Do TNF Blockers Increase Cancer Risk in Children?

Laurie Barclay, MD

July 30, 2010

July 30, 2010 ( UPDATED August 3, 2010 ) — Children treated with tumor necrosis factor-alpha (TNF-alpha) blockers may have increased cancer risk, but cancer cases reported in these children were confounded by underlying illnesses and concomitant use of immunosuppressants, according to the results of a study reported in the August issue of Arthritis & Rheumatism.

"The risk of malignancy should be considered prior to initiating TNF blockers in children," lead author Peter Diak, PharmD, MPH, safety evaluator from the US Food and Drug Administration (FDA), told Medscape Medical News. "Clinicians should be aware of the increased risk of malignancy in children receiving these products, and they should discuss the benefits and risks with their patients and the patient's caregivers."

Three TNF blockers are FDA-approved for pediatric use: etanercept and adalimumab for juvenile idiopathic arthritis (JIA), and infliximab for Crohn's disease. Although previous evidence suggests that some adult subgroups may be at an increased risk for malignancy with TNF-alpha blockers, the extent to which immunosuppression from TNF-alpha blockers may also increase the risk for malignancy in children is still undetermined.

"For the time being, and in the absence of a reasonable chance to evaluate whether the reported cases represent a higher-than-expected number of cancers in a population of patients with JIA severe enough to merit treatment with TNF-antagonists, this finding will remain a concern, rather than a fact," said Johan Askling, MD, PhD, an associate professor of epidemiology and rheumatology at the Karolinska Institutet in Stockholm, Sweden, told Medscape Medical News when asked for independent comment.

Dr. Askling recommended that the finding "be communicated together with other potential adverse events but not preclude treatment with these drugs when indicated."

In children from birth to 19 years, cancer is very rare, with the US annual incidence rate being 16.8/100,000 vs 469.7/100,000 for all cancers in adults. These rates were derived from the National Cancer Institute's Surveillance, Epidemiology, and End-Results database and age-adjusted to the 2000 US population.

Study Design and Findings

Because reports of cancer in children treated with TNF-alpha blockers have raised concerns about potentially greater risk, the investigators assessed postmarketing reports of malignancy in this population by searching the Adverse Event Reporting System database. Cancers in children who started treatment with infliximab, etanercept, and adalimumab between the ages of 0 and 18 years were identified, and reporting rates were compared with the background rate of cancers in the general pediatric population.

Of 48 reports of cancers in children identified through the Adverse Event Reporting System, 31 occurred after infliximab use, 15 after etanercept use, and 2 after adalimumab use. Median duration of TNF-alpha blocker treatment was 30 months, suggesting that these cases were not preexisting cancers misdiagnosed as a rheumatic disease. Lymphomas, including both Hodgkin's and non-Hodgkin's lymphoma, accounted for half of the reported malignancies, and leukemia, melanoma, solid-organ cancers, and other malignancies accounted for the remainder.

Use of other immunosuppressants concomitantly with TNF blockers was reported in 88% of cases.

  • Infliximab: Compared with background rates in the general pediatric population, reporting rates with infliximab were consistently higher for lymphomas and all malignancies. The reporting rate for lymphomas was 44/100,000 patient years — 18 times the background rate, or 9 times the background rate if cases of hepatosplenic T-cell lymphoma were excluded.

  • Etanercept: For etanercept, reporting rates were 5 times the background rate for lymphomas (11/100,000 patient years), but for all malignancies, reporting rates were similar to background (22/100,000 patient years). Reporting rates were not calculated for adalimumab because there were only 2 reported cases, and both involved previous use of other TNF blockers.

"There is evidence that treatment with TNF-a blockers in children may increase the risk of malignancy," the study authors write. "However, the cases were confounded by the potential risk of malignancy associated with underlying illnesses and the use of concomitant immunosuppressants; therefore, a clear causal relationship could not be established."

Study Strengths and Limitations

Limitations of this study include possible underestimation of the true incidence rate for these events in the pediatric population because of substantial underreporting to FDA's MedWatch program, long latency period for malignancy, and confounding as noted earlier.

Dr. Askling calls this study a "numerator only" study, in which a strength is that spontaneously reported cases are presented in some detail. However, the denominator, detailing how many cases would have been expected, and with what characteristics, is largely missing, which is a limitation.

"This type of study is important, as they may generate new safety signals, but they are inherently limited when it comes to evaluation of the signals generated, which adds to the concern they generate," Dr. Askling said. "This conundrum is indeed pointed out by the authors."

He said that the study looks at cancer occurrence among patients with a broad range of diseases treated with TNF-alpha inhibitors and compares it with the expected cancer occurrence rate in the general pediatric population.

"Strictly speaking, the study result may reflect an association with TNF-alpha inhibitors, with the disease in itself (in particular, severe disease, such as is the case in adults), or with the accompanying [immunosuppressive] medications, and we cannot tell which is true," Dr. Askling continued. "Had there been information on the occurrence of cancer in patients with pediatric [inflammatory bowel disease] and JIA, things would have been much easier."

In an accompanying editorial, Thomas J.A. Lehman, MD, chief of pediatric rheumatology at Hospital for Special Surgery, and professor of clinical pediatrics at Weill Medical College of Cornell University, in New York City, notes that 25 of the 48 cases of malignancies were reported in children with inflammatory bowel disease who were also receiving immunosuppressants. He therefore suggests that the study findings are not directly relevant to children with JIA.

"The FDA study used a reporting system and collected cases from around the world," Dr. Lehman told Medscape Medical News. "This makes it very difficult to assess the accuracy of the reports, the full nature of the compounding medications, and the size of the population from which the reported cases have been drawn. In many cases, the children had been on multiple medications associated with an increased risk of malignancy, and the role of the TNF blocker cannot be clearly assessed."

Clinical Implications

When asked which therapeutic options clinicians should now consider for children with conditions usually treated with TNF-alpha blockers, Dr. Diak noted that clinicians continue to have all therapeutic options previously available, including TNF inhibitors.

"Therapeutic alternatives to TNF inhibitors differ depending on the condition being treated," Dr. Diak said. "The decision of which product(s) provide the optimal benefit vs risk balance is one that is best made on an individual level, by clinicians in conjunction with patients and their caregivers, taking into account variables such as the severity and impact of the underlying disease."

Although Dr. Diak said that there are no specific monitoring recommendations at this time for children treated with TNF blockers, patients and their families should tell their physicians whether they have had any type of cancer or a family history of cancer, and they should discuss with their physicians any need to adjust the medicines that they may be taking.

"The clinical implications of the study regarding the safety of TNF-alpha blockers in children with arthritis are very limited," Dr. Lehman said. "Although it appears that extra concern is warranted in treating children with inflammatory bowel disease, there is no evidence that the incidence of malignancies in children with arthritis was increased above that due to the disease. We have long known that there is an increased incidence of malignancies associated with the use of methotrexate, yet this drug has been a mainstay of our therapy for children with arthritis. The FDA report is a cause of concern for parents and children alike but is not a justification for withholding TNF blockers from the children."

Dr. Lehman and Dr. Askling agree that based on evidence to date, no additional clinical monitoring seems warranted for children who have been on TNF-alpha blockers. Dr. Askling recommends additional research on the occurrence of cancer in biologics-naive as well as in biologics-treated patients with pediatric rheumatic conditions.

"Luckily, such data from several different populations are in the pipeline," Dr. Askling said. "In the near future we will be much better equipped to evaluate this risk."

"Additional research areas would include determining the background and incidence rates of malignancy in pediatric patients with immunological diseases, JIA and Crohn's, with and without biologic and other concomitant immunosuppressive therapies," Dr. Diak concluded.

The study authors have disclosed no relevant financial relationships. Dr. Diak and Dr. Askling have disclosed no relevant financial relationships. Dr. Lehman has received speaking fees from Abbott, Amgen, Pfizer, and Wyeth (less than $10,000 each).

Arthritis Rheum. 2010;62:2183-2184, 2517-2524 .


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.