• autoimmunity
• autoimmune diseases
• treatment

The risk of immune-related adverse events (irAEs) secondary to cancer immunotherapy with checkpoint inhibitors has provided both a challenge and an opportunity for rheumatologists to engage meaningfully in the care and investigation of such patients. Our viewpoint is that not only do rheumatologists have much to add in terms of diagnostic and management skills for patients developing typical rheumatic complications (ie, arthritis, myositis, sicca syndrome, vasculitis) but we as a profession are, among all medical specialties, perhaps the best-equipped to manage the more complex strata of irAEs that can be life-threatening, involve multiple organ systems and at times evolve into chronic diseases.

By way of background, the six currently approved immune checkpoint inhibitors have added meaningfully to the care of patients with cancer; these agents have a broad range of activity demonstrating response rates of 15%–90% in more than 10 cancer types with the capacity to induce durable responses in select patients.1 While the benefit to risk ratio of these agents is well established, the antitumour effects come at a cost: the generation of a unique spectrum of toxicities referred to as irAEs, including dermatologic, hepatic, endocrine, pulmonary and rheumatic complications.2–5 It is not uncommon for checkpoint-treated patients to develop more than one irAE involving multiple organ systems.6 While these complications are believed to arise due to off-target immunoenhancement of effector pathways,7 they pose unique challenges and frequently require immunosuppressive regimens grounded in initial glucocorticoids, followed often by the addition of other non-biological or biological immunomodulators depending on the level of severity or persistence.8 Moving ahead, it will also be critical to assess whether more aggressive immunosuppressive regimens will have deleterious effects on the antitumoural response for as of now some evidence suggests that the presence of irAEs in general and the use of glucocorticoids have not been deleterious.9 10 Furthermore, in the absence of comparative effectiveness data in the overall management of irAEs, clinicians often craft specific regimens based on complex patient to patient considerations (ie, end organ involvement with tumour, previous immunosuppressive chemotherapy history, comorbidities and drug–drug interactions, etc). In general, these complications have been managed by oncologists guided by a general approach to toxicity management designed for ipilimumab (an anti-CTLA4), which was the first approved checkpoint inhibitor11 and later by consensus opinion such as the USA-European collaborative for PD-1 agents.12 Recently, a single centre has proposed a management algorithm for patients with immune-mediated inflammatory arthritis.13 It is important to note that to date no randomised controlled trials in irAE management to determine if one strategy is superior to the other has been published. With this in mind, rheumatologists would appear well positioned to add meaningfully to the overall approach to irAEs, particularly in the following settings.

One of the greatest challenges in the growth of checkpoint therapy surrounds the question of whether patients with pre-existing autoimmune diseases, including conditions such as rheumatoid arthritis, other forms of inflammatory arthritis and various connective tissue diseases, are candidates for cancer immunotherapy; these patients were censored from early clinical trials. Given that in the USA alone, there may be upward of 50 million patients with some form of autoimmunity; this is a formidable problem.14 To date, several small studies and anecdotal case reports examining the effects of checkpoint inhibition in patients with underlying autoimmunity, including rheumatic diseases, suggest that these types of patients can be effectively treated but flares of underlying diseases can and do occur in perhaps 1/3 of patients.15–17 This issue is of particular importance as it can lead to an interruption in cancer treatment: when patients with underlying autoimmunity and cancer are treated with checkpoint inhibitors and experience a flare of their underlying disease, they may, as a result, require increased immunosuppression; this added immunosuppression may prevent them from continuing their cancer immunotherapy until the underlying disease can be controlled. Clearly, maintaining their eligibility for continuing cancer immunotherapy must be the first priority, and rheumatologists can aid in this effort by collaborating on strategies that will allow this. Surely, from the patient’s perspective, it can be reassuring to know that their specialist for their underlying autoimmune disease is actively engaged with their oncologist in the team’s desire to defeat their cancer.

Another area of interest is predicting irAEs before they occur. While the early clinical series3 18 of rheumatic irAEs failed to demonstrate the presence of autoantibodies in patients with inflammatory polyarthritis, several recent reports suggest that pre-existing autoantibodies such as anticitrullinated protein antibodies5 and diabetes-associated islet antibodies19 may be predictive of new-onset rheumatoid arthritis and type 1 diabetes. Thus, efforts to profile risk for incident irAEs prior to checkpoint therapy need to be explored and specialists such as rheumatologists may be logical partners to monitor and manage such complications before they occur.

Obviously, other specialties must be involved in the specific management of some of these IrAEs: neurologists, endocrinologists, internal medicine specialists, intensive care specialists, etc. We think that rheumatologists may be best equipped to orchestrate the management of these IrAEs since they are the specialists most closely and historically associated to immunology and mainly to immune treatment of systemic inflammatory diseases. Rheumatologists may also contribute meaningfully on the treatment team for the management of rare and occasionally life-threatening complications of critical end organs resulting in myocarditis20 or central nervous system vasculitis21 among others where prompt and effective immunosuppression is needed. Treatment algorithms11 12 for common adverse events such as skin, bowel and liver endorse first-line glucocorticoids and drugs such as azathioprine, mycophenolate and short-term infliximab but do not explore other biological agents and strategies that may offer targeted therapy as our understanding of the underlying immunopathogenesis increases. A recent case report from the New England Journal of Medicine 22 exemplifies the need for more complex approaches to therapy, as they described a patient flaring with both Crohn’s disease and psoriasis after treatment with a PD1 targeting agent who was treated with an anti-interleukin (IL)17 leading to exacerbation of the cancer. A more rational approach perhaps would have selected an antitumour necrosis factor (anti-TNF) agent with dual effects for both comorbidities. Of note, in a recent series of three cases, it was suggested that tocilizumab, an anti-IL6 receptor antibody, might be an effective alternative to corticosteroids or TNFα inhibitors for the treatment of arthritis irAEs.23

We believe this report merely demonstrates the need for more a sophisticated approach to the expanding spectrum of irAEs.

Finally, in terms of rheumatic complications such as inflammatory arthritis and connective tissue diseases, from early experience with small numbers of patients, it appears that these complications may be distinctive, for unlike virtually all other irAEs there is a high rate of chronicity with perhaps the majority of patients requiring ongoing therapy with biological and non-biological disease-modifying antirheumatic drugs. Thus, rheumatologists will inevitably be involved with the management of increasing numbers of oncology patients and engaging interprofessionally with oncologists moving forward.

The management of patients with cancer is undergoing a frame shift change with the introduction and expansion of checkpoint therapy which up until recently was only established in academic centres but now is rapidly spreading to community practices. New interprofessional relationships are needed to optimally manage increasing numbers of patients with irAEs and these oncology–rheumatology collaborations will be vital. We conclude by noting that numerous issues now challenge the rheumatology community in relationship to our evolving role in caring for patients with irAEs but an immediate challenge is how will we educate ourselves about this new area of medicine and new area of autoimmune disease? Merely encouraging rheumatologists to participate in the care of such patients is inadequate unless we strategise to educate ourselves on the rapidly expanding base of declarative and procedural knowledge that is essential to bring informed care to the patients and a new research agenda to address the evolving field.