Heart failure associated with ustekinumab therapy for treatment of Crohn’s Disease

Abstract

A man in his 60s with penetrating ileocolonic Crohn’s disease (CD), recently started on ustekinumab therapy, presented with new onset dyspnoea, paroxysmal nocturnal dyspnoea and dependent oedema. He was diagnosed with heart failure (HF) 10 months after starting ustekinumab therapy. His symptoms resolved with discontinuation of ustekinumab and he had recovery of his cardiac function. Though initial studies that led to the U.S Food and Drug Administration (FDA)approval for ustekinumab did not detect a signal for HF, postmarketing surveillance has detected rare cases of HF after initiation of the medication. This is one of the few reported cases of HF associated with ustekinumab therapy for CD.

Background

Ustekinumab is a monoclonal antibody to the p40 subunit of interleukin (IL)-12 and IL-23,¹ FDA approved for the treatment of Crohn’s disease (CD). Initial studies that led to FDA approval did not detect a signal for heart failure (HF). We present a rare case of HF after initiation of ustekinumab for CD with subsequent recovery of cardiac function after discontinuation of therapy and appropriate guideline-directed therapy for HF.

Case presentation

A man in his 60s with a history of Barrett’s oesophagus, stage 4 chronic kidney disease, recurrent deep vein thrombosis, chronic neuropathy and spondyloarthritis, presented with an enterocutaneous fistula and was diagnosed with penetrating ileocolonic CD. Medications at the time included pantoprazole, warfarin, duloxetine, pregabalin and nortriptyline.

Treatment was initiated with corticosteroids and anti-tumour necrosis factor (anti-TNF) adalimumab in combination with weight-based azathioprine. The patient achieved clinical remission on therapy but endoscopic evaluation showed continued inflammation and ulcerations in the colon. An adalimumab panel showed a trough of 12 µg/mL and no anti-adalimumab antibodies. Additional testing showed undetectable thiopurine metabolites, suggesting medication non-adherence. Ultimately, it was felt that he had a secondary loss of response and needed to transition to an alternate class of therapy. Ustekinumab was selected over the gut-selective vedolizumab given his comorbid spondyloarthritis. He remained in clinical remission on ustekinumab but developed dyspnoea on exertion, orthopnoea and dependent oedema 10 months into therapy (figure 1). A metabolic panel showed no electrolyte derangements. Baseline ECG showed no arrhythmias. Transthoracic echocardiography showed severe diastolic dysfunction with an ejection fraction (EF) of 26%. Cardiac catheterisation showed no coronary artery disease, most consistent with non-ischaemic dilated cardiomyopathy (NIDCM). He was diagnosed with HF with reduced EF secondary to NIDCM. He had no history of infectious symptoms or substance abuse as a potential aetiology and thyroid function was normal. Imaging to evaluate for infiltrative disorder was limited by chronic renal disease. Given the time course of his development of symptoms, ustekinumab was suspected as a possible aetiology. In addition to initiation of metoprolol, furosemide and losartan, the decision was made to transition to vedolizumab for treatment of his CD. Repeat echocardiography 6 months later showed a recovered EF of 45%.

Figure 1.

Timeline of events from diagnosis of Crohn’s disease to the development of heart failure with reduced ejection fraction (HFrEF).

Discussion

Crohn’s disease is an inflammatory condition that can affect any portion of the gastrointestinal tract with clinical presentations ranging from isolated intestinal symptoms to systemic, extraintestinal manifestations. The pathophysiology is not fully understood but is thought to be related to the upregulation of proinflammatory cytokines.^{2 3} Several modalities exist to target different parts of the inflammatory cascade in the treatment of CD, but they come with potential side effects and must be tailored to individual patients based on disease characteristics and patient demographics. Our patient was initially treated with steroids and a combination of anti-TNF/thiopurine therapy, but subsequently developed pharmacodynamic failure and required switching to an alternate class of therapy. Ustekinumab was selected given our patient’s comorbid spondyloarthritis, which vedolizumab may have been less effective in managing.

Evaluation for other causes for HF in this patient was unrevealing. His thyroid function was normal. Given his cardiac recovery, a chronic infiltrative disorder was unlikely. The patient was not using alcohol or illicit substances to suggest substance-induced HF and had no history of cardiovascular disease. While viral causes of cardiomyopathy were considered, retrospective data suggest that 25% of patients with active myocarditis have elevated white blood cell count (WBC), 65% have elevated C reactive protein (CRP) and ECG abnormalities are common.⁴ Our patient did not have any symptoms of a previous infection, WBC and CRP were normal, and ECG showed no abnormalities. Because clinical suspicion for viral aetiology was low and confirmatory testing with endomyocardial biopsy is invasive and not always diagnostic,⁵ further testing with serology, MRI or biopsy was not obtained. Anti-TNF therapy has been associated with HF but the timing of the onset of his HF after stopping adalimumab made this less likely as well.^{6 7} While pregabalin has been linked to peripheral oedema and case reports do describe a possible association with HF, this is rare.⁸ Our patient’s pregabalin was continued and his EF improved, making pregabalin a less likely cause. In a case report, duloxetine therapy has been noted to worsen pre-existing HF but is not causative.⁹ Ultimately, the timing between the initiation of ustekinumab and the onset of HF was felt to be correlated. The patient’s recovery of cardiac function after ustekinumab was discontinued supports this correlation, though concurrent guideline-directed therapy may have aided his recovery as well. Our suspicion for ustekinumab causing HF is also supported by a Naranjo algorithm score of 4, which suggests that this is possible.¹⁰

Dilated cardiomyopathy is characterised by dilation and impaired contraction of one or both ventricles, defined as EF less than 40%.¹¹ Those with impaired systolic function may develop HF as a result. HF is associated with an upregulation of cytokines such as TNF-α, IL-6 and IL-1β.^{12 13} The prolonged upregulation of these proinflammatory cytokines is linked to maladaptive responses in cardiac myocytes including hypertrophy, contractile dysfunction and remodelling, leading to the development of chronic HF.¹⁴ While not proven, IL therapy such as ustekinumab has the potential to affect this remodelling. Ten months into therapy, our patient developed HF with subsequent recovery of cardiac function after discontinuation of ustekinumab and appropriate guideline-directed therapy for HF for 6 months. While his EF recovery can be explained by treatment with guideline-directed therapy, a study showed that only 9% of patients with non-ischaemic HF on guideline-directed therapy recovered with EF>40% at 1 year.¹⁵

The incidence of HF is not observed in clinical trials for ustekinumab because a history of HF is often specified as an exclusion criterion. During phase III randomised control trials for CD and ulcerative colitis, UNITI-1, IM-UNITI and UNIFI over 1600 patients were evaluated but no HF incidents were attributed to the medication.^{16 17} Analysis of 11 cases of HF in 4458 patients treated with ustekinumab in clinical trials concluded that these events were unrelated to ustekinumab.¹⁸ A meta-analysis of psoriatic arthritis patients receiving IL-12/23 therapy concluded that compared with placebo, there was an increased rate of major adverse cardiac events (MACEs). However, ustekinumab independently was not associated with MACEs, which was categorised and driven by briakinumab, which did not receive FDA approval.¹⁹ There was a case report which presented a patient with psoriasis who developed HF after ustekinumab therapy. Though the case did not specify the timing of HF after starting ustekinumab, similar to our case, the patient recovered after discontinuation of the drug and administration of guideline-directed therapy.²⁰ A case-time-control study concluded that ustekinumab is associated with increased risks of severe cardiovascular events (SCEs), defined as an acute coronary syndrome, ischaemic stroke and cardiovascular deaths.²¹

Anti-IL-23 agents are widely used to treat various autoimmune conditions like CD, ulcerative colitis, psoriasis and ankylosing spondylitis. Compared with anti-IL-23-p19 agents, anti-IL12/23-p40 agents are associated with more adverse effects and more frequently lead to cardiovascular events.²² Although the relationship between anti-IL-12/23-p40 agents and SCEs remains unclear, this rare case of HF after initiation of ustekinumab serves as an important data point in determining that relationship; clinicians should consider this in the differential while this relationship continues to be explored.

Learning points.

• Ustekinumab is a monoclonal antibody to the p40 subunit of interleukin (IL)-12 and IL-23, FDA approved for the treatment of Crohn’s disease.

• Initial studies that led to FDA approval of ustekinumab did not detect a signal for heart failure (HF). This is a rare case of HF associated with ustekinumab.

• The relationship between anti-IL-12/23 p40 agents and severe cardiac events such as HF is infrequently described in the literature, this rare case of HF after initiation of ustekinumab serves as another important data point in determining that relationship.

Ethics statements

Patient consent for publication

Consent obtained directly from patient(s).