To the Editor:

The review by Solomon et al about malignancies related to tumor necrosis factor (TNF) inhibitor use based on data from observational studies in rheumatoid arthritis (1) raises important issues. The authors make the point that in randomized trials, evidence of emergence of adverse events such as neoplasms often is not revealed because the trials are frequently of short duration and the patients are selected for lack of comorbid conditions. These well-recognized considerations likely lead to fewer cases of cancer occurring during a clinical trial as compared to real-life situations.

On the other hand, the important debate about whether TNF inhibitors cause or increase the risk of cancer arose after a major meta-analysis of controlled trials provided rather clear evidence of an increased incidence of cancer (2), and another indicated a potential association (3). Solomon and colleagues quote 3 other meta-analyses (4–6) of controlled trials in which no increased evidence of cancer was found in the treatment arms. However, they fail to mention that in one of these studies (4) the authors concluded “[d]espite a reassuring overall short-term risk, we could neither refute nor verify that individual anti-TNF therapies affect the short-term clinical emergence of cancer.” In the second study, no malignancy data were available for >50% of the patients included in the overall meta-analysis (5), and in the third (6), the comparators were several general-population cancer registries and not, as they should have been, the non–TNF inhibitor arms of the studies included in the meta-analysis. As related to the last study, one of us recently illustrated the danger of comparing the frequency of cancer in the general population to that in a subset of the population (7). The issue at hand was the frequency of cancer in a patient registry, but this selection bias, we propose, would equally apply to a group of patients enrolled in a trial: those patients in the parent population who develop cancer and die never get the chance to be enrolled in a registry or in a drug trial. Quite interestingly, this important bias is also probably present in the mortality data presented in the meta-analysis by Burmester et al (6). There, the comparator to the adalimumab arms was World Health Organization data on general-population mortality for specific countries. The study demonstrated that mortality decreased during the trials of adalimumab in rheumatoid arthritis, psoriatic arthritis, psoriasis, and Crohn's disease, with no deaths reported during trials of the drug in juvenile rheumatoid arthritis or ankylosing spondylitis. However, death due to any cause in the general population would exclude the possibility of the deceased being enrolled in a drug study.

Currently, with perhaps no real need for further efficacy studies of TNF inhibitors in most rheumatologic conditions, and with the ethical impossibility of conducting a controlled study addressing harm (especially when that harm is cancer), and considering the built-in methodologic problems with the available observational studies, we seem to be at an impasse on further assessment of the problem of cancer associated with TNF inhibitor use. A novel approach might be a meticulous tabulation of associated diagnoses in all cancer cases, including the cases in which no such comorbidities are present, in a reputable cancer registry or even in several such registries. The primary outcome of this survey would be the relative representation of rheumatoid arthritis (and perhaps of the other main diseases for which TNF inhibitor treatment is indicated, e.g., ankylosing spondylitis) in this registry before and during the TNF inhibitor era.


Dr. H. Yazici has received travel support, speaking fees, and/or honoraria from Abbott, Merck, Pfizer, and Roche (less than $10,000 each). Dr. Y. Yazici has received consulting fees, speaking fees, and/or honoraria from Bristol-Myers Squibb, Genentech, Janssen, Takeda, Abbott, Merck, and Pfizer (less than $10,000 each).

  • 1
    Solomon DH, Mercer E, Kavanaugh A. Observational studies on the risk of cancer associated with tumor necrosis factor inhibitors in rheumatoid arthritis: a review of their methodologies and results [review]. Arthritis Rheum 2012; 64: 2132.
  • 2
    Bongartz T, Sutton AJ, Sweeting MJ, Buchan I, Matteson EL, Montori V. Anti-TNF antibody therapy in rheumatoid arthritis and the risk of serious infections and malignancies: systematic review and meta-analysis of rare harmful effects in randomized controlled trials. JAMA 2006; 295: 227585.
  • 3
    Bongartz T, Warren FC, Mines D, Matteson EL, Abrams KR, Sutton AJ. Etanercept therapy in rheumatoid arthritis and the risk of malignancies: a systematic review and individual patient data meta-analysis of randomised controlled trials. Ann Rheum Dis 2009; 68: 117783.
  • 4
    Askling J, Fahrbach K, Nordstrom B, Ross S, Schmid CH, Symmons D. Cancer risk with tumor necrosis factor α (TNF) inhibitors: meta-analysis of randomized controlled trials of adalimumab, etanercept, and infliximab using patient level data. Pharmacoepidemiol Drug Saf 2011; 20: 11930.
  • 5
    Leombruno JP, Einarson TR, Keystone EC. The safety of anti-tumour necrosis factor treatments in rheumatoid arthritis: meta and exposure-adjusted pooled analyses of serious adverse events. Ann Rheum Dis 2009; 68: 113645.
  • 6
    Burmester GR, Mease P, Dijkmans BA, Gordon K, Lovell D, Panaccione R, et al. Adalimumab safety and mortality rates from global clinical trials of six immune-mediated inflammatory diseases. Ann Rheum Dis 2009; 68: 18639.
  • 7
    Yazici H, Tascilar K, Kiroglu G, Erar A, Duransoy ML. Mortality bias in reporting cancer incidence in disease registries: comment on the article by Chen et al [letter]. Arthritis Rheum 2011; 63: 25434.

Hasan Yazici MD*, Yusuf Yazici MD†, * University of Istanbul, Istanbul, Turkey, † New York University Hospital for Joint Diseases, New York, NY.