anxiety (grade 1), insomnia (grade 2), tinnitus (grade 1), somnolence (grade 2), cognitive disturbance (grade 1), decreased libido (grade 1), and headache (grade 1) were the neuropsychiatric adverse events noted. The grades of the aforementioned adverse events were the maximum noted. The patients with grade 2 and grade 3 adverse events recovered after either symptomatic treatment or regimen interruption. Depression and abnormal dreams were not observed among patients on the treatment arms containing mefloquine. However, patients with a history of psychosis/schizophrenia were excluded from the treatment arms containing mefloquine. For the adverse event of dizziness, which resulted in a dose reduction of memantine, we feel confident regarding the origin of this symptom because these patients recovered after dose reduction. It remains uncertain which was the offending agent causing dizziness in the treatment arms containing both mefloquine and memantine that did not require dose reduction of either of the 2 agents. Drug-to-drug and/or pharmacokinetic interactions between memantine and mefloquine, which to the best of our knowledge currently are unknown, may be responsible for dizziness and/or other noted neuropsychiatric adverse events. Future studies with mefloquine treatment and larger sample sizes will provide us with additional information regarding the safety of mefloquine as treatment of patients with glioblastoma. No specific funding was disclosed. The authors made no disclosures.