Chronic inflammation, anti-inflammatory treatment and risk of malignant lymphomas

Abstract

The aim of this thesis was to investigate the link between malignant lymphomas and chronic inflammation, in particular the associations between inflammatory activity, immune-modulatory treatment and risk of lymphoma development. To address this aim, we used Swedish population-based registers, mandatory cancer reporting and nation-wide clinical registers including the Swedish Rheumatology Register (SRQ).

In study I we investigated whether patients with rheumatoid arthritis (RA) are at increased risk of lymphoma or cancer overall before clinical onset of RA, and if this risk increases with time since RA diagnosis. A cohort of 6,745 incident RA patients (1997-2006) was identified using data from the SRQ. Each cohort member was matched to five population comparator subjects. Relative risks of lymphoma and cancer overall before and after diagnosis of RA were estimated. We observed no increased risks before diagnosis of RA whereas during the first ten years following diagnosis, the lymphoma risk was almost doubled with a non-significant trend of increasing risks with longer RA disease duration.

In study II with a larger study population and with longer follow-up, we set out to assess whether the lymphoma risk remains elevated in RA patients diagnosed in the last few years and to explore potential risk predictors in relation to disease characteristics and therapy, focusing on the first year following RA diagnosis. Through SRQ, a cohort of 10,367 incident RA patients (1997-2010) was assembled. Each patient was matched to five population comparator subjects. The risk of lymphoma in recently diagnosed RA patients remained increased to the same magnitude as that reported from historical RA cohorts without any change in risk in recent years. The results indicated that inflammatory activity during the 1st year following RA diagnosis correlated with higher lymphoma risk, whereas oral glucocorticoids were associated with a reduced risk. Exposure to tumor necrosis factor inhibitor (TNFi) during the study period was not associated with increased lymphoma risk. Through histopathological classification, we further noted an increased proportion of Hodgkin lymphoma and diffuse large B-cell lymphoma and a decreased proportion of chronic lymphocytic leukemia in RA patients compared with the general population.

In study III the relationship between treatment with glucocorticoids and lymphoma risk in RA patients was investigated in a case-control study. From a nationwide cohort of 74,651 prevalent RA patients 1964-1994, 378 RA patients with lymphoma and 378 matched RA controls were identified. Different aspects of glucocorticoid treatment were abstracted from medical records. Treatment with glucocorticoids, oral as well as intra-articular, was associated with a reduced lymphoma risk.

In study IV the underlying risks of lymphoma in patients with ankylosing spondylitis (AS) and psoriatic arthritis (PsA) were investigated, including risks in relation to disease characteristics and treatment. Nationwide prevalent cohorts of AS (n=8,707) and PsA (n=19,283) 2001-2010 were assembled. Each cohort member was matched to five population comparator subjects. On average, we observed no increased risks of lymphoma in AS or in PsA, although there was a risk increase in PsA patients treated with methotrexate and/or sulfasalazine. Based on small numbers there was no evidence of increased lymphoma incidence following therapy with TNFi in AS or in PsA patients.