When should we consider chronic patients as non-responders to monoclonal antibodies targeting the CGRP pathway?

Abstract

The shut-down of the calcitonin gene-related peptide (CGRP) pathway by monoclonal antibodies (mAbs) has transformed migraine care, representing the first specific way of prevention, proved highly effective by randomized clinical trials [1] and real-world experience [2]. However, due to their high cost, some national drug agencies posed strict restrictions for reimbursement. The Italian Medicines Agency allows it for patients with eight or more migraine days/month, ≥ 3 prophylactic class failures, and a Migraine Disability Assessment (MIDAS) score > 11. Moreover, the reimbursement is maintained only if the MIDAS score reduces by at least 50% after the first 3 months of treatment. In the GARLIT (GalcanezumAb in Real Life in ITaly) study, we showed the 6-month real-life outcomes in patients receiving galcanezumab on direct hospital dispensation before the Italian Medicines Agency had defined reimbursement criteria [2]. Among chronic migraine (CM) patients, 83/130 (63.5%) presented a 50% responder rate (RR) in monthly headache days (MHDs) of at least moderate intensity. Notably, only ten patients discontinued galcanezumab because of dissatisfaction. This is a post-hoc analysis of the data collected in the GARLIT study [2], aiming at reporting the trends of CM patients who presented an MHDs RR < 50% and thus considered as non-responders. Statistical analyses (SPSS version 25.0, SPSS Inc., Chicago, IL, USA) were performed for continuous variables (expressed as means with SD) with paired t test to assess within-group variable changes over time and with one-way ANOVA to compare change between groups (responders with non-responders. In 50%RR nonresponders, although the initial reduction in MHDs gradually became less substantial, at 6 months, MHDs were fewer than baseline (p < 0.001, paired t test, Fig. 1). Besides, while the absolute reduction in pain severity (numerical rating scale, NRS) and HIT-6 score was consistent but significantly less remarkable in non-responders (p = 0.002 and p < 0.001, respectively; one-way ANOVA), the decrease in symptomatic drugs (monthly painkillers intake, MPI (19.5 SD 19.0 vs. 19.3 SD 40.3) was comparable (p = 0.355). Most importantly, Fig. 2 shows the comparison of CM patients with (3mMIDAS responders) and without (3mMIDAS non-responders) ≥ 50% reduction in MIDAS score at 3 months. Although at the sixth month 3mMIDAS responders showed a statistically greater reduction in MHDs (− 14, SD 9.1; RR − 59.7%; p = 0.028) and HIT-6 score (− 13.2 SD 9.7; p = 0.045) than 3mMIDAS non-responders, the latter had a clinically meaningful reduction in both parameters (MHD: − 9, SD 9.6; RR − 33.9%; HIT-6: − 7.9 SD 11.9), and the reduction in pain severity (p = 0.977) and MPI (p = 0.223) was comparable in the two groups. To note, the proportion of 3mMIDAS non-responders showing a ≥ 50%, ≥ 75% and 100% MHDs RR at 6 months was 45.7%, 17.1% and 5.7% respectively. In the non-specific preventives era, physicians used to discontinue treatments, balancing efficacy with tolerability. As a paradox, now we can offer specific and well-tolerated migraine therapy with a convenient monthly injection, but we are forced to stop it even when the patients have a The GARLIT Study Group collaborators members and their respective affiliations are added in acknowledgements section.