Brainomaly: Unsupervised Neurologic Disease Detection Utilizing Unannotated T1-weighted Brain MR Images.

Harnessing the power of deep neural networks in the medical imaging domain is challenging due to the difficulties in acquiring large annotated datasets, especially for rare diseases, which involve high costs, time, and effort for annotation. Unsupervised disease detection methods, such as anomaly detection, can significantly reduce human effort in these scenarios. While anomaly detection typically focuses on learning from images of healthy subjects only, real-world situations often present unannotated datasets with a mixture of healthy and diseased subjects. Recent studies have demonstrated that utilizing such unannotated images can improve unsupervised disease and anomaly detection. However, these methods do not utilize knowledge specific to registered neuroimages, resulting in a subpar performance in neurologic disease detection. To address this limitation, we propose Brainomaly, a GAN-based image-to-image translation method specifically designed for neurologic disease detection. Brainomaly not only offers tailored image-to-image translation suitable for neuroimages but also leverages unannotated mixed images to achieve superior neurologic disease detection. Additionally, we address the issue of model selection for inference without annotated samples by proposing a pseudo-AUC metric, further enhancing Brainomaly's detection performance. Extensive experiments and ablation studies demonstrate that Brainomaly outperforms existing state-of-the-art unsupervised disease and anomaly detection methods by significant margins in Alzheimer's disease detection using a publicly available dataset and headache detection using an institutional dataset. The code is available from https://github.com/mahfuzmohammad/Brainomaly.

Brain structural and functional abnormalities associated with acute post-traumatic headache: iron deposition and functional connectivity.

BACKGROUND: The purpose of this study was to interrogate brain iron accumulation in participants with acute post-traumatic headache (PTH) due to mild traumatic brain injury (mTBI), and to determine if functional connectivity is affected in areas with iron accumulation. We aimed to examine the correlations between iron accumulation and headache frequency, post-concussion symptom severity, number of mTBIs, and time since most recent TBI. METHODS: Sixty participants with acute PTH and 60 age-matched healthy controls (HC) underwent 3T magnetic resonance imaging including quantitative T2* maps and resting-state functional connectivity imaging. Between group T2* differences were determined using T-tests (p < 0.005, cluster size threshold of 90 voxels). For regions with T2* differences, two analyses were conducted. First, the correlations with clinical variables including headache frequency, number of lifetime mTBIs, time since most recent mTBI, and Sport Concussion Assessment Tool (SCAT) symptom severity scale scores were investigated using linear regression. Second, the functional connectivity of these regions with the rest of the brain was examined (significance of p < 0.05 with family wise error correction for multiple comparisons). RESULTS: The acute PTH group consisted of 60 participants (22 male, 38 female) with average age of 42 ± 14 years. The HC group consisted of 60 age-matched controls (17 male, 43 female, average age of 42 ± 13). PTH participants had lower T2* values compared to HC in the left posterior cingulate and the bilateral cuneus. Stronger functional connectivity was observed between bilateral cuneus and right cerebellar areas in PTH compared to HC. Within the PTH group, linear regression showed negative associations of T2* in the left posterior cingulate with SCAT symptom severity score (p = 0.05) and T2* in the left cuneus with headache frequency (p = 0.04). CONCLUSIONS: Iron accumulation in posterior cingulate and cuneus was observed in those with acute PTH relative to HC; stronger functional connectivity was detected between the bilateral cuneus and the right cerebellum. The correlations of decreased T2* (suggesting higher iron content) with headache frequency and post mTBI symptom severity suggest that the iron accumulation that results from mTBI might reflect the severity of underlying mTBI pathophysiology and associate with post-mTBI symptom severity including PTH.

Functional impairment of chronic migraine with medication overuse: Secondary analysis from the Medication Overuse Treatment Strategy (MOTS) trial.

BACKGROUND: Chronic migraine exerts substantial negative impacts on daily functioning. Efforts to manage impaired functioning may result in medication overuse, which contributes to the worsening profile and chronification of migraine. The Migraine Functional Impact Questionnaire (MFIQ) is a recently developed measure assessing the impact of migraine on physical, social, and emotional function. OBJECTIVE: The objective of this analysis was to assess changes in MFIQ scores following initiation or modification of migraine preventive medication and determine if changes in function are associated with changes in other aspects of migraine burden, such as headache frequency, headache intensity, and symptoms of anxiety and depression. METHODS: This is a secondary analysis of data from the Medication Overuse Treatment Strategy (MOTS) trial, a prospective pragmatic clinical trial that investigated two treatment strategies for those with chronic migraine and medication overuse. Data from both treatment arms were pooled and analyzed using a pre-post design. Prior to and 12 weeks following initiation or modification of migraine preventive medication, participants completed a series of questionnaires that captured migraine characteristics, medication use, migraine-related physical impairment (MFIQ), anxiety (Generalized Anxiety Disorder-7), and depression (Patient Health Questionnaire 9 [PHQ-9]) symptoms. Changes from baseline in all measures were assessed using the paired t-test. Relationships between changes in MFIQ scores and other measures were assessed using linear regression. Multivariable modeling was performed to determine which additional variables contributed to the change in MFIQ beyond that already explained by an individual variable. Model terms were selected by using elastic net regularization. Only those participants who completed the baseline and 12-week MFIQ were included in this analysis. RESULTS: Of the 537 patients, 88.2% were female, and the average age was 45 years (standard deviation 13). The mean frequency of days with moderate-to-severe headache improved 39.2% from 13.5 per 30 days at baseline to 8.1 per 30 days at week 12. The mean MFIQ Usual Activities Global score improved by 15.0 points (on a 100-point scale). All five domains (Usual Activities Global, Usual Activities, Social Function, Emotional Function, Physical Function) of the MFIQ improved by a mean of at least 13.0 points. Changes in PHQ-9 score, followed by changes in headache frequency, had the strongest associations with change in all domains of the MFIQ. CONCLUSIONS: The negative impact of chronic migraine with medication overuse on physical, social, and emotional functioning substantially lessened following initiation or modification of migraine preventive medication. Improved functioning, as measured by the MFIQ, was most strongly associated with reductions in depression scores and headache frequency, highlighting the importance of recognizing and monitoring changes in depressive symptoms, in addition to headache frequency and functional impairment, when evaluating response to preventive medications.

Physical impairment during and between migraine attacks: A daily diary study of patients with chronic migraine.

OBJECTIVE: While a substantial body of research describes the disabling impacts of migraine attacks, less research has described the impacts of migraine on physical functioning between migraine attacks. The objective of this study is to describe physical impairment during and between migraine attacks as a dimension of burden experienced by people living with chronic migraine. METHODS: The physical impairment domain of the Migraine Physical Function Impact Diary was recorded in headache diaries from the Medication Overuse Treatment Strategy trial. Days with moderate to severe headache were used to approximate migraine attacks. Factor analysis and regression analysis were used to describe associations between migraine and physical impairment. RESULTS: 77,662 headache diary entries from 720 participants were analyzed, including 25,414 days with moderate to severe headache, 19,149 days with mild headache, and 33,099 days with no headache. Mean physical impairment score was 41.5 (SD = 26.1) on days with moderate to severe headache, 12.8 (SD = 15.0) on days with mild headache, and 5.2 (SD = 13.1) on days with no headache. Physical impairment on days with mild headache and days with no headache was significantly associated with days since last moderate to severe headache, physical impairment with last moderate to severe headache, mild headache (compared to no headache), depression, hypersensitivities and cranial autonomic symptoms. CONCLUSIONS: Physical impairment occurs on migraine and non-migraine days. Study participants with frequent headaches, symptoms of depression, hypersensitivities and cranial autonomic symptoms experience physical impairment at a higher rate on days with no headache and days with mild headache.Clinical Trial Registration: ClinicalTrials.gov (NCT02764320).

Posttraumatic Headache.

OBJECTIVE: This article provides an overview of the epidemiology, diagnosis, clinical presentation, pathophysiology, prognosis, and treatment of posttraumatic headache attributed to mild traumatic brain injury (mTBI). LATEST DEVELOPMENTS: The International Classification of Headache Disorders, Third Edition requires that posttraumatic headache begin within 7 days of the inciting trauma. Although posttraumatic headache characteristics and associated symptoms vary, most commonly there is substantial overlap with symptoms of migraine or tension-type headache. New insights into posttraumatic headache pathophysiology suggest roles for neuroinflammation, altered pain processing and modulation, and changes in brain structure and function. Although the majority of posttraumatic headache resolves during the acute phase, about one-third of individuals have posttraumatic headache that persists for at least several months. Additional work is needed to identify predictors and early markers of posttraumatic headache persistence, but several potential predictors have been identified such as having migraine prior to the mTBI, the total number of TBIs ever experienced, and the severity of initial symptoms following the mTBI. Few data are available regarding posttraumatic headache treatment; studies investigating different treatments and the optimal timing for initiating posttraumatic headache treatment are needed. ESSENTIAL POINTS: Posttraumatic headache begins within 7 days of the causative injury. The characteristics of posttraumatic headache most commonly resemble those of migraine or tension-type headache. Posttraumatic headache persists for 3 months or longer in about one-third of individuals. Additional studies investigating posttraumatic headache treatment are needed.

Photophobia is associated with lower sleep quality in individuals with migraine: results from the American Registry for Migraine Research (ARMR).

BACKGROUND: Patients with migraine often have poor sleep quality between and during migraine attacks. Furthermore, extensive research has identified photophobia as the most common and most bothersome symptom in individuals with migraine, second only to headache. Seeking the comfort of darkness is a common strategy for managing pain during an attack and preventing its recurrence between episodes. Given the well-established effects of daily light exposure on circadian activity rhythms and sleep quality, this study aimed to investigate the relationship between photophobia symptoms and sleep quality in a cohort of patients with migraine. METHODS: A cross-sectional observational study was conducted using existing data extracted from the American Registry for Migraine Research (ARMR). Participants with a migraine diagnosis who had completed the baseline questionnaires (Photosensitivity Assessment Questionnaire (PAQ), Generalized Anxiety Disorder-7 (GAD-7), Patient Health Questionnaire-2 (PHQ-2)), and selected questions of the ARMR Sleep questionnaire were included. Models were created to describe the relationship of photophobia and photophilia with various sleep facets, including sleep quality (SQ), sleep disturbance (SDis), sleep onset latency (SOL), sleep-related impairments (SRI), and insomnia. Each model was controlled for age, sex, headache frequency, anxiety, and depression. RESULTS: A total of 852 patients meeting the inclusion criteria were included in the analysis (mean age (SD) = 49.8 (13.9), 86.6% (n = 738) female). Those with photophobia exhibited significantly poorer sleep quality compared to patients without photophobia (p < 0.001). Photophobia scores were associated with SQ (p < 0.001), SDis (p < 0.001), SOL (p = 0.011), SRI (p = 0.020), and insomnia (p = 0.005) after controlling for age, sex, headache frequency, depression, and anxiety, signifying that higher levels of photophobia were associated with worse sleep-related outcomes. Conversely, photophilia scores were associated with better sleep-related outcomes for SQ (p < 0.007), SOL (p = 0.010), and insomnia (p = 0.014). CONCLUSION: Results suggest that photophobia is a significant predictor of poor sleep quality and sleep disturbances in migraine. These results underscore the necessity for comprehensive and systematic investigations into the intricate interplay between photophobia and sleep to enhance our understanding and develop tailored solutions for individuals with migraine.

Brain Structural and Functional Abnormalities Associated with Acute Post-Traumatic Headache: Iron Deposition and Functional Connectivity.

Background: The purpose of this study was to interrogate brain iron accumulation in participants with acute post-traumatic headache (PTH) due to mild traumatic brain injury (mTBI), and to determine if functional connectivity is affected in areas with iron accumulation. We aimed to examine the correlations between iron accumulation and headache frequency, post-concussion symptom severity, number of mTBIs and time since most recent TBI. Methods: Sixty participants with acute PTH and 60 age-matched healthy controls (HC) underwent 3T magnetic resonance imaging including quantitative T2* maps and resting-state functional connectivity imaging. Between group T2* differences were determined using T-tests (p < 0.005, cluster size threshold of 10 voxels). For regions with T2* differences, two analyses were conducted. First, the correlations with clinical variables including headache frequency, number of lifetime mTBIs, time since most recent mTBI, and Sport Concussion Assessment Tool (SCAT) symptom severity scale scores were investigated using linear regression. Second, the functional connectivity of these regions with the rest of the brain was examined (significance of p < 0.05 with family wise error correction for multiple comparisons). Results: The acute PTH group consisted of 60 participants (22 male, 38 female) with average age of 42 ± 14 years. The HC group consisted of 60 age-matched controls (17 male, 43 female, average age of 42 ± 13). PTH participants had lower T2* values compared to HC in the left posterior cingulate and the bilateral cuneus. Stronger functional connectivity was observed between bilateral cuneus and right cerebellar areas in PTH compared to HC. Within the PTH group, linear regression showed negative associations of T2* and SCAT symptom severity score in the left posterior cingulate (p = 0.05) and with headache frequency in the left cuneus (p = 0.04). Conclusions: Iron accumulation in posterior cingulate and cuneus was observed in those with acute PTH relative to HC; stronger functional connectivity was detected between the bilateral cuneus and the right cerebellum. The correlations of decreased T2* (suggesting higher iron content) with headache frequency and post mTBI symptom severity suggest that the iron accumulation that results from mTBI might reflect the severity of underlying mTBI pathophysiology and associate with post-mTBI symptom severity including PTH.

A large language model-based generative natural language processing framework fine-tuned on clinical notes accurately extracts headache frequency from electronic health records.

OBJECTIVE: To develop a natural language processing (NLP) algorithm that can accurately extract headache frequency from free-text clinical notes. BACKGROUND: Headache frequency, defined as the number of days with any headache in a month (or 4 weeks), remains a key parameter in the evaluation of treatment response to migraine preventive medications. However, due to the variations and inconsistencies in documentation by clinicians, significant challenges exist to accurately extract headache frequency from the electronic health record (EHR) by traditional NLP algorithms. METHODS: This was a retrospective cross-sectional study with patients identified from two tertiary headache referral centers, Mayo Clinic Arizona and Mayo Clinic Rochester. All neurology consultation notes written by 15 specialized clinicians (11 headache specialists and 4 nurse practitioners) between 2012 and 2022 were extracted and 1915 notes were used for model fine-tuning (90%) and testing (10%). We employed four different NLP frameworks: (1) ClinicalBERT (Bidirectional Encoder Representations from Transformers) regression model, (2) Generative Pre-Trained Transformer-2 (GPT-2) Question Answering (QA) model zero-shot, (3) GPT-2 QA model few-shot training fine-tuned on clinical notes, and (4) GPT-2 generative model few-shot training fine-tuned on clinical notes to generate the answer by considering the context of included text. RESULTS: The mean (standard deviation) headache frequency of our training and testing datasets were 13.4 (10.9) and 14.4 (11.2), respectively. The GPT-2 generative model was the best-performing model with an accuracy of 0.92 (0.91, 0.93, 95% confidence interval [CI]) and R2 score of 0.89 (0.87, 0.90, 95% CI), and all GPT-2-based models outperformed the ClinicalBERT model in terms of exact matching accuracy. Although the ClinicalBERT regression model had the lowest accuracy of 0.27 (0.26, 0.28), it demonstrated a high R2 score of 0.88 (0.85, 0.89), suggesting the ClinicalBERT model can reasonably predict the headache frequency within a range of ≤ ± 3 days, and the R2 score was higher than the GPT-2 QA zero-shot model or GPT-2 QA model few-shot training fine-tuned model. CONCLUSION: We developed a robust information extraction model based on a state-of-the-art large language model, a GPT-2 generative model that can extract headache frequency from EHR free-text clinical notes with high accuracy and R2 score. It overcame several challenges related to different ways clinicians document headache frequency that were not easily achieved by traditional NLP models. We also showed that GPT-2-based frameworks outperformed ClinicalBERT in terms of accuracy in extracting headache frequency from clinical notes. To facilitate research in the field, we released the GPT-2 generative model and inference code with open-source license of community use in GitHub. Additional fine-tuning of the algorithm might be required when applied to different health-care systems for various clinical use cases.

Guidelines of the International Headache Society for controlled trials of pharmacological preventive treatment for persistent post-traumatic headache attributed to mild traumatic brain injury.

BACKGROUND: Persistent headache attributed to traumatic injury to the head is divided into two subtypes, one attributed to moderate or severe traumatic injury and another attributed to mild traumatic injury (i.e., concussion). The latter is much more prevalent, in part because more than 90% of cases with traumatic brain injury are classified as mild. The pathophysiology of persistent post-traumatic headache is poorly understood and the underlying mechanisms are likely multifactorial. There is currently no approved treatment specifically for persistent post-traumatic headache, and management strategies rely on medications used for migraine or tension-type headache. Therefore, high-quality trials are urgently needed to support clinical decision-making and optimize management strategies. International guidelines can facilitate appropriate trial design and ensure the acquisition of high-quality data evaluating the efficacy, tolerability, and safety of available and novel pharmacological therapies for the preventive treatment of persistent post-traumatic headache. METHODS: The development of this guideline was based on a literature review of available studies in MEDLINE, Embase, and the Cochrane Central Register of Controlled Trials, along with a review of previously published guidelines for controlled trials of preventive treatment for episodic and chronic migraine. The identified literature was critically appraised, and due to the scarcity of scientific evidence, recommendations were primarily based on the consensus of experts in the field. OBJECTIVE: To provide guidelines for designing state-of-the-art controlled clinical trials aimed at evaluating the effectiveness of preventive treatments for persistent post-traumatic headache attributed to mild traumatic brain injury.

Posttraumatic headache is a distinct headache type from migraine.

PURPOSE OF REVIEW: Posttraumatic headache (PTH), a headache that develops within 7 days of a causative injury, is one of the most common secondary headaches, mostly attributed to mild traumatic brain injury (mTBI). Because presence of preinjury headache is a risk factor for developing PTH and PTH symptoms often resemble migraine or tension-type headache, the association between PTH and primary headaches has attracted attention from clinicians and scientists. RECENT FINDINGS: Recent studies on epidemiological aspects, headache features, risk factors, imaging characteristics, and response to treatment, suggest overlapping features and distinct objective findings in PTH compared to migraine. SUMMARY: We argue that PTH is distinct from migraine. Therefore, PTH epidemiology, pathophysiology, diagnosis, treatment, and prognosis should continue to be investigated separately from migraine.

Comparing the Efficacy and Safety of Galcanezumab Versus Rimegepant for Prevention of Episodic Migraine: Results from a Randomized, Controlled Clinical Trial.

INTRODUCTION: There have been no prior trials directly comparing the efficacy of different calcitonin gene-related peptide (CGRP) antagonists for migraine prevention. Reported are the results from the first head-to-head study of two CGRP antagonists, galcanezumab (monoclonal antibody) versus rimegepant (gepant), for the prevention of episodic migraine. METHODS: In this 3-month, double-blind, double-dummy study, participants were randomized (1:1) to subcutaneous (SC) galcanezumab 120 mg per month (after a 240 mg loading dose) and a placebo oral disintegrating tablet (ODT) every other day (q.o.d.) or to rimegepant 75 mg ODT q.o.d. and a monthly SC placebo. The primary endpoint was the proportion of participants with a ≥ 50% reduction in migraine headache days per month from baseline across the 3-month double-blind treatment period. Key secondary endpoints were overall mean change from baseline in: migraine headache days per month across 3 months and at month 3, 2, and 1; migraine headache days per month with acute migraine medication use; Migraine-Specific Quality of Life Questionnaire Role Function-Restrictive domain score at month 3; and a ≥ 75% and 100% reduction from baseline in migraine headache days per month across 3 months. RESULTS: Of 580 randomized participants (galcanezumab: 287, rimegepant: 293; mean age: 42 years), 83% were female and 81% Caucasian. Galcanezumab was not superior to rimegepant in achieving a ≥ 50% reduction from baseline in migraine headache days per month (62% versus 61% respectively; P = 0.70). Given the pre-specified multiple testing procedure, key secondary endpoints cannot be considered statistically significant. Overall, treatment-emergent adverse events were reported by 21% of participants, with no significant differences between study intervention groups. CONCLUSIONS: Galcanezumab was not superior to rimegepant for the primary endpoint; however, both interventions demonstrated efficacy as preventive treatments in participants with episodic migraine. The efficacy and safety profiles observed in galcanezumab-treated participants were consistent with previous studies. TRIAL REGISTRATION: ClinTrials.gov-NCT05127486 (I5Q-MC-CGBD).

Galcanezumab and rimegepant are preventive treatments for episodic migraine. The goal of this study was to compare the efficacy of galcanezumab and rimegepant in reducing the number of monthly migraine headaches and to determine if galcanezumab was better than rimegepant. The study provides important information to doctors and their patients when making treatment decisions.People with episodic migraine were assigned to the galcanezumab (given as an injection under the skin) or rimegepant (given as a tablet that dissolves in the mouth) group and treated for 3 months. The doctor and the patient did not know which group they were assigned to, and to keep it unknown to both, people in the galcanezumab group got an injection with real medicine and a fake tablet, and people in the rimegepant group got a tablet with real medicine and a fake injection. The researchers wanted to know how many people in each group had at least a 50% reduction in their monthly migraine headaches.Of the 580 people in the study, 287 were assigned to galcanezumab and 293 to rimegepant. In both groups, most were female and white. After 3 months of treatment, 62% of the people in the galcanezumab group and 61% of people in the rimegepant group had at least a 50% reduction in monthly migraine headaches. Both treatments were effective, but galcanezumab was not better than rimegepant. About 20% of the people in each treatment group had a side effect from the medication, and most were mild or moderate in severity.

Identification of brain areas in mice with peak neural activity across the acute and persistent phases of post-traumatic headache.

BACKGROUND: Post-traumatic headache is very common after a mild traumatic brain injury. Post-traumatic headache may persist for months to years after an injury in a substantial proportion of people. The pathophysiology underlying post-traumatic headache remains unknown but is likely distinct from other headache disorders. Identification of brain areas activated in acute and persistent phases of post-traumatic headache can provide insights into the underlying circuits mediating headache pain. We used an animal model of mild traumatic brain injury-induced post-traumatic headache and c-fos immunohistochemistry to identify brain regions with peak activity levels across the acute and persistent phases of post-traumatic headache. METHODS: Male and female C57BL/6 J mice were briefly anesthetized and subjected to a sham procedure or a weight drop closed-head mild traumatic brain injury . Cutaneous allodynia was assessed in the periorbital and hindpaw regions using von Frey filaments. Immunohistochemical c-fos based neural activity mapping was then performed on sections from whole brain across the development of post-traumatic headache (i.e. peak of the acute phase at 2 days post- mild traumatic brain injury), start of the persistent phase (i.e. >14 days post-mild traumatic brain injury) or after provocation with stress (bright light). Brain areas with consistent and peak levels of c-fos expression across mild traumatic brain injury induced post-traumatic headache were identified and included for further analysis. RESULTS: Following mild traumatic brain injury, periorbital and hindpaw allodynia was observed in both male and female mice. This allodynia was transient and subsided within the first 14 days post-mild traumatic brain injury and is representative of acute post-traumatic headache. After this acute post-traumatic headache phase, exposure of mild traumatic brain injury mice to a bright light stress reinstated periorbital and hindpaw allodynia for several hours - indicative of the development of persistent post-traumatic headache. Acute post-traumatic headache was coincident with an increase in neuronal c-fos labeling in the spinal nucleus of the trigeminal caudalis, primary somatosensory cortex, and the nucleus accumbens. Neuronal activation returned to baseline levels by the persistent post-traumatic headache phase in the spinal nucleus of the trigeminal caudalis and primary somatosensory cortex but remained elevated in the nucleus accumbens. In the persistent post-traumatic headache phase, coincident with allodynia observed following bright light stress, we observed bright light stress-induced c-fos neural activation in the spinal nucleus of the trigeminal caudalis, primary somatosensory cortex, and nucleus accumbens. CONCLUSION: Examination of mild traumatic brain injury-induced changes in peak c-fos expression revealed brain regions with significantly increased neural activity across the acute and persistent phases of post-traumatic headache. Our findings suggest mild traumatic brain injury-induced post-traumatic headache produces neural activation along pain relevant pathways at time-points matching post-traumatic headache-like pain behaviors. These observations suggest that the spinal nucleus of the trigeminal caudalis, primary somatosensory cortex, and nucleus accumbens may contribute to both the induction and maintenance of post-traumatic headache.

Ubrogepant for the treatment of migraine attacks during the prodrome: a phase 3, multicentre, randomised, double-blind, placebo-controlled, crossover trial in the USA.

BACKGROUND: Ubrogepant is a calcitonin gene-related peptide (CGRP) receptor antagonist that is approved for acute treatment of migraine. The prodrome is the earliest phase of a migraine attack and is characterised by non-aura symptoms that precede headache onset. The aim of this trial was to evaluate the efficacy, safety, and tolerability of ubrogepant 100 mg compared with placebo for the acute treatment of migraine when administered during the prodrome. METHODS: This PRODROME trial was a phase 3, multicentre, randomised, double-blind, placebo-controlled, crossover trial of ubrogepant 100 mg conducted at 75 research centres and headache clinics in the USA. Eligible participants were adults aged 18-75 years who had at least a 1-year history of migraine with or without aura and a history of two to eight migraine attacks per month with moderate to severe headache in each of the 3 months before screening. Eligible participants were randomly assigned (1:1) to either receive placebo to treat the first qualifying prodrome event and ubrogepant 100 mg to treat the second qualifying prodrome event or to receive ubrogepant 100 mg to treat the first qualifying prodrome event and placebo to treat the second qualifying prodrome event. An automated interactive web-response system used permuted blocks of four to manage randomisation. All people giving interventions and assessing outcomes were masked to group assignment during the study. People doing data analysis, which occurred after study completion, were not masked to group assignment. During the double-blind treatment period, each participant was instructed to orally take two tablets of the study drug at the onset of each qualifying prodrome event. The primary endpoint was absence of moderate or severe intensity headache within 24 h after study-drug dose; efficacy analyses were conducted with the modified intention-to-treat (mITT) population, defined as all randomly assigned participants with at least one headache assessment within 24 h after taking the study drug during the treatment period. The safety population included all treated participants who took at least one administration of study drug. The trial is registered with ClinicalTrials.gov (NCT04492020). FINDINGS: Between Aug 21, 2020, and April 19, 2022, 518 participants were randomly assigned to double-blind crossover treatment. The safety population included 480 participants and the mITT population included 477 participants; 421 (88%) of 480 participants were female and 59 (12%) were male. Absence of moderate or severe headache within 24 h after a dose occurred after 190 (46%) of 418 qualifying prodrome events that had been treated with ubrogepant and after 121 (29%) of 423 qualifying prodrome events that had been treated with placebo (odds ratio 2·09, 95% CI 1·63-2·69; p<0·0001). Adverse events that occurred within 48 h after study-drug administration were reported after 77 (17%) of 456 qualifying prodrome events that had been treated with ubrogepant and after 55 (12%) of 462 events that had been treated with placebo. INTERPRETATION: Ubrogepant was effective and well tolerated for the treatment of migraine attacks when taken during the prodrome. FUNDING: AbbVie.

Rethinking headache as a global public health case model for reaching the SDG 3 HEALTH by 2030.

The 2030 Agenda for Sustainable Development sets out, through 17 Sustainable Development Goals (SDGs), a path for the prosperity of people and the planet. SDG 3 in particular aims to ensure healthy lives and promote well-being for all at all ages and includes several targets to enhance health. This review presents a "headache-tailored" perspective on how to achieve SDG 3 by focusing on six specific actions: targeting chronic headaches; reducing the overuse of acute pain-relieving medications; promoting the education of healthcare professionals; granting access to medication in low- and middle-income countries (LMIC); implementing training and educational opportunities for healthcare professionals in low and middle income countries; building a global alliance against headache disorders. Addressing the burden of headache disorders directly impacts on populations' health, as well as on the possibility to improve the productivity of people aged below 50, women in particular. Our analysis pointed out several elements, and included: moving forward from frequency-based parameters to define headache severity; recognizing and managing comorbid diseases and risk factors; implementing a disease management multi-modal management model that incorporates pharmacological and non-pharmacological treatments; early recognizing and managing the overuse of acute pain-relieving medications; promoting undergraduate, postgraduate, and continuing medical education of healthcare professionals with specific training on headache; and promoting a culture that favors the recognition of headaches as diseases with a neurobiological basis, where this is not yet recognized. Making headache care more sustainable is an achievable objective, which will require multi-stakeholder collaborations across all sectors of society, both health-related and not health-related. Robust investments will be needed; however, considering the high prevalence of headache disorders and the associated disability, these investments will surely improve multiple health outcomes and lift development and well-being globally.

A Large Language Model-Based Generative Natural Language Processing Framework Finetuned on Clinical Notes Accurately Extracts Headache Frequency from Electronic Health Records.

Background: Headache frequency, defined as the number of days with any headache in a month (or four weeks), remains a key parameter in the evaluation of treatment response to migraine preventive medications. However, due to the variations and inconsistencies in documentation by clinicians, significant challenges exist to accurately extract headache frequency from the electronic health record (EHR) by traditional natural language processing (NLP) algorithms. Methods: This was a retrospective cross-sectional study with human subjects identified from three tertiary headache referral centers- Mayo Clinic Arizona, Florida, and Rochester. All neurology consultation notes written by more than 10 headache specialists between 2012 to 2022 were extracted and 1915 notes were used for model fine-tuning (90%) and testing (10%). We employed four different NLP frameworks: (1) ClinicalBERT (Bidirectional Encoder Representations from Transformers) regression model (2) Generative Pre-Trained Transformer-2 (GPT-2) Question Answering (QA) Model zero-shot (3) GPT-2 QA model few-shot training fine-tuned on Mayo Clinic notes; and (4) GPT-2 generative model few-shot training fine-tuned on Mayo Clinic notes to generate the answer by considering the context of included text. Results: The GPT-2 generative model was the best-performing model with an accuracy of 0.92[0.91 - 0.93] and R2 score of 0.89[0.87, 0.9], and all GPT2-based models outperformed the ClinicalBERT model in terms of the exact matching accuracy. Although the ClinicalBERT regression model had the lowest accuracy 0.27[0.26 - 0.28], it demonstrated a high R2 score 0.88[0.85, 0.89], suggesting the ClinicalBERT model can reasonably predict the headache frequency within a range of ≤ ± 3 days, and the R2 score was higher than the GPT-2 QA zero-shot model or GPT-2 QA model few-shot training fine-tuned model. Conclusion: We developed a robust model based on a state-of-the-art large language model (LLM)- a GPT-2 generative model that can extract headache frequency from EHR free-text clinical notes with high accuracy and R2 score. It overcame several challenges related to different ways clinicians document headache frequency that were not easily achieved by traditional NLP models. We also showed that GPT2-based frameworks outperformed ClinicalBERT in terms of accuracy in extracting headache frequency from clinical notes. To facilitate research in the field, we released the GPT-2 generative model and inference code with open-source license of community use in GitHub.

Real-world persistence and costs among patients with chronic migraine treated with onabotulinumtoxinA or calcitonin gene-related peptide monoclonal antibodies.

BACKGROUND: Chronic migraine (CM) is a common neurologic disorder that imposes substantial burden on payers, patients, and society. Low rates of persistence to oral migraine preventive medications have been previously documented; however, less is known about persistence and costs associated with innovative nonoral migraine preventive medications. OBJECTIVE: To evaluate real-world persistence and costs among adults with CM treated with onabotulinumtoxinA (onabotA) or calcitonin gene-related peptide monoclonal antibodies (CGRP mAbs). METHODS: This was a retrospective, longitudinal, observational study analyzing the IBM MarketScan Commercial and Medicare databases. The study sample included adults with CM initiating treatment with either onabotA or a CGRP mAb on or after January 1, 2018. Persistence and costs over 12 months after treatment initiation were evaluated using chi-square and Student's t-tests. Persistence to onabotA was compared with CGRP mAbs as a weighted average of the class and by individual CGRP mAbs. Mean pharmacy (acute and preventive), medical (inpatient, emergency department, and outpatient), and total costs are reported. Multivariate regression analyses were conducted to generate adjusted estimates of persistence and costs after controlling for potential confounders (age, sex, region, insurance type, number of baseline comorbidities, Charlson Comorbidity Index, and number of previously used oral migraine preventive medications). RESULTS: Of 66,303 individuals with onabotA or CGRP mAb claims, 2,697 with CM met the inclusion/exclusion criteria. In the total population, individuals were primarily female (85.5%), lived in the South (48.5%), and had a mean (SD) age of 44 (12) years, which was consistent across the onabotA and CGRP mAb cohorts. Common comorbid conditions included anxiety (23.9%), depression (18.2%), hypertension (16.5%), and sleep disorders (16.9%). After adjusting for potential confounding variables, persistence to onabotA during the 12-month follow-up period was 40.7% vs 27.8% for CGRP mAbs (odds ratio [OR] = 0.683; 95% CI = 0.604-0.768; P < 0.0001). Persistence to erenumab, fremanezumab, and galcanezumab was 25.5% (OR = 0.627; 95% CI = 0.541-0.722; P < 0.0001), 30.3% (OR = 0.746; 95% CI = 0.598-0.912; P = 0.0033), and 33.7% (OR = 0.828; 95% CI = 0.667-1.006; P = 0.058). All-cause ($18,292 vs $18,275; P = 0.9739) and migraine-related ($8,990 vs $9,341; P = 0.1374) costs were comparable between the onabotA and CGRP mAb groups. CONCLUSIONS: Among adults with CM receiving onabotA and CGRP mAbs, individuals initiating onabotA treatment had higher persistence compared with those receiving CGRP mAbs. Total all-cause and migraine-related costs over 12 months were comparable between those receiving onabotA and CGRP mAbs. DISCLOSURES: This study was sponsored by Allergan (prior to its acquisition by AbbVie), they contributed to the design and interpretation of data and the writing, reviewing, and approval of final version. Writing and editorial assistance was provided to the authors by Dennis Stancavish, MS, of Peloton Advantage, LLC, an OPEN Health company, Parsippany, NJ, and was funded by AbbVie. The opinions expressed in this article are those of the authors. The authors received no honorarium/fee or other form of financial support related to the development of this article. Dr Schwedt serves on the Board of Directors for the American Headache Society and the American Migraine Foundation. Within the prior 12 months he has received research support from Amgen, Henry Jackson Foundation, Mayo Clinic, National Institutes of Health, Patient-Centered Outcomes Research Institute, SPARK Neuro, and US Department of Defense. Within the past 12 months, he has received personal compensation for serving as a consultant or advisory board member for AbbVie, Allergan, Axsome, BioDelivery Science, Biohaven, Collegium, Eli Lilly, Ipsen, Linpharma, Lundbeck, and Satsuma. He holds stock options in Aural Analytics and Nocira. He has received royalties from UpToDate. Dr Lee and Ms Shah are employees of AbbVie and may hold AbbVie stock. Dr Gillard was an employee of AbbVie and may hold AbbVie stock. Dr Knievel has served as a consultant for AbbVie, Amgen, Eli Lilly, and Biohaven; conducted research for AbbVie, Amgen, and Eli Lilly; and is on speaker programs for AbbVie and Amgen. Dr McVige has served as a speaker and/or received research support from Allergan (now AbbVie Inc.), Alder, Amgen/Novartis, Avanir, Biohaven, Eli Lilly, Lundbeck, and Teva. Ms Wang and Ms Wu are employees of Genesis Research, which provides consulting services to AbbVie. Dr Blumenfeld, within the past 12 months, has served on advisory boards for Allergan, AbbVie, Aeon, Alder, Amgen, Axsome, BDSI, Biohaven, Impel, Lundbeck, Lilly, Novartis, Revance, Teva, Theranica, and Zosano; as a speaker for Allergan, AbbVie, Amgen, BDSI, Biohaven, Lundbeck, Lilly, and Teva; as a consultant for Allergan, AbbVie, Alder, Amgen, Biohaven, Lilly, Lundbeck, Novartis, Teva, and Theranica; and as a contributing author for Allergan, AbbVie, Amgen, Biohaven, Novartis, Lilly, and Teva. He has received grant support from AbbVie and Amgen. AbbVie is committed to responsible data sharing regarding the clinical trials we sponsor. This includes access to anonymized, individual, and trial-level data (analysis data sets), as well as other information (eg, protocols, clinical study reports, or analysis plans), as long as the trials are not part of an ongoing or planned regulatory submission. This includes requests for clinical trial data for unlicensed products and indications. These clinical trial data can be requested by any qualified researchers who engage in rigorous, independent scientific research, and will be provided following review and approval of a research proposal and Statistical Analysis Plan and execution of a Data Sharing Agreement. Data requests can be submitted at any time after approval in the United States and Europe and after acceptance of this manuscript for publication. The data will be accessible for 12 months, with possible extensions considered. For more information on the process, or to submit a request, visit the following link: https://www.abbvieclinicaltrials.com/hcp/data-sharing/.

Healthcare costs and resource utilization in patients with migraine treated with erenumab: A retrospective, non-interventional study using claims data from the United States.

OBJECTIVE: To assess healthcare costs and healthcare resource utilization (HCRU) among adult patients who newly initiated erenumab in the United States. METHODS: This retrospective, non-interventional analysis included adult patients (aged ≥18 years) newly initiating erenumab and who had three consecutive monthly claims for erenumab (11/1/2017-9/1/2019) from the Komodo Health database. Outcomes included migraine-related and all-cause costs, use of other preventive/acute migraine medications, and HCRU. All outcomes were compared during the 180-day pre- versus the 180-day post-index periods. Cost outcomes were also assessed for longer periods including post-index Days 91-270 and monthly mean post-index costs for the longest time of continuous insurance enrollment. RESULTS: Overall, 1839 patients with migraine were included for analysis. Compared to the 180-day pre-index period, an increase in total migraine-related costs (+$2639; p < 0.0001), migraine-related prescription costs (+$3435, p < 0.0001), all-cause total costs (+$2977; p < 0.001), and all-cause prescription costs (+$4102; p < 0.0001) were observed during the 180-day post-index period after adjusting for covariates. Conversely, reduction in migraine-related medical costs (-$896; p < 0.0001), and significantly lower odds of migraine-related emergency room visits (odds ratio [OR] 0.60, 95% confidence interval [CI] 0.44-0.82; p = 0.001), migraine-related office visits (OR 0.58, 95% CI 0.53-0.64; p < 0.0001), and migraine-related neurologist visits (OR 0.69, 95% CI 0.63-0.75; p < 0.0001) were observed during the 180-days post-index period. There were significant decreases in the odds of having overall preventive migraine medications (OR 0.81, 95% CI 0.75-0.87; p < 0.0001), acute-migraine medications (OR 0.92, 95% CI 0.85-1.00; p = 0.038), and triptan (OR 0.79, 95% CI 0.73-0.85; p < 0.0001) during the 180-day post-index period. Sensitivity analyses on cost outcomes found no statistically significant differences in pre-index migraine-related costs compared to post-index migraine-related costs when assessing longer post-index follow-up periods. CONCLUSION: Initiation of therapy with a novel treatment is often associated with an increase in overall healthcare costs due to the entrance costs associated with novel therapy. For a chronic condition such as migraine, cost versus health benefits should be evaluated over a long period (e.g., ≥2 years) to better understand the true benefits of therapy. Data from this study suggest that the entrance cost for erenumab, the primary driver of the high post-index prescription costs gets mitigated by reduced medical costs over long-term follow-up. The results indicate better disease management in adult patients with migraine, which should be an important consideration for both patients and payors, as these findings have shown an offset between migraine-related prescription and medical costs.

Photophobia Contributes to Migraine-Associated Disability and Reduced Work Productivity: Results From the American Registry for Migraine Research (ARMR).

BACKGROUND: Photosensitivity, often called "photophobia" in the migraine literature, is a common and bothersome symptom for most people during their migraine attacks. This study aimed to investigate the association of photophobia severity with work productivity, activity impairment, and migraine-associated disability using data from a large cohort of patients with migraine who were enrolled into the American Registry for Migraine Research (ARMR). METHODS: This study used Photosensitivity Assessment Questionnaire (PAQ) scores to investigate the relationship between photophobia severity with work productivity and activity impairment (using the Work Productivity and Activity Impairment [WPAI] questionnaire) and migraine-related disability (using the Migraine Disability Assessment [MIDAS]) among those with migraine. Summary statistics are presented as means and standard deviations for variables that were normally distributed and as medians and interquartile ranges for variables that were not normally distributed. Multiple linear regression models were developed to measure the relationships between photophobia scores with work productivity and activity impairment and migraine-associated disability, controlling for age, sex, headache frequency, headache intensity, anxiety (using the generalized anxiety disorder [GAD-7]), and depression (using the Patient Health Questionnaire [PHQ-2]). RESULTS: One thousand eighty-four participants were included. Average age was 46.1 (SD 13.8) years, 87.2% (n = 945) were female, average headache frequency during the previous 90 days was 44.3 (SD 29.9), average headache intensity was 5.9 (SD 1.7), median PHQ-2 score was 1 (IQR 0-2), and median GAD-7 was 5 (IQR 2-8). Mean PAQ score was 0.47 (SD 0.32), and median MIDAS score was 38 (IQR 15.0-80.0). Among the 584 employed participants, 47.4% (n = 277) reported missing work in the past week because of migraine, mean overall work impairment was 42.8% (SD 26.7), mean activity impairment was 42.5% (SD 26.2), mean presenteeism score was 38.4% (SD 24.4), and median absenteeism was 0 (IQR 0-14.5). After controlling for age, sex, headache frequency, average headache intensity, PHQ-2 score, and GAD-7 score, there was a statistically significant association between photophobia scores with: a) MIDAS scores (F[7,1028] = 127.42, P < 0.001, R2 = 0.461, n = 1,036); b) overall work impairment (F[7,570] = 29.23, P < 0.001, R2 = 0.255, n = 578); c) activity impairment (F[7,570] = 27.42, P < 0.001, R2 = 0.243, n = 578); d) presenteeism (F[7,570] = 29.17, P < 0.001, R2 = 0.255, n = 578); and e) absenteeism for the zero-inflated (P = 0.003) and negative binomial (P = 0.045) model components (P < 0.001, n = 578). CONCLUSIONS: In those with migraine, severe photophobia is associated with reduced work productivity and higher presenteeism, absenteeism, activity impairment, and migraine-related disability.