Pain Ther (2021) 10:637–650 https://doi.org/10.1007/s40122-021-00253-0 ORIGINAL RESEARCH Early Management of OnabotulinumtoxinA Treatment in Chronic Migraine: Insights from a Real-Life European Multicenter Study Raffaele Ornello . Fayyaz Ahmed . Andrea Negro . Anna Maria Miscio . Antonio Santoro . Alicia Alpuente . Antonio Russo . Marcello Silvestro . Sabina Cevoli . Nicoletta Brunelli . Fabrizio Vernieri . Licia Grazzi . Carlo Baraldi . Simona Guerzoni . Anna P. Andreou . Giorgio Lambru . Ilaria Frattale . Katharina Kamm . Ruth Ruscheweyh . Marco Russo . Paola Torelli . Elena Filatova . Nina Latysheva . Anna Gryglas-Dworak . Marcin Straburzynski . Calogera Butera . Bruno Colombo . Massimo Filippi . Patricia Pozo-Rosich . Paolo Martelletti . Simona Sacco Received: January 13, 2021 / Accepted: March 6, 2021 / Published online: March 28, 2021 Ó The Author(s) 2021 ABSTRACT Methods: We performed a retrospective analy- sis on real-life prospectively collected data in 16 Introduction: OnabotulinumtoxinA (BT-A) European headache centers. All the centers quarterly was the first treatment approved provided data on patients treated with BT-A for specifically for chronic migraine (CM). It is CM over the first three cycles of treatment. For unclear whether three cycles are better than two each treatment cycle we defined patients as to assess early BT-A response. ‘‘good responders’’ if reporting a C 50% reduc- tion in monthly headache days compared with the three months before starting BT-A, ‘‘partial responders’’ if reporting a 30–49% reduction in Supplementary Information The online version monthly headache days, and ‘‘non-responders’’ contains supplementary material available at https:// doi.org/10.1007/s40122-021-00253-0. if reporting a \ 30% reduction in monthly R. Ornello  S. Sacco (&) A. Alpuente  P. Pozo-Rosich Neuroscience Section, Department of Applied Headache Unit, Department of Neurology, Vall Clinical Sciences and Biotechnology, University of D’Hebron University, Barcelona, Spain L’Aquila, Via Vetoio 1 Coppito, 67100 L’Aquila, Italy A. Alpuente  P. Pozo-Rosich e-mail: simona.sacco@univaq.it Headache and Neurological Pain Research Group, Vall D’Hebron Institute of Research (VHIR), F. Ahmed Department of Medicine, Universitat Autonoma de Department of Neurosciences, Hull University Barcelona, Barcelona, Spain Teaching Hospitals, Hull, UK A. Russo  M. Silvestro A. Negro  P. Martelletti Headache Center, Department of Medical, Surgical, Department of Clinical and Molecular Medicine, Neurological, Metabolic, and Aging Sciences, Faculty of Medicine and Psychology, Sant’Andrea University of Campania ‘‘Luigi Vanvitelli’’, Naples, Hospital, Sapienza University, 00189 Rome, RM, Italy Italy S. Cevoli A. M. Miscio  A. Santoro IRCCS Istituto Delle Scienze Neurologiche Di Headache Center, Unit of Neurology, , Fondazione Bologna, Bologna, Italy IRCCS ‘‘Casa Sollievo Della Sofferenza’’, San Giovanni Rotondo, Foggia, Italy 638 Pain Ther (2021) 10:637–650 headache days or stopping the treatment before the third cycle. Key Summary Points Results: We included 2879 patients. Seven hundred and eighty-four (64.6%) of the 1213 Why carry out this study? patients reporting a good response during the first and/or the second cycle had a good OnabotulinumtoxinA treatment for response during the third cycle; 309 (49.3%) of chronic migraine is very effective in many the 627 patients reporting a partial response most patients; however, it is unclear (but no good response) during the first and/or whether the onset of its efficacy should be the second cycle had a good response during assessed after two or three cycles. the third cycle; only 65 (6.3%) of the 1039 We performed a large, retrospective, patients who did not respond during both the multicenter study to assess the evolution first two cycles achieved a good response during of response to onabotulinumtoxinA over the third cycle. Multivariate analyses showed the first three cycles of treatment. that partial or good response during the first or the second cycle were independently associated What was learned from this study? with good response during the third cycle. We found that most patients with chronic Conclusions: Our data suggest that patients migraine reporting at least a partial with CM responding to BT-A during the first response to any of the first two cycles of two cycles will likely benefit from the third onabotulinumtoxinA reported a [ 50% cycle of treatment, while the probability that decrease in monthly headache days non-responders to the first two cycles start during the third cycle. responding during the third cycle is low. These results can help guide the individual decision to Patients not responding to both the first stop or continue treatment after the second two cycles are unlikely to respond during cycle in patients who have not responded to the the third cycle. first two cycles. Keywords: Chronic migraine; Multicenter study; OnabotulinumtoxinA; Real-life evidence M. Russo N. Brunelli  F. Vernieri Headache Center, Neurology Unit, Neuromotor and Headache and Neurosonology Unit, Campus Bio- Rehabilitation Department, Azienda USL-IRCCS Di Medico University Hospital, Rome, Italy Reggio Emilia, Reggio Emilia, Italy L. Grazzi P. Torelli Headache Center, Neuroalgology Department, Headache Center, University of Parma, Parma, Italy IRCCS Foundation ‘‘Carlo Besta’’ Neurological Institute, via Celoria, 11, 20133 Milan, Italy E. Filatova  N. Latysheva Department of Neurology, Institute for Postgraduate C. Baraldi  S. Guerzoni Education, Sechenov First Moscow State Medical Department of Biomedical, Metabolic and Neural University (Sechenov University), Moscow, Russia Sciences, Medical Toxicology, Headache and Drug Abuse Research Center, University of Modena and A. Gryglas-Dworak Reggio Emilia, Modena, Italy Headache Center Wroclaw, Wrocław, Poland A. P. Andreou  G. Lambru M. Straburzynski Headache Service, Guy’s and St Thomas’ NHS Headache Clinic, Terapia Neurologiczna Foundation Trust, London, UK Samodzielni, Maurycego Mochnackiego 10, 02-042 Warsaw, Poland K. Kamm  R. Ruscheweyh Department of Neurology, Ludwig Maximilians University München, Munich, Germany Pain Ther (2021) 10:637–650 639 DIGITAL FEATURES and in which to proceed with the third one. Furthermore, the National Institute for Health This article is published with digital features, and Care Excellence (NICE) UK guidance for the including a summary slide, to facilitate under- use of BT-A in CM recommends discontinuing standing of the article. To view digital features the treatment after two cycles if patients do not for this article go to https://doi.org/10.6084/ reach at least 30% reduction in headache days m9.figshare.14170229. [16]. The therapeutic dilemma of negative stop- ping criteria for BT-A treatment has gained INTRODUCTION importance with the advent of specific thera- peutic alternatives for CM, i.e. calcitonin gene- OnabotulinumtoxinA (BT-A) is the first treat- related peptide (CGRP)-targeted treatments ment approved specifically for chronic migraine [17, 18]. Besides, patients with CM have sur- (CM); its efficacy and safety have been proven prisingly low persistence to oral preventive in randomized clinical trials [1–5] and con- treatments, possibly due to their perceived firmed in real-life studies [6–8]. The response to ineffectiveness or to tolerability issues [19, 20]. BT-A varies in degree of efficacy and time of For those reasons, providing reliable estimates onset amongst patients. Many patients report a of the onset and the early evolution of response favorable response to BT-A treatment during the to BT-A might help avoid prolonged ineffective first cycle of treatment with BT-A, whereas other treatment and favor patient compliance. patients who do not report any meaningful In order to improve the management of improvement during the first cycle start patients with CM who start BT-A treatment, we responding to the treatment during the second aimed to provide data to clarify how the clinical or third cycle. Unfortunately, other patients do response develops and evolves over the first not respond even after prolonged treatment [9]. three treatment cycles. However, despite the wealth of studies assessing the possible predictors of response to BT-A [10–14], none of them have been commonly METHODS accepted. The European Headache Federation (EHF) Inclusion Criteria guidelines for the use of BT-A in CM suggest stopping the treatment in the presence of \ We performed a retrospective analysis of 30% reduction in monthly headache days prospectively collected data from European compared to baseline after 2–3 cycles of treat- centers treating CM with BT-A. All the centers ment [15]. However, it is unclear in which met the following inclusion criteria: patients to stop treatment at the second cycle 1. Having already performed a real-life prospective data collection on patients with C. Butera  M. Filippi CM, diagnosed according to the Interna- Neurophysiology Service, IRCCS San Raffaele Scientific Institute, Milan, Italy tional Classification of Headache Disorders (ICHD) criteria, aged C 18 years, treated B. Colombo  M. Filippi with BT-A 155–195 units quarterly accord- Neurology Unit, IRCCS San Raffaele Scientific Institute, Milan, Italy ing to the Phase 3 REsearch Evaluating Migraine Prophylaxis Therapy (PREEMPT) M. Filippi protocol [1] Neurorehabilitation Unit, IRCCS San Raffaele 2. Study approval by the local ethics commit- Scientific Institute, Milan, Italy tees and informed consent obtained from I. Frattale patients, if necessary, according to the local Child Neurology and Psychiatry Unit, Systems regulations Medicine Department, Tor Vergata University, Rome, Italy 3. Ability to share a database of collected data 640 Pain Ther (2021) 10:637–650 4. Planned follow-up of at least 9 months for treatment before the third cycle and those lost all patients, irrespective of treatment to follow-up were considered as non-respon- discontinuation ders. Variables were considered for the analyses, if available, for at least two thirds of patients; no Centers were selected following a literature imputation was done for missing data. We per- search including the terms ‘‘botulinum toxin’’ formed the main analyses on the whole data set, and ‘‘migraine’’ or by personal contacts. Pre-se- comparing the characteristics of good respon- lected centers were contacted by e-mail. Nine- ders at the third BT-A cycles with those of par- teen centers were contacted; two centers did not tial or non-responders. We then repeated the meet the inclusion criteria, while one center analyses after excluding patients with partial declined to participate in the study; 16 centers response during the third BT-A cycle, to better agreed to participate in the study. Inclusion differentiate patient subgroups according to BT- periods varied across centers and ranged from A response. We also performed sensitivity 2010 to 2020. All centers collected data by using analyses after the exclusion of patients not prospective headache diaries. receiving the third BT-A dose to correct for the effect of negative stopping rules adopted in Ethical Aspects some centers [16]. No sample size calculation was performed, as the sample size was based on The present analysis was approved by the the available data. Internal Review Board of the University of We computed the proportions of good L’Aquila with protocol number 23/2020. responders, partial responders, and non-re- Patients did not have to sign an additional sponders after 3, 6, and 9 months from BT-A informed consent, as no additional data were treatment initiation. Descriptive statistics were required for the present analyses. reported as numbers and proportions and means and standard deviations (SDs), as Definition of Treatment Response appropriate. We performed univariate compar- isons between patients reporting and not According to response to BT-A, patients were reporting a good response to BT-A using the chi- defined as follows: square test or Student’s t test for independent samples, as appropriate. The variables reaching 1. Good responders patients achieving a C 50% a P value \ 0.1 for significance were included in reduction in headache days from baseline— a multinomial stepwise logistic regression i.e. the 3 months prior to BT-A treatment model to identify the factors independently initiation—to the respective 3-month cycle associated with response during the third cycle. 2. Partial responders patients achieving a Two-tailed P for significance was set at \ 0.05. 30–49% reduction in headache days from Statistical analyses were conducted using SPSS baseline to the respective 3-month cycle version 20 software. 3. Non-responders patients achieving a \ 30% reduction in headache days from baseline to the respective 3-month cycle. The non- RESULTS responder group also included patients lost to follow-up or discontinuing treatment Baseline Characteristics before the third cycle. During the study period, we included 2879 patients, 81.7% female, with a mean age of Statistical Analysis 46.6 ± 12.3 years and a mean CM duration of 8.0 ± 8.0 years. Table 1 summarizes the All analyses used the intent-to-treat population, patients’ characteristics at baseline. Table S1 in including all patients who started BT-A treat- the electronic supplementary material reports ment. As reported earlier, those discontinuing the contribution of each center. Pain Ther (2021) 10:637–650 641 Table 1 Baseline characteristics of the 2879 included Predictors of Response subjects Variable Univariate comparisons showed that good responders during the third cycle had a higher Age (years), mean ± SD 46.6 ± 12.3 proportion of females (83.5% vs 80.4%; Sex (female), n (%) 2351 (81.7) P = 0.020) and medication overuse (78.8% vs 63.3%; P \ 0.001), a lower number of monthly Medication overuse, n (%) 2055 (71.4) headache days (23.5 ± 5.6 vs 24.0 ± 6.2; Baseline headache days, mean ± SD 23.8 ± 5.9 P \ 0.001) and a higher number of monthly medication days (20.8 ± 8.0 vs CM duration (years), mean ± SD 8.0 ± 8.0 18.3 ± 9.6 years; P \ 0.001) at baseline com- Acute medication days (number), 19.4 ± 9.1 pared with partial responders or non-respon- mean ± SD ders; good responders during the third cycle also had a higher proportion of good or partial response during the first two cycles (Table 2). However, after multivariate adjustments, the Treatment Response only factor independently associated with good response during the third cycle was partial or good response during the first or the second During the first BT-A cycle, 774 (26.9%) patients cycle (Table 2). were good responders, 534 (18.5%) partial After the exclusion of partial responders at responders, and 1571 (54.6%) non-responders the third cycle, the multivariate analyses (Fig. 1). The proportion of good responders showed that medication overuse and partial or increased through the second (946 good good response during the first or the second responders; 32.9%; 552 partial responders, cycle were independently associated with good 19.2%; 1381 non-responders, 47.9%) and the response during the third cycle (see Table S2 in third cycle (1158 good responders; 40.2%; 337 the electronic supplementary material). partial responders, 11.7%; 1384 non-respon- ders, 48.1%; Fig. 1). Of the 1213 patients reporting a good Sensitivity Analysis response during the first and/or the second cycle, 784 (64.6%) reported a good response and After the exclusion of patients not receiving the 109 (9.0%) a partial response during the third third BT-A dose because of a negative stopping cycle; notably, of the 507 patients who were rule or physician/patient decision, 2080 good responders during both the first and sec- patients were left. The proportions of good ond cycle, 389 (76.7%) maintained the good responders, partial responders, and non-re- responder status during the third cycle. A partial sponders were similar to those of the overall response during either the first or the second group (see Fig. S1 in the electronic supplemen- cycle, without a good response during the first tary material). Sixty-five (11.5%) of the 565 two cycles, was attained by 627 patients, 309 of patients maintaining the non-responder status whom (49.3%) had a good response and 130 during both the first two cycles achieved a good (20.7%) a partial response during the third response during the third cycle, while 98 cycle. Only 65 (6.3%) of the 1039 patients (17.3%) had a partial response. maintaining the non-responder status during Univariate comparisons showed that good both the first two cycles achieved a good responders during the third cycle had a higher response during the third cycle, while 98 (9.4%) prevalence of medication overuse (78.8% vs had a partial response (see Fig. 2). 74.2%; P = 0.004) and a higher number of acute medication days (20.8 ± 8.0 vs 19.6 ± 8.9; P = 0.003) compared with the remaining patients (see Table S3 in the electronic 642 Pain Ther (2021) 10:637–650 Fig. 1 Overall response to onabotulinumtoxinA during the first three cycles in the study population supplementary material). Good responders DISCUSSION during the third cycle also had a higher pro- portion of good and partial response during the In our study, a good or at least partial response first two cycles compared with the remaining during the first two BT-A cycles was consistently patients (see Table S3 in the electronic supple- associated with a good response during the mentary material). The multivariate analyses third cycle across the main and several sensi- also confirmed that good or partial response to tivity analyses. Our study focused on the first BT-A during the second cycle in this subgroup three cycles of BT-A treatment, as it aimed to was independently associated with good provide guidance for the decision of whether to response during the third cycle, together with continue or to stop the treatment after the younger age (see Table S3 in the electronic second cycle, according to the initial response. supplementary material). Our data suggest that (1) patients showing early response to BT-A tend to maintain their response; (2) a relevant proportion of patients Pain Ther (2021) 10:637–650 643 Fig. 2 Detailed flowchart of response to onabotulinum- response; red numbers, no response. Numbers in paren- toxinA across the first three treatment cycles. Green theses indicate proportions over total (N = 2879) numbers indicate good response; yellow numbers, partial showing even partial response to BT-A during developed a new-onset response and patients the first two cycles of treatment will become who lost their response over time. This was good responders during the third cycle; and 3) possible because of the large number of inclu- the probability of good clinical response to BT-A ded patients selected from international real-life is lower in patients not responding during both practice; all patients would have fulfilled the the first two cycles. recently issued definition of ‘‘resistant The added value of the present study is pro- migraine’’ [21]. As investigators were not aware viding the detail of response to BT-A during of the study aim a priori, we can exclude any each cycle (Fig. 1), considering patients who influence on treatment disposition according to 644 Pain Ther (2021) 10:637–650 Table 2 Comparison between patients with good response and patients with partial or no response during the third onabotulinumtoxinA treatment cycle Status at the third cycle Good Partial or non- P value OR (95% CI) P value responders responders (univariate) (multivariate) (n = 1158) (n = 1721) Baseline characteristics Female, n (%) 967 (83.5) 1384 (80.4) 0.020 1.09 0.540 (0.83–1.43) Age, mean ± SD 46.3 ± 11.7 46.9 ± 12.7 0.200 – Medication overuse, n 913 (78.8) 1142 (63.3) \ 0.001 1.40 0.074 (%) (0.97–2.02) Years of CM history, 7.7 ± 7.6 8.3 ± 8.4 0.110 – mean ± SD Baseline headache days, 23.5 ± 5.6 24.0 ± 6.2 0.013 0.99 0.475 mean ± SD (0.97–1.01) Baseline acute medication 20.8 ± 8.0 18.3 ± 9.6 \ 0.001 1.02 0.080 days, mean ± SD (1.00–1.04) Response during the first \ 0.001 cycle, n (%) Good 458 (39.6) 316 (18.4) 1.36 0.024 (1.04–1.79) Partial 291 (25.1) 243 (14.1) 1.43 0.011 (1.09–1.88) None 409 (35.3) 1162 (67.5) 1.00 (Ref.) Response during the \ 0.001 second cycle, n (%) Good 715 (61.7) 231 (13.4) 24.72 \ 0.001 (18.70–32.66) Partial 323 (27.9) 229 (13.3) 15.51 \ 0.001 (11.73–20.50) None 120 (10.4) 1261 (73.3) 1.00 (Ref.) personal expectations. On the other hand, we English, \ 50% according to the Australian cri- have to consider that regulations and physician teria) or personal beliefs, physicians discontin- attitude have implications for the results. In ued treatment in non-responders after the some countries and in some centers, because of second cycle. This may have led to an under- regulatory constraints [8, 16, 22] (\ 30% estimation of the benefit of BT-A in non-re- decrease in headache days according to the sponders. To control for that potential bias, we Pain Ther (2021) 10:637–650 645 repeated the analyses excluding patients who patients develop a new-onset response to BT-A did not receive the third BT-A dose because of a during the third or fourth cycle [29]. Response negative stopping rule or physician/patient to BT-A may fluctuate over time, and not all decision; those sensitivity analyses still showed responders show a sustained response to the that partial or good response during the second drug [29, 30]. Fluctuations in response to BT-A BT-A cycle can help in predicting response might be common in the long term if we con- during the third cycle. sider each single patient. In our study, the proportion of patients Our data have further limitations. Although reporting a C 50% reduction in monthly head- the data were collected prospectively, the study ache days compared with baseline, i.e. ‘‘good had a retrospective design, leading to hetero- responders,’’ was lower (26.9%, 32.9%, and geneity of data collection across centers. We 40.2% during the first, second, and third BT-A minimized the collected data in order to limit cycle, respectively) as compared with those of missing information and inaccuracy. We rec- the PREEMPT trials, in which 49.0% of patients ognize that important information, such as the had a C 50% response during the first cycle presence of cutaneous allodynia and the num- [23]. This difference might be due to the dif- ber of prior preventive treatment failures, was ference in study design and setting, as real-life not available from most centers. For the same patients have a higher disease burden than the reason, we could not consider the effect of BT-A highly selected patients included in clinical dosage, which is likely higher with increasing trials. Similarly to the available trials, in our BT-A doses [31]. A further study limitation is population the proportion of good responders represented by the heterogeneity in the quan- increased during the first three cycles (Fig. 1). titative contribution of different centers (see Together with good responders, our study also Table S1 in the electronic supplementary considered ‘‘partial responders,’’ reporting a material); large centers might have had a greater 30–49% reduction in monthly headache days impact on the study than the smaller ones. We compared with baseline. In fact, a 30% decrease also did not distinguish migraine days from in monthly headache days was suggested to be headache days, which could have added further acceptable in some patients with CM [24, 25]. detail to outcome assessment. Furthermore, we Our study suggests that the condition of partial could not distinguish patients discontinuing responder to BT-A is relevant, as it may predict BT-A because of ineffectiveness from those who the future condition of good responder. discontinued because of noncompliance. It is worth mentioning that our definition of the responder status was based only on the Implications for Clinical Practice decrease in monthly headache days compared in the New Treatment Era with baseline. The efficacy assessment of CM treatments should consider not only headache Our findings suggest that it is not reasonable to days, but also migraine days, moderate/severe discontinue the treatment with BT-A after only headache days, medication use, conversion one cycle, otherwise a relevant proportion of from CM to episodic migraine, termination of patients may miss a valid therapeutic opportu- medication overuse, and patient-reported out- nity; in our study, 532 (33.8%) of the 1571 non- comes and scales. These outcomes are suggested responders during the first cycle achieved a by guidelines and reported by real-life studies good or partial response during the second [15, 26–28]. cycle, while 590 (37.6%) achieved a good or A further point which needs to be considered partial response during the third cycle. Con- is that we focused on the first three BT-A cycles. sidering the findings of this study, continuing Evidence suggests that about one quarter of BT-A treatment is advisable in patients report- patients treated with BT-A who develop a ing even a partial response within the second response to the drug within the first two cycles cycle, while stopping treatment might be con- might lose their response during the fourth sidered in patients who are non-responders cycle, while a comparable proportion of 646 Pain Ther (2021) 10:637–650 during the first two treatment cycles, because CONCLUSION the probability of a later response is low. Our data can help in establishing clearer rules for According to our data, the first two treatment BT-A discontinuation, which according to cur- cycles of BT-A can help determine the response rent guidelines, is reasonable if the patient to the drug during the third cycle in patients ‘‘does not respond’’ to the first ‘‘2–3’’ treatment with CM. Patients reporting at least a 30% cycles [15]. Our data suggest that we should decrease in monthly headache days within the wait for a 30% reduction in monthly headache first two cycles compared with baseline are days over the first two treatment cycles to likely to respond to the third cycle of BT-A. determine the efficacy of BT-A. Many patients Conversely, patients not reporting at least a with CM continue treatment with BT-A because partial response to the first two cycles are unli- of a decrease in headache duration and/or kely to respond to the subsequent cycle and intensity, even in the absence of a decrease in might be considered for treatment discontinu- monthly headache days, as they still perceive a ation. Our data are limited by the small number benefit. This aspect of BT-A benefit is usually of outcomes assessed; therefore, they cannot assessed by patient-reported outcomes, such as constitute a guideline and cannot replace clin- standardized scales, for which the present study ical judgment. Nevertheless, we think that our does not provide enough data. study can help inform clinical practice. The possibility of early treatment switch from BT-A is important, given the long duration (3 months) of a treatment cycle. Establishing criteria for early treatment change from BT-A is ACKNOWLEDGEMENTS of increasing importance in light of the rise in CGRP-targeted treatments to treat CM [18, 32]; Funding. No funding or sponsorship was the new BT-A treatment algorithms should received for this study or publication of this include the possibility of early switch to other article. effective agents [28]. The evidence on the combination of BT-A with CGRP-targeted Authorship. All named authors meet the treatment is only preliminary [33, 34], and International Committee of Medical Journal some European regulatory authorities prohibit Editors (ICMJE) criteria for authorship for this such a combination. If CGRP-targeted treat- article, take responsibility for the integrity of ments can be offered to patients with insuffi- the work as a whole, and have given their cient response to BT-A, stopping BT-A as early as approval for this version to be published. after the first two cycles and switching to a CGRP-targeted treatment might be a viable Medical Writing, Editorial, and Other option. In contrast, if a patient starts BT-A Assistance. The Authors wish to thank Vı́ctor treatment after failing CGRP-targeted treat- José Gallardo López from the Vall d’Hebron ments, it might be reasonable to prolong the Institute of Research (VHIR), Barcelona, Spain, treatment for more than 3–4 cycles before for statistical advice. addressing the clinical response, especially if the patients have already tried the other available Prior Presentation. Preliminary results of preventive options. Future personalized the present work were presented in poster form approaches to treatment, including pharmaco- at the 34th congress of the Italian Society for genetics [35, 36], will likely provide an objective the Study of Headaches (SISC), virtual edition, basis for the prediction of response to BT-A as 23–26 September, 2020, and at the 51st congress well as to other migraine preventive treatments. of the Italian Neurological Society (SIN), virtual edition, 20–30 November, 2020. Disclosures. Raffaele Ornello reports speaker fees from Novartis and Eli Lilly, travel grants Pain Ther (2021) 10:637–650 647 from Teva, and support for publication from Ricerca per la SLA). Elena Filatova reports Novartis and Allergan. Simona Sacco reports speaker fees from Novartis and Teva. Nina personal fees from Allergan-AbbVie, AstraZe- Latysheva reports speaker fees from Novartis neca, Abbott, Teva, Novartis, and Eli Lilly. Alicia and Allergan. Carlo Baraldi has received travel Alpuente has received honoraria for educational grants and honoraria from Allergan, Novartis, projects from Allergan-Abbvie, Novartis, and Teva, and Ely Lilly. Licia Grazzi has received Teva. Patricia Pozo-Rosich has received hono- consultancy and advisory fees from Allergan, raria as a consultant and speaker for Allergan- electroCore LLC, Novartis, and Eli Lilly, and Abbvie, Almirall, Amgen, Biohaven, Chiesi, Eli collaborates in trials sponsored by Eli Lilly, Lilly, Medscape, Neurodiem, Novartis, and Teva, and PharmInd. Simona Guerzoni has Teva; her research group has received research received travel grants and honoraria from grants from Allergan, and has received funding Allergan, Eli Lilly, Teva, and Novartis. Sabina for clinical trials from Alder, Allergan-Abbvie, Cevoli has received honoraria as speaker, for electroCore, Eli Lilly, Novartis, and Teva; she is participating on advisory boards, or for clinical a trustee member of the board of the Interna- investigation studies from Teva, Allergan, tional Headache Society, member of the Coun- Novartis, Lilly, and Ibsa. Fayyaz Ahmed has cil of the European Headache Society; she is on received honorarium from Allergan for being on the editorial board of Revista de Neurologia, their advisory board and is paid to the charita- associate editor of Cephalalgia, Headache, Neu- ble organizations. Giorgio Lambru has received rologia, and Frontiers in Neurology, and advisor speaker honoraria and funding for travel, and for The Journal of Headache and Pain; she is a has received honoraria for participation on member of the Clinical Trials Standing Com- advisory boards sponsored by Allergan, Novar- mittee of the International Headache Society; tis, Eli Lilly, and Teva; he has also received she has edited the Guidelines for the Diagnosis speaker honoraria and funding for travel from and Treatment of Headache of the Spanish electroCore, Nevro Corp., and Autonomic Neurological Society; she is the founder of Technologies. Anna P. Andreou has received www.midolordecabeza.org; she does not own speaker honoraria and funding for travel from stocks from any pharmaceutical company. Allergan, Eli Lilly, and eNeura, honoraria for Andrea Negro has received speaking honoraria participation on advisory boards sponsored by and has served on the advisory boards of Aller- Allergan and Eli Lilly, sponsorship for educa- gan, Eli Lilly and Company, and Novartis. Paolo tional purposes from eNeura, Allergan, Auto- Martelletti has received speaker’s honoraria and nomic Technologies, and Novartis, and an has served on the advisory boards of Allergan, equipment grant from eNeura. Antonio Russo Eli Lilly and Company, Novartis, and Teva has received speaker honoraria from Allergan, Pharmaceuticals; he is also an EMA Expert and Lilly, Novartis, and Teva, and serves as an has served as Editor-in-Chief of The Journal of associate editor of Frontiers in Neurology (Head- Headache and Pain and as Editor-in-Chief of SN ache Medicine and Facial Pain session). Mar- Comprehensive Clinical Medicine; he also declares cello Silvestro has received speaker honoraria royalties received from Springer Nature. Mas- from Lilly, Novartis, and Teva. Ruth Rusche- simo Filippi is Editor-in-Chief of the Journal of weyh has received travel grants and/or hono- Neurology and Associate Editor of Neurological raria from Allergan, Hormosan, Lilly, Novartis, Sciences; received compensation for consulting and Teva. Fabrizio Vernieri has received travel services and/or speaking activity from Bayer, grants, honoraria for advisory boards, and Biogen Idec, Merck Serono, Novartis, Roche, speaker panels from Allergan, Angelini, Lilly, Sanofi Genzyme, Takeda, and Teva Pharma- Novartis, and Teva. Marcin Straburzynski has ceutical Industries; and receives research sup- received personal fees from Novartis and Teva. port from Biogen Idec, Merck Serono, Novartis, Katharina Kamm has received travel grants and/ Teva Pharmaceutical Industries, Roche, Italian or honoraria from Lilly, Novartis, and Teva. Ministry of Health, Fondazione Italiana Sclerosi Paola Torelli has received travel grants and/or Multipla, and ARiSLA (Fondazione Italiana di honoraria from Allergan, Lilly, Novartis, and 648 Pain Ther (2021) 10:637–650 Teva. Anna Gryglas-Dworak has received 2010;30(7):793–803. https://doi.org/10.1177/ honoraria as a speaker and travel grants from 0333102410364676. Allergan and honoraria as a speaker from 2. Aurora SK, Winner P, Freeman MC, Spierings EL, Novartis. Ilaria Frattale, Anna Maria Miscio, Heiring JO, DeGryse RE, et al. OnabotulinumtoxinA Antonio Santoro, Nicoletta Brunelli, Bruno for treatment of chronic migraine: pooled analyses Colombo, Calogera Butera, and Marco Russo of the 56-week PREEMPT clinical program. Head- ache. 2011;51(9):1358–73. https://doi.org/10.1111/ declare no conflicts of interest. j.1526-4610.2011.01990.x. Compliance with Ethics Guidelines. The 3. Aurora SK, Dodick DW, Diener HC, DeGryse RE, present analysis was approved by the Internal Turkel CC, Lipton RB, et al. OnabotulinumtoxinA for chronic migraine: efficacy, safety, and tolera- Review Board of the University of L’Aquila with bility in patients who received all five treatment protocol number 23/2020. Patients did not have cycles in the PREEMPT clinical program. Acta to sign an additional informed consent as no Neurol Scand. 2014;129(1):61–70. https://doi.org/ additional data were required for the present 10.1111/ane.12171. analyses. 4. Diener HC, Dodick DW, Aurora SK, Turkel CC, DeGryse RE, Lipton RB, et al. OnabotulinumtoxinA Data Availability. The data sets analyzed for treatment of chronic migraine: results from the during the current study are available from the double-blind, randomized, placebo-controlled participant centers upon reasonable request. phase of the PREEMPT 2 trial. Cephalalgia. 2010;30(7):804–14. https://doi.org/10.1177/ 0333102410364677. Open Access. This article is licensed under a Creative Commons Attribution-NonCommer- 5. Dodick DW, Turkel CC, DeGryse RE, Aurora SK, cial 4.0 International License, which permits Silberstein SD, Lipton RB, et al. Onabotulinumtox- any non-commercial use, sharing, adaptation, inA for treatment of chronic migraine: pooled results from the double-blind, randomized, pla- distribution and reproduction in any medium cebo-controlled phases of the PREEMPT clinical or format, as long as you give appropriate credit program. Headache. 2010;50(6):921–36. https:// to the original author(s) and the source, provide doi.org/10.1111/j.1526-4610.2010.01678.x. a link to the Creative Commons licence, and 6. Blumenfeld AM, Stark RJ, Freeman MC, Orejudos A, indicate if changes were made. The images or Manack AA. Long-term study of the efficacy and other third party material in this article are safety of OnabotulinumtoxinA for the prevention included in the article’s Creative Commons of chronic migraine: COMPEL study. J Headache licence, unless indicated otherwise in a credit Pain. 2018;19(1):13. https://doi.org/10.1186/ s10194-018-0840-8. line to the material. If material is not included in the article’s Creative Commons licence and 7. Ahmed F, Gaul C, Garcı́a-Moncó JC, Sommer K, your intended use is not permitted by statutory Martelletti P, Investigators RP. An open-label regulation or exceeds the permitted use, you prospective study of the real-life use of onabo- tulinumtoxinA for the treatment of chronic will need to obtain permission directly from the migraine: the REPOSE study. J Headache Pain. copyright holder. To view a copy of this licence, 2019;20(1):26. https://doi.org/10.1186/s10194-019- visit http://creativecommons.org/licenses/by- 0976-1. nc/4.0/. 8. Andreou AP, Trimboli M, Al-Kaisy A, Murphy M, Palmisani S, Fenech C, et al. Prospective real-world analysis of OnabotulinumtoxinA in chronic migraine post-National Institute for Health and REFERENCES Care Excellence UK technology appraisal. Eur J Neurol. 2018;25(8):1069-e83. https://doi.org/10. 1111/ene.13657. 1. Aurora SK, Dodick DW, Turkel CC, DeGryse RE, Silberstein SD, Lipton RB, et al. Onabotulinumtox- 9. Sarchielli P, Romoli M, Corbelli I, Bernetti L, Verz- inA for treatment of chronic migraine: results from ina A, Brahimi E, et al. Stopping onabotulinum the double-blind, randomized, placebo-controlled treatment after the first two cycles might not be phase of the PREEMPT 1 trial. Cephalalgia. justified: results of a real-life monocentric prospec- tive study in chronic migraine. Front Neurol. Pain Ther (2021) 10:637–650 649 2017;8:655. https://doi.org/10.3389/fneur.2017. 19. Hepp Z, Dodick DW, Varon SF, Gillard P, Hansen 00655. RN, Devine EB. Adherence to oral migraine-pre- ventive medications among patients with chronic 10. Alpuente A, Gallardo VJ, Torres-Ferrus M, Alvarez- migraine. Cephalalgia. 2015;35(6):478–88. https:// Sabin J, Pozo-Rosich P. Early efficacy and late gain doi.org/10.1177/0333102414547138. in chronic and high-frequency episodic migraine with onabotulinumtoxinA. Eur J Neurol. 2019. 20. Hepp Z, Dodick DW, Varon SF, Chia J, Matthew N, https://doi.org/10.1111/ene.14028. Gillard P, et al. Persistence and switching patterns of oral migraine prophylactic medications among 11. Alpuente A, Gallardo VJ, Torres-Ferrús M, Álvarez- patients with chronic migraine: A retrospective Sabin J, Pozo-Rosich P. Short and mid-term predic- claims analysis. Cephalalgia. 2017;37(5):470–85. tors of response to onabotulinumtoxina: real-life https://doi.org/10.1177/0333102416678382. experience observational study. Headache. 2020;60(4):677–85. https://doi.org/10.1111/head. 21. Sacco S, Braschinsky M, Ducros A, Lampl C, Little P, 13765. van den Brink AM, et al. European headache fed- eration consensus on the definition of resistant and 12. di Cola SF, Caratozzolo S, Liberini P, Rao R, Pado- refractory migraine: developed with the endorse- vani A. Response predictors in chronic migraine: ment of the European Migraine & Headache Alli- medication overuse and depressive symptoms neg- ance (EMHA). J Headache Pain. 2020;21(1):76. atively impact onabotulinumtoxin-A treatment. https://doi.org/10.1186/s10194-020-01130-5. Front Neurol. 2019;10:678. https://doi.org/10. 3389/fneur.2019.00678. 22. Stark C, Stark R, Limberg N, Rodrigues J, Cordato D, Schwartz R, et al. Real-world effectiveness of 13. Cernuda-Morollón E, Ramón C, Larrosa D, Alvarez onabotulinumtoxinA treatment for the prevention R, Riesco N, Pascual J. Long-term experience with of headaches in adults with chronic migraine in onabotulinumtoxinA in the treatment of chronic Australia: a retrospective study. J Headache Pain. migraine: what happens after one year? Cephalal- 2019;20(1):81. https://doi.org/10.1186/s10194-019- gia. 2015;35(10):864–8. https://doi.org/10.1177/ 1030-z. 0333102414561873. 23. Silberstein SD, Dodick DW, Aurora SK, Diener HC, 14. Domı́nguez C, Pozo-Rosich P, Torres-Ferrús M, DeGryse RE, Lipton RB, et al. Per cent of patients Hernández-Beltrán N, Jurado-Cobo C, González- with chronic migraine who responded per onabo- Oria C, et al. OnabotulinumtoxinA in chronic tulinumtoxinA treatment cycle: PREEMPT. J Neurol migraine: predictors of response. A prospective Neurosurg Psychiatry. 2015;86(9):996–1001. multicentre descriptive study. Eur J Neurol. https://doi.org/10.1136/jnnp-2013-307149. 2018;25(2):411–6. https://doi.org/10.1111/ene. 13523. 24. Khalil M, Zafar HW, Quarshie V, Ahmed F. Prospective analysis of the use of Onabotulinum- 15. Bendtsen L, Sacco S, Ashina M, Mitsikostas D, toxinA (BOTOX) in the treatment of chronic Ahmed F, Pozo-Rosich P, et al. Guideline on the use migraine; real-life data in 254 patients from Hull, U. of onabotulinumtoxinA in chronic migraine: a K. J Headache Pain. 2014;15:54. https://doi.org/10. consensus statement from the European Headache 1186/1129-2377-15-54. Federation. J Headache Pain. 2018;19(1):91. https:// doi.org/10.1186/s10194-018-0921-8. 25. Ruscheweyh R, Förderreuther S, Gaul C, Gendolla A, Holle-Lee D, Jürgens T, et al. Treatment of 16. (NICE) NIfCE: Botulinum toxin type A for the pre- chronic migraine with botulinum neurotoxin A: vention of headaches in adults with chronic Expert recommendations of the German Migraine migraine. In.; 2012. and Headache Society. Nervenarzt. 2018;89(12): 1355–64. https://doi.org/10.1007/s00115-018- 17. Tiseo C, Ornello R, Pistoia F, Sacco S. How to inte- 0534-0. grate monoclonal antibodies targeting the calci- tonin gene-related peptide or its receptor in daily 26. Tassorelli C, Diener HC, Dodick DW, Silberstein SD, clinical practice. J Headache Pain. 2019;20(1):49. Lipton RB, Ashina M, et al. Guidelines of the https://doi.org/10.1186/s10194-019-1000-5. International Headache Society for controlled trials of preventive treatment of chronic migraine in 18. De Matteis E, Guglielmetti M, Ornello R, Spuntarelli adults. Cephalalgia. 2018;38(5):815–32. https://doi. V, Martelletti P, Sacco S. Targeting CGRP for org/10.1177/0333102418758283. migraine treatment: mechanisms, antibodies, small molecules, perspectives. Expert Rev Neurother. 27. Tassorelli C, Tedeschi G, Sarchielli P, Pini LA, Grazzi 2020. https://doi.org/10.1080/14737175.2020. L, Geppetti P, et al. Optimizing the long-term 1772758. management of chronic migraine with onabo- tulinumtoxinA in real life. Expert Rev Neurother. 650 Pain Ther (2021) 10:637–650 2018;18(2):167–76. https://doi.org/10.1080/ 32. Sacco S, Bendtsen L, Ashina M, Reuter U, Terwindt 14737175.2018.1419867. G, Mitsikostas DD, et al. European headache feder- ation guideline on the use of monoclonal anti- 28. Sacco S, Russo A, Geppetti P, Grazzi L, Negro A, bodies acting on the calcitonin gene related peptide Tassorelli C, et al. What is changing in chronic or its receptor for migraine prevention. J Headache migraine treatment? An algorithm for onabo- Pain. 2019;20(1):6. https://doi.org/10.1186/s10194- tulinumtoxinA treatment by the Italian chronic 018-0955-y. migraine group. Expert Rev Neurother. 2020. https://doi.org/10.1080/14737175.2020.1825077. 33. Armanious M, Khalil N, Lu Y, Jimenez-Sanders R. Erenumab and ONABOTULINUMTOXINA combi- 29. Vernieri F, Paolucci M, Altamura C, Pasqualetti P, nation therapy for the prevention of Mastrangelo V, Pierangeli G, et al. Onabotulinum- intractable chronic migraine without aura: a retro- toxin-A in chronic migraine: should timing and spective analysis. J Pain Palliat Care Pharmacother. definition of non-responder status be revised? sug- 2020. https://doi.org/10.1080/15360288.2020. gestions from a real-life Italian multicenter experi- 1829249. ence. Headache. 2019;59(8):1300–9. https://doi. org/10.1111/head.13617. 34. Martelletti P. Combination therapy in migraine: asset or issue? Expert Rev Neurother. 2020;20(10): 30. Ornello R, Guerzoni S, Baraldi C, Evangelista L, 995–6. https://doi.org/10.1080/14737175.2020. Frattale I, Marini C, et al. Sustained response to 1821655. onabotulinumtoxin A in patients with chronic migraine: real-life data. J Headache Pain. 35. Borro M, Guglielmetti M, Simmaco M, Martelletti P, 2020;21(1):40. https://doi.org/10.1186/s10194-020- Gentile G. The future of pharmacogenetics in the 01113-6. treatment of migraine. Pharmacogenomics. 2019;20(16):1159–73. https://doi.org/10.2217/pgs- 31. Negro A, Curto M, Lionetto L, Martelletti P. A two 2019-0069. years open-label prospective study of Onabo- tulinumtoxinA 195 U in medication overuse head- 36. Pomes LM, Guglielmetti M, Bertamino E, Simmaco ache: a real-world experience. J Headache Pain. M, Borro M, Martelletti P. Optimising migraine 2015;17:1. https://doi.org/10.1186/s10194-016- treatment: from drug-drug interactions to person- 0591-3. alized medicine. J Headache Pain. 2019;20(1):56. https://doi.org/10.1186/s10194-019-1010-3.