EXPERT ANALYSIS FROM THE WINTER RHEUMATOLOGY SYMPOSIUM
SNOWMASS, COLO. – Physicians can expect to encounter more and more patients with inflammatory arthritis and other rheumatic adverse events induced by immune checkpoint inhibitors as a result of anticipated exponential growth in the use of these drugs to treat an expanding list of cancers, Clifton O. Bingham III, MD, said at the Winter Rheumatology Symposium sponsored by the American College of Rheumatology.
These cancer immunotherapy–induced rheumatic diseases may superficially look like the classic forms of familiar autoimmune diseases, but they have highly atypical features that will affect treatment decisions.
Dr. Clifton O. Bingham III
“What we’ve seen consistently is that the normal doses of prednisone we would use to treat an inflammatory arthritis are really ineffective in most of these patients. We’ve had to use super doses – up to 120 mg/day – for initial control, and then 7.5-40 mg daily for maintenance of response,” according to Dr. Bingham, professor of medicine and director of the Johns Hopkins Arthritis Center in Baltimore.
To date, only limited data from case series are available on rheumatic IRAEs. There are no prospective patient registries logging accurate data on the incidence of these rheumatic adverse events among cancer patients treated with immune checkpoint inhibitors (ICIs). These IRAEs, which lie at the intersection of rheumatology and oncology, are of special interest to Dr. Bingham – he and his coinvestigators have published five articles on the topic over the course of a single year.
In a wide-ranging talk at the symposium, he touched on the phenotypic spectrum of rheumatologic IRAEs, his conviction that they are greatly underdiagnosed, why physicians can expect to encounter them much more frequently, rheumatologic IRAE treatment issues, and the risks of prescribing ICIs in patients with known preexisting rheumatologic disease.
Inflammatory arthritis is the most common form of rheumatologic IRAE, followed by sicca syndrome. At the Johns Hopkins Arthritis Center, Dr. Bingham and his coworkers have 25 well-characterized patients with inflammatory arthritis resulting from an ICI, only 1 of whom is HLA-B27-positive.
“Also, just one is autoantibody-positive, even though they all look for all the world as though they have rheumatoid arthritis,” the rheumatologist observed.
This ICI-induced inflammatory arthritis initially presents most commonly in the midsize and large joints – knees, ankles, elbows – then expands to include small joints such as the wrists, proximal interphalangeal joints, and the metacarpophalangeal joints.
Notably, the Hopkins group also has three patients with classic reactive arthritis marked by conjunctivitis, urethritis, arthritis, and dactylitis.
“I don’t know about you, but, in our general rheumatology practice, we see maybe one case of reactive arthritis in several years, so this is something that has struck us as really quite interesting,” said Dr. Bingham, who is also director of research in the division of rheumatology at Johns Hopkins.
The arthritis center is also managing a group of patients with ICI-induced sicca syndrome, which is uniformly extremely severe and treatment resistant, as well as a couple of patients with myositis IRAE, one with polymyalgia rheumatica, and two with crystal disease that is highly inflammatory in nature, difficult to treat, and includes an inflammatory polyarthritis component not typical in patients with crystal arthritis.
ICIs have dramatically transformed the treatment of selected advanced-stage cancers. For example, whereas patients with metastatic melanoma historically had a 2-year survival rate of 5%, combination therapy with the ICIs ipilimumab (Yervoy) and nivolumab (Opdivo) resulted in a 60% rate of partial or complete remission in a landmark clinical trial.
The basis of cancer immunotherapy is the discovery that, in order for cancer cells to thrive, they emit blocking signals that downregulate the native ability of T cells to recognize and kill them. This is true for both solid tumors and hematologic malignancies. The ICIs inhibit these blocking signals, which include cytotoxic T-lymphocyte–associated protein 4 (CTLA4), programmed death-1 (PD-1), and programmed death ligand-1 (PDL-1), thereby freeing up the T cells for tumor fighting.
Because of the nonspecific mechanism of this T-cell activation, however, ICIs have, as their main toxicities, T-cell–mediated autoimmune inflammatory tissue damage, which gets lumped under the umbrella term IRAEs. It can affect almost every organ system. Skin rashes are the most common, colitis second. Other commonly encountered IRAEs include thyroiditis, hypophysitis, hepatitis, peripheral neuropathy, and pneumonitis.
In addition to the four currently approved ICIs – ipilimumab, nivolumab, pembrolizumab (Keytruda), and atezolizumab (Tecentriq) – investigational ICIs targeting CTLA4, PD-1, and/or PDL-1 are in development. Plus, new ICIs targeting other blocking signals, including lymphocyte activation gene-3, CD137, and T-cell immunoglobulin and mucin domain-3, are now in clinical trials.
Clinical trials aimed at expanding the indications of existing ICIs and using ICIs in earlier-stage cancers in an effort to improve rates of lasting remission are also underway.
All told, probably at least 400 clinical trials of ICIs are ongoing worldwide, the rheumatologist estimated.
“More people will be exposed to these drugs, and we’ll see more and more of these rheumatologic IRAEs,” Dr. Bingham predicted.
Back in the pre-ICI days, Dr. Bingham was coauthor of a major study which concluded that clinical trialists in oncology consistently downgrade the severity of rheumatologic adverse events, often by 1 or 2 grades (J Rheumatol. 2007 Jun;34[6]:1401-14).
Unpublished details of ICI clinical trials in melanoma that he obtained from Bristol-Myers Squibb suggest that the true rate of rheumatologic IRAEs is about 20%, or roughly double that reported in the studies. That’s because the adverse events–grading system used in oncology undercalls the severity of arthritis and autoimmune disorders.
Indeed, the National Cancer Institute’s Common Terminology Criteria for Adverse Events, used in oncology clinical trials, is confusing on the topic of musculoskeletal and connective tissue disorders as treatment-emergent adverse events, according to Dr. Bingham. He noted that an oncologist can code a swollen joint in three different ways – joint effusion, arthritis, or arthralgia – and it takes disabling interference with self-care in activities of daily living for that swollen joint to rise to the level of a Grade 3 adverse event. From a rheumatology trialist’s perspective, that would be a Grade 4 disability.
Plus, neither the product labeling nor the patient information guides for the approved immunotherapy drugs mention the importance of monitoring for rheumatologic IRAEs or their management.
“There is poor awareness of musculoskeletal and rheumatic IRAEs in the general oncology community,” Dr. Bingham asserted. “But, if you talk with any oncology nurses who work in a clinical trial, they will tell you they’re seeing these events with significant frequency and severity.”
It’s critical to gain control of rheumatologic IRAEs quickly so that patients can get on with their cancer immunotherapy. Dr. Bingham uses intra-articular steroid injections for patients with oligoarthritis and high-dose oral prednisone for polyarticular disease. He starts methotrexate and/or leflunomide early because the conventional disease-modifying antirheumatic drugs have roughly a 2-month delay in onset of action. He has had several patients who are unable to taper steroids despite background methotrexate.
In the most severely affected patients, he has turned to biologic agents in consultation with their oncologists. Tumor necrosis factor (TNF) inhibitors are the ones he and other rheumatologists have used most often.
“Notably, we have not been able to taper down very well. We have patients who are out more than 2 years now who still require their TNF inhibitor to treat their inflammatory arthritis, and these are patients on conventional disease–modifying antirheumatic drugs as well. As soon as it’s tapered, the arthritis begins to come back,” according to Dr. Bingham.
In marked contrast, colitis as an IRAE typically clears in response to just one or two doses of a TNF inhibitor.
One audience member related that she’d encountered a roadblock in trying to get authorization for a TNF inhibitor for a patient with a rheumatologic IRAE secondary to ICI treatment for metastatic melanoma because the labeling states these agents are relatively contraindicated in melanoma patients. Dr. Bingham offered a tip: Collaborate with the patient’s oncologist.
“In most cases, oncologists can get infliximab for these patients and administer it in their infusion centers. They are able to get things authorized with very little trouble,” he said.
Besides, most of these patients with severe inflammatory arthritis meet conventional criteria for TNF inhibitor therapy, based on their number of infected joints and elevated acute phase reactants for longer than 6 weeks, Dr. Bingham noted.
“We’ve had some very interesting conversations with patients. It’s impressive to see the impact arthritis can have on people. A lot of patients have said, ‘I don’t care if I die. Get me functional right now.’ That’s pretty profound. Quality of life is still very important for people, even when dealing with life-threatening diseases,” he observed.
Oncologists are actually eager for their patients to get on steroid-sparing therapy because of concern that high doses of steroids may reduce the efficacy of cancer immunotherapy. That’s not an issue with the TNF inhibitors, the rheumatologist continued.
Turning to the utility of other classes of biologic agents, Dr. Bingham advised avoiding abatacept (Orencia) because its mechanism of action is likely to cause interference with the cancer immunotherapy. Rituximab (Rituxan) takes too long to act. Anakinra (Kineret), tofacitinib (Xeljanz), and tocilizumab (Actemra), on the other hand, are agents he is interested in using as alternatives to TNF inhibitors, although he hasn’t done so yet.
The experience here is entirely anecdotal, since such patients have been excluded from ICI clinical trials, but the available evidence suggests physicians should be prepared for higher rheumatologic IRAE rates in this setting. Investigators at Vanderbilt University reported that 8 of 30 cancer patients with known preexisting autoimmune disease experienced flares of that disease when treated with ipilimumab, and 10 developed a new IRAE (Therap Adv Gastroenterol. 2016 Jul;9[4]:457-62).
The Hopkins group has three patients with preexisting rheumatoid arthritis and two with preexisting scleroderma who have received ICIs. All three rheumatoid arthritis patients flared. Rheumatologists are trying to manage these flares so the patients can continue on their ICI. One of the scleroderma patients experienced no change in that disease while on an ICI, while the other showed a definite improvement in scleroderma symptoms.
“I think the jury’s still out in terms of what you do about ICI therapy in patients with preexisting autoimmunity. The data would say that there’s maybe a 50-50 chance of the autoimmune disease becoming worse, but, if patients have an otherwise fatal cancer, I think it’s probably worth the chance,” Dr. Bingham said.
Anecdotal reports suggest that more severe IRAEs may be a favorable prognostic sign in terms of cancer eradication, but a lot more patient experience will be needed in order to be sure, the rheumatologist said.
Dr. Bingham reported serving as a consultant to Bristol-Myers Squibb.
For the first time, new-onset connective tissue disease has been reported in patients who were treated with anti-PD1/PDL-1 agents, according to...
