Practical Considerations for the Treatment of Patients with Psoriatic Arthritis and Comorbid Inflammatory Bowel Disease

INTRODUCTION

Psoriatic arthritis is a subtype of spondyloarthritis which may feature several overlapping immune-mediated disease manifestations including peripheral and/or axial arthritis, cutaneous and/or nail psoriasis, enthesitis, dactylitis, uveitis and inflammatory bowel disease (IBD). Subclinical gut inflammation is found in up to 61% of patients with spondyloarthritis¹ and ongoing research efforts are focused on the potential role of underlying intestinal inflammation in the pathogenesis of inflammatory arthritis in this population. Patients with psoriasis or PsA have a two-fold increased risk of developing IBD compared to the general population^2-4 with an overall IBD prevalence of 2-4% in patients with PsA⁵. While the incidence of comorbid IBD, psoriasis, and inflammatory arthritis may occur in any order in a given patient with PsA, recent mendelian randomization studies suggest that IBD, particularly the Crohn’s subtype, may be causally associated with the development of psoriasis and PsA, but not vice versa⁶.

There are several key considerations when approaching the treatment of a patient with PsA and comorbid IBD. While many drugs have overlapping disease indications, their effectiveness often differs for inflammatory arthritis, psoriasis, IBD and other comorbid disease manifestations. Clinicians should be further aware that drug induction and maintenance dosing may differ depending on the disease indication (Table 1). Additionally, certain medications should be prescribed with caution or altogether avoided in patients with IBD. Ultimately, close interdisciplinary management between the treating rheumatologist and gastroenterologist is key to successful holistic treatment.

Table 1.

Comparison of Induction and Maintenance Dosing for Medications Used Across PsA, Psoriasis, and IBD

	Psoriasis		Psoriatic Arthritis		Inflammatory Bowel Disease
Drug	Induction	Maintenance	Induction	Maintenance	Induction	Maintenance
infliximab	5-10mg/kg IV wks 0,2,6	5-10mg/kg IV q4-8 wks	5-10mg/kg IV wks 0,2,6	5mg/kg IV q8 wks	5-10mg/kg IV wks 0,2,6	5-10mg/kg IV q4-8 wks
adalimumab	80mg SQ single dose (wk 0)	40mg SQ q2 wks (starting wk 1)	N/A	40mg SQ q2 wks	160mg SQ wk 0, 80mg wk 2	40-80mg SQ q1-2 wks
certolizumab	400mg SQ q2 wks	400mg q2 wks SQ; if <90 kg consider 200mg dose	400mg SQ wks 0,2,4	200mg SQ q2 wks	400mg SQ wks 0,2,4	400mg SQ q4 wks
golimumab	N/A	N/A	IV: 2mg/kg wks 0,4 SQ: 50mg qmonth	IV: 2mg/kg q8 wks SQ: 50mg qmonth	200mg SQ wk 0, 100mg wk 2	100mg SQ q4 wks
ustekinumab	<100 kg: 45mg SQ wks 0,4 >100 kg: 90mg SQ wks 0,4	<100 kg: 45mg SQ q12 wks; may require 90mg or q8 wks >100 kg: 90mg SQ q12 wks; may require q8 wks	45mg SQ wks 0,4	45mg SQ q12 wks; may require 90mg or q8 wks	260/390/520mg IV single dose (weight-based)	90mg SQ q8 wks
risankizumab	150mg SQ wks 0,4	150mg SQ q12 wks	150mg SQ wks 0,4	150mg SQ q12 wks	CD: 600mg IV wks 0,4,8 UC: 1200mg IV wks 0,4,8	180 or 360mg SQ q8 wks (starting wk 12)
guselkumab	100mg SQ wks 0,4	100mg SQ q8 wks	100mg SQ wks 0, 4	100mg SQ q8 wks	UC: 200mg IV wks 0,4,8	200mg SQ q4 or 100mg q8 wks
tofacitinib	N/A	N/A	N/A	5mg PO BID or 11mg daily (ER)	10mg PO BID	5-10mg PO BID
upadacitinib	N/A	N/A	N/A	15mg PO daily	CD: 45mg PO daily x12 wks UC: 45mg PO daily x8 wks	15 - 30mg PO daily

Legend: PsA: psoriatic arthritis; IBD: inflammatory bowel disease; q: every; wk(s): week(s); IV: intravenous; SQ: subcutaneous; CD: Crohn’s disease; UC: ulcerative colitis; PO: oral; N/A: not applicable; BID: twice daily; ER: extended-release

DISCUSSION

Considerations by Drug Class

A tabular summary of treatment considerations by diagnosis and comorbidity is shown in Table 2. The use of combined advanced immunosuppressive therapies (e.g., combination biologic therapies) is an area of active study in both IBD and PsA^7-9 and may be implemented in the most complex or refractory cases. However, for simplicity and relevance to the majority of patient cases, a detailed description of treatment considerations by individual drug class follows.

Table 2.

Treatment indications for patients with psoriatic arthritis and associated comorbidities

	Peripheral arthritis	Axial arthritis	Pso	Nail Pso	Enthesitis	Dactylitis	IBD	Uveitis
NSAIDs							?
MTX
TNFi							Not ETN	Not ETN
IL-17i
IL-12i/IL-23i
IL-23i		?
JAKi								?
PDE4i
CTLA4i
Integrin inhibitors

Legend: Green=effective, yellow=moderately effective, gray=ineffective, red=caution. Pso: Psoriasis, CD: Crohn’s disease, ETN: etanercept, IBD: inflammatory bowel disease, NSAIDs: nonsteroidal anti-inflammatory drugs, MTX: methotrexate, TNFi: tumor necrosis alpha inhibitor, IL: interleukin, JAKi: janus kinase inhibitor, PDE4i: Phosphodiesterase-4 inhibitor, CTLA4i: cytotoxic T-lymphocyte antigen 4 inhibitor

Tumor necrosis alpha inhibitors (TNFi)

Most TNFi are effective and FDA-indicated for treatment of IBD. Infliximab and adalimumab are approved for the treatment of both Crohn’s disease (CD) and ulcerative colitis (UC). Golimumab is approved for the treatment of UC only while certolizumab is only approved for use in CD. TNFi are also preferred agents for the treatment of more aggressive fistulizing CD^10,11. Etanercept is considered ineffective in the treatment of IBD¹² due to its primary binding of soluble TNF in the bloodstream and thus poorer ability to induce T cell apoptosis in the gut mucosa¹³. There are additional reports of paradoxical provocation of IBD in patients exposed to etanercept, further supporting avoidance in this condition¹⁴.

Methotrexate

Methotrexate is considered moderately effective for the treatment of IBD, but is less commonly used as a primary therapeutic in the biologic era. Specifically, methotrexate has shown modest efficacy in induction and maintenance of CD^15,16 but has less consistent benefit in UC^17,18. Methotrexate is more commonly used for anti-drug antibody prevention in patients on TNFi and has also been shown to reduce the frequency of infliximab infusion reactions¹⁹. The 2014 COMMIT trial did not show any increased efficacy in maintenance of remission in patients with CD on infliximab versus those on combination infliximab and methotrexate²⁰. However, patients in the combination arm were significantly less likely to develop anti-drug antibodies and had higher trough infliximab concentrations. There is ongoing controversy as to the utility of routine methotrexate use for the prevention of anti-TNFi antibodies with differing guidance from rheumatology and gastroenterology treatment guidelines^21,22. The decision to use low-dose methotrexate should be shared between the patient, their rheumatologist and gastroenterologist.

Interleukin(IL)-12i/IL-23i and IL-23i

Both IL-12i/IL-23i and IL-23i are approved for the treatment of IBD, psoriasis and psoriatic arthritis but are considered generally ineffective for axial arthritis. Ustekinumab is FDA-approved for the treatment of both moderate-to-severe CD and UC^23,24. Amongst the IL-23 inhibitors, risankizumab is currently the only drug with an indication for psoriasis and PsA that is also approved for use in both CD and UC^25,26. A recent head-to-head open-label randomized controlled trial in patients with CD showed risankizumab to be noninferior to ustekinumab in inducing clinical remission at 24 weeks but superior for endoscopic remission at week 48²⁶. These data along with the superiority of risankizumab over ustekinumab for treatment of moderate-to-severe plaque psoriasis^27,28 suggest that risankizumab may be preferable in patients with comorbid psoriasis and CD. Guselkumab is only approved for the treatment of moderate-to-severe UC²⁹ but initial phase II studies in CD are promising^30,31 and the first head-to-head trial of guselkumab versus ustekinumab was recently completed but not yet published at the time of this review. Mirikizumab is another IL-23i approved for the treatment of UC and CD but does not have approval for treatment of psoriasis or psoriatic arthritis despite two positive phase III trials in plaque psoriasis^32,33.

IL-17i

IL-17i were initially felt to have promise in the treatment of IBD based on animal models suggesting a protective role of IL-17A in IBD as well as genome-wide association studies linking the IL-23-IL-17 axis to the pathogenesis of IBD^34-38. However, a proof-of-concept double-blind randomized placebo-controlled controlled trial of secukinumab showed inefficacy relative to placebo in the treatment of severe CD and that a sizable minority of patients receiving secukinumab experienced a worsening of their intestinal disease³⁹. Subgroup analyses revealed that those patients with “inflammatory CD”, evidenced by a baseline serum C-reactive protein ≥10mg/L and/or fecal calprotectin ≥200ng/mL, were at the highest risk for both inefficacy and exacerbation of their underlying Crohn’s. Further studies have revealed this class of medications to be associated with an overall small risk (0.5-1%) of provoking new IBD or precipitating flares in patients with existing IBD^40-42, with the highest risk within 6 months of drug initiation⁴³. As such, avoidance of IL17i is generally recommended in patients with known IBD and especially in those patients with a more inflammatory phenotype. In patients with comorbid IBS or frequent diarrhea, IL-17i should be used with some caution. In these cases, counseling of the patient is warranted regarding the recognition of changes in symptoms, and low threshold for medication cessation and gastroenterologist evaluation in case of symptom progression. If a patient has otherwise experienced significant clinical improvement (e.g., arthritis) on IL-17i and gastrointestinal symptoms are mild, then continued treatment with IL-17i can be considered on a case-by-case basis with the addition of an intestinally-targeted drug as directed by the gastroenterologist.

Janus kinase inhibitors (JAKi)

Tofacitinib and upadacitinib are both approved for the treatment of UC but only upadacitinib is approved for the treatment of CD. An additional JAKi, filgotinib, has regulatory approval in Europe but not the United States. Two phase II trials of tofacitinib in CD did not meet their primary endpoint^44,45 but it is not clear whether this failure was due to tofacitinib’s pan-JAK inhibition (compared to other, more targeted JAKi) or issues with study design, as evidenced by the high rates of placebo response. Subsequent real world studies have shown inconsistent signal for benefit in CD⁴⁶. Given these data and the slightly better performance of upadacitinib relative to other JAKi in psoriasis and psoriatic arthritis trials⁴⁷, upadacitinib may be the preferred JAKi to use in patients with psoriasis and comorbid IBD. It should be noted that JAKi dosing is generally higher for IBD than for psoriasis or PsA, but it is not known whether this higher dosing is more effective for psoriasis or arthritis. Finally, given that JAKi are orally-administered, special consideration needs to be taken in patients with a history of significant bowel resection (e.g., colectomy). In these patients, there may not be full enteral absorption of the drug and patients may even report seeing pill passage in the stool. If this is the case and there is ongoing disease activity, additional immunosuppression or substitution for a parenteral agent may be warranted.

Integrin inhibitors

While approved for the treatment of both CD and UC, this class of medications (e.g., vedolizumab) is generally ineffective for the treatment of inflammatory arthritis. Post-hoc analyses from the GEMINI trial found no significant increase in sustained resolution of arthritis/arthralgia in patients with IBD receiving vedolizumab versus placebo⁴⁸. However, in patients with well-controlled arthritis but active IBD, integrin inhibitors can be added to background immunosuppressive therapy in consultation with the treating gastroenterologist.

Phosphodiesterase-4 (PDE4i) and cytotoxic T-lymphocyte antigen 4 inhibitors (CTLA4i)

Apremilast and abatacept are not currently FDA-approved for use in IBD. A single phase II double-blinded randomized control trial of apremilast in patients with UC did not meet the primary endpoint of clinical remission by week 12 but there was a numerical improvement in clinical and endoscopic features in the apremilast arm⁴⁹. Phase III trials of abatacept did not reveal efficacy for the treatment of CD or UC⁵⁰.

Nonsteroidal anti-inflammatory drugs (NSAIDs)

The use of NSAIDs in patients with IBD is controversial. The most recent American College of Gastroenterology IBD treatment guidelines make strong recommendations against the use of NSAIDs in this patient population^21,51. However, a recent meta-analysis did not show any increased risk of IBD exacerbation, noting a large heterogeneity in prior studies addressing this question with varied outcome definitions and study designs⁵². In general, however, COX-2 selective inhibitors are associated with relatively less gastric toxicity and should be preferentially prescribed over nonselective inhibitors after accounting for other patient comorbidities. This is supported by two randomized placebo-controlled trials of COX-2 selective inhibitors (celecoxib, etoricoxib) in IBD which did not show any significant risk of precipitating intestinal disease flare^53,54. As such, short courses of COX-2 selective NSAIDs may be reasonable on a case-by-case basis for patients in intestinal remission, but interdisciplinary discussion with gastroenterology is advisable. Ultimately, further safety studies of NSAID use in IBD are needed to better delineate the true risk profile of this drug class and identify patient subgroups in whom the risk/benefit ratio may be favorable.

Case Study

The following are a selection of real-life cases that highlight key considerations in the treatment of patients with comorbid PsA and IBD.

Case 1

A 64 year old man with a history of low grade bladder and prostate cancer diagnosed 10 years ago (surgically resected with no prior radiation or systemic chemotherapy) and prior history of psoriasis presented with acute onset of pain and swelling of the bilateral feet. The patient was diagnosed with psoriatic arthritis and initially treated with apremilast with minimal improvement and ultimately lost to follow up. About 1 year later he presented to another rheumatologist with hip and shoulder pain, prolonged morning stiffness, and elevated markers of inflammation. He was diagnosed with polymyalgia rheumatica and started on prednisone 10 mg daily with improvement in hip and shoulder pain, but ongoing hand and foot pain without relief. Over a period of two years, he received periodic corticosteroid tapers with partial resolution of joint symptoms, but ongoing psoriasis and recurrence of joint pain upon tapering.

Upon initial evaluation in our clinic, the patient was noted to have synovitis involving the bilateral knees, distal interphalangeal joints (DIPs), proximal interphalangeal joints (PIPs), metacarpophalangeal joints (MCPs), and right wrist with 68 joint count demonstrating 26 swollen and 24 tender joints. There was active psoriasis involving the scalp, elbows, and knees with total body surface area of 2%. Laboratory evaluation showed elevated erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP), negative rheumatoid factor, and negative anti-cyclic citrullinated peptide antibody.

A diagnosis of psoriatic arthritis was made. In light of his multiple prior malignancies, a shared decision was made with the patient to start the IL-17i, secukinumab. After the initial five loading doses, the patient reported a complete resolution of psoriasis and dramatic improvement in joint symptoms, with swollen joint count of 1 on follow-up examination. He had prolonged remission of symptoms for 2 years, but developed secondary failure of secukinumab with recurrent arthritis and psoriasis. He was switched to ixekizumab and was again in prolonged remission for 18 months.

At a routine follow up, the patient noted continued resolution of articular and cutaneous symptoms, but newly reported a six month history of decreased stool caliber, loose stools, and fecal urgency with incontinence. He had recently undergone colonoscopy which revealed patchy erythema in the colon, with biopsies showing non-specific vascular congestion and reactive epithelial changes with submucosal perivascular lymphocytic inflammation. Workup for enteric pathogens was negative and a fecal calprotectin was elevated at 398.

Gastroenterology was consulted and felt that the clinical picture was suggestive of IBD, possibly triggered by IL-17i therapy. Given the mild gastrointestinal symptoms and otherwise excellent response to ixekizumab, an interdisciplinary decision was made to continue ixekizumab and add mesalamine with close monitoring of his gastrointestinal symptoms and plan for repeat colonoscopy. The patient’s bowel symptoms resolved and fecal calprotectin normalized. He remains in remission at this time with continued use of ixekizumab and mesalamine.

Case 2

A 42 year old man with a prior diagnosis of celiac disease (diagnosed at age 12), psoriasis, and Crohn’s disease presented to clinic with a complaint of foot pain. His Crohn’s disease was diagnosed ten years prior to presentation in the setting of bloody stools and abdominal pain with evidence of an enterocutaneous fistula and associated small-bowel enteritis on magnetic resonance enterography. He was initially treated with 6-mercaptopurine followed by infliximab with improvement in his bowel symptoms and no recurrent enteritis or fistulas. For his celiac disease, he has remained completely adherent to a gluten-free diet.

Approximately six years later, the patient developed new-onset neck, jaw, and lower back pain associated with severe pain in the plantar aspect of the bilateral feet, not limited to the heel. He was seen by a rheumatologist and was started on a prednisone taper with some improvement. Methotrexate was added but the patient developed elevated liver enzymes and so was stopped. Infliximab was switched to adalimumab but with inadequate improvement in musculoskeletal symptoms and therefore another switch to certolizumab was made. Despite ongoing gastrointestinal remission on certolizumab, the patient’s back and foot pain became progressively disabling. Leflunomide was added with some improvement but the patient again developed elevated liver enzymes on high-dose leflunomide and recurrent musculoskeletal symptoms on lower doses. He presented to our clinic on a regimen of certolizumab 200mg every 2 weeks and leflunomide 20mg daily.

On initial exam, the patient did not have clear evidence of synovitis but had possible enthesitis of the bilateral plantar fascial insertions and Achilles tendon insertions. No dactylitis was noted. Paraspinous tenderness was noted throughout the lumbar spine and a FABER test was positive. A small patch of psoriasis was noted behind the bilateral ears. Subsequent workup demonstrated normal markers of inflammation. Magnetic resonance imaging (MRI) of the sacroiliac joints and lumbar spine showed no evidence of spondylitis or sacroiliitis, and minimal degenerative changes of the lumbar spine. MRI and subsequent ultrasound of the bilateral feet demonstrated no clear evidence of enthesitis.

An initial decision was made to continue certolizumab for the patient’s Crohn’s disease but to hold leflunomide given persistently elevated liver transaminases. However, shortly after the initial visit, the patient developed acute, severe left eye pain, photophobia, and redness. He was evaluated urgently by ophthalmology and found to have acute anterior uveitis (AAU) of the left eye.

Given prior insufficient response to two TNFi and new AAU on a third TNFi, an interdisciplinary decision was made with ophthalmology, gastroenterology, rheumatology, and the patient to trial the JAKi, upadacitinib 15mg daily. Within a couple of months his ocular symptoms gradually resolved with no further evidence of uveitis on ophthalmologic examination. His gastrointestinal symptoms remained stable and he experienced significant improvement in his multifocal musculoskeletal symptoms.

SUMMARY

In summary, IBD is common in patients with PsA and has important implications for the overarching therapeutic approach. Notably, not all drugs with an indication for the treatment of PsA may be effective for IBD and in certain instances are associated with a risk of worsening intestinal disease. Furthermore, dosing may differ for drugs with overlapping treatment indications. Ultimately, close coordination of care with the treating gastroenterologist is necessary for the optimal care of this patient population.

CLINICS CARE POINTS

• Close coordination between the rheumatologist and gastroenterologist is critical for the optimal treatment of patients with comorbid PsA and IBD.

• Drug dosing may be higher for IBD than for psoriasis or PsA and thus should be discussed with a clinical pharmacist and/or gastroenterologist prior to initiating therapy.

• Select therapies that are effective for psoriasis or PsA (e.g., apremilast) are ineffective for IBD while other drugs (e.g., IL-17i and possibly non-selective NSAIDs) may precipitate or worsen underlying gastrointestinal disease and thus should be avoided or used with caution on a case-by-case basis.

Key points:

• Close coordination between the rheumatologist and gastroenterologist is critical for the optimal treatment of patients with comorbid psoriatic arthritis (PsA) and inflammatory bowel disease (IBD).

• Drug dosing may differ for IBD than for psoriasis or PsA.

• Select therapies that are effective for psoriasis or PsA are ineffective for IBD while other drugs may precipitate or worsen underlying gastrointestinal disease.

Synopsis:

Inflammatory bowel disease is a common comorbidity in patients with psoriasis and psoriatic arthritis. Several key considerations are needed to provide optimal treatment of this unique patient population.