Among the year's most noteworthy findings in rheumatology, includes a systemic review and meta-analysis that provides a better understanding of how common infections and cancer really are in patients treated with interleukin (IL) inhibitors. In this article, Dr. Murray offers his personal perspectives on the findings.
In October, Jawad Bilal, M.D., of the University of Arizona, published a systemic review and meta-analysis in JAMA Network Open that caused quite a bit of interest.
The review, “Risk of Infections and Cancer in Patients With Rheumatologic Diseases Receiving Interleukin Inhibitors, a Systematic Review and Meta-Analyses,” found that treatment with interleukin (IL) inhibitors appeared to be associated with an increased risk of serious infections, opportunistic infections and cancer in rheumatology patients, as compared to patients treated with placebo. To date, there is little documented evidence on the association between IL inhibitors and infections or cancer.
Dr. Bilal also reported that “the duration of drug use was significantly associated with the effect size for cancer outcome. However, there were no significant associations of duration of drug use with serious or opportunistic infections.” Plus, “the Egger regression test for small-study effect was statistically significant for serious infections but not for opportunistic infections or cancer.”
It was previously known that these drugs could cause opportunistic infections but the data on cancer and serious infections were lacking. The authors correctly speculated that due to the small number of adverse events seen in each individual trial, well this made it difficult to determine with certainty that there was a connection between interleukin inhibitors with cancer and serious infection. By grouping several trials together, the authors were able to document trends.
The report included 74 studies and 29,214 rheumatology patients who were being treated for rheumatoid arthritis, ankylosing spondylitis, familial Mediterranean fever, gout, juvenile idiopathic arthritis, psoriatic arthritis, juvenile rheumatoid arthritis, systemic sclerosis, primary Sjogren syndrome, giant cell arteritis, and systemic lupus erythematosus. They were treated with one of the following interleukin inhibitors tocilizumab (IL-6), secukinumab (IL-17), anakinra (IL-1), ixekizumab (IL-17), rilonacept (IL-1), sarilumab (IL-6), sirukumab (IL-6), ustekinumab (IL-12/23), brodalumab (IL-17), guselkumab (IL-23), clazakizumab (IL-6), and canakinumab (IL-1).
Serious infections were defined as those being serious enough to cause hospital admission, use of antibiotics or death. The following opportunistic infections were reported: 23 oral candidiasis, nine herpes zoster, four esophageal candidiasis, one unspecified candidiasis, two Mycobacterium tuberculosis, two atypical mycobacterial infections, one histoplasmosis, and six unspecified.
Oddly enough, the types of cancers that occurred and the patients receiving interleukin inhibitors had a higher risk of serious infections (odds ratio [OR] = 1.97; 95% CI, 1.58-2.44; P<0.001), opportunistic infections (OR=2.35; 95% CI, 1.09-5.05; P=0.03), and cancer (OR=1.52; 95% CI, 1.05-2.19; P=0.03).
Overall, though, the number needed to harm (NNH) was calculated to be 67 for one additional serious infection within a median follow-up of 24 weeks. The NNH for cancer was 250 (median follow-up, 28 weeks) and, for opportunistic infections, 250 (median follow-up, 54 weeks). The number to harm with cancer decreased substantially the longer the drugs were taken. By way of comparison the number needed to cause one major bleed with aspirin is 261 over 7 years and the number needed to cause new-onset diabetes with statins ranges from 125-250 if taken for one year. The one-year risk of dying from opioids, both legal and illegal is 7,569, from automobile accidents 8,000 over one year, and the one-year risk of dying by a firearm is 22,399. The numbers given for interleukin inhibitors were not risks from dying but from suffering any harm. I suspect if we used car accidents, gun shots, and opioids the number needed to harm would be much greater.
THE FINDINGS IN PERSPECTIVE
This would suggest that the taking interleukin inhibition is much riskier than other common interventions considered to be major health threats. One confounding variable is that most of the diseases treated are associated with higher risks of cancer to begin with. Various studies have suggested an increased risk of malignancy among patients with rheumatoid arthritis (RA), and malignant diseases contribute significantly to the morbidity and mortality of the disease.
Lymphoproliferative disorders occur with increased frequency in patients with rheumatoid arthritis (RA); incidence and mortality rates due to leukemia or lymphoma are approximately twofold higher than expected. The lymphoma incidence increases as active RA persists and correlates with the severity of disease activity. Non-Hodgkin lymphoma, particularly diffuse large B-cell lymphoma, is the most common type.
Psoriasis is known to increase the risk of prostate and lung cancer, as well as various skin cancers. Inflammatory bowel disease increases colon cancer risk. It would have been helpful to know what cancers were reported with each IL inhibitor. It may be possible that these agents don’t cause cancer, but rather bring it out at an earlier and possibly more treatable stage similar the effect of exogenous testosterone on prostate cancer.
Patients with immune-mediated disorders often receive other medications concomitantly with TNF-alpha inhibitors that increase the risk of cancer. As an example, methotrexate (MTX), commonly used with TNF-alpha inhibitors in the treatment of RA, is associated with an elevated risk of lymphoma in some, but not all studies. Similar considerations apply for cyclophosphamide, chlorambucil, azathioprine, and other conventional immunosuppressive medications.The cancer risk may have been underestimated by the short duration of the trials which may take many years to develop.
The current guidelines for rheumatoid arthritis stress early use of DMARDs but not necessarily with biologic agents. They recommend methotrexate or other small molecules as first line, along with nonsteroidal and prednisone when needed. They recommend that biologic drugs be reserved for more difficult cases to prevent the joint injury that comes early in the course of rheumatoid arthritis. Some reports have shown complete remission of rheumatoid arthritis if biologicals are started early. This makes this a very complicated topic and one that must be discussed with the patients.
It would be helpful to know what cancers occurred because if they were basal cell cancers, one might be less reluctant to take the risk, than if these were lung or colon cancers. Sales of these drugs are in the billions of dollars and have consistently been in the top 10 of money makers for pharma. It remains to be seen if rheumatologists change prescribing behavior after considering these findings. Obviously, the package inserts need to be revised to include cancer, and guidelines on dosing strategies, length of treatment, or combinations of drugs need to be developed.
No one takes biologic agents, or prescribes them, because we like to. The decision to use these medications has just gotten more complicated.
It is also clear that we still don’t have a complete understanding of how biologic agents work, because their manipulation can cause confounding and unpredictable results in some cases. Nonetheless there are thousands of new compounds in development because of the huge potential for profit, and because, in many cases, results have been dramatic. We must be cautious in how we apply these new technologies as this meta-analysis implies.
Simon D. Murray, M.D., is a practicing internist in New Jersey. He serves as chief medical officer for MJH Life Sciences, the parent company of Rheumatology Network.
REFERENCE: Jawad Bilal, MD, Adam Berlinberg, MD, Irbaz Bin Riaz, MD, et al. "Risk of Infections and Cancer in Patients With Rheumatologic Diseases Receiving Interleukin Inhibitors, a Systematic Review and Meta-Analysis."JAMA Network Open. October 18, 2019. DOI:10.1001/jamanetworkopen.2019.13102