Advancing toward precision migraine treatment: Predicting responses to preventive medications with machine learning models based on patient and migraine features.

OBJECTIVE: To develop machine learning models using patient and migraine features that can predict treatment responses to commonly used migraine preventive medications. BACKGROUND: Currently, there is no accurate way to predict response to migraine preventive medications, and the standard trial-and-error approach is inefficient. METHODS: In this cohort study, we analyzed data from the Mayo Clinic Headache database prospectively collected from 2001 to December 2023. Adult patients with migraine completed questionnaires during their initial headache consultation to record detailed clinical features and then at each follow-up to track preventive medication changes and monthly headache days. We included patients treated with at least one of the following migraine preventive medications: topiramate, beta-blockers (propranolol, metoprolol, atenolol, nadolol, timolol), tricyclic antidepressants (amitriptyline, nortriptyline), verapamil, gabapentin, onabotulinumtoxinA, and calcitonin gene-related peptide (CGRP) monoclonal antibodies (mAbs) (erenumab, fremanezumab, galcanezumab, eptinezumab). We pre-trained a deep neural network, "TabNet," using 145 variables, then employed TabNet-embedded data to construct prediction models for each medication to predict binary outcomes (responder vs. non-responder). A treatment responder was defined as having at least a 30% reduction in monthly headache days from baseline. All model performances were evaluated, and metrics were reported in the held-out test set (train 85%, test 15%). SHapley Additive exPlanations (SHAP) were conducted to determine variable importance. RESULTS: Our final analysis included 4260 patients. The responder rate for each medication ranged from 28.7% to 34.9%, and the mean time to treatment outcome for each medication ranged from 151.3 to 209.5 days. The CGRP mAb prediction model achieved a high area under the receiver operating characteristics curve (AUC) of 0.825 (95% confidence interval [CI] 0.726, 0.920) and an accuracy of 0.80 (95% CI 0.70, 0.88). The AUCs of prediction models for beta-blockers, tricyclic antidepressants, topiramate, verapamil, gabapentin, and onabotulinumtoxinA were: 0.664 (95% CI 0.579, 0.745), 0.611 (95% CI 0.562, 0.682), 0.605 (95% CI 0.520, 0.688), 0.673 (95% CI 0.569, 0.724), 0.628 (0.533, 0.661), and 0.581 (95% CI 0.550, 0.632), respectively. Baseline monthly headache days, age, body mass index (BMI), duration of migraine attacks, responses to previous medication trials, cranial autonomic symptoms, family history of headache, and migraine attack triggers were among the most important variables across all models. A variable could have different contributions; for example, lower BMI predicts responsiveness to CGRP mAbs and beta-blockers, while higher BMI predicts responsiveness to onabotulinumtoxinA, topiramate, and gabapentin. CONCLUSION: We developed an accurate prediction model for CGRP mAbs treatment response, leveraging detailed migraine features gathered from a headache questionnaire before starting treatment. Employing the same methods, the model performances for other medications were less impressive, though similar to the machine learning models reported in the literature for other diseases. This may be due to CGRP mAbs being migraine-specific. Incorporating medical comorbidities, genomic, and imaging factors might enhance the model performance. We demonstrated that migraine characteristics are important in predicting treatment responses and identified the most crucial predictors for each of the seven types of preventive medications. Our results suggest that precision migraine treatment is feasible.

Immunisation rates and predictors of undervaccination in infants with CHD.

Vaccination coverage for infants with CHD is unknown, yet these patients are at high risk for morbidity and mortality associated with vaccine-preventable illnesses. We determined vaccination rates for this population and identified predictors of undervaccination. We prospectively enrolled infants with CHD born between 1 January, 2012 and 31 December, 2015, seen in a single-centre cardiology clinic between 15 February, 2016 and 28 February, 2017. We assessed vaccination during the first year of life. Subjects who by age 1 year received all routine immunisations recommended during the first 6 months of life were considered fully vaccinated. We also evaluated influenza vaccination during subjects' first eligible influenza season. We obtained immunisation histories from primary care providers and collected demographic and clinical data via a parent survey and chart review. We used multivariable logistic regression to identify predictors of undervaccination. Among 260 subjects, only 60% were fully vaccinated. Vaccination rates were lowest for influenza (64.6%), rotavirus (71.1%), and Haemophilus influenzae type b (79.3%). Cardiac surgery with cardiopulmonary bypass during the first year of life was associated with undervaccination (51.5% versus 76.4% fully vaccinated, adjusted odds ratio 2.1 [95% confidence interval 1.1-3.9]). Other predictors of undervaccination were out-of-state primary care (adjusted odds ratio 2.7 [1.5-4.9]), multiple comorbidities (≥2 versus 0-1, adjusted odds ratio 2.0 [1.1-3.6]), and hospitalisation for >25% of the first year of life (>25% versus ≤25%, adjusted odds ratio 2.1 [1.1-3.9]). Targeted quality improvement initiatives focused on improving vaccination coverage for these infants, especially surrounding cardiac surgery, are needed.

Real-World Patient Experience of CGRP-Targeting Therapy for Migraine: a Narrative Review.

PURPOSE OF REVIEW: To summarize available calcitonin gene-related peptide (CGRP)-targeting therapies for migraine and discuss their use in real-world populations. BACKGROUND: CGRP has long been a topic of interest in migraine pathophysiology, with new therapies targeting CGRP since 2018 for both the preventive and acute treatment of migraine. METHODS: We searched PubMed using keywords including "migraine," "CGRP," "real-world," "erenumab," "galcanezumab," "fremanezumab," "eptinezumab," "ubrogepant," "rimegepant," and "atogepant." We reviewed all pertinent studies and summarized main findings. We also compiled detailed patient characteristics (e.g., migraine diagnoses, medication overuse, prior treatment failures) and treatment outcome measures, such as 50% responder rates, reduction in migraine days, and adverse event rates in several tables. Overall, studies reporting real-world patient experiences of CGRP-targeting therapies suggested meaningful effectiveness for migraine treatment with response rates comparable to the numbers reported in clinical trials. Furthermore, studies suggested benefit in patients with multiple prior unsuccessful treatment trials, medication overuse, and complex medical comorbidities. In some studies, adverse event rates have been notably higher than reported in clinical trials. Additional long-term data is needed to further evaluate sustained efficacy, predictors of treatment response, and adverse events.

Using a Mock Rounds Model and Neurology Patients to Teach Neurological Exam Skills in a Medical Neurobiology Course.

First-year medical student groups rotated through classrooms, each containing a Neurology patient and physician, as a "Neuro Day" event to make direct clinical connections with the basic sciences. Inspired by post-graduate Clerkships, this event provided timely first-hand experiences focusing on pathological neurologic exam findings. Qualitative and quantitative data were collected from end-of-course surveys. The results show how the event served to reinvigorate enthusiasm for learning Clinical Neurobiology outside the traditional lecture environment and could empower patients as educators within a teaching community.

Utilization of a Parental Approach to Informed Consent in Intravenous Tissue Plasminogen Activator Administration Decision-Making: Patient Preference and Ethical Considerations.

OBJECTIVE: While administration of intravenous tissue plasminogen activator (IV-tPA) is the standard of care in acute ischemic stroke and has been shown to have statistically significant benefit, there can also be potentially life-threatening complications; however, there is no standard informed consent approach. The purpose of this study was to present a parental, technical, and general model of informed consent for IV-TPA and to determine which approach was preferred. METHODS: Survey respondents were asked to hypothetically decide whether or not to provide consent for their family member to receive IV-tPA. Respondents were presented with 3 informed consent models: one emphasizing parental qualities, one emphasizing statistical data, and one representing a general consent statement. After being presented each model, the respondents had to select their preferred consent model, as well as rate their level of agreeability toward their family member receiving the medication following each approach. RESULTS: The results of 184 surveys showed respondents were equally as likely to give consent for their family member to receive IV-TPA following all three approaches; however, respondents were significantly more likely to prefer the parental approach compared to a technical or general approach. CONCLUSION: Our results indicate that while paternalism is generally discouraged in the medical community, some degree of parental language may be preferred by patients in tough decision-making situations toward consent to receive medical interventions.

Discharge Day: A Case-Based Interprofessional Exercise About Team Collaboration in Pediatrics.

Introduction: Interprofessional education, which gives medical students the opportunity to learn from, with, and about other health professionals, is an essential component of the undergraduate medical education curriculum. Nonetheless, deliberate and sustained integration of interprofessional education into the undergraduate medical learning experience can be challenging, especially within the clinical setting. Methods: We implemented a 75-minute, interactive, collaborative, case-based conference focusing on an interprofessional clinical challenge in a pediatric setting. Medical students on their pediatrics core rotation and trainees within social work, nursing, pharmacy, and nutrition explored the concept of a team, reflected on roles, and considered how interprofessional collaboration could influence patient outcomes. Results: One hundred ninety-two health professions students participated in 15 sessions at three sites over a 10-month period (September 2017-July 2018). After each session, participants completed a session evaluation. They gave high ratings to the effectiveness and relevance of the experience and the case vignette. Responses to open-ended questions revealed that students had learned the importance of leveraging the expertise of team members and had resolved to speak up when faced with an interprofessional challenge in the future. Discussion: This case-based session is a logistically feasible and positively received opportunity for health professions students to discuss interprofessional collaboration. It could be adapted for a variety of learner populations and academic environments and could be incorporated into existing curricula.

Lower limit of normal based spirometric abnormalities associated with radiographic abnormality in an elderly cohort at low risk for exposure.

BACKGROUND: Although the relation between radiographic abnormalities and spirometric impairment in people with asbestosis has been studied extensively, the extent of spirometric impairment associated with milder radiographic abnormalities is not established. OBJECTIVE: To test associations between mild radiographic abnormalities and Lower Limit of Normal (LLN)-based spirometry interpretation. METHODS: Spirometry and CXRs were collected for 1,026 at low risk of exposure to pneumoconiotic agents participants in a medical screening program. RESULTS: Individuals with each type of isolated or combined International Labour Organization (ILO) abnormalities had up to over sixfold statistically significant increase in odds of LLN-based restrictive pattern physiology (OR = 1.96, 95%CI 1.03-3.73 for parenchymal to OR = 6.09, 95%CI 1.94-19.10 for parenchymal and pleural) compared to those with normal films. CONCLUSIONS: The findings from this study confirm the association of mild profusion abnormalities with clinically relevant, LLN-based lung function abnormalities.

A novel pathway spatiotemporally activates Rac1 and redox signaling in response to fluid shear stress.

Hemodynamic forces regulate embryonic organ development, hematopoiesis, vascular remodeling, and atherogenesis. The mechanosensory stimulus of blood flow initiates a complex network of intracellular pathways, including activation of Rac1 GTPase, establishment of endothelial cell (EC) polarity, and redox signaling. The activity of the nicotinamide adenine dinucleotide phosphate (NADPH) oxidase can be modulated by the GTP/GDP state of Rac1; however, the molecular mechanisms of Rac1 activation by flow are poorly understood. Here, we identify a novel polarity complex that directs localized Rac1 activation required for downstream reactive oxygen species (ROS) production. Vav2 is required for Rac1 GTP loading, whereas, surprisingly, Tiam1 functions as an adaptor in a VE-cadherin-p67phox-Par3 polarity complex that directs localized activation of Rac1. Furthermore, loss of Tiam1 led to the disruption of redox signaling both in vitro and in vivo. Our results describe a novel molecular cascade that regulates redox signaling by the coordinated regulation of Rac1 and by linking components of the polarity complex to the NADPH oxidase.

The structure of the Tiam1 PDZ domain/ phospho-syndecan1 complex reveals a ligand conformation that modulates protein dynamics.

PDZ (PSD-95/Dlg/ZO-1) domains are protein-protein interaction modules often regulated by ligand phosphorylation. Here, we investigated the specificity, structure, and dynamics of Tiam1 PDZ domain/ligand interactions. We show that the PDZ domain specifically binds syndecan1 (SDC1), phosphorylated SDC1 (pSDC1), and SDC3 but not other syndecan isoforms. The crystal structure of the PDZ/SDC1 complex indicates that syndecan affinity is derived from amino acids beyond the four C-terminal residues. Remarkably, the crystal structure of the PDZ/pSDC1 complex reveals a binding pocket that accommodates the phosphoryl group. Methyl relaxation experiments of PDZ/SCD1 and PDZ/pSDC1 complexes reveal that PDZ-phosphoryl interactions dampen dynamic motions in a distal region of the PDZ domain by decoupling them from the ligand-binding site. Our data are consistent with a selection model by which specificity and phosphorylation regulate PDZ/syndecan interactions and signaling events. Importantly, our relaxation data demonstrate that PDZ/phospho-ligand interactions regulate protein dynamics and their coupling to distal sites.

Distinct ligand specificity of the Tiam1 and Tiam2 PDZ domains.

Guanine nucleotide exchange factor proteins of the Tiam family are activators of the Rho GTPase Rac1 and critical for cell morphology, adhesion, migration, and polarity. These proteins are modular and contain a variety of interaction domains, including a single post-synaptic density-95/discs large/zonula occludens-1 (PDZ) domain. Previous studies suggest that the specificities of the Tiam1 and Tiam2 PDZ domains are distinct. Here, we sought to conclusively define these specificities and determine their molecular origin. Using a combinatorial peptide library, we identified a consensus binding sequence for each PDZ domain. Analysis of these consensus sequences and binding assays with peptides derived from native proteins indicated that these two PDZ domains have overlapping but distinct specificities. We also identified residues in two regions (S(0) and S(-2) pockets) of the Tiam1 PDZ domain that are important determinants of ligand specificity. Site-directed mutagenesis of four nonconserved residues in these two regions along with peptide binding analyses confirmed that these residues are crucial for ligand affinity and specificity. Furthermore, double mutant cycle analysis of each region revealed energetic couplings that were dependent on the ligand being investigated. Remarkably, a Tiam1 PDZ domain quadruple mutant had the same specificity as the Tiam2 PDZ domain. Finally, analysis of Tiam family PDZ domain sequences indicated that the PDZ domains segregate into four distinct families based on the residues studied here. Collectively, our data suggest that Tiam family proteins have highly evolved PDZ domain-ligand interfaces with distinct specificities and that they have disparate PDZ domain-dependent biological functions.