What is the Risk of Lymphoma With Treatments?

Issue: December 2016

A: During the past 2 decades, the treatment of inflammatory bowel disease has undergone a paradigm shift. Whereas previous treatment was largely reactive, with drugs being used in response to symptoms, current treatment aims require a more aggressive approach in an attempt to alter the natural history of the disease.

Accordingly, the use of immunomodulating drugs has increased rapidly and it is therefore unsurprising that this has resulted in an increase in treatment-related morbidity. Of the multitude of adverse events that can occur in association with drugs used in the management of inflammatory bowl disease (IBD), malignancy is probably the one that creates the greatest anxiety in both patients and physicians.

Is There an Underlying Increase in the Risk of Lymphoma in People With IBD?

Data from the rheumatological literature have suggested that patients with rheumatoid arthritis (RA) are at increased risk of developing lymphoma, with disease activity a major risk factor. However, studies investigating whether IBD itself increases the risk of developing lymphoma have failed to show a clear relationship.

Although several single-center cohort studies have described an increased rate of lymphoma in IBD, such studies are open to bias. In contrast to this, population-based studies have, in general, shown no increase in lymphoma rate in patients with Crohn’s disease (CD) or ulcerative colitis (UC). For example, a large primary care study compared the rate of Hodgkin’s and non-Hodgkin’s lymphoma (NHL) among almost 17,000 patients with IBD in the United Kingdom, and more than 60,000 age- and sex-matched controls. Average follow-up was approximately 4 years, and the relative risk (RR) of lymphoma among patients with IBD was not found to be increased (risk ratio, 1.20). Subanalysis confirmed this was true when patients with CD and UC were examined separately. Furthermore, patients who developed lymphoma had similar rates of exposure to medication and need for surgery as those who did not develop lymphoma, suggesting lymphoma risk was not clearly influenced by disease activity. Nevertheless, studies specifically examining whether disease activity increases lymphoma risk, as it appears to in RA, have yet to be performed.

Treatment of IBD and Lymphoma Risk

The link between immunosuppression and an increased risk of malignancy, including lymphoma, has been increasingly recognized in recent years. The evolution of immunosuppressive therapy to allow for organ transplantation heralded an appreciation of the risks of such treatment and the development of post-transplant lymphoproliferative disease (PTLD), along with an understanding of its etiopathogensis. Such knowledge has informed our understanding of some of the risks associated with the use of immunosuppressants in IBD.

Thiopurines

Whereas the population-based study previously described showed no increased risk of lymphoma in patients receiving thiopurines, other studies have shown the risk of developing lymphoma is increased in people taking thiopurines. A meta-analysis of pooled data from six cohort studies described a relative risk of 4.18 of developing lymphoma for people with IBD taking thiopurines compared with people with IBD not taking azathioprine (AZA) or 6-mercaptopurine (6-MP).

Subsequently, a large French multicenter study (Cancers et Sur-Risque Associé aux Maladies Inflammatoires Chroniques Intestinales en France [CESAME]) addressed this question and found remarkably similar findings (hazard ratio, 5.28). As with previous studies, the overall incidence of lymphoma was extremely low (0.9/1000 per patient-years). However, it was possible in this study to stratify the risk by age (Table). Although the risk of developing lymphoma was 0.37/1000 patient-years in people younger than 50 years, this rose to 5.41/1000 patient-years in those older than 65 years. Another useful finding was the increased risk of lymphoma did not persist after the thiopurine was discontinued. This information has particular relevance for the communication of treatment plans and risks.

Distinguishing whether the increased risk of lymphoma identified in these studies related to active disease or to thiopurine exposure could not be addressed. However, the pathogenesis of PTLD is potentially instructive. Unlike most lymphomas that develop in people who are not immunosuppressed, PTLD is largely driven by Epstein-Barr virus (EBV). In a series of patients with IBD who developed lymphomas, more than 80% of those who received thiopurines had positive staining for EBV in biopsies taken from the lymphoma compared with 20% of those taken from people who did not receive immunosuppression. Accordingly, it seems likely most lymphomas that develop in people treated with thiopurines are a result of their immunosuppressive effects.

PAGE BREAK

In recent years, concern has arisen about a rare but generally fatal lymphoma, hepatosplenic T-cell lymphoma (HSTCL), which has been seen in patients with IBD receiving thiopurines. Although the importance of HSTCL in IBD first came to light following the description of its development in a small number of young people with CD treated with a combination of anti-tumor necrosis factor (TNF) therapy and thiopurines, it has become apparent it also can occur in the context of thiopurine monotherapy. In addition, in the CESAME cohort, two cases of fatal lymphoproliferative disease occurred in young men following primary EBV infection; this highlights another possible association between AZA and aggressive lymphoproliferative disorders.

Summarizing the evidence, there appears to be a convincing link between the use of thiopurines and the development of lymphoma, largely driven by EBV. The risk is probably greater in men than in women and is somewhere in the region of three fold- to four-fold over the background risk; therefore, the absolute risk varies markedly with age. The risk is probably constant over time and, while evidence to support this is lacking, probably relates to the degree of immunosuppression.

Methotrexate

In comparison with its use in RA, methotrexate (MTX) is not commonly used in IBD. Similarly, thiopurine prescription greatly outweighs MTX use in IBD. Accordingly, quantifying the risk of MTX use in IBD in terms of lymphoma development is difficult. Looking at the rheumatological literature, concern was raised by reports of increased incidence of lymphoma among people being treated with MTX, particularly because some were EBV positive. However, consistent evidence with regard to this is lacking, and evidence among the IBD cohort is almost nonexistent. Thus, although it is reasonable to suppose that MTX does not greatly increase the risk of lymphoma, it is not possible to exclude a risk associated with its use.

Anti-TNF Agents

Quantifying the risk of anti-TNF therapy in patients with IBD has been clouded by the fact that most patients also have been exposed to immunomodulators, often at the same time they are receiving anti-TNF. However, the issue was brought into sharp focus with the appearance of a small number of cases of HSTCL in association with combined anti-TNF and thiopurine therapy. Initially, this condition was reported largely within the pediatric population; however, its occurrence has subsequently been reported in a wider age range. Fortunately, despite increasing use of anti-TNF agents, the incidence of HSTCL does not appear to be increasing in parallel. Of note, cases are also being reported in patients treated with thiopurines alone, such that it is not clear whether anti-TNF is involved in the etiology of HSTCL or whether it relates to thiopurine use alone. In any event, although it is difficult to quantify the risk of developing HSTCL in association with thiopurine monotherapy or anti-TNF/thiopurine combination therapy, it seems reasonable to believe its incidence is significantly lower than that of immunosuppression-associated NHL.

Although initial concerns about the risks of anti-TNF therapy and lymphoma arose from increased reports of lymphoma incidence in patients with RA treated with infliximab, this was not reflected in a number of case series and registries comprosed of patients with IBD. However, a recent meta-analysis attempted to quantify the risk of NHL associated with anti-TNF use in IBD. In 26 studies including nearly 9,000 patients and more than 20,000 patient-years of follow-up, the lymphoma rate was approximately six per 10,000 patient-years. This equated to a standardized incidence ratio of 3.23. Most the patients also had been exposed to immunomodulators, so it was not possible to identify whether anti-TNF monotherapy is associated with an increased risk of lymphoma. However, in this analysis, the risk of lymphoma was higher in patients exposed to anti-TNF (mostly combination therapy) than thiopurines alone, suggesting that there is probably an added risk of anti-TNF therapy on top of thiopurines.

Conclusion

There is little doubt the use of thiopurines is associated with an increase in risk of developing NHL. Although the magnitude of this increase is approximately four times the background risk, the absolute incidence of lymphoma remains extremely low. However, the risks are probably greater with increasing age and in men, but, importantly, regress to baseline after discontinuing treatment. The picture for both MTX and anti-TNF therapy is less clear, which is also the case for the factors associated with the development of HSTCL, although the use of thiopurines seems highly likely to be implicated.

However, although immunosuppression increases the risk of lymphoproliferative disorders, the benefits of treatment often outweigh the risks. Therefore, it is essential the risks of therapy are communicated clearly to patients in terms they can understand, in comparison with the risks of undertreatment of active disease, and in the context of life, which is, after all, a risky business.

References:
Beaugerie L, et al. Lancet. 2009;doi:10.1016/50140-6736(09)61302-7.
Dayharsh GA, et al. Gastroenterology. 2002;122:72-77.
Dias C, et al. Nat Rev Rheumatol. 2011;doi:10.1038/hrrheum.2011.62.
Jones JL, et al. Inflamm Bowel Dis. 2007;13:1299-1307.
Kandiel A, et al. Gut. 2005;54:1121-1125.
Kamel OW, et al. Hum Pathol. 1994;25(7):638-643.
Kotlyar DS, et al. Clin Gastroenterol Hepatol. 2011;doi:10.1016/j.cgh.2010.09.016.
Lewis JD, et al. Gastroenterology. 2001;121:1080-1087.
Lichtenstein GR, et al. Am J Gastroenterol. 2014;doi:10.1038/ajg.2013.441.
Siegel CA, et al. Clin Gastroenterol Hepatol. 2009;doi:10.1016/j.cgh.2009.01.004.