Background Lymphoma risk in RA remains a concern, in particular the long-term risks with biological therapies.
Objectives To extend assessments of relative risks, and of distribution of subtypes, of malignant lymphomas in patients with RA starting biological therapies
Methods Data from the Swedish Biologics Register (ARTIS) were cross-linked with nation-wide population-based registers: the Patient Register, the Cancer Register, and the Total Population Register. Incidences of a first malignant lymphoma in patients starting a first anti-TNF therapy 1998 through 2010 (n=10,998) within a nation-wide cohort of patients with RA (n=42,230, censored at first start of anti-TNF therapy), and in a matched general population referent cohort (n=170,649) were compared using Cox’ regression (hazard ratios (HR)), taking age, sex, calendar time, selected co-morbidities, and family history of lymphoma into account. Patients were considered at risk since start of first anti-TNF therapy. HRs were assessed overall and per age and accumulated time of active anti-TNF therapy. Following histopathological review, the distribution of lymphoma subtypes was compared to previously reported distributions of biologics-naive RA-lymphomas 1964-1995 from our group (1) and to the distribution in general lymphoma patients ( National Swedish Lymphoma Register 2000-2006).
Results Whereas both anti-TNF treated (45 lymphomas/56,493 person-years) and anti-TNF naive RA patients (185 lymphomas/197,056 person-years) were at increased risk compared to the general population (407 lymphomas/873,326 person-years), HR=2.2 (95% CI 1.6-2.9) and HR=1.9 (95%CI 1.6-2.3), respectively, the incidence of lymphoma following start of anti-TNF therapy was not higher than in anti-TNF naive RA, HR=1.1 (95%CI 0.8-1.6, Table). There was no obvious trend in HRs with accumulated time on active anti-TNF therapy, nor any appreciable difference in HRs across attained age during follow-up. The overall distribution of B-cell, T-cell and Hodgkin lymphoma was similar among anti-TNF treated RA, anti-TNF naive RA, and the general population lymphoma patients. Diffuse large B-cell lymphoma was numerically more frequent in anti-TNF treated RA (34%) than in general lymphoma patients (24%), but was lower than in biologics-naive RA from the pre-biologic era (48%). Of all anti-TNF treated lymphomas, 75 % had been treated with only one biologic.
Conclusions In this study, the already increased lymphoma risk in patients with RA does not seem to increase further following anti-TNF therapy, at least not following up to 5-10 years of active treatment, nor did we find any treatment-related excess risk in younger or older adults. The distribution of lymphoma subtypes was not markedly different from that in general lymphoma patients.
References Baecklund et al, Arthritis Rheum, 2006
Disclosure of Interest None Declared