Long-Term Steroid Treatment may Decrease Lymphoma Risk in Rheumatoid Arthritis, Results from a Swedish Case-Control Study of 400 RA-Lymphomas
Authors: Baeklund, Iliadou, Hellgren, Ekbom, Backlin, Sundstrom, Feltelius, Klareskog, Askling Uppsala and Stockholm, Sweden
Background:
The incidence of lymphoma is increased in RA, with a risk highest in RA patients with the greatest cumulative inflammation. Non-biologic DMARDs do not appear to increase this risk, in general. However, whether corticosteroids are associated with lymphoma risk had not been established.
Methods:
Hospitalized patients with RA identified through the Swedish Hospital Inpatient Register were identified and linked to the Swedish Cancer Registry. RA patients with lymphoma were identified and individually matched to an RA control patient without a history of lymphoma. Retrospective clinical data were obtained through medical records. Pathologic specimens of all lymphomas were obtained and systematically reclassified. The association of corticosteroid use and lymphoma was assessed by comparing corticosteroid exposure in cases with lymphoma vs. matched controls without lymphoma.
Results:
378 RA subjects with lymphoma (cases) were compared to 378 matched RA subjects without lymphoma (controls). Among these, 48% of the cases vs. 57% of the controls had been treated with oral steroids (of duration greater than 4 weeks) and 44% of the cases vs. 63% of the controls had been treated with intra-articular steroids.
Compared to subjects with no steroid exposure, subjects receiving steroid therapy for durations less than 2 years showed no difference in the risk of lymphoma (RR 0.9 (95%CI 0.5 – 1.5)). However, subjects with steroid exposures greater than 2 years demonstrated a significant protective effect against lymphoma compared to subjects never treated with steroids (RR 0.4 (95% CI 0.2 – 0.7)). Increased steroid duration protected the greatest against diffuse large B-cell lymphoma (the most common subtype in RA). Subjects receiving steroid treatment early (within the first 5 years of RA diagnosis) compared to those receiving treatment later in their disease course (after 5 years of disease) showed similar protection against lymphoma compared to subjects never receiving steroids. There was no difference in cumulative disease activity for subjects with and without a history of steroid use.
Conclusions:
Increased duration of corticosteroid use (more than 2 years) was associated with a reduced risk of lymphoma (particularly large B-cell lymphoma) in RA patients.
Editorial Comment:
These are interesting findings that suggest that the protective effect of corticosteroids on lymphoma risk may be independent of the effects of steroids on controlling RA disease activity. However, caution must be taken when interpreting the results, as it is impossible to adjust for the effects of treatment allocation in this type of study. It is interesting, though, that subjects with the highest inflammatory burden (and thus at greatest risk for lymphoma) are typically those that require supplemental treatment with prolonged doses of steroids. However, whether that relationship is true for this population was not presented. It is important to note that the overall risk of lymphoma is quite low in RA patients, and that any benefit of prolonged steroid treatment on the prevention of lymphoma must be balanced against known detrimental side effects (i.e. ischemic heart disease, insulin resistance, osteoporosis, etc…) that are more common and may result in greater group morbidity and mortality than lymphoma. In this way, an analysis of number needed to benefit vs. number needed to harm would be needed before the use of prolonged corticosteroids for protection against lymphoma could be advocated.