Individuals with IBD do not seem to incur an intrinsic risk of developing lymphoma. However, there is evidence to suggest that either alone or in combination with anti-TNF, the use of thiopurines increases the risk of lymphoma, albeit rarely. Many of these reported lymphomas occur concomitantly with primary EBV infection. Furthermore, hepatosplenic T-cell lymphoma (HSTCL) is a rare but concerning issue in young adult males with IBD who have been exposed to at least 2 years of thiopurines. Examining the rheumatoid arthritis (RA) literature, it is interesting to note that exposure to anti-TNF therapy with or without MTX does not seem to carry the same risk of development of lymphomas.[68–70]
Lewis et al. concluded that the relative risk of lymphoma for CD/UC compared with matched controls was 1.2 with standardized incidence ratio 1.32 (0.78–2.10) and that there was no evidence of increased risk of lymphoma in patients with IBD compared with the general population. In this study cohort, immunomodulators (AZA or 6-mercaptopurine) were only used by 13% of CD patients and 6% of UC patients.[71]
Several studies have suggested that there is an increased risk of lymphoma in IBD patients both on thiopurine alone and in combination with anti-TNF.[72,73] Beaugerie et al. studied a large prospective cohort of adults with IBD with median follow-up of 35 months.[74] The three groups included: thiopurine continued (5867) with 9% also on anti-TNF; thiopurine discontinued (2809) with 12% also on anti-TNF; and never on thiopurines (10,810) with <1% on anti-TNF. Duration of disease ranged from 7.5 to 10.5 years among the three groups. Twenty three lymphomas were reported (14 non-Hodgkin's lymphoma [NHL] and one Hodgkin's lymphoma) on continued thiopurines compared with two NHL in discontinued thiopurines and six NHL in never thiopurines. The youngest case was in a 20-year-old male. The multivariate adjusted hazard ratio was 5.26 (2.2–12.6) in those continuing thiopurines compared with others. Furthermore, ten out of 15 lymphoma cases in the continuing thiopurine group exhibited post-transplant lymphoproliferative disorder with EBV positivity. Only two out of eight cases were EBV positive in the naive thiopurine group. The authors concluded that the thiopurines increased the risk of developing lymphoproliferative disease with risk factors including EBV presence, older age, male gender and longer duration of IBD.[74]
Dayharsh et al. described a similar association between EBV and the risk of lymphoma.[75] In a retrospective review of adults at the Mayo clinic, 18 patients with lymphoma were identified: EBV infection was found in five out of six patients on thiopurine for a median of 3.5 years, versus two out of 12 patients never receiving thiopurines. The authors hypothesized that post-transplant lymphoproliferative disorder probably occurs because of reduced cell-mediated immune surveillance allowing EBV-infected lymphocytes (especially B cells) to proliferate.[75]
In the pediatric literature, Ashworth et al. performed a retrospective review at a single pediatric hospital to examine the relationship between IBD and lymphoma.[76] Among the 1374 patients, 58% had CD, 34% had UC and 3% were IBD undetermined. Two lymphoma cases were observed among 2574 patient-years of taking thiopurine (2 per 4441 past years of ever taking thiopurine). Both patients were male, aged 12 and 18 years old, with CD and UC, respectively. Both patients had been taking thiopurines (22 and 28 months thiopurine use) but were EBV negative. A third patient had hemophagocytic lymphohistiocytosis on thiopurines. Compared with the Surveillance Epidemiology and End Results Program data (0.58 per 10,000 patient-years) the rate of lymphoma was three out of 10,000 patient-years in total and 4.5 lymphoma per 10,000 patient-years on thiopurines. Interestingly, no malignancies were observed in 1140 patient-years of combination anti-TNF and thiopurine therapy or 694 patient-years of MTX therapy.[76]
HSTC lymphoma is a rare but concerning issue for medical teams caring for young adults and children with IBD. Parakkal et al. examined the US FDA adverse event reporting system database for lymphomas and use of biologic agents.[77] They found 25 cases of HSTCL. Twenty two (88%) had IBD; 19 had CD and three had RA. Ninety six percent (24/25) were on anti-TNF and immunomodulation. Seventeen out of the 25 patients were less then 35 years old, on combination thiopurine and anti-TNF therapy and 16 out of 17 were male. Three elderly patients were never exposed to thiopurine (two on anti-TNF and MTX and one on anti-TNF alone).[77] Kotlyar et al. reviewed published reports and MedWatch for reports of HSTCL. Thirty six patients were identified with HSTCL; 20 were on IFX and thiopurine and 16 were on thiopurine alone. The median duration of thiopurine was 5.5 years in the combination group and 6 years in the thiopurine monotherapy group. Twenty seven out of 30 were less than 35 years old and 28 out of 30 were male. All but one of the IBD patients with HSTCL had been taking thiopurines for at least 2 years before developing lymphoma.[78] HSTCL is a very rare but concerning issue particularly in young adult males with IBD, most commonly with at least a 2-year exposure to thiopurines and often in combination with anti-TNF therapy.
An alternative practice, substituting the immunomodulator MTX for a thiopurine in combination with biologic therapy, warrants consideration. In a recent review examining the long-term safety of MTX therapy in RA, Salliot and van der Heijde concluded that there did not seem to be an increased risk of lymphoma in adults receiving MTX monotherapy with mean dose of 10.5 mg weekly (range: 4.6–18 mg).[79] The majority of experience of combination biologic therapy with MTX arises from the rheumatology literature, where a biologic is almost always used in combination with MTX. Although the clinical trial literature has suggested a potential link between anti-TNF therapy and lymphoma,[80] longer-term RA studies suggest there is no association. Wolfe and Michaud led a large observational study of lymphoma in 19,562 adult patients, which included over 89,710 person-years of follow-up including anti-TNF exposure in 10,815 patients. This study found no evidence supporting a link between anti-TNF therapy, MTX or combination therapy and the risk of lymphoma in routine clinical practice.[70] This study is in agreement with an analysis of the linked cancer registry reports from adults with RA in Sweden by Askling et al..[68] Pediatric juvenile idiopathic arthritis (JIA) observational studies provide further reassurance.[69,81] Beukelman et al. observed 7812 children with JIA with 12,614 person-years follow-up. The standardized incidence ratio (SIR) for probable and highly probable malignancies was 4.4 (95% CI: 1.8–9.0) in children with JIA versus children without JIA. The treatment for JIA, including MTX monotherapy (SIR: 3.9 [95% CI: 0.4–14]) and TNF inhibitors (SIR: 0 [95% CI: 0–9.7]) did not appear to be significantly associated with the development of malignancy.[69] Most of the studies in the rheumatologic literature do not account for RA severity. This could be important because natural history studies in RA have exhibited an increased risk of lymphoma in comparison with the general population, primarily thought to be linked to high levels of disease activity.[68–70,80,82–84] In summary, large recent population-based studies primarily in adults with RA provide us with reasonable reassurance of the lack of an association between MTX and anti-TNF and lymphoma. However, analyses controlling for disease severity in adults and children with RA and JIA have been more difficult to conduct.
