• Decrease font size
  • Return font size to normal
  • Increase font size
U.S. Department of Health and Human Services


  • Print
  • Share
  • E-mail

Questions and Answers - TNF Blockers 8/25/2009


The U.S. Food and Drug Administration (FDA) recently communicated about increased risk of malignancies associated with the use of TNF blockers in children and adolescents Tumor Necrosis Factor (TNF) Blockers (marketed as Remicade, Enbrel, Humira, Cimzia, and Simponi). This question and answer sheet provides additional information concerning FDA’s analysis of these cases. 

Q1: What is the denominator we used in our calculations?

A: For the reporting rate denominator, we used the estimated number of patient treatment years for the domestic use of infliximab and etanercept, supplied by the manufacturers.  Patient treatment years were estimated for infliximab to be 22,645 treatment years for ages 0-16 years from 2003 through 2007, and for etanercept to be 26,800 treatment years for ages 0-17 years from 1998 through 2007.  The reporting rate numerator, for these respective dates and ages, was 15 domestic cases of pediatric malignancy for infliximab (including hepatosplenic T-cell lymphomas in patients with inflammatory bowel disease) and 6 domestic cases of pediatric malignancy for etanercept.  However, the malignancy reporting rates for the different TNF blockers cannot be directly compared, because the drugs are largely prescribed for patients with different diseases on different background medications.  For example, the primary use of infliximab in children has been for the treatment of inflammatory bowel disease (IBD) while the majority of etanercept use has been for the treatment of juvenile idiopathic arthritis (JIA).  The age and date cutoffs were different between products, when calculating reporting rates, due to our available drug use data.  Up to this date, limited population based drug use data for parenterally administered drugs have been available to the FDA.

Q2: Over how many years were the 48 malignancies reported? What were the precise malignancies and number of each type?

A: The FDA received the 48 reports of pediatric malignancy over an eight year period (2001-2008; see Table 1).  Note, the 10 pediatric cases of anti-TNF-associated hepatosplenic T-cell lymphoma were reported in IBD patients also treated with 6-MP or azathioprine.

Table 1. Types of pediatric malignancy reported to FDA from 2001-2008.
Type of Malignancy# cases Type of Malignancy

# cases

Hepatosplenic T-cell lymphoma


 Renal cell carcinoma


Non-Hodgkin’s lymphoma


 Hepatic malignancy


Hodgkin’s lymphoma


 Metastatic hepatocellular cancer




 Malignant mastocytosis


Malignant melanoma




Thyroid cancer


 Colorectal cancer


Basal cell carcinoma


 Yolk sac tumor


Lymphoma and AML






 Bladder cancer







Q3: How many cancers were in children with inflammatory bowel disease and how many with JIA? How many of the children with IBD were also on 6-MP (rarely used by rheumatologists)?

A: The number of malignancy cases by indication are presented in Table 2. Thirteen cases (11- Crohn’s disease, 2-ulcerative colitis) also reported concomitant 6-MP exposure.

Table 2. Number of malignancy cases by indication
Indication of TNF blocker therapy# cases
Ankylosing spondylitis


Crohn's disease


In utero exposure




Psoriatic arthritis




Ulcerative colitis





Q4: How many children with JIA have been exposed to anti-TNF agents?

A: The drug use data supplied by the manufacturers of the three TNF blockers did not include unique patient counts by indication. However, the data estimated that 14,837 children ages 0-18 years received infliximab through February 2008 (source – Centocor),  9,200 children ages 0-17 years received etanercept through December 2007 (source – Amgen), and 2,636 children ages 0-16 years received adalimumab for the two year period of 2006-2007 (Source: IMS Health, NPA Plus™, 2006-2007.  Information derived by Abbott from data provided by Wolters Kluwer Health, Source Lx, 2006-2007).

Q5: What is the background incidence of malignancy in children with JIA? In children with JIA on methotrexate?

A: The background incidence of malignancy in children with JIA is not well defined.  It is unclear if the risk of malignancy is increased in children with JIA as the data are more limited than the data from adults indicating an increased risk of lymphoma in adults with RA.  Furthermore, the magnitude of this potential risk in adults or children is difficult to estimate from the epidemiological studies due to multiple biases of case-control and cohort studies (for example: sampling bias and differential measurement bias).

Q6: In the HCP we provided a comparison of the reporting rates for Remicade and for Enbrel in comparison to the background rates in the general population.  Can we provide confidence intervals around these ratios?

A: We provided point estimates of reporting rates but not confidence intervals.  Because of the limitations of AERS, in particular the unknown extent of underreporting, we believe that calculating confidence intervals would convey a degree of precision which we believe to be lacking.

Q7: Did FDA announcement include only U.S. or both U.S. and non-U.S. cases?

A: The FDA announcement of the 48 cases of pediatric malignancy included both U.S. and non-U.S. cases.  There were 32 U.S. cases and 16 non-U.S. cases of pediatric malignancy reported to the FDA.

Q8: In the reported cases how many on etanercept at any point?  How many on adalimumab at any point? How many on infliximab at any point? 

A: Thirty-one cases of malignancy in children taking infliximab were reported, including 10 cases of hepatosplenic T-cell lymphoma in IBD; 15 cases of malignancy in children receiving etanercept were reported; two cases of malignancy in children receiving adalimumab were reported.

Q9: In the 20 cases of children treated for rheumatic conditions how many had received etanercpet?  How many infliximab? How many adalimumab?

A: Five cases of malignancy in children taking infliximab for rheumatic conditions were reported; 14 cases of malignancy in children receiving etanercept for rheumatic conditions were reported; one case of malignancy in a child receiving adalimumab for rheumatic conditions was reported.

Q10: What were the numbers of patients in the various JIA subtypes (JIA is not RA and is quite diverse in its genetic subtypes)?

A: This level of clinical detail was not reported in the spontaneous AERS reports.

Q11: Is there a dose association with malignancy? 
A: No, the 48 cases of pediatric malignancy did not confirm a dose association with malignancy.