WASHINGTON – Tumor necrosis factor inhibitors don’t appear to confer any additional cancer risk upon children with juvenile idiopathic arthritis above the increased incidence of cancer that comes hand in hand with the disease itself.
In 2009, the drugs came under suspicion of boosting the already-known increased cancer risk in these patients, Timothy G. Beukelman, MD, said at the annual meeting of the American College of Rheumatology. But the large database review that he conducted with his colleagues doesn’t validate those fears.
“I feel fairly confident now that I can stand in front of parents and say that we can treat their child effectively without putting that child at an even higher risk of a malignancy,” he said in a video interview.
The video associated with this article is no longer available on this site. Please view all of our videos on the MDedge YouTube channel.
Thomas J.A. Lehman, MD, chief of pediatric rheumatology at the Hospital for Special Surgery, New York, and professor of clinical pediatrics at Weill Cornell Medical College in New York, agreed.
“This study again indicates that anti-TNF therapy does not increase the risk of cancer for children with arthritis,” he said in an interview. “Although children with rheumatic diseases have a small increased background risk of malignancies, this is independent of the use of anti-TNF therapies. For physicians who have cared for children in the era when we did not have anti-TNF therapies available, it is clear that any minor risks associated with these medications are far outweighed by their dramatic benefits.”
In the last few years, five large studies have found that children with juvenile idiopathic arthritis (JIA) have a two- to sixfold increased malignancy risk, compared with the general pediatric population. However, only two of those studies included children taking TNF inhibitors, who comprised just 2% and 9% of those study populations.
In 2009, based on voluntary adverse event reporting, the Food and Drug Administration issued a black box warning on TNF inhibitors, citing a possibly increased risk of cancer in children and adolescents who received the drugs for JIA, inflammatory bowel diseases, and other inflammatory diseases.
Shortly thereafter, a report identified a fivefold increase in the risk of childhood lymphoma associated with the medications (Arthritis Rheum. 2010 Aug;62[8]:2517-24). Other studies have not borne this out, but the boxed warning stands.
To further explore the association, Dr. Beukelman of the University of Alabama, Birmingham, and his associates examined billing data from two large national billing databases: the National U.S. Truven MarketScan claims database and Medicaid billing records. Together, the databases contained information on 27,000 children with JIA who received a prescription for a TNF inhibitor any time during 2000-2014. Cancer rates in this population were compared with those seen in a cohort of 2.64 million children with attention-deficit/hyperactivity disorder who were included in the national Surveillance, Epidemiology, and End Results (SEER) database. The investigators chose individuals with ADHD as a control group because of ADHD’s chronicity and lack of any association with cancer risk.
Dr. Beukelman also performed a within-group analysis on the JIA patients, comparing cancer rates among those treated with a TNF inhibitor and with methotrexate. The mean follow-up for patients who took TNF inhibitors was 4 years (median of 1.4 years), but there were a full 14 years of data for some patients.
Among the controls, with more than 4 million person-years of follow-up, there were 727 cases of any malignancy – a standardized incident rate (SIR) of 1.03. Among all children with JIA, with more than 52,000 person-years of follow-up, there were 20 malignancies. The SEER database predicted eight among a sex- and age-matched cohort of healthy children. This translated to an SIR of 2.4. This represents the baseline increased risk of cancer conferred by JIA alone.
Nine malignancies occurred in the subgroup of children with JIA who took no medications. The SEER expectation among this group was 3.8 cancers, also translating to an SIR of 2.4
One malignancy occurred in the group treated with methotrexate only. Among these children, the SEER expected number was 1.9; the SIR in this group was 0.53.
Seven malignancies occurred among children who took TNF inhibitors, translating to an SIR of 2.9. Six occurred in children who took a TNF inhibitor in combination with or without methotrexate – an SIR of 3.0.
A final group consisted of children who took a wide range of other medications used in JIA (abatacept, anakinra, canakinumab, rilonacept, rituximab, tocilizumab, ustekinumab, tofacitinib, azathioprine, cyclosporine, gold, leflunomide, mycophenolate mofetil, tacrolimus, thalidomide, lenalidomide). This group also included patients who may or may not have taken methotrexate or a TNF inhibitor. Among these, there were four cancers when the SEER expected number was 0.7. This translated to an SIR of almost 6 – a surprising finding, Dr. Beukelman said. But since there were only four cancers and the group was exposed to so many different medications, it’s tough to know what that means, if anything, Dr. Beukelman said.
“There’s a lot to unpack here. The treatment paradigm for JIA is methotrexate followed by a TNF inhibitor if that’s ineffective. So these kids were on all of these more uncommon drugs,” suggesting that neither TNF inhibition nor methotrexate worked. “Some of these patients might actually have had systemic arthritis, Still’s disease, which is a completely separate thing, and we don’t know anything about the risk of malignancy in that. They might have an even higher rate of malignancies at baseline due to having worse disease, or uncontrolled inflammation. It is concerning, but I think it probably speaks to the fact that these patients are difficult to treat and probably at higher risk.”
Dr. Beukelman didn’t specifically break out the types and numbers of cancer, except to say that 3 of the 20 were lymphomas. The rest were leukemias and brain cancers – a finding that reflects the general pattern of childhood malignancies.
“Unfortunately, the most common childhood cancers are lymphomas, leukemias, and brain cancers, and that is what we saw in this study as well,” he said.
The study was supported by the U.S. Agency for Healthcare Research and Quality. Dr. Beukelman noted that he has received consulting fees from Novartis, Genetech/Roche, and UCB.
msullivan@frontlinemedcom.com
On Twitter @alz_gal