Objectives Immune checkpoint inhibitors (ICIs) targeting cytotoxic T-lymphocyte-associated protein 4 and programmed cell death protein 1 (PD-1) have demonstrated improved survival for multiple cancers. However, these new drug classes have led to increased immune-related adverse events (IrAE). Rheumatic IrAEs have not been well described in clinical trials. We report here cases of rheumatoid arthritis (RA) and polymyalgia rheumatica (PMR) occurring after ICI treatment.
Methods This was a retrospective study of patients receiving an ICI in whom symptoms of arthritis or arthralgia developed and revealed a diagnosis of RA or PMR.
Results In 10 patients who received ICI therapy (all anti-PD-1 or anti-PDL1 antibodies), RA or PMR developed at a median of 1 month (1 to 9) after exposure. No patient had pre-existing rheumatic or autoimmune disease. RA developed in six patients; all six were positive for anti-cyclic citrullinated peptide (anti-CCP) antibodies and four for rheumatoid factor. Anti-CCP antibodies were detected in two out of three patients tested before immunotherapy. Disease-modifying antirheumatic drugs were needed for three patients; the three others received corticosteroids or non-steroid anti-inflammatory drugs. PMR was diagnosed in four patients, all responded to corticosteroids. Despite these IrAEs, immunotherapy was pursued for all but one patient until cancer progression.
Conclusions This is the first description of RA occurring after ICI therapy for cancer. PMR can also occur after ICI, particularly after anti-PD-1 therapy. All cases responded to corticosteroids or with immunosuppressive therapy. Collaboration between rheumatologists and oncologists is crucial and could lead to better recognition and care of these patients.
- immune-checkpoint inhibitors
- rheumatoid arthritis
- polymyalgia rheumatica
Statistics from Altmetric.com
In recent years, immunotherapy has dramatically transformed the prognosis of several cancers, including principally metastatic melanoma and non-small cell lung cancer (NSCLC). Ipilimumab, an immune checkpoint inhibitor (ICI) targeting cytotoxic T-lymphocyte-associated protein 4 (CTLA-4), has been approved by the US Food and Drug Administration (FDA) and European Medicines Agency (EMA) for treating metastatic melanoma. The ICIs nivolumab and pembrolizumab, targeting programmed cell death protein 1 (PD-1), or atezolizumab, targeting its ligand PDL1, have been FDA and EMA approved for treating metastatic melanoma and NSCLC.1 2
Adverse events (AEs) observed with ICI therapy are related to their mechanisms of action, although the precise pathophysiology needs to be better understood. Immune-related AEs (IrAEs), such as colitis, autoimmune thyroid disease and vitiligo, have been described with ipilimumab and anti-PD-1 therapy. Others seem to be more specific, such as hypophysitis with ipilimumab and pneumonitis with anti-PD-1 antibodies. Some can be life threatening, such as pneumonitis.3–5 Relapse or flare of pre-existing autoimmune diseases has been reported,6 but the occurrence of new autoimmune diseases seems to be less frequent. However, a series of 21 patients with psoriasis induced by anti-PD-1 therapy was recently published.7
Little is known about rheumatic AEs with ICI therapy.8 Indeed, in phase III studies, arthralgia, which includes all musculoskeletal disorders, was present in about 5% of patients receiving ipilimumab for melanoma, 9%–20% with pembrolizumab and 5%–16% with nivolumab for melanoma or NSCLC versus <1% with placebo.4 However, these AEs may be underestimated, and no clinical description was provided. Recently, a series of 13 patients with non-classified rheumatic IrAEs was published9: non-specific inflammatory arthritis developed in 9 patients without autoantibodies and 4 presented sicca symptoms but did not fulfil criteria for Sjögren syndrome.
In this article, we report a series of 10 patients in whom seropositive rheumatoid arthritis (RA) or polymyalgia rheumatica (PMR) developed after ICI treatment.
The Gustave Roussy Cancer Center (Villejuif, France) has established a national pharmacovigilance registry called Registre des Effets Indésirables Sévères des Anticorps Monoclonaux Immunomodulateurs en Cancérologie dedicated to collecting immunotherapy AEs. We also performed a retrospective multicentre collection of observations through the Club Rhumatismes et Inflammation (CRI) network, a section of the French Society of Rheumatology: between September 2016 and January 2017, all rheumatologists and internal medicine practitioners registered on the CRI website (http://www.CRI-net.com), almost 2400 physicians all over France, were contacted by successive newsletters over 6 months to report cases.
Patients were included if they had received ipilimumab, nivolumab, pembrolizumab or another ICI in development and thereafter had a diagnosis of RA according to the 2010 American College of Rheumatology/European League Against Rheumatism (ACR/EULAR) criteria9 or PMR according to EULAR/ACR criteria 2012.10 All patients underwent a rheumatological assessment by a trained rheumatologist or internal medicine specialist with biological and radiological evaluations. Patients had no pre-existing rheumatic disease or arthritis based on the rheumatologist’s questioning.
Ten cases of rheumatic inflammatory diseases following ICI therapy were collected: four from Gustave Roussy and six from other French centres (tables 1 and 2 and supplementary file). All patients received an anti-PD-1 antibody (nivolumab or pembrolizumab) or anti-PDL1 antibody (one patient in an open phase I clinical study). One received ipilimumab in addition to nivolumab for four cycles and then nivolumab alone. The cancers treated were metastatic melanoma, endometrial adenocarcinoma, mesothelioma, lung adenocarcinoma, squamous cell carcinoma of the vagina, gastric adenocarcinoma and colon adenocarcinoma.
In six patients, seropositive RA developed and in four, PMR developed. The mean age of patients was 65 years and 60% were male. The median time to IrAE after ICI exposure was 1 month (range 1–9 months). Only one patient showed other types of IrAEs before rheumatic IrAE: a rash with anti-PD-1 antibody (patient 2). Disease-modifying anti-rheumatic drugs (DMARDs) were needed for three patients with RA (hydroxychloroquine for two and methotrexate for one), the three others patients with RA received only corticosteroids or non-steroid anti-inflammatory drugs (NSAIDs). The four patients with PMR responded to corticosteroids. The ICI therapy was continued for all patients but one until cancer progression. Nivolumab was stopped in only one patient (patient 6) who showed no improvement with corticosteroids and because cancer was stable, methotrexate being introduced at the same time for RA.
To our knowledge, this is the first series describing seropositive RA occurring in patients receiving ICI therapy. Only one published series described non-specific rheumatic IrAEs but no case of seropositive RA.11
In our series, the six patients with RA fulfilled the 2010 ACR/EULAR criteria9 and were seropositive (anti-cyclic citrullinated peptide (CCP) positivity in all six and rheumatoid factor (RF) in four). Anti-CCP antibodies could be detected in 2/3 patients (patients 1, 4 and 5) with available serum samples before immunotherapy. Three patients were negative for RF before ICI therapy and one showed weak positivity (18 U/mL) afterwards.
The short time between ICI treatment and the development of joint symptoms and anti-CCP positivity before ICI therapy in 2/3 patients suggested that some of these patients had a pre-RA status and that the treatment with ICI may have triggered the clinical disease. Indeed, studies have shown that antibodies (anti-CCP and RF) may be present several years before RA onset.12 Nevertheless, no patient presented arthralgia before ICI treatment. All cases of RA occurred after anti-PD-1 treatment.
The observation that PD-1 inhibition can trigger RA may suggest an important role of the PD-1/PDL1 pathway in RA pathogenesis. PD-1 has been found important for self-tolerance, because PD1−/− mice showed spontaneous autoimmune disease development.13 PD-1 polymorphisms have been associated with increased susceptibility to RA, and membrane and soluble PD-1 expression is decreased in patients with RA.14 A recent case report showed RA recurrence in a patient with pre-existing RA in remission after receiving nivolumab for NSCLC.15
Two cases of PMR with ipilimumab have been reported,16 but the PD-1/PDL1-2 system could also be important in the pathophysiology of PMR. Indeed, two cases were previously described17 and we now report four additional cases. Interestingly, giant cell arteritis was recently found to involve a deficiency in the PD-1 immune checkpoint: vessel-wall dendritic cells fail to express PDL1, which leaves lesional T cells unchecked.18
The fact that our 10 cases occurred only after anti-PD-1 and not anti-CTLA-4 antibody treatment (except one patient who received both during four cycles) suggests a contrast between the IrAE occurring after blockade with CTLA-4 (colitis, endocrine disorders, skin rashes) and PD-1/PDL1 like non-specific arthritis, sicca syndrome11 RA, PMR (this report) and connective tissue diseases.19
Pre-existing autoimmune disease (AID) was an exclusion criterion in clinical trials of ICI. In real life, these patients are frequently not excluded from ICI treatment because the priority is obviously the treatment of cancer. Because rheumatic IrAEs are closely linked to the mechanisms of action of ICI on immune cells, we can expect many rheumatic IrAEs or exacerbations of rheumatic IrAEs considering the future worldwide prescription of ICI. In one study, about one-third of 30 patients (n=8) with known AID who received ipilimumab showed a flare of AID and another third (n=10) showed a new IrAE.6 In another study of 52 patients with pre-existing AID who received anti-PD-1 therapy (52% with rheumatic diseases), 20 (38%) had a flare of the AID after the first anti-PD-1 dose (median time 38 days). Flares were mild and were managed with corticosteroids and steroid-sparing agents such as methotrexate. Flares occur mainly in patients with rheumatic disorders. Only two patients discontinued anti-PD-1 therapy because of exacerbation of the AID.20
CTLA-4 is a target of both autoimmunity and cancer. Drugs inhibiting this immune checkpoint have been developed for cancer (ipilimumab) and drugs activating this inhibitory signal have been developed for autoimmunity (abatacept) or transplantation (belatacept). Even if no signal for a possible increased risk of cancer has been described with abatacept, the development of RA or other types of IrAEs with ICI therapy is a strong reminder to rheumatologists that long-term studies need to be continued in real-life patients receiving immunosuppressive drugs in general, to be sure of no increased risk of cancer with time. In addition, the possible balance between cancer and autoimmunity is illustrated by a possible link between an IrAE and response to the immune treatment of the cancer. This suggestion was raised with the treatment of melanoma with interferon21 and has been recently reemphasised in patients receiving pembrolizumab, for whom vitiligo occurrence, a clinically visible IrAE, could be associated with response to treatment.22
In our series, all patients were referred to a rheumatologist or an internist. However, the number of rheumatic IrAE cases could have been underestimated because mild arthralgia could have been managed with NSAIDs or corticosteroids prescribed by oncologists and then may not be referred to rheumatologists. The combined expertise of oncologists, immunologists and rheumatologists is crucial for successful management of these patients. In most cases, ICIs may be pursued with symptomatic treatment of the RA or PMR. As for ICI-induced colitis, prescription of anti-tumour necrosis factor antibodies or other DMARDs may be possible if needed with this multidisciplinary expertise.
In conclusion, patients with cancer treated with immunotherapy and who develop autoimmune or other reactions should be managed by multidisciplinary care units (MCU). Cancer immunotherapy is very promising and the number of available treatments is raising; thus, an implementation of MCU could be of great value for patients with cancer.
We would like to express our appreciation to all the authors, the patients and the Club Rhumatismes et Inflammation network.
Disclaimer: This is a summary of a scientific article written by a medical professional (“the Original Article”). The Summary is written to assist non medically trained readers to understand general points of the Original Article. It is supplied “as is” without any warranty. You should note that the Original Article (and Summary) may not be fully relevant nor accurate as medical science is constantly changing and errors can occur. It is therefore very important that readers not rely on the content in the Summary and consult their medical professionals for all aspects of their health care and only rely on the Summary if directed to do so by their medical professional. Please view our full Website Terms and Conditions.
Copyright © 2017 BMJ Publishing Group Ltd & European League Against Rheumatism. Medical professionals may print copies for their and their patients and students non commercial use. Other individuals may print a single copy for their personal, non commercial use. For other uses please contact our Rights and Licensing Team.
Contributors All the authors contributed to the manuscript: conception and design (RB and XM), collection of data (all authors). All the authors read and validated the final version of the manuscript.
Funding The authors received no financial support for the research, authorship and/or publication of this article.
Competing interests AM reports being one of the principal investigator for clinical trials from Roche/Genentech, BMS, Merck (MSD), Pfizer, Lytix pharma, Eisai, Astra Zeneca/Medimmune, Bayer, Celgene. AH reports advisory boards for Amgen and Lilly. OL reports consulting for MSD, BMS and Genzyme and received travel support from LFB and CSL Behring. All the others authors declared no conflict of interest for this work.
Patient consent Informed consent was obtained from all subjects, and the study was approved by the local ethics committee.
Ethics approval local ethics committee.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.