Association of Rituximab With Risk of Long-term Cardiovascular and Metabolic Outcomes in Patients With Pemphigus

Key Points

Question

What is the risk of long-term cardiovascular and metabolic outcomes and all-cause mortality in patients with pemphigus managed by rituximab vs azathioprine/mycophenolate mofetil?

Findings

In this cohort study of 1922 patients with pemphigus, compared with conventional first-line corticosteroid-sparing adjuvants (azathioprine and mycophenolate mofetil), treatment with rituximab was associated with a lower risk of developing cardiovascular and metabolic comorbidities.

Meaning

The study results suggest that rituximab might be particularly preferred in individuals with cardiovascular and metabolic risk factors, for whom corticosteroid-related adverse events must be strictly avoided.

Abstract

Importance

The association of different therapeutic approaches with long-term cardiovascular and metabolic outcomes in patients with pemphigus remains to be precisely evaluated.

Objective

To assess the risk of long-term cardiovascular and metabolic outcomes and all-cause mortality in patients with pemphigus managed by rituximab compared with those receiving treatment with first-line corticosteroid-sparing agents (azathioprine and mycophenolate mofetil [MMF]).

Design, Setting, and Participants

A global population–based retrospective cohort study compared 961 patients with pemphigus that was managed with rituximab with those treated with azathioprine or MMF (n = 961) regarding the risk of several cardiovascular and metabolic outcomes. Propensity score matching was performed to optimize comparability. Patients were enrolled from the Global Collaborative Network of TriNetX platform.

Main Outcomes and Measures

Risk of myocardial infarction, stroke, peripheral vascular disease, pulmonary embolism, hypertension, hyperlipidemia, type 2 diabetes, obesity, osteoporosis, and avascular bone necrosis.

Results

Of 1602 participants, 855 (53.4%) were women and 747 (46.6%) were men; the mean (SD) age was 54.8 (16.6) years for those treated with rituximab and 54.4 (18.2) years for those treated with azathioprine or MMF. Compared with those treated by azathioprine/MMF, patients treated with rituximab experienced a lower risk of myocardial infarction (relative risk [RR], 0.45; 95% CI, 0.24-0.86; P = .01), stroke (RR, 0.42; 95% CI, 0.26-0.69; P < .001), peripheral vascular disease (RR, 0.47; 95% CI, 0.28-0.79; P = .003), hypertension (RR, 0.48; 95% CI, 0.38-0.63; P < .001), hyperlipidemia (RR, 0.45; 95% CI, 0.32-0.64; P < .001), type 2 diabetes (RR, 0.63; 95% CI, 0.51-0.77; P < .001), obesity (RR, 0.49; 95% CI, 0.34-0.72; P < .001), and osteoporosis (RR, 0.46; 95% CI, 0.30-0.71; P < .001). The all-cause mortality was comparable between patients in both groups (hazard ratio, 0.94; 95% CI, 0.62-1.43; log-rank P = .77).

Conclusions and Relevance

The results of this cohort study suggest that rituximab was associated with protection against long-term cardiovascular and metabolic outcomes compared with conventional immunosuppressants. This agent might be particularly preferred in individuals with preexisting cardiovascular and metabolic risk factors.

This cohort study examines the risk of long-term cardiovascular and metabolic outcomes and all-cause mortality in patients with pemphigus managed by rituximab compared with those receiving treatment with azathioprine/mycophenolate mofetil.

Introduction

Pemphigus is a potentially life-threatening autoimmune bullous disease that manifests with erosions and blisters in the skin and mucosal surfaces.¹ Pathogenically, it is mediated by pathogenic autoantibodies that target desmoglein 1 and desmoglein 3 adhesion molecules of the epidermis that are responsible for the cohesion between keratinocytes in the skin and mucosa.^2,3,4,5 Despite the substantial drop in the 1-year mortality rates of patients with pemphigus who receive proper treatment, they remain more susceptible to death compared with the general population.^6,7,8,9

Systemic corticosteroids embodied the cornerstone of management of pemphigus for decades. All recent guidelines and consensus statements still define systemic corticosteroids as first-line therapy for pemphigus owing to their role in remission induction and consolidation.^{10,11,12,13,14} However, the main challenge in treating pemphigus is to maintain long-term remission with the smallest dose of systemic corticosteroids for the shortest time, aiming to mitigate their life-threatening adverse events.^10,12,15 Introducing adjuvant agents provided a favorable corticosteroid-sparing effect and contributed to the reduction of the cumulative dose of systemic corticosteroids.^10,11,12,15 However, evidence attributing superiority to adjuvants vs corticosteroid monotherapy emerged only recently for rituximab.¹⁶ Very limited evidence supports the use of conventional immunosuppressants as first-line therapy for pemphigus.^17,18

Rituximab is a monoclonal chimeric mouse/human antibody directed against CD20-expressing B lymphocytes.¹⁹ Rituximab recently gained widespread acceptance as a first-line therapy for moderate and severe pemphigus following an open-label clinical trial that confirmed its superiority (in conjunction with low-dose and short-term corticosteroids) vs systemic corticosteroids monotherapy and mycophenolate mofetil (MMF), both in efficacy and safety measures.^16,20

Azathioprine is a prodrug that converts nonenzymatically to 6-mercaptopurine after oral administration. This drug antagonizes purine metabolism and blocks the synthesis of DNA, RNA, and proteins.^21,22 Mycophenolate mofetil is a drug that impairs the immune response via selectively inhibiting inosine monophosphate dehydrogenase, which is associated with inhibition of the de novo pathway of purine synthesis in T and B lymphocytes.^15,23 Azathioprine and MMF were proved efficient as corticosteroid-sparing agents^24,25,26 and are considered as reliable adjuvant/corticosteroid-sparing agents alongside systemic corticosteroid therapy in moderate to severe pemphigus vulgaris (PV).^10,11,14

A recent randomized clinical trial (RCT) assigned 67 and 68 patients with PV to receive rituximab and MMF, respectively, in addition to an oral corticosteroid administered on the same tapering schedule.²⁰ Rituximab was superior to MMF in producing sustained complete remission and achieving a lower cumulative dose of corticosteroids, but more patients receiving treatment with rituximab experienced severe adverse events.²⁰ While this study illuminated the safety profile of these drugs, our knowledge about the long-term complications of rituximab, MMF, and azathioprine remains sparse. The aim of the current study is to evaluate the long-term cardiovascular and metabolic comorbidities and mortality rates of patients with pemphigus treated with rituximab compared with those treated with azathioprine or MMF.

Methods

Study Design and Database

We performed a population-based, retrospective cohort study that relied on the computerized database of TriNetX. TriNetX is a global federated health research network that enables access to electronic medical records from approximately 117.5 million patients from 86 health care organizations (HCOs) worldwide. For the current study, we investigated a set of HCOs grouped into a network called the Global Collaborative Network, which includes 86 HCOs. TriNetX, LLC is compliant with the US Health Insurance Portability and Accountability Act, which protects the privacy and security of health care data, and any additional data privacy regulations applicable to the contributing HCO. Because this study used only deidentified patient records and did not involve the collection, use, or transmittal of individually identifiable data, this study was exempted from institutional review board approval and granted a waiver of informed consent.

Study Population and Definition of Eligible Patient

All 117.5 million individuals with health insurance in the Global Collaborative Network were screened for the presence of a diagnostic code that was compatible with pemphigus. To investigate the role exerted by different therapeutic approaches on the cardiovascular and metabolic outcomes of patients with pemphigus, 2 treatment groups were allocated. The first group comprised patients with pemphigus who were exposed at least once to rituximab (two 1-g intravenous infusions separated by 2 weeks). Before inclusion in this group, we ascertained that all patients had no preexisting or subsequent treatment with azathioprine, MMF, mycophenolic acid, or cyclophosphamide. The second group comprised patients with pemphigus that was managed with azathioprine or MMF. To be eligible for inclusion in the second group, patients had to be naive to treatment with rituximab or cyclophosphamide. The eTable in the Supplement further details as to the definition of the study populations.

Eligible patients in both groups underwent propensity score matching based on demographic variables (age, sex, and race), body mass index at baseline, and the following confounding comorbidities (before initiation of the investigated drug treatments): congestive heart failure, ischemic heart disease, atrial fibrillation, personal history of cardiac arrest, smoking, chronic kidney disease, respiratory disorders, family history of ischemic heart disease and other diseases of the circulatory system, family history of diabetes, family history of stroke, and family history of osteoporosis. Moreover, the propensity score matching included baseline levels of hemoglobin A_1C, triglycerides, low-density lipoprotein cholesterol, and high-density lipoprotein cholesterol. The propensity score matching ensured a uniform distribution of confounders between groups, thus improving between-group comparability.²⁷ Participants in both groups were followed up longitudinally to assess the risk of the following outcomes: myocardial infarction, stroke, peripheral vascular disease, pulmonary embolism, hypertension, hyperlipidemia, type 2 diabetes, obesity, osteoporosis, and avascular bone necrosis. The eTable in the Supplement describes the definition of each of the study outcome measures. The index date on which the follow-up after study participants began was the date in which treatment with rituximab or azathioprine/MMF (first of which) was initiated.

Statistical Analysis

Baseline characteristics were described by means and standard deviations for continuous variables and numbers and percentages for dichotomous variables. Continuous variables were compared using the Student t test and dichotomous variables by Pearson χ² test.

Relative risk (RR), also known as risk ratio, is a proportion representing the risk of developing an outcome in patients who were receiving treatment with rituximab divided by the risk of developing the same outcome in patients receiving treatment with azathioprine/MMF during the study period. Survival analyses were conducted by the Kaplan-Meier method. A log-rank test was run to determine if there were differences in the survival distribution for patients in the 2 investigated groups. Hazard ratios (HRs) for the risk of death were obtained by the use of the Cox regression model. Nelson-Aalen plots were used to test the proportional hazards assumption. A 2-tailed P value of less than .05 was considered statistically significant.

Results

The current study encompassed 1602 patients with pemphigus, of whom 801 were treated with rituximab and 801 with azathioprine or MMF. The baseline demographic characteristics of study participants are described in Table 1.

Table 1. Baseline Characteristics of Study Participants.

Characteristic	No. (%)		P value
	Rituximab (n = 801)	Azathioprine or MMF (n = 801)
Age at the initiation of the drug treatment, mean (SD), y	54.8 (16.6)	54.4 (18.2)	.62
Sex
Female	418 (52.2)	437 (54.6)	.34
Male	383 (47.8)	364 (45.4)	.34
Race
Racial and ethnic minority groups	382 (47.7)	393 (49.1)	.58
White	419 (52.3)	408 (50.9)
BMI, mean (SD)a	28.8 (6.8)	28.5 (7.7)	.60
Baseline laboratory values
Hemoglobin A1C, %b	6.1 (1.4)	6.1 (1.2)	.77
Triglyceride, mg/dLc	120 (65.7)	133 (74.5)	.13
Low-density lipoprotein cholesterol, mg/dLd	99.0 (36.5)	96.0 (40.2)	.53
High-density lipoprotein cholesterol, mg/dLe	54.1 (19.4)	51.5 (20.2)	.28
Comorbidities
Congestive heart failure	34 (4.2)	29 (3.6)	.52
Ischemic heart disease	52 (6.5)	43 (4.1)	.03
Atrial fibrillation	27 (3.4)	22 (2.7)	.47
Personal history of cardiac arrest	10 (1.3)	0	.02
Chronic kidney disease	37 (4.6)	35 (4.4)	.81
Other respiratory disorders	54 (6.7)	43 (5.4)	.25
Smoking	47 (5.9)	34 (4.2)	.14
Major depression disorder	60 (7.5)	59 (7.4)	.92
Family history of ischemic heart disease and other diseases of the circulatory system	34 (4.3)	31 (3.9)	.70
Family history of diabetes	20 (2.5)	18 (2.2)	.74
Family history of stroke	10 (1.3)	10 (1.3)	>.99
Family history of osteoporosis	0	0	>.99

Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); MMF, mycophenolate mofetil.

SI conversion factor: to convert hemoglobin A_1c to the proportion of total hemoglobin, multiply by 0.01; for high-density and low-density lipoprotein cholesterol to mmol/L, multiply by 0.0259; for triglycerides to mmol/L, multiply by 0.0113.

^a This variable was available for 287 and 265 patients receiving treatment with rituximab and azathioprine/MMF, respectively.

^b This variable was available for 129 and 111 patients receiving treatment with rituximab and azathioprine/MMF, respectively.

^c This variable was available for 154 and 135 patients receiving treatment with rituximab and azathioprine/MMF, respectively.

^d This variable was available for 142 and 121 patients receiving treatment with rituximab and azathioprine/MMF, respectively.

^e This variable was available for 144 and 126 patients receiving treatment with rituximab and azathioprine/MMF, respectively.

Table 2 demonstrates the risk of different cardiovascular, cerebrovascular, and metabolic outcomes among patients with pemphigus that was managed with rituximab compared with those treated with azathioprine or MMF. Patients treated with rituximab were found to experience a lower risk of myocardial infarction (RR, 0.45; 95% CI, 0.24-0.86; P = .01), stroke (RR, 0.42; 95% CI, 0.26-0.69; P < .001), and peripheral vascular disease (RR, 0.47; 95% CI, 0.28-0.79; P = .003). However, the risk of pulmonary embolism was comparable between the 2 investigated groups (RR, 0.63; 95% CI, 0.33-1.19; P = .15).

Table 2. Risk of Cardiovascular, Cerebrovascular, and Metabolic Outcomes Among Patients With Pemphigus Treated With Rituximab Compared With Those Treated With Azathioprine or MMF.

Disease	Rituximab			Azathioprine or MMF			Risk difference, % (95% CI)	Risk ratio (95% CI)	P value
	No. of eligible participantsa	No. of outcomes	Risk, %	No. of eligible participantsa	No. of outcomes	Risk, %
Cardiovascular outcomes
Myocardial infarction	783	13	1.7	785	29	3.7	−2.0 (−3.6 to −0.4)	0.45 (0.24 0.86)	.01
Stroke	777	21	2.7	779	50	6.4	−3.7 (−6.3 to −1.2)	0.42(0.26 to 0.69)	<.001
Peripheral vascular disease	783	20	2.6	786	43	5.5	−2.9 (−4.9 to −1.0)	0.47 (0.28 to 0.79)	.003
Pulmonary embolism	787	15	1.9	792	24	3.0	−1.1 (−2.7 to 0.4)	0.63 (0.33 to 1.19)	.15
Metabolic outcomes
Hypertension	596	72	12.1	613	153	25.0	−13.0 (−17.0 to −9.0)	0.48 (0.38 to 0.63)	<.001
Hyperlipidemia	699	43	6.2	704	96	13.6	−7.5 (−10.6 to −4.4)	0.45 (0.32 to 0.64)	<.001
Type 2 diabetes	801	121	15.1	801	192	24.0	−8.9 (−12.7 to −5.0)	0.63 (0.51 to 0.77)	<.001
Obesity	748	37	4.9	752	76	10.1	−5.2 (−7.8 to −2.5)	0.49 (0.34 to 0.72)	<.001
Osteoporosis	767	29	3.8	769	63	8.2	−4.4 (−6.8 to −2.0)	0.46 (0.30 to 0.71)	<.001
Avascular bone necrosis	796	10	1.3	797	10	1.3	0.0 (−1.1 to 1.1)	1.00 (0.42 to 2.39)	>.99

Abbreviation: MMF, mycophenolate mofetil.

^a Patients who had the investigated outcome before the initiation of the drug treatment were excluded from the analysis.

We then compared the risk of metabolic outcomes in the 2 different study groups. Patients prescribed rituximab were less susceptible to develop hypertension (RR, 0.48; 95% CI, 0.38-0.63; P < .001), hyperlipidemia (RR, 0.45; 95% CI, 0.32-0.64; P < .001), type 2 diabetes (RR, 0.63; 95% CI, 0.51-0.77; P < .001), obesity (RR, 0.49; 95% CI, 0.34-0.72; P < .001), and osteoporosis (RR, 0.46; 95% CI, 0.30-0.71; P < .001). The risk of avascular necrosis of the bone was not statistically different between the groups (RR, 1.00; 95% CI, 0.42-2.39; P > .99).

The last end point of the current study was to evaluate the all-cause mortality among the 2 investigated groups (Figure). Overall, 37 deaths occurred in the rituximab group compared with 60 in the azathioprine/MMF group. The risk of mortality was statistically similar (HR, 0.94; 95% CI, 0.62-1.43; log-rank P = .77).

Figure. Kaplan-Meier Survival Curves of Survival of Patients With Pemphigus That Was Managed With Rituximab Compared With Those Treated With Azathioprine/Mycophenolate Mofetil (MMF).

The log-rank test was insignificant (hazard ratio, 0.94; 95% CI, 0.62-1.43; P = .77), signifying that survival was comparable between these 2 groups.

Discussion

The current global population–based cohort study illuminated the risk of cardiovascular and metabolic outcomes of patients with pemphigus that was managed with rituximab compared with those treated with azathioprine/MMF. Patients treated with rituximab were significantly less susceptible to develop most of the investigated outcomes, including myocardial infarction, stroke, peripheral vascular disease, hypertension, hyperlipidemia, type 2 diabetes, obesity, and osteoporosis. However, the all-cause mortality rate was comparable between the 2 groups of interest.

Before the advent of systemic corticosteroids, pemphigus was lethal within the initial 2 years following presentation.¹² In the subsequent decades, rapid tapering of corticosteroids after consolidation due to the introduction of novel adjuvant immunosuppressants was associated with a further decrease in morbidity and mortality.^1,6 One of the main unmet needs treating pemphigus is to achieve and maintain long-term remission with the lowest cumulative dose of systemic corticosteroids. Owing to the poor safety profile of systemic corticosteroids, it is advantageous to minimize exposure to these drugs by using adjuvant drugs that exert a corticosteroid-sparing effect and enable a tapering of corticosteroids more rapidly.

Azathioprine and MMF were found to have a steroid-sparing role in several RCTs that explored patients with pemphigus.^24,26,28 Although they were not superior to systemic corticosteroids plus placebo in many efficacy outcomes, MMF and azathioprine were associated with a lower cumulative dose of systemic corticosteroids.^24,26,28 In a landmark clinical trial, rituximab with low-dose and short-term prednisone was superior to high-dose and long-term prednisone monotherapy, both in efficacy and safety outcomes.¹⁶ Taken together, rituximab, MMF, and azathioprine are associated with a decrease in the reliance on systemic corticosteroids and as well as reduced cardiovascular and metabolic adverse events.

However, to our knowledge, it remains to be established how these therapeutic options are associated with the long-term risk of developing subsequent cardiovascular and metabolic comorbidities. In a recent RCT comparing 67 patients with PV who were receiving treatment with rituximab and 68 patients receiving treatment with MMF for 52 weeks, the latter group experienced less frequent serious adverse events (15% vs 22%). When stratifying adverse events, grade 3 or higher corticosteroid-related adverse events were more frequently encountered among those treated with MMF (7% vs 1%). Both groups were administered oral corticosteroids on the same tapering schedule. Although novel, prospective, and valid in delineating the safety profile of MMF vs rituximab, this study was hampered by a relatively low sample size and short length of follow-up. The sparsity of the current literature necessitates large-scale real-world studies that follow patients with pemphigus for long durations to elucidate the long-term cardiovascular and metabolic complications of different therapeutic regimens.

The comparison of rituximab with first-line corticosteroid-sparing agents (azathioprine/MMF) yielded findings in favor of rituximab. Rituximab was associated with significant protection against the development of several long-term cardiovascular (myocardial infarction, stroke, and peripheral vascular disease) and metabolic outcomes (hypertension, hyperlipidemia, type 2 diabetes, obesity, and osteoporosis) in pemphigus. While we were not able to directly assess the cumulative dose of systemic corticosteroids in each group, it is tempting to postulate that rituximab enabled the administration of a lower cumulative dose of systemic corticosteroids, in accordance with the findings of the landmark practice-changing trials.^16,20 This might account for the lower risk of corticosteroid-related cardiovascular and metabolic phenomena.^29,30

Another interpretation of our findings might lie in the emerging atherogenic role of B2 cells in promoting atherosclerosis and cardiovascular diseases.^31,32 Congruently, rituximab, a B cell–depleting agent, proved protective against atherosclerotic cardiovascular disease among patients after kidney transplant.³³ Hsue et al³⁴ attested that rituximab was associated with improved endothelial function and reduced inflammation in patients with rheumatoid arthritis. These observations support the assumption that by destroying atherogenic B2 cells (that embody a major population of the adult human B-cell repertoire), rituximab plays an atheroprotective role. Further research is necessary to better delineate the effect of B cell subsets in atherogenesis and eliciting cardiovascular outcomes.

Despite the significant differences in most investigated outcomes, the survival of patients in the 2 treatment groups was not statistically different. Correspondingly, the RCT that compared patients who were receiving treatment with rituximab with those receiving MMF reported no significant intergroup differences in mortality.²⁰ It is possible that this study was underpowered to detect significant differences.

The current study uses a large-scale population-based study to illuminate a clinically relevant topic. The large sample size provides adequate power to exclude chance as the basis for the findings. The comprehensive and global nature of TrinetX argues against the presence of meaningful selection bias. This study evaluates the risk of several cardiovascular and metabolic outcomes within 2 treatment approaches for pemphigus. This potentially provides clinicians with a wide perspective about the long-term safety of rituximab compared with conventional immunosuppressants.

Limitations

This study had some limitations. Because the definition of pemphigus relied on International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) codes, there was no firm guarantee for the immunopathological validation of eligible patients. However, misclassification is highly improbable, because the likelihood of prescribing these immune-modulating/immunosuppressive agents in uncertain cases is negligible. Second, the retrospective data collection embodies another methodological limitation. Third, there is a lack of clinical data about the immunological and clinical features of the disease, which interfered with investigating the efficacy of treatment. Moreover, this study was not able to adjust for multiple treatment courses of rituximab. The association of secular trends in cardiovascular trends with the study findings could not be evaluated. We were unable to quantify the cumulative exposure to systemic corticosteroids, thus leaving the possibility that corticosteroids potentially mediate the observed association unrefuted. Although the use of propensity score matching reduces the likelihood of meaningful differences between groups, there is still a potential for residual unmeasured confounding. Due to insufficient statistical power, we were unable to conduct subgroup analyses that examined MMF and azathioprine individually.

Conclusions

The results of this cohort study suggest that patients with pemphigus that was managed with rituximab were less susceptible to develop myocardial infarction, stroke, peripheral vascular disease, hypertension, hyperlipidemia, type 2 diabetes, obesity, and osteoporosis compared with those treated with azathioprine/MMF. All-cause mortality, pulmonary embolism, and avascular necrosis of the bone were not statistically different between the 2 groups. In accordance with a large body of evidence suggesting a favorable short-term safety profile, the current study suggested that rituximab features a good long-term safety profile. Clinicians treating patients with pemphigus should be aware of the study findings. Rituximab might be considered positively in individuals with cardiovascular and metabolic risk factors, for whom corticosteroid-related adverse events must be avoided.

Supplement.

eTable. Definition of study participants