Does Thiopurine Therapy Increase Cancer Risk?

Patients with ulcerative colitis (UC) have a 4-fold increase in risk of lymphoma during treatment with thiopurines, compared to UC patients who have not been treated with these drugs, according to a nationwide cohort study in the November issue of Gastroenterology. The risk increases gradually with successive years of therapy, but decreases when patients stop taking the drugs.

UC is a chronic inflammatory disorder; it is commonly treated with corticosteroids, which cause a number of side effects (weight gain, mood disorders, and osteoporosis) and often lack long-term efficacy. Patients are therefore often treated with other drugs, such as thiopurines (azathioprine and 6-mercaptopurine)— purine anti-metabolites with immunosuppressive properties.

Although these drugs spare patients from the potential adverse effects of corticosteroids, they have their side effects, increasing the risk of lymphoma.

Thiopurines are also used to treat patients with lymphoproliferative disorders, including lymphoma, following organ transplantation. They are reported to increase the risk of lymphoproliferative diseases by up to 20- and 200-fold in kidney and heart transplant recipients, respectively. However, the level of risk for patients with inflammatory bowel disease (IBD) treated with thiopurines is not clear—studies have produced conflicting results.

The Veterans Affairs (VA) administration has the largest integrated health care system in the United States, serving approximately 8.3 million veterans each year. Nabeel Khan et al. used the VA population-based database to estimate the risk of lymphoma among 4734 patients with UC treated with thiopurines for a median time of 1 year.

They found that incidence rates of lymphoma were 0.60 per 1000 person-years among patients who had not been treated with thiopurines, 2.31 among patients who were treated with thiopurines, and 0.28 among patients who had discontinued treatment with thiopurines. Treatment with thiopurines was associated with an approximate 4-fold increase in the risk of lymphoma relative to patients with UC who were not treated with thiopurines (see below figure).

Incidence rate of lymphoma, stratified by age and use of thiopurines.

Khan et al. showed that risk increased with the duration of treatment—the incidence rates of lymphoma during the first year, second year, third year, fourth year, and >4 years of thiopurine therapy were 0.9, 1.6, 1.6, 5, and 8.9 per 1000 person-years, respectively. Furthermore, stopping thiopurine treatment reduced the risk of lymphoma during a median follow-up time of 3.5 years.

After adjusting for age, race, and use of thiopurines, men had a nonsignificant trend toward a higher incidence of lymphoma.

How might thiopurines contribute to development of this cancer? Decreased immune surveillance of Epstein–Barr virus (EBV)-infected B cells is one way (thiopurines been implicated in the development of EBV-positive lymphomas). They also destabilize DNA by incorporating thiopurine nucleotides during replication, which interferes with replication and repair mechanisms to cause mutations.

Khan et al. state that in the absence of a comparable therapy with a lower risk of malignancy, patients with corticosteroid-dependent IBD will still be treated with thiopurines. However, there are several approaches to reduce cancer risk. Candidates for thiopurine therapy should be selected with caution—especially older men. Patients should be assessed for thiopurine methyltransferase genotype or enzyme activity, and weight-based dose adjustments can be made. Finally, stopping thiopurine therapy after 3 years should be considered, especially for patients in long-term remission.

In an editorial that accompanies the article, Laurent Beaugerie reminds us that in patients with longstanding extensive colitis, the reduction of colorectal cancer risk may outweigh the excess risk of lymphoma from thiopurines.

Khan et al. warn that although the study is based on the largest integrated nationwide health care system in the US, it is not a population-based cohort, due to the differences between the demographics of the VA population and the general US population (specifically in age, race, and sex distribution). The VA population is predominantly composed of middle-age to older white men, which can limit the external validity of the study.

Beaugerie says that randomized, mid- or long-term strategy and benefit–risk studies on the prolonged use of thiopurines beyond 5 years, based on age, sex, and IBD phenotype, are needed.