Journal of neurology, neurosurgery, and psychiatry · 2025
Сопоставимость, безопасность и стойкость на этапе III клинических испытаний при множественном склерозе.
СтатьяМетаанализНазвание и аннотация: Машинный перевод · не проверен
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Salter Amber, Lancia Samantha, Kowalec Kaarina, Fitzgerald Kathryn, Marrie Ruth Ann
Аннотация
СВЯЗЬ МЕЖДУ коморбидностью и снижением стойкости к лечению различных видов склероза (MS) с целью изменения заболеваний (DMT). Имеется ограниченная информация о связи совместимости с результатами в области безопасности. Цель исследования состояла в том, чтобы оценить связь между совместимостью и результатами в области безопасности и стойкостью. Мы провели two- метаанализ данных об отдельных участниках, полученных в ходе клинических испытаний на этапе III МНТ. Индивидуальные совпадающие и совпадающие нагрузки, определяемые как сумма всех совпадающих факторов (n=15), были осмотрены. Были проанализированы результаты в области безопасности, которые были определены с использованием данных о неблагоприятных явлениях (АЭ), с тем чтобы выявить конкретные ПВ, представляющие интерес, включая инфекцию; аутоиммунное заболевание, связанное с лечением; рак; повышенные трансминазы и лимфопопию. Мы также изучили вопрос о досрочном прекращении судебного разбирательства. Результаты: Мы включили 17 клинические испытания, представляющие собой 16 794 Участники MS. Больше а 2-year :: после принятия последующих мер объединенная доля ОЭ 64% (В тыс.95% CI 59.4% в адрес 68.9%) и большинство из них были инфекциями AE. Увеличение бремени комординации было связано с повышением коэффициента АЭ (коэффициент процентной доли)95% CI) 1:: 1.13 (В тыс.1.09 в адрес 1.17); 2:: 1.19 (В тыс.1.14 в адрес 1.23); ≥3:: 1.25 (В тыс.1.18 в адрес 1.33)) по сравнению с теми, кто не имеет комординации. В тех случаях, когда судебное разбирательство проводится в одном и том же месте, досрочное прекращение 17% участников95% CI 13.8% в адрес 20.9%). Более высокий риск прекращения судебного разбирательства связан с более высоким уровнем комординации (см.2:: 1.23 (В тыс.1.07 в адрес 1.42); ≥3:: 1.19 (В тыс.1.01 в адрес 1.40)) по сравнению с теми, кто не имеет комординации. Психиатрические расстройства были связаны с прекращением судебных процессов. ВЫВОДЫ: Увеличение бремени коморбидности связано с повышением риска получения результатов в области безопасности и ранним прекращением деятельности ДМТ среди лиц, имеющих МЗС, участвующих в клинических испытаниях МЗМР, с уделением особого внимания важной роли комординаций в обеспечении безопасности и стойкости ДМТ.
Оригинал публикации
Comorbidities, safety and persistence in phase III clinical trials in multiple sclerosis.
BACKGROUND: Associations between comorbidity and reduced persistence to disease-modifying therapies (DMTs) in multiple sclerosis (MS) have been identified. Limited information is available regarding the association of comorbidity with safety outcomes. The study objective was to evaluate the association of comorbidities with safety outcomes and persistence.
METHODS: We conducted a two-stage meta-analysis of individual participant data from phase III clinical trials of MS DMTs. Individual comorbidities and comorbidity burden, defined as the sum of all comorbidities (n=15), were examined. Safety outcomes, defined using adverse event (AE) data, were reviewed to identify specific AEs of interest, including infection; treatment-emergent autoimmune disease; cancer; elevated transaminases and lymphopenia. We also examined any early trial discontinuation.
RESULTS: We included 17 clinical trials representing 16 794 MS participants. Over a 2-year follow-up, the pooled proportion of AEs was 64% (95% CI 59.4% to 68.9%) and the majority were infection AEs. Increasing comorbidity burden was associated with an increased rate of AEs (rate ratio (95% CI) 1: 1.13 (1.09 to 1.17); 2: 1.19 (1.14 to 1.23); ≥3: 1.25 (1.18 to 1.33)) compared with those with no comorbidity. When pooled across trials, early discontinuation affected 17% of participants (95% CI 13.8% to 20.9%). A higher risk of trial discontinuation was associated with higher comorbidity burden (2: 1.23 (1.07 to 1.42); ≥3: 1.19 (1.01 to 1.40)) compared with those with no comorbidity. Psychiatric disorders were associated with trial discontinuation.
CONCLUSIONS: Higher comorbidity burden is associated with increased risk of experiencing safety outcomes and early DMT discontinuation among individuals with MS enrolled in clinical trials of MS-DMTs, highlighting the important role of comorbidities in the safety and persistence of DMTs.
Перевод полного текста: Перевод в очереди
Читать найденный полный текст
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/ .
WHAT IS ALREADY KNOWN ON THIS TOPIC
WHAT THIS STUDY ADDS
A greater number of comorbidities was associated with elevated rates of adverse events, infection adverse events, serious adverse events and early trial discontinuation. A higher burden of psychiatric disorders and individual psychiatric disorders were associated with an increased rate of early trial discontinuation, as were diabetes, functional cardiovascular disease, cerebrovascular disease, peripheral vascular disease and migraine.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Given the increasing use of disease-modifying therapies in persons with multiple sclerosis (MS) who tend to be older and have a higher prevalence of comorbidity, the effect of comorbidity warrants further study in the broader MS population to understand the net benefit of treatment in those with comorbidity. Earlier discontinuation of therapy for those with comorbidity suggests there is a need to understand the mechanisms behind early discontinuations of disease-modifying therapy and provide support to patients from the onset.
Introduction
Treatment of multiple sclerosis (MS) has changed substantially over the last two decades with over 20 treatments approved. Several of the newer therapies are more effective than the first-generation disease-modifying therapies (DMTs), but they are also associated with more serious adverse events (SAEs) such as progressive multifocal leukoencephalopathy and infections. 1 2 Comorbid conditions can affect safety and tolerability of some DMTs. Comorbid migraine, for example, is associated with greater difficulty tolerating interferon-beta owing to worsening headache profiles. 3 4 Comorbid diabetes is associated with an increased risk of fingolimod-associated macular oedema. 5 Observational studies also suggest comorbidity is associated with increased risk of intolerance. 6 Yet, the association of comorbidity with the safety of DMTs has not been evaluated systematically. This is important because of the high prevalence of comorbidity in MS.
Although participants with comorbidities in clinical trials are often excluded because of safety concerns, the consequence is a lack of knowledge about the safety, tolerability and efficacy of DMT in persons with MS living with comorbidities. The International Advisory Committee on Clinical Trials in MS recommended that the comorbidity status of enrolled clinical trial populations should be clearly and consistently described; and recommended evaluating safety and tolerability in clinical trials according to comorbidity status. 7 Recently, we found that the pooled prevalence of at least one comorbidity among participants from 17 phase III clinical trials for MS DMTs was 45% 8 ; suggesting the feasibility of examining the safety of DMTs in persons with MS and comorbidity.
Therefore, we aimed to evaluate the association between comorbidity and safety and persistence in phase III clinical trials of MS DMTs. We hypothesised that a greater number of comorbidities would be associated with a greater likelihood of AEs and early discontinuation from the clinical trial.
Methods
Study design
We performed a two-stage meta-analysis using individual participant data from phase III clinical trials of MS DMTs for the first stage and pooling summary estimates across trials in the second stage. Sponsors provided individual participant data which was harmonised across trials using the Maelstrom retrospective harmonisation guidelines. 9
We followed the Strengthening the Reporting of Observational Studies in Epidemiology reporting guidelines. 10
Trials and Populations
All phase III clinical trials of MS DMTs with individual participant data accessible ( online supplemental etable 1 ; NCT00027300 , NCT00030966 , NCT00211887 , NCT00289978 , NCT00355134 , NCT00213135 , NCT00340834 , NCT00530348 , NCT00548405 , NCT00451451 , NCT00420212 , NCT00906399 , NCT00731692 , NCT01247324 , NCT01412333 , NCT01194570 , NCT01416181 , NCT02792218 , NCT02792231 ) were requested from trial sponsors via sponsor-specific mechanisms. The trials used in the analysis and description of the harmonisation approach have been described previously. 8 11 Trials with at least 2 years of follow-up were included. We used the intention-to-treat population from all trial arms (treatment, placebo and/or active comparator) in each trial for this analysis.
Comorbidities
Classification of comorbidities based on the individual participant data from the clinical trials has been described. 8 Briefly, we classified chronic conditions recommended by the International Advisory Committee on Clinical Trials in MS 7 and additional comorbidities of interest based on medical history data reported at trial enrolment only ( online supplemental etable 2 ). Chronic conditions included depression, anxiety, hypertension, hyperlipidaemia, migraine, diabetes and chronic lung disease (including asthma and chronic obstructive pulmonary disease), autoimmune thyroid disease, other psychiatric disorders (eg, bipolar disorder, personality disorder), cerebrovascular diseases, functional cardiovascular diseases (CVDs), peripheral vascular disease (PVD), other miscellaneous autoimmune conditions and skin conditions. For each participant, we summed the number of comorbidities (comorbidity burden) present (total=15) and categorised the total number present as 0, 1, 2 or 3 or more. We also summed the number of cardiometabolic (hypertension, hyperlipidaemia, functional cardiovascular, ischaemic, PVD, diabetes) and psychiatric (depression, anxiety and other psychiatric disorders) conditions for each participant and categorised the number present in each as 0, 1 or 2 or more.
Safety outcomes were defined using AE data captured for the duration of follow-up from an individual trial. AE data were reviewed to identify specific events of interest including (1) infection; (2) treatment-emergent autoimmune disease; (3) cancer; (4) elevated transaminases and (5) lymphopenia. The presence of any of these events (AEs of special interest) was the primary safety outcome. We also considered the presence of infection AEs only, SAE, and the reporting of any AE and SAE during the trial as secondary safety outcomes. Persistence was determined using study completion status on the participant disposition form indicating whether the participant completed the study or discontinued early for any reason. The time to discontinuation was based on the treatment initiation and end of treatment date.
Covariates
Covariates considered in the analysis included demographic and clinical characteristics at baseline (sex, age, disability level (measured using the Expanded Disability Status Scale, EDSS), treatment assignment (intervention vs placebo/active comparator), disease duration and the number of relapses in the prior year). Several trials reported age only in 5-year increments, so age was treated as categorical. 12 , 14 We based disease duration on year of symptom onset when reported, and diagnosis year otherwise. Body mass index (BMI) was included in the model as a sensitivity analysis where available.
Analysis
The association of the comorbidity (burden or individual) with safety outcomes and early trial discontinuation was evaluated using a two-stage meta-analysis approach. In the first stage, we fit generalised linear models with a Poisson distribution to evaluate the association of comorbidity burden and each comorbidity with each outcome for each trial. The natural log of person-years (from enrolment to the date the AE is first reported) was included as an offset in the model to account for variable duration of follow-up. We adjusted for age, sex, treatment assignment, disease duration, number of relapses 1 year prior and baseline EDSS. In the second stage, the adjusted rate ratios (RRs) and 95% CIs for each trial were pooled using random-effects meta-analysis with the Der-Simonian and Laird estimator. 15 The pooled adjusted RR (95% CI) for models is reported. The heterogeneity of findings across studies was assessed using the I 2 statistic where values of 0%–50% indicated low heterogeneity and 51%–100% indicated high heterogeneity. 16 Missing data were not imputed. Sensitivity analyses included examining the results for trials with BMI data (continuous, n=16 trials), and stratifying the meta-analysis for AEs, infection AEs and early discontinuation by placebo, low and high efficacy therapies. The stratified meta-analyses were conducted by examining the association of comorbidity and AEs, infection AEs and early discontinuation within a treatment group and meta-analysing the results by treatment category (placebo, low (dimethyl fumarate, fingolimod, glatiramer acetate, interferon β-1a, pegylated interferon) and high (natalizumab, alemtuzumab, ocrelizumab, cladribine tablets)). 17 Analyses were conducted using SAS V.9.4 (SAS Institute).
Data availability
Investigators may request access to anonymised individual patient data and redacted trial documents from each sponsor based on their individual processes. Before data use, proposals must be approved by the sponsor or an independent review panel, as appropriate, and a data sharing agreement obtained. The investigators will provide the analysis code for replication of results on request.
Results
Trials
19 trials were accessed from sponsors ( NCT00027300 , NCT00030966 , NCT00211887 , NCT00289978 , NCT00355134 , NCT00213135 , NCT00340834 , NCT00530348 , NCT00548405 , NCT00451451 , NCT00420212 , NCT00906399 , NCT00731692 , NCT01247324 , NCT01412333 , NCT01194570 , NCT01416181 , NCT02792218 , NCT02792231 ). Two trials were excluded due to having only 1 year of follow-up ( NCT00340834 , NCT00906399 ), leaving 17 trials for the analysis ( online supplemental etable 1 describes key trial information). 12 , 1618 The harmonised demographic and clinical characteristics of each trial are described in online supplemental etable 3 .
Adverse events
Considering the AEs of interest, the pooled prevalence was 64.0% (95% CI 59.4% to 68.9%) and the pooled prevalence of infection AEs was 60.7% (95% CI 55.8% to 66.1%). Increasing comorbidity burden was associated with an increased rate of AEs ( table 1 ). All comorbidities except for functional CVDs were associated with an increased rate of AEs ( table 1 ). Specifically examining infection AEs, similar associations were observed ( table 1 ). Compared with having no comorbidities, a greater number of comorbidities was associated with an elevated rate of infection AEs (1 comorbidity: 1.14 (1.09 to 1.19); 2 comorbidities: 1.21 (1.15 to 1.27); ≥3 comorbidities: 1.29 (1.19 to 1.40)). Individual comorbidities, except for autoimmune thyroid conditions, were associated with an elevated rate of infection AEs, with comorbidities having a 10%–20% higher rate of infection AEs compared with those without the comorbidity. Sensitivity analyses including BMI were similar with minimal attenuation of the effect ( online supplemental etable 4 ). Across DMT categories, placebo and high efficacy therapies were observed to have nominally higher rates of AEs and infection AEs compared with lower efficacy therapies ( figure 1 , online supplemental etable 5 ).
Table 1 Association of comorbidity with safety outcomes and persistence in MS clinical trial populations Pooled adjusted rate ratios (95% CI) Comorbidity * * AE Infection AE SAE Discontinuation All comorbidities 0 Reference Reference Reference Reference 1 1.13 (1.09, 1.17) 1.14 (1.09, 1.19) 1.19 (0.96, 1.47) 1.01 (0.92, 1.11) 2 1.19 (1.14, 1.23) 1.21 (1.15, 1.27) 1.54 (1.19, 1.99) 1.23 (1.07, 1.42) ≥3 1.25 (1.18, 1.33) 1.29 (1.19, 1.40) 1.88 (1.42, 2.50) 1.19 (1.01, 1.40) Cardiometabolic 0 Reference Reference Reference Reference 1 1.10 (1.06, 1.14) 1.10 (1.05, 1.14) 1.50 (1.18, 1.91) 0.99 (0.90, 1.08) ≥2 1.15 (1.10, 1.21) 1.17 (1.10, 1.24) 2.03 (1.44, 2.85) 0.96 (0.81, 1.14) Psychiatric 0 Reference Reference Reference Reference 1 1.13 (1.08, 1.19) 1.15 (1.09, 1.20) 1.39 (1.13, 1.71) 1.19 (1.07, 1.32) ≥2 1.15 (1.08, 1.23) 1.17 (1.08, 1.27) 1.91 (1.38, 2.64) 1.38 (1.11, 1.72) Hyperlipidaemia 1.15 (1.09, 1.22) 1.15 (1.06, 1.25) 1.61 (1.24, 2.09) 0.92 (0.82, 1.03) Hypertension 1.08 (1.04, 1.11) 1.08 (1.03, 1.13) 1.37 (1.09, 1.73) 0.96 (0.86, 1.06) Diabetes 1.14 (1.07, 1.21) 1.16 (1.07, 1.26) 4.03 (2.62, 6.19) 1.37 (1.06, 1.77) Ischaemic heart disease 1.32 (1.26, 1.39) 1.36 (1.29, 1.43) 1.29 (0.24, 7.05) 1.57 (0.98, 2.52) Functional CVD 1.08 (0.97, 1.21) 1.08 (1.00, 1.18) 2.13 (1.27, 3.55) 1.22 (1.04, 1.44) Cerebrovascular 1.50 (1.32, 1.71) 1.52 (1.27, 1.82) 11.6 (4.87, 27.63) 2.90 (2.01, 4.17) PVD 1.19 (1.04, 1.37) 1.24 (1.08, 1.43) 3.90 (1.78, 8.52) 1.93 (1.28, 2.90) Depression 1.13 (1.07, 1.19) 1.15 (1.08, 1.22) 1.36 (1.11, 1.67) 1.20 (1.06, 1.37) Anxiety 1.12 (1.06, 1.19) 1.11 (1.05, 1.18) 1.62 (1.21, 2.17) 1.29 (1.10, 1.52) Other psychiatric 1.11 (1.03, 1.18) 1.10 (1.03, 1.17) 2.09 (1.48, 2.95) 1.31 (1.07, 1.62) Lung 1.15 (1.10, 1.19) 1.17 (1.12, 1.22) 1.72 (1.26, 2.34) 1.11 (0.99, 1.25) Migraine 1.14 (1.09, 1.20) 1.17 (1.10, 1.24) 1.26 (0.94, 1.68) 1.15 (1.04, 1.27) Skin condition 1.15 (1.11, 1.20) 1.17 (1.12, 1.23) 1.22 (0.78, 1.92) 1.08 (0.93, 1.26) Autoimmune thyroid 1.11 (1.01, 1.21) 1.06 (0.95, 1.18) 2.99 (1.26, 7.12) 1.25 (0.91, 1.70) Miscellaneous autoimmune 1.36 (1.25, 1.48) 1.33 (1.18, 1.50) 8.06 (4.39, 14.79) 1.64 (0.89, 3.04) * Models adjusted for age, sex, treatment assignment, baseline EDSS, disease duration and number of relapses in the year prior to enrolment. Separate models were conducted for each burden of comorbidity and individual comorbidity. AE, adverse event; CVD, cardiovascular disease; EDSS, Expanded Disability Status Scale; MS, multiple sclerosis; PVD, peripheral vascular disease; SAE, serious AE.
Association of comorbidity with safety outcomes and persistence in MS clinical trial populations
Models adjusted for age, sex, treatment assignment, baseline EDSS, disease duration and number of relapses in the year prior to enrolment. Separate models were conducted for each burden of comorbidity and individual comorbidity.
AE, adverse event; CVD, cardiovascular disease; EDSS, Expanded Disability Status Scale; MS, multiple sclerosis; PVD, peripheral vascular disease; SAE, serious AE.
Association of comorbidity with infection adverse events by disease-modifying therapy class in MS clinical trial populations. Separate models were conducted for each burden of comorbidity and individual comorbidity. *Models adjusted for age, sex, treatment assignment, baseline EDSS, disease duration and number of relapses in the year prior to enrolment. EDSS, Expanded Disability Status Scale; PVD, peripheral vascular disease.
The prevalence of any AE (secondary outcome) across trials was 88.7% (95% CI 86.1% to 91.4%). The rate of any AE increased with greater comorbidity burden (1 comorbidity: 1.06 (1.04 to 1.08), 2 comorbidities: 1.07 (1.04 to 1.10), ≥3 comorbidities: 1.11 (1.07 to 1.15); table 2 . Comorbidities were associated with an elevated rate of any AE, although the magnitude was smaller, with the risk ranging between 3% and 11% among the different conditions compared with those without the comorbidity ( table 2 ).
Table 2 Association of comorbidity with any adverse event and any serious adverse event in MS clinical trial populations Pooled adjusted rate ratios (95% CI) Comorbidity * Any AE Any SAE All comorbidities 0 Reference Reference 1 1.06 (1.04, 1.08) 1.12 (1.06, 1.17) 2 1.07 (1.04, 1.10) 1.20 (1.05, 1.37) ≥3 1.11 (1.07, 1.15) 1.81 (1.45, 2.25) Cardiometabolic 0 Reference Reference 1 1.03 (1.01, 1.05) 1.33 (1.17, 1.52) ≥2 1.05 (1.03, 1.07) 1.70 (1.33, 2.17) Psychiatric 0 Reference Reference 1 1.07 (1.04, 1.09) 1.13 (1.07, 1.20) ≥2 1.07 (1.04, 1.11) 1.47 (1.18, 1.83) Hyperlipidaemia 1.04 (1.02, 1.06) 1.31 (1.11, 1.55) Hypertension 1.03 (1.01, 1.04) 1.39 (1.19, 1.61) Diabetes 1.04 (1.03, 1.06) 1.70 (1.20, 2.42) Ischaemic heart disease 1.08 (1.05, 1.10) 1.75 (1.02, 3.00) Functional CVD 1.05 (1.02, 1.08) 1.22 (1.02, 1.46) Cerebrovascular 1.11 (1.07, 1.15) 3.07 (1.51, 6.27) PVD 1.08 (1.05, 1.12) 2.55 (1.45, 4.47) Depression 1.07 (1.04, 1.09) 1.19 (1.08, 1.31) Anxiety 1.06 (1.03, 1.09) 1.28 (1.09, 1.50) Other psychiatric 1.04 (1.02, 1.07) 1.49 (1.18, 1.87) Lung 1.04 (1.03, 1.05) 1.37 (1.15, 1.64) Migraine 1.07 (1.04, 1.09) 1.14 (0.99, 1.32) Skin condition 1.08 (1.05, 1.11) 1.04 (0.95, 1.15) Autoimmune thyroid 1.04 (1.03, 1.05) 1.33 (1.00, 1.78) Miscellaneous autoimmune 1.09 (1.06, 1.13) 2.42 (1.30, 4.49) * Models adjusted for age, sex, treatment assignment, baseline EDSS, disease duration and number of relapses in the year prior to enrolment. Separate models were conducted for each burden of comorbidity and individual comorbidity. AE, adverse event; CVD, cardiovascular disease; EDSS, Expanded Disability Status Scale; MS, multiple sclerosis; PVD, peripheral vascular disease; SAE, serious AE.
Association of comorbidity with any adverse event and any serious adverse event in MS clinical trial populations
Models adjusted for age, sex, treatment assignment, baseline EDSS, disease duration and number of relapses in the year prior to enrolment. Separate models were conducted for each burden of comorbidity and individual comorbidity.
AE, adverse event; CVD, cardiovascular disease; EDSS, Expanded Disability Status Scale; MS, multiple sclerosis; PVD, peripheral vascular disease; SAE, serious AE.
Serious AEs
For SAEs of interest, the pooled prevalence across trials was 3.2% (95% CI 2.6% to 4.0%). We observed a higher rate of SAEs in those with 2 or ≥3 comorbidities compared with those with no comorbidity (1 comorbidity: 1.19 (0.96 to 1.47); 2 comorbidities: 1.54 (1.19 to 1.99); ≥3 comorbidities: 1.88 (1.42 to 2.50), table 1 ). The number of cardiometabolic and psychiatric conditions was also associated with an elevated rate of SAEs. Most comorbidities were associated with an elevated rate of SAEs except for ischaemic heart disease, migraine and skin conditions ( table 1 ). After further adjusting for BMI, results were similar ( online supplemental etable 4 ).
Similarly, for the secondary outcome of any SAE, an increased rate of SAEs was identified with comorbidity ( table 2 ). Compared with those with no comorbidity, a greater comorbidity burden was associated with an elevated rate of SAEs (1 comorbidity: 1.12 (1.06 to 1.17); 2 comorbidities: 1.20 (1.05 to 1.37); ≥3 comorbidities: 1.81 (1.45 to 2.25), table 2 ). The presence of most comorbidities was associated with an elevated rate of SAE compared with those without the condition.
Persistence
The pooled prevalence of early trial discontinuation was 17.0% (13.8% to 20.9%). The presence of two or three or more comorbidities was associated with an increased rate of early trial discontinuation from a DMT after controlling for covariates (2 comorbidities: 1.22 (1.07 to 1.42), ≥3 comorbidities: 1.19 (1.01 to 1.40), table 1 ). Psychiatric comorbidity burden was associated with higher rates of early trial discontinuation in a dose-response fashion (1 comorbidity: 1.19 (1.07 to 1.32), ≥2 comorbidities: 1.38 (1.11 to 1.72), table 1 ) and was also observed for those with depression, anxiety, other psychiatric disorders ( figure 2 ). In contrast, a greater number of cardiometabolic comorbidities was not associated with early trial discontinuation; however, a subgroup of these comorbidities was associated with greater early discontinuation including diabetes, functional CVD, cerebrovascular disease and PVD ( figure 2 , table 1 ). Those with migraine when pooled had a higher rate of early trial discontinuation (1.15 (1.04 to 1.27)). These results were consistent with small attenuations when further adjusting for BMI ( online supplemental etable 4 ). A trend for higher efficacy therapies to have higher rates of early trial discontinuation for those with greater total comorbidity burden, psychiatric comorbidity burden and individual psychiatric comorbidity ( online supplemental etable 6 ).
Association of comorbidity with early trial discontinuation. CVD, cardiovascular disease; PVD, peripheral vascular disease.
Discussion
In this large study involving 16 794 participants from 17 clinical trials, we used a meta-analytical approach to evaluate the association between comorbidity and safety of DMTs, as well as early trial discontinuation. We found that a greater number of comorbidities were associated with elevated rates of AEs, infection AEs, SAEs and early trial discontinuation after adjusting for multiple demographic and clinical factors. A higher burden of psychiatric disorders and individual psychiatric disorders were associated with an increased rate of early discontinuation, as were diabetes, functional CVD, cerebrovascular disease, PVD and migraine.
Most of the AEs of interest in this study were infection-related AEs. MS is associated with an increased risk of infections, and those infections may have clinically significant consequences. 28 , 31 In a large study using administrative data from four Canadian provinces, some DMTs were associated with a modest increase in infection-related physician visits. 1 The association of comorbidity with infection-related AEs suggests comorbidity status may need to be considered in future studies investigating infection risk. The Canadian study also found a higher hazard of acute liver injury and abnormal haematologic findings for some DMTs. 1 However, the CIs for those associations were wide due to a small number of events. In our study, we were unable to evaluate these specific events due to their low frequency among the trials. Future studies using longitudinal data with a longer follow-up and greater sample sizes are likely needed to better examine the association of these events with comorbidity.
The association of comorbidities with safety outcomes is critical to understanding the net benefit of treatment in those with comorbidity. Prior meta-analyses on safety outcomes for MS DMTs have been conducted, but those studies did not consider comorbidity status. 32 33 Information on comorbidity and AEs is limited and primarily has focused on treatment-specific AEs for relevant comorbidities. 3 , 534 Our study systematically investigated comorbidity burden and individual comorbidities for multiple safety outcomes using data pooled across 17 trials. The robust association of higher comorbidity and elevated AE rates highlights the need for reporting of comorbidities in clinical trial participants as recommended by the International Advisory Committee on Clinical Trials. 7
In contrast to safety outcomes, the association of early DMT discontinuation with comorbidity has been examined previously. Persons with comorbidities at diagnosis have been shown to have lower rates of initiating and persisting on treatment and higher rates of switching when compared with those with no comorbidity. 6 35 36 However, a cohort study of 1464 individuals with MS in the Canadian province of Nova Scotia did not find an association between general comorbidity burden and time to discontinuation of DMTs. 37 Differences in our findings may be associated with differences in study designs and a number of comorbidities considered. Our study found an increased rate of early trial discontinuation for those with greater psychiatric comorbidity burden as well as individual psychiatric disorders. The Nova Scotia cohort study identified an increased hazard of discontinuing DMT (1.22 (1.03 to 1.44)) and a Swedish MS Registry study showed reduced persistence for specific DMTs in participants with depression or those using antidepressants. 37 38 As we used comorbidity reported at enrolment, our study design should preclude a DMT from causing the psychiatric disorder. Additionally, a systematic review did not find associations between DMTs and adverse psychiatric effects. 39
Cardiometabolic comorbidity burden and individual comorbidities were associated with increased rates of AEs and SAEs, but not early trial discontinuation. Similar to our findings, the Nova Scotia cohort study found no association of hyperlipidaemia, hypertension and ischaemic heart disease with DMT discontinuation, 37 but contrary to their study findings, we did find an association between diabetes and increased rates of early discontinuation. An Italian study found an association between increasing Framingham risk score, a cardiovascular health measure that incorporates multiple factors such as smoking, age, BMI, blood pressure and diabetes, and DMT escalation. 40 Another study found comorbidity was associated with increased hazard of switching to intolerance. 6 Information regarding cardiometabolic conditions and safety outcomes and persistence is limited, but given the increasing prevalence of cardiometabolic conditions with age, the effect of these conditions on safety and persistence in the broader MS populations warrants further study and will be relevant to trials in progressive MS populations who tend to be older.
Our study has strengths and limitations. We have systematically studied a broad range of comorbidities and DMTs in both relapsing and progressive MS populations. Our analysis included robust adjustment for sex, age, disability and other clinical factors in 17 clinical trials. Clinical trial data are less subject to surveillance bias compared with other observational study designs and ensure that comorbidity exposure preceded the introduction of DMT. However, the relatively short duration of DMT clinical trials limits our ability to examine longer-term safety concerns, such as cancer and hepatic dysfunction. These trials were not powered to detect these outcomes, and a low prevalence of some comorbidities may also reduce the ability to detect associations. Clinical trial populations are highly selected with respect to comorbidity, and while we identified participants with comorbidities, the severity of these comorbidities is unknown. We may be underestimating the association of safety and tolerability if only milder comorbidity is included. Moreover, the effects may be underestimated for the broader MS population where comorbidity burden and severity are likely greater. There is a need for studies evaluating these outcomes in real world MS populations.
In summary, comorbidity plays an important role in the safety and persistence of DMTs in persons with MS. The increased rate of safety outcomes in those with comorbidities can inform clinical care decisions for those with comorbidity, especially given the more severe AEs observed with the introduction of newer therapies. Studies with longer follow-up are still needed to improve our understanding of long-term safety outcomes. We showed greater early trial discontinuation, indicating lower persistence on therapy, for those with more than two comorbidities. This suggests there is a need to understand the mechanisms behind early discontinuations of DMT and provide greater support to patients with comorbidities.
Supplementary material
Acknowledgements
Amber Salter is supported (in part) by a Biostatistics/Informatics Junior Faculty Award (BI-2105-37656) from the National Multiple Sclerosis Society. This publication is based on research using data from Biogen provided by Biogen, Cambridge, USA. This publication is based on research using data provided by Merck KGaA, (CrossRef Funder ID: 10.13039/100009945). This publication is based on research using data from Novartis Pharma AG and Sanofi-Aventis Groupe that has been made available through www.ClinicalStudyDataRequest.com . This publication is based on research using data from Roche that has been made available through Vivli. Vivli has not contributed to or approved and is not in any way responsible for, the contents of this publication. CombiRx ( NCT00211887 ) was funded by the NIH, National Institute of Neurologic Disorders and Stroke (phase III study: UO1NS045719, Planning Grant R21NS41986) and the study agents and matched placebo were kindly provided by their manufacturers, Biogen and Teva Pharmaceutical Sponsors (Merck (CrossRef Funder ID: 10.13039/100009945), Biogen, Novartis Pharma AG, Sanofi-Aventis Groupe and Roche) reviewed this manuscript for medical accuracy only before journal submission.
The authors are fully responsible for the content of this manuscript, and the views and opinions described in the publication reflect solely those of the authors.
Footnotes
Funding: This work is supported by the Consortium of Multiple Sclerosis Centers Pilot award. Partial support was provided by the Department of Defense through the Multiple Sclerosis Research Program under Award No. W81XWH-21-1-0721. Opinions, interpretations conclusions and recommendations are those of the author and are not necessarily endorsed by the Department of Defense.
Provenance and peer review: Not commissioned; externally peer reviewed.
Patient consent for publication: Not applicable.
Ethics approval: The UT Southwestern institutional review board designated this study as non-human subjects’ research. Trial sponsors obtained study consents and reviewed data access requests to ensure they were consistent with consents obtained for the trials.
Data availability statement
Data may be obtained from a third party and are not publicly available.
References
Associated Data
Supplementary Materials
Data Availability Statement
Investigators may request access to anonymised individual patient data and redacted trial documents from each sponsor based on their individual processes. Before data use, proposals must be approved by the sponsor or an independent review panel, as appropriate, and a data sharing agreement obtained. The investigators will provide the analysis code for replication of results on request.
Data may be obtained from a third party and are not publicly available.
WHAT THIS STUDY ADDS
A greater number of comorbidities was associated with elevated rates of adverse events, infection adverse events, serious adverse events and early trial discontinuation. A higher burden of psychiatric disorders and individual psychiatric disorders were associated with an increased rate of early trial discontinuation, as were diabetes, functional cardiovascular disease, cerebrovascular disease, peripheral vascular disease and migraine.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Given the increasing use of disease-modifying therapies in persons with multiple sclerosis (MS) who tend to be older and have a higher prevalence of comorbidity, the effect of comorbidity warrants further study in the broader MS population to understand the net benefit of treatment in those with comorbidity. Earlier discontinuation of therapy for those with comorbidity suggests there is a need to understand the mechanisms behind early discontinuations of disease-modifying therapy and provide support to patients from the onset.
Introduction
Treatment of multiple sclerosis (MS) has changed substantially over the last two decades with over 20 treatments approved. Several of the newer therapies are more effective than the first-generation disease-modifying therapies (DMTs), but they are also associated with more serious adverse events (SAEs) such as progressive multifocal leukoencephalopathy and infections. 1 2 Comorbid conditions can affect safety and tolerability of some DMTs. Comorbid migraine, for example, is associated with greater difficulty tolerating interferon-beta owing to worsening headache profiles. 3 4 Comorbid diabetes is associated with an increased risk of fingolimod-associated macular oedema. 5 Observational studies also suggest comorbidity is associated with increased risk of intolerance. 6 Yet, the association of comorbidity with the safety of DMTs has not been evaluated systematically. This is important because of the high prevalence of comorbidity in MS.
Although participants with comorbidities in clinical trials are often excluded because of safety concerns, the consequence is a lack of knowledge about the safety, tolerability and efficacy of DMT in persons with MS living with comorbidities. The International Advisory Committee on Clinical Trials in MS recommended that the comorbidity status of enrolled clinical trial populations should be clearly and consistently described; and recommended evaluating safety and tolerability in clinical trials according to comorbidity status. 7 Recently, we found that the pooled prevalence of at least one comorbidity among participants from 17 phase III clinical trials for MS DMTs was 45% 8 ; suggesting the feasibility of examining the safety of DMTs in persons with MS and comorbidity.
Therefore, we aimed to evaluate the association between comorbidity and safety and persistence in phase III clinical trials of MS DMTs. We hypothesised that a greater number of comorbidities would be associated with a greater likelihood of AEs and early discontinuation from the clinical trial.
Methods
Study design
We performed a two-stage meta-analysis using individual participant data from phase III clinical trials of MS DMTs for the first stage and pooling summary estimates across trials in the second stage. Sponsors provided individual participant data which was harmonised across trials using the Maelstrom retrospective harmonisation guidelines. 9
We followed the Strengthening the Reporting of Observational Studies in Epidemiology reporting guidelines. 10
Trials and Populations
All phase III clinical trials of MS DMTs with individual participant data accessible ( online supplemental etable 1 ; NCT00027300 , NCT00030966 , NCT00211887 , NCT00289978 , NCT00355134 , NCT00213135 , NCT00340834 , NCT00530348 , NCT00548405 , NCT00451451 , NCT00420212 , NCT00906399 , NCT00731692 , NCT01247324 , NCT01412333 , NCT01194570 , NCT01416181 , NCT02792218 , NCT02792231 ) were requested from trial sponsors via sponsor-specific mechanisms. The trials used in the analysis and description of the harmonisation approach have been described previously. 8 11 Trials with at least 2 years of follow-up were included. We used the intention-to-treat population from all trial arms (treatment, placebo and/or active comparator) in each trial for this analysis.
Comorbidities
Classification of comorbidities based on the individual participant data from the clinical trials has been described. 8 Briefly, we classified chronic conditions recommended by the International Advisory Committee on Clinical Trials in MS 7 and additional comorbidities of interest based on medical history data reported at trial enrolment only ( online supplemental etable 2 ). Chronic conditions included depression, anxiety, hypertension, hyperlipidaemia, migraine, diabetes and chronic lung disease (including asthma and chronic obstructive pulmonary disease), autoimmune thyroid disease, other psychiatric disorders (eg, bipolar disorder, personality disorder), cerebrovascular diseases, functional cardiovascular diseases (CVDs), peripheral vascular disease (PVD), other miscellaneous autoimmune conditions and skin conditions. For each participant, we summed the number of comorbidities (comorbidity burden) present (total=15) and categorised the total number present as 0, 1, 2 or 3 or more. We also summed the number of cardiometabolic (hypertension, hyperlipidaemia, functional cardiovascular, ischaemic, PVD, diabetes) and psychiatric (depression, anxiety and other psychiatric disorders) conditions for each participant and categorised the number present in each as 0, 1 or 2 or more.
Safety outcomes were defined using AE data captured for the duration of follow-up from an individual trial. AE data were reviewed to identify specific events of interest including (1) infection; (2) treatment-emergent autoimmune disease; (3) cancer; (4) elevated transaminases and (5) lymphopenia. The presence of any of these events (AEs of special interest) was the primary safety outcome. We also considered the presence of infection AEs only, SAE, and the reporting of any AE and SAE during the trial as secondary safety outcomes. Persistence was determined using study completion status on the participant disposition form indicating whether the participant completed the study or discontinued early for any reason. The time to discontinuation was based on the treatment initiation and end of treatment date.
Covariates
Covariates considered in the analysis included demographic and clinical characteristics at baseline (sex, age, disability level (measured using the Expanded Disability Status Scale, EDSS), treatment assignment (intervention vs placebo/active comparator), disease duration and the number of relapses in the prior year). Several trials reported age only in 5-year increments, so age was treated as categorical. 12 , 14 We based disease duration on year of symptom onset when reported, and diagnosis year otherwise. Body mass index (BMI) was included in the model as a sensitivity analysis where available.
Analysis
The association of the comorbidity (burden or individual) with safety outcomes and early trial discontinuation was evaluated using a two-stage meta-analysis approach. In the first stage, we fit generalised linear models with a Poisson distribution to evaluate the association of comorbidity burden and each comorbidity with each outcome for each trial. The natural log of person-years (from enrolment to the date the AE is first reported) was included as an offset in the model to account for variable duration of follow-up. We adjusted for age, sex, treatment assignment, disease duration, number of relapses 1 year prior and baseline EDSS. In the second stage, the adjusted rate ratios (RRs) and 95% CIs for each trial were pooled using random-effects meta-analysis with the Der-Simonian and Laird estimator. 15 The pooled adjusted RR (95% CI) for models is reported. The heterogeneity of findings across studies was assessed using the I 2 statistic where values of 0%–50% indicated low heterogeneity and 51%–100% indicated high heterogeneity. 16 Missing data were not imputed. Sensitivity analyses included examining the results for trials with BMI data (continuous, n=16 trials), and stratifying the meta-analysis for AEs, infection AEs and early discontinuation by placebo, low and high efficacy therapies. The stratified meta-analyses were conducted by examining the association of comorbidity and AEs, infection AEs and early discontinuation within a treatment group and meta-analysing the results by treatment category (placebo, low (dimethyl fumarate, fingolimod, glatiramer acetate, interferon β-1a, pegylated interferon) and high (natalizumab, alemtuzumab, ocrelizumab, cladribine tablets)). 17 Analyses were conducted using SAS V.9.4 (SAS Institute).
Data availability
Investigators may request access to anonymised individual patient data and redacted trial documents from each sponsor based on their individual processes. Before data use, proposals must be approved by the sponsor or an independent review panel, as appropriate, and a data sharing agreement obtained. The investigators will provide the analysis code for replication of results on request.
Results
Trials
19 trials were accessed from sponsors ( NCT00027300 , NCT00030966 , NCT00211887 , NCT00289978 , NCT00355134 , NCT00213135 , NCT00340834 , NCT00530348 , NCT00548405 , NCT00451451 , NCT00420212 , NCT00906399 , NCT00731692 , NCT01247324 , NCT01412333 , NCT01194570 , NCT01416181 , NCT02792218 , NCT02792231 ). Two trials were excluded due to having only 1 year of follow-up ( NCT00340834 , NCT00906399 ), leaving 17 trials for the analysis ( online supplemental etable 1 describes key trial information). 12 , 1618 The harmonised demographic and clinical characteristics of each trial are described in online supplemental etable 3 .
Adverse events
Considering the AEs of interest, the pooled prevalence was 64.0% (95% CI 59.4% to 68.9%) and the pooled prevalence of infection AEs was 60.7% (95% CI 55.8% to 66.1%). Increasing comorbidity burden was associated with an increased rate of AEs ( table 1 ). All comorbidities except for functional CVDs were associated with an increased rate of AEs ( table 1 ). Specifically examining infection AEs, similar associations were observed ( table 1 ). Compared with having no comorbidities, a greater number of comorbidities was associated with an elevated rate of infection AEs (1 comorbidity: 1.14 (1.09 to 1.19); 2 comorbidities: 1.21 (1.15 to 1.27); ≥3 comorbidities: 1.29 (1.19 to 1.40)). Individual comorbidities, except for autoimmune thyroid conditions, were associated with an elevated rate of infection AEs, with comorbidities having a 10%–20% higher rate of infection AEs compared with those without the comorbidity. Sensitivity analyses including BMI were similar with minimal attenuation of the effect ( online supplemental etable 4 ). Across DMT categories, placebo and high efficacy therapies were observed to have nominally higher rates of AEs and infection AEs compared with lower efficacy therapies ( figure 1 , online supplemental etable 5 ).
Table 1 Association of comorbidity with safety outcomes and persistence in MS clinical trial populations Pooled adjusted rate ratios (95% CI) Comorbidity * * AE Infection AE SAE Discontinuation All comorbidities 0 Reference Reference Reference Reference 1 1.13 (1.09, 1.17) 1.14 (1.09, 1.19) 1.19 (0.96, 1.47) 1.01 (0.92, 1.11) 2 1.19 (1.14, 1.23) 1.21 (1.15, 1.27) 1.54 (1.19, 1.99) 1.23 (1.07, 1.42) ≥3 1.25 (1.18, 1.33) 1.29 (1.19, 1.40) 1.88 (1.42, 2.50) 1.19 (1.01, 1.40) Cardiometabolic 0 Reference Reference Reference Reference 1 1.10 (1.06, 1.14) 1.10 (1.05, 1.14) 1.50 (1.18, 1.91) 0.99 (0.90, 1.08) ≥2 1.15 (1.10, 1.21) 1.17 (1.10, 1.24) 2.03 (1.44, 2.85) 0.96 (0.81, 1.14) Psychiatric 0 Reference Reference Reference Reference 1 1.13 (1.08, 1.19) 1.15 (1.09, 1.20) 1.39 (1.13, 1.71) 1.19 (1.07, 1.32) ≥2 1.15 (1.08, 1.23) 1.17 (1.08, 1.27) 1.91 (1.38, 2.64) 1.38 (1.11, 1.72) Hyperlipidaemia 1.15 (1.09, 1.22) 1.15 (1.06, 1.25) 1.61 (1.24, 2.09) 0.92 (0.82, 1.03) Hypertension 1.08 (1.04, 1.11) 1.08 (1.03, 1.13) 1.37 (1.09, 1.73) 0.96 (0.86, 1.06) Diabetes 1.14 (1.07, 1.21) 1.16 (1.07, 1.26) 4.03 (2.62, 6.19) 1.37 (1.06, 1.77) Ischaemic heart disease 1.32 (1.26, 1.39) 1.36 (1.29, 1.43) 1.29 (0.24, 7.05) 1.57 (0.98, 2.52) Functional CVD 1.08 (0.97, 1.21) 1.08 (1.00, 1.18) 2.13 (1.27, 3.55) 1.22 (1.04, 1.44) Cerebrovascular 1.50 (1.32, 1.71) 1.52 (1.27, 1.82) 11.6 (4.87, 27.63) 2.90 (2.01, 4.17) PVD 1.19 (1.04, 1.37) 1.24 (1.08, 1.43) 3.90 (1.78, 8.52) 1.93 (1.28, 2.90) Depression 1.13 (1.07, 1.19) 1.15 (1.08, 1.22) 1.36 (1.11, 1.67) 1.20 (1.06, 1.37) Anxiety 1.12 (1.06, 1.19) 1.11 (1.05, 1.18) 1.62 (1.21, 2.17) 1.29 (1.10, 1.52) Other psychiatric 1.11 (1.03, 1.18) 1.10 (1.03, 1.17) 2.09 (1.48, 2.95) 1.31 (1.07, 1.62) Lung 1.15 (1.10, 1.19) 1.17 (1.12, 1.22) 1.72 (1.26, 2.34) 1.11 (0.99, 1.25) Migraine 1.14 (1.09, 1.20) 1.17 (1.10, 1.24) 1.26 (0.94, 1.68) 1.15 (1.04, 1.27) Skin condition 1.15 (1.11, 1.20) 1.17 (1.12, 1.23) 1.22 (0.78, 1.92) 1.08 (0.93, 1.26) Autoimmune thyroid 1.11 (1.01, 1.21) 1.06 (0.95, 1.18) 2.99 (1.26, 7.12) 1.25 (0.91, 1.70) Miscellaneous autoimmune 1.36 (1.25, 1.48) 1.33 (1.18, 1.50) 8.06 (4.39, 14.79) 1.64 (0.89, 3.04) * Models adjusted for age, sex, treatment assignment, baseline EDSS, disease duration and number of relapses in the year prior to enrolment. Separate models were conducted for each burden of comorbidity and individual comorbidity. AE, adverse event; CVD, cardiovascular disease; EDSS, Expanded Disability Status Scale; MS, multiple sclerosis; PVD, peripheral vascular disease; SAE, serious AE.
Association of comorbidity with safety outcomes and persistence in MS clinical trial populations
Models adjusted for age, sex, treatment assignment, baseline EDSS, disease duration and number of relapses in the year prior to enrolment. Separate models were conducted for each burden of comorbidity and individual comorbidity.
AE, adverse event; CVD, cardiovascular disease; EDSS, Expanded Disability Status Scale; MS, multiple sclerosis; PVD, peripheral vascular disease; SAE, serious AE.
Association of comorbidity with infection adverse events by disease-modifying therapy class in MS clinical trial populations. Separate models were conducted for each burden of comorbidity and individual comorbidity. *Models adjusted for age, sex, treatment assignment, baseline EDSS, disease duration and number of relapses in the year prior to enrolment. EDSS, Expanded Disability Status Scale; PVD, peripheral vascular disease.
The prevalence of any AE (secondary outcome) across trials was 88.7% (95% CI 86.1% to 91.4%). The rate of any AE increased with greater comorbidity burden (1 comorbidity: 1.06 (1.04 to 1.08), 2 comorbidities: 1.07 (1.04 to 1.10), ≥3 comorbidities: 1.11 (1.07 to 1.15); table 2 . Comorbidities were associated with an elevated rate of any AE, although the magnitude was smaller, with the risk ranging between 3% and 11% among the different conditions compared with those without the comorbidity ( table 2 ).
Table 2 Association of comorbidity with any adverse event and any serious adverse event in MS clinical trial populations Pooled adjusted rate ratios (95% CI) Comorbidity * Any AE Any SAE All comorbidities 0 Reference Reference 1 1.06 (1.04, 1.08) 1.12 (1.06, 1.17) 2 1.07 (1.04, 1.10) 1.20 (1.05, 1.37) ≥3 1.11 (1.07, 1.15) 1.81 (1.45, 2.25) Cardiometabolic 0 Reference Reference 1 1.03 (1.01, 1.05) 1.33 (1.17, 1.52) ≥2 1.05 (1.03, 1.07) 1.70 (1.33, 2.17) Psychiatric 0 Reference Reference 1 1.07 (1.04, 1.09) 1.13 (1.07, 1.20) ≥2 1.07 (1.04, 1.11) 1.47 (1.18, 1.83) Hyperlipidaemia 1.04 (1.02, 1.06) 1.31 (1.11, 1.55) Hypertension 1.03 (1.01, 1.04) 1.39 (1.19, 1.61) Diabetes 1.04 (1.03, 1.06) 1.70 (1.20, 2.42) Ischaemic heart disease 1.08 (1.05, 1.10) 1.75 (1.02, 3.00) Functional CVD 1.05 (1.02, 1.08) 1.22 (1.02, 1.46) Cerebrovascular 1.11 (1.07, 1.15) 3.07 (1.51, 6.27) PVD 1.08 (1.05, 1.12) 2.55 (1.45, 4.47) Depression 1.07 (1.04, 1.09) 1.19 (1.08, 1.31) Anxiety 1.06 (1.03, 1.09) 1.28 (1.09, 1.50) Other psychiatric 1.04 (1.02, 1.07) 1.49 (1.18, 1.87) Lung 1.04 (1.03, 1.05) 1.37 (1.15, 1.64) Migraine 1.07 (1.04, 1.09) 1.14 (0.99, 1.32) Skin condition 1.08 (1.05, 1.11) 1.04 (0.95, 1.15) Autoimmune thyroid 1.04 (1.03, 1.05) 1.33 (1.00, 1.78) Miscellaneous autoimmune 1.09 (1.06, 1.13) 2.42 (1.30, 4.49) * Models adjusted for age, sex, treatment assignment, baseline EDSS, disease duration and number of relapses in the year prior to enrolment. Separate models were conducted for each burden of comorbidity and individual comorbidity. AE, adverse event; CVD, cardiovascular disease; EDSS, Expanded Disability Status Scale; MS, multiple sclerosis; PVD, peripheral vascular disease; SAE, serious AE.
Association of comorbidity with any adverse event and any serious adverse event in MS clinical trial populations
Models adjusted for age, sex, treatment assignment, baseline EDSS, disease duration and number of relapses in the year prior to enrolment. Separate models were conducted for each burden of comorbidity and individual comorbidity.
AE, adverse event; CVD, cardiovascular disease; EDSS, Expanded Disability Status Scale; MS, multiple sclerosis; PVD, peripheral vascular disease; SAE, serious AE.
Serious AEs
For SAEs of interest, the pooled prevalence across trials was 3.2% (95% CI 2.6% to 4.0%). We observed a higher rate of SAEs in those with 2 or ≥3 comorbidities compared with those with no comorbidity (1 comorbidity: 1.19 (0.96 to 1.47); 2 comorbidities: 1.54 (1.19 to 1.99); ≥3 comorbidities: 1.88 (1.42 to 2.50), table 1 ). The number of cardiometabolic and psychiatric conditions was also associated with an elevated rate of SAEs. Most comorbidities were associated with an elevated rate of SAEs except for ischaemic heart disease, migraine and skin conditions ( table 1 ). After further adjusting for BMI, results were similar ( online supplemental etable 4 ).
Similarly, for the secondary outcome of any SAE, an increased rate of SAEs was identified with comorbidity ( table 2 ). Compared with those with no comorbidity, a greater comorbidity burden was associated with an elevated rate of SAEs (1 comorbidity: 1.12 (1.06 to 1.17); 2 comorbidities: 1.20 (1.05 to 1.37); ≥3 comorbidities: 1.81 (1.45 to 2.25), table 2 ). The presence of most comorbidities was associated with an elevated rate of SAE compared with those without the condition.
Persistence
The pooled prevalence of early trial discontinuation was 17.0% (13.8% to 20.9%). The presence of two or three or more comorbidities was associated with an increased rate of early trial discontinuation from a DMT after controlling for covariates (2 comorbidities: 1.22 (1.07 to 1.42), ≥3 comorbidities: 1.19 (1.01 to 1.40), table 1 ). Psychiatric comorbidity burden was associated with higher rates of early trial discontinuation in a dose-response fashion (1 comorbidity: 1.19 (1.07 to 1.32), ≥2 comorbidities: 1.38 (1.11 to 1.72), table 1 ) and was also observed for those with depression, anxiety, other psychiatric disorders ( figure 2 ). In contrast, a greater number of cardiometabolic comorbidities was not associated with early trial discontinuation; however, a subgroup of these comorbidities was associated with greater early discontinuation including diabetes, functional CVD, cerebrovascular disease and PVD ( figure 2 , table 1 ). Those with migraine when pooled had a higher rate of early trial discontinuation (1.15 (1.04 to 1.27)). These results were consistent with small attenuations when further adjusting for BMI ( online supplemental etable 4 ). A trend for higher efficacy therapies to have higher rates of early trial discontinuation for those with greater total comorbidity burden, psychiatric comorbidity burden and individual psychiatric comorbidity ( online supplemental etable 6 ).
Association of comorbidity with early trial discontinuation. CVD, cardiovascular disease; PVD, peripheral vascular disease.
Discussion
In this large study involving 16 794 participants from 17 clinical trials, we used a meta-analytical approach to evaluate the association between comorbidity and safety of DMTs, as well as early trial discontinuation. We found that a greater number of comorbidities were associated with elevated rates of AEs, infection AEs, SAEs and early trial discontinuation after adjusting for multiple demographic and clinical factors. A higher burden of psychiatric disorders and individual psychiatric disorders were associated with an increased rate of early discontinuation, as were diabetes, functional CVD, cerebrovascular disease, PVD and migraine.
Most of the AEs of interest in this study were infection-related AEs. MS is associated with an increased risk of infections, and those infections may have clinically significant consequences. 28 , 31 In a large study using administrative data from four Canadian provinces, some DMTs were associated with a modest increase in infection-related physician visits. 1 The association of comorbidity with infection-related AEs suggests comorbidity status may need to be considered in future studies investigating infection risk. The Canadian study also found a higher hazard of acute liver injury and abnormal haematologic findings for some DMTs. 1 However, the CIs for those associations were wide due to a small number of events. In our study, we were unable to evaluate these specific events due to their low frequency among the trials. Future studies using longitudinal data with a longer follow-up and greater sample sizes are likely needed to better examine the association of these events with comorbidity.
The association of comorbidities with safety outcomes is critical to understanding the net benefit of treatment in those with comorbidity. Prior meta-analyses on safety outcomes for MS DMTs have been conducted, but those studies did not consider comorbidity status. 32 33 Information on comorbidity and AEs is limited and primarily has focused on treatment-specific AEs for relevant comorbidities. 3 , 534 Our study systematically investigated comorbidity burden and individual comorbidities for multiple safety outcomes using data pooled across 17 trials. The robust association of higher comorbidity and elevated AE rates highlights the need for reporting of comorbidities in clinical trial participants as recommended by the International Advisory Committee on Clinical Trials. 7
In contrast to safety outcomes, the association of early DMT discontinuation with comorbidity has been examined previously. Persons with comorbidities at diagnosis have been shown to have lower rates of initiating and persisting on treatment and higher rates of switching when compared with those with no comorbidity. 6 35 36 However, a cohort study of 1464 individuals with MS in the Canadian province of Nova Scotia did not find an association between general comorbidity burden and time to discontinuation of DMTs. 37 Differences in our findings may be associated with differences in study designs and a number of comorbidities considered. Our study found an increased rate of early trial discontinuation for those with greater psychiatric comorbidity burden as well as individual psychiatric disorders. The Nova Scotia cohort study identified an increased hazard of discontinuing DMT (1.22 (1.03 to 1.44)) and a Swedish MS Registry study showed reduced persistence for specific DMTs in participants with depression or those using antidepressants. 37 38 As we used comorbidity reported at enrolment, our study design should preclude a DMT from causing the psychiatric disorder. Additionally, a systematic review did not find associations between DMTs and adverse psychiatric effects. 39
Cardiometabolic comorbidity burden and individual comorbidities were associated with increased rates of AEs and SAEs, but not early trial discontinuation. Similar to our findings, the Nova Scotia cohort study found no association of hyperlipidaemia, hypertension and ischaemic heart disease with DMT discontinuation, 37 but contrary to their study findings, we did find an association between diabetes and increased rates of early discontinuation. An Italian study found an association between increasing Framingham risk score, a cardiovascular health measure that incorporates multiple factors such as smoking, age, BMI, blood pressure and diabetes, and DMT escalation. 40 Another study found comorbidity was associated with increased hazard of switching to intolerance. 6 Information regarding cardiometabolic conditions and safety outcomes and persistence is limited, but given the increasing prevalence of cardiometabolic conditions with age, the effect of these conditions on safety and persistence in the broader MS populations warrants further study and will be relevant to trials in progressive MS populations who tend to be older.
Our study has strengths and limitations. We have systematically studied a broad range of comorbidities and DMTs in both relapsing and progressive MS populations. Our analysis included robust adjustment for sex, age, disability and other clinical factors in 17 clinical trials. Clinical trial data are less subject to surveillance bias compared with other observational study designs and ensure that comorbidity exposure preceded the introduction of DMT. However, the relatively short duration of DMT clinical trials limits our ability to examine longer-term safety concerns, such as cancer and hepatic dysfunction. These trials were not powered to detect these outcomes, and a low prevalence of some comorbidities may also reduce the ability to detect associations. Clinical trial populations are highly selected with respect to comorbidity, and while we identified participants with comorbidities, the severity of these comorbidities is unknown. We may be underestimating the association of safety and tolerability if only milder comorbidity is included. Moreover, the effects may be underestimated for the broader MS population where comorbidity burden and severity are likely greater. There is a need for studies evaluating these outcomes in real world MS populations.
In summary, comorbidity plays an important role in the safety and persistence of DMTs in persons with MS. The increased rate of safety outcomes in those with comorbidities can inform clinical care decisions for those with comorbidity, especially given the more severe AEs observed with the introduction of newer therapies. Studies with longer follow-up are still needed to improve our understanding of long-term safety outcomes. We showed greater early trial discontinuation, indicating lower persistence on therapy, for those with more than two comorbidities. This suggests there is a need to understand the mechanisms behind early discontinuations of DMT and provide greater support to patients with comorbidities.
Supplementary material
Acknowledgements
Amber Salter is supported (in part) by a Biostatistics/Informatics Junior Faculty Award (BI-2105-37656) from the National Multiple Sclerosis Society. This publication is based on research using data from Biogen provided by Biogen, Cambridge, USA. This publication is based on research using data provided by Merck KGaA, (CrossRef Funder ID: 10.13039/100009945). This publication is based on research using data from Novartis Pharma AG and Sanofi-Aventis Groupe that has been made available through www.ClinicalStudyDataRequest.com . This publication is based on research using data from Roche that has been made available through Vivli. Vivli has not contributed to or approved and is not in any way responsible for, the contents of this publication. CombiRx ( NCT00211887 ) was funded by the NIH, National Institute of Neurologic Disorders and Stroke (phase III study: UO1NS045719, Planning Grant R21NS41986) and the study agents and matched placebo were kindly provided by their manufacturers, Biogen and Teva Pharmaceutical Sponsors (Merck (CrossRef Funder ID: 10.13039/100009945), Biogen, Novartis Pharma AG, Sanofi-Aventis Groupe and Roche) reviewed this manuscript for medical accuracy only before journal submission.
The authors are fully responsible for the content of this manuscript, and the views and opinions described in the publication reflect solely those of the authors.
Footnotes
Funding: This work is supported by the Consortium of Multiple Sclerosis Centers Pilot award. Partial support was provided by the Department of Defense through the Multiple Sclerosis Research Program under Award No. W81XWH-21-1-0721. Opinions, interpretations conclusions and recommendations are those of the author and are not necessarily endorsed by the Department of Defense.
Provenance and peer review: Not commissioned; externally peer reviewed.
Patient consent for publication: Not applicable.
Ethics approval: The UT Southwestern institutional review board designated this study as non-human subjects’ research. Trial sponsors obtained study consents and reviewed data access requests to ensure they were consistent with consents obtained for the trials.
Data availability statement
Data may be obtained from a third party and are not publicly available.
References
Associated Data
Supplementary Materials
Data Availability Statement
Investigators may request access to anonymised individual patient data and redacted trial documents from each sponsor based on their individual processes. Before data use, proposals must be approved by the sponsor or an independent review panel, as appropriate, and a data sharing agreement obtained. The investigators will provide the analysis code for replication of results on request.
Data may be obtained from a third party and are not publicly available.