Rheumatological Toxicities of Immunotherapy

Introduction

■

The introduction of immune checkpoint inhibitor (ICI) therapy to treat cancer has changed the treatment paradigm for many malignancies. These agents exploit suppressor and regulatory pathways to boost integrated immunity against tumors but are associated with a unique spectrum of immune-related adverse events (irAEs) related to their untoward autoinflammatory and off-target effects. The field of irAEs from checkpoint inhibitor therapy is growing at a rapid pace and many unanswered questions remain. irAEs have been described in nearly every organ system, most commonly involving the gastrointestinal tract, skin, and endocrine system, and can range from mild and self-limiting to severe and life-threatening, often causing significant patient morbidity and even mortality. ^,Sometimes these irAEs require cessation of the checkpoint inhibitor.
■

In addition to some of the more commonly reported irAEs, a variety of rheumatic manifestations have been described including arthralgia, inflammatory arthritis, polymyalgia rheumatica, myositis, sicca symptoms, rare reports of vasculitis, and others. Rheumatic irAEs are infrequently reported in clinical trials and their descriptions are largely limited to case reports and small series. Rheumatic irAEs remain one of the most poorly understood irAEs and their true prevalence is unknown. Their natural history is also poorly defined. Whereas the more common irAEs often appear to be self-limited or resolve with effective immunosuppressive therapy, rheumatic irAEs may have a more persistent course, with some requiring treatment long after discontinuation of checkpoint inhibitor therapy. Despite numerous case reports and series, the natural history of rheumatic irAEs is only starting to be elucidated. In this chapter, the current literature on rheumatic irAEs is summarized, including prevalence, clinical spectrum, diagnosis, and treatment.

Mechanism of Action

■

The immune system is comprised of a system of checks and balances with many complex factors at play to maintain equilibrium. Immunological checkpoints are crucial for immune homeostasis and are involved in immune activation and deactivation. There are hundreds of known immunological checkpoints in the human immune system but the currently US Food and Drug Administration (FDA)–approved ICIs target two main immune checkpoints: cytotoxic T-lymphocyte-associated protein-4 (CTLA-4) and the programmed cell death-1 (PD-1)/PD-ligand-1 (PD-L1) pathway. CTLA-4 is primarily responsible for attenuating early activation of naïve memory T-cells in lymphoid organs. Following immune activation there is a rapid expansion of T-cells which requires two signals: (1) cognate antigen presentation by major histocompatibility complex (MHC) and (2) a co-stimulatory signal provided by engagement of CD28 on the naive T-cell with its ligand CD80/86 on the antigen-presenting cell. When the immune response is no longer needed, CTLA-4 is expressed on the T-cell surface and competes for CD80/86 binding with CD28, attenuating T-cell activation. The PD-1/PD-L1 pathway, on the other hand, functions in the periphery to inhibit T-cell activation at the effector stage. In certain cancers, tumor cells manipulate these pathways to evade immunosurveillance. By blocking these immunological checkpoints, ICIs harness the patient’s immune system to target tumor cells. ICIs work by exploiting these suppressor and regulatory pathways to boost the integrated immune response against tumors—essentially “releasing the breaks” on immune activation and reinvigorating the immune system. This results in nonspecific immune activation that can lead to irAEs.
■

The exact pathophysiology of rheumatic irAEs, as well as irAEs involving other systems, has yet to be fully defined, as there have been few investigations into possible immunopathogenic mechanisms. It remains unclear whether rheumatic irAEs represent de novo events versus indicative of underlying immune-mediated disease. At least four candidate mechanisms have been proposed, including generalized immune activating secondary to checkpoint neutralization, direct off-target effects of checkpoint inhibitors, preexisting asymptomatic autoimmunity, and off-target effects of T-cell mediated immunity. Further investigations into immunopathogenic mechanisms of rheumatic irAEs is urgently needed.

The Agents

The first ICI to be approved was ipilimumab in 2011, which blocks CTLA-4 and is indicated for metastatic melanoma. Since then, six more agents have been approved for an ever-growing list of indications, including melanoma, renal cell carcinoma (RCC), non–small-cell lung cancer (NSCLC), Hodgkin’s lymphoma, hepatocellular carcinoma, and many others, including a tumor agnostic indication for pembrolizumab, which is approved for treatment of microsatellite instability high or mismatch repair deficient solid tumors. Agents targeting many other immunological checkpoints are currently being investigated in clinical trials for a variety of cancers and at varying stages. Ipilimumab is approved for treatment in combination with nivolumab for melanoma, NSCLC, and RCC. ^,

Prevalence

■

irAEs of any type are quite common, occurring in up to 90% of patients receiving CTLA-4 therapy and 70% of patients receiving anti–PD-1/PD-L1 therapy, ^,but vary depending on tumor type. Incidence of irAEs increases when the drugs are used in combination.
■

Compared with other system irAEs, the true incidence of rheumatic irAEs remains less well characterized for a variety of reasons, with the most important issue being that the Common Terminology Criteria for Adverse Events (CTCAE) used to grade adverse events is poorly applicable to rheumatic irAEs. The grading system ranges from grade 1 through grade 5 (grade 1 = mild, grade 2 = moderate, grade 3 = severe, grade 4 = life-threatening, grade 5 = death). The majority of clinical trials report only events of grade 3 or higher, which generally require hospitalization. Rheumatic events such as inflammatory arthritis or polymyalgia rheumatica rarely lead to hospitalization and therefore are not captured in clinical trial data. Additionally, inflammatory arthritis does not have to require hospitalization to be severely debilitating for patients. Another reason for underreporting is that there is no standardization for coding rheumatic symptoms such that the same symptoms may be coded differently by different practitioners. For example, a swollen knee may be coded as joint pain, knee pain, knee swelling, knee effusion, or arthralgia. The creation of a more comprehensive grading system for rheumatic irAEs will facilitate earlier recognition and referral to rheumatologists, as well as more accurate reporting of their incidence and prevalence.
■

The most commonly reported rheumatic irAE in clinical trials is arthralgia. A systematic review of the literature that examined reporting of irAEs secondary to ICI found that 1% to 43% of participants across 33 clinical trials developed arthralgia. The true incidence of arthritis with ICI is less clear, as in that arthritis was only reported in 5 clinical trials, occurring in 1% to 7% of patients, and it is unclear how arthritis was defined in these trials. Some information on incidence of rheumatic irAEs can be gained from observational studies, including one single-center study that reported the incidence of inflammatory arthritis to be 5.1% in melanoma patients treated with anti–PD-1 therapy. Other less-commonly described rheumatic irAEs such as myositis, Sjögren’s syndrome, and polymyalgia rheumatica are not routinely reported in clinical trials and their incidence is largely based on case series, reports, and small retrospective studies. No large prospective studies have been conducted to help answer these questions. A systematic review and meta-analysis of irAEs was published wherein rates of certain irAEs from anti–PD-1/PD-L1 therapy in primary clinical trial data were reported. Investigators found reporting of musculoskeletal events to be inconsistent. Out of 13 studies examined, 3 made no mention of musculoskeletal problems. Of the studies that did provide data, only two reported arthritis, with the remainder providing rates of arthralgia, back pain, musculoskeletal pain, and myalgia. When reported, rates of arthralgia varied across studies from 10% to 26% and 2% to 12% for myalgia.
■

In general, rheumatic irAEs have been reported to occur within 12 weeks of initiation of ICI ; however, timing may vary from onset after the first dose to after therapy has been discontinued. ^,

Specific Rheumatic Immune-Related Adverse Events

INFLAMMATORY ARTHRITIS

ICI–related inflammatory arthritis has been increasingly described. One series included 30 patients with rheumatologist-confirmed inflammatory arthritis from either PD-1/PD-L1 monotherapy or anti–CTLA-4/PD-1 combination therapy. They observed three major clinical phenotypes of inflammatory arthritis: a polyarticular form resembling rheumatoid arthritis with small joint involvement, a reactive arthritis-like phenotype, and a large-joint predominant seronegative spondyloarthropathy. In another case series including seven cases of ICI–related inflammatory arthritis, the median time to onset of irAE was 7.3 weeks after starting ICI, with the exception of two patients who experienced irAEs over 1 year after starting immunotherapy. Rheumatic irAE led to withholding/discontinuing ICI in half of these patients. All patients required at least glucocorticoid therapy and several required further immunosuppression to treat their rheumatic irAE. Some still remained symptomatic months after their last ICI infusion. The vast majority of cases of ICI–related inflammatory arthritis are seronegative for rheumatoid factor or anti–cyclic citrullinated peptide (CCP) antibodies, although there have been rare reports of patients who developed anti–CCP-positive inflammatory arthritis who were found to be anti–CCP-positive prior to receiving ICI. The most notable feature of ICI–related inflammatory arthritis is that it may persist after cessation of ICI in a large proportion of patients, oftentimes requiring long-term treatment with targeted therapies and continued follow-up with rheumatologists.

SICCA SYNDROME

An entity resembling Sjögren’s syndrome has been described in the setting of ICI, with atypical features including abrupt onset of severe salivary hypofunction, with xerophthalmia reported less frequently, and absence of traditional Sjögren’s-related autoantibodies. Cappelli et al. reported four cases of severe salivary gland hypofunction with all patients experiencing more prominent dry mouth symptoms than dry eyes. One of these patients had positive anti-La/SS-B antibodies and had imaging evidence of parotitis. Another series reported five cases of sicca (two isolated cases and three in conjunction with other rheumatic irAEs) and two patients were positive for antinuclear antibodies (ANA), one in high titer (1:1280) with positive anti-Ro/SS-A antibodies.

POLYMYALGIA RHEUMATICA

An entity resembling polymyalgia rheumatica (PMR) has been increasingly described in the setting of ICI therapy. In a large series of ICI-related PMR, the authors sought to identify how many cases fulfilled the 2012 European League Against Rheumatism (EULAR)/American College of Rheumatology (ACR) Criteria for PMR ( Table 24.1 ). Of 49 cases, 37 (75%) had sufficient data to reliably apply the criteria, and of these 28 (75%) fulfilled the criteria for PMR. ICI–related PMR cases share several atypical features that are often not seen in de novo PMR, including presence of inflammatory arthritis overlap features in other joints, and requirement for higher doses of glucocorticoids than typically used to treat PMR.

TABLE 24.1■

Provisional Classification Criteria for PMR Scoring Algorithm

	Points Without Ultrasound (0–6)	Points With Ultrasound (0–8)
Morning stiffness lasting >45 min	2	2
Hip pain or limited range of motion	1	1
Absence of RF or ACPA	2	2
Absence of other joint involvement	1	1
≥1 shoulder with subdeltoid bursitis and/or biceps tenosynovitis and/or glenohumeral synovitis and ≥1 hip with synovitis and/or trochanteric bursitis	n/a	1
Both shoulders with subdeltoid bursitis, biceps tenosynovitis or glenohumeral synovitis	n/a	1