Intended for healthcare professionals


Disease activity and risk of lymphoma in patients with rheumatoid arthritis: nested case-control study

BMJ 1998; 317 doi: (Published 18 July 1998) Cite this as: BMJ 1998;317:180
  1. Eva Baecklund (Eva.Baecklund{at}, rheumatologista,
  2. Anders Ekbom, associate professorb,
  3. Pôr Sparén, statisticianb,
  4. Nils Feltelius, associate professora,
  5. Lars Klareskog, professorc
  1. aDepartment of Medicine, University Hospital, S-751 85 Uppsala, Sweden
  2. bDepartment of Medical Epidemiology, Karolinska Institute, Box 281, S-171 11 Stockholm, Sweden
  3. cDepartment of Medicine, Karolinska Hospital, S-171 76 Stockholm, Sweden
  1. Correspondence to: Dr Baecklund
  • Accepted 27 January 1998

Inflammatory diseases such as rheumatoid arthritis and their treatment may have a pathogenic relation with cancer. The possible relation also has practical implications for the care and control of rheumatic patients.

Patients with rheumatoid arthritis have been shown to have an increased risk of developing lymphomas. 1 2 The underlying mechanisms for this association are unclear. We performed a study to assess to what extent disease activity, various secondary manifestations of rheumatic disease, and drug treatment were independent risk factors for the development of lymphoma in patients with rheumatoid arthritis.

Subjects, methods, and results

We performed a nested case-control study using a previously described population based cohort of all patients admitted to hospital with rheumatoid arthritis during 1965-83 in Uppsala health care region, Sweden.3 We identified 42 cases of lymphoma in the 11 683 patients with rheumatoid arthritis through record linkages with the Swedish cancer registry. Cases were individually matched to three controls from the same rheumatoid arthritis cohort. All medical records were reviewed and data, including disease manifestations and treatment from the first symptoms compatible with rheumatoid arthritis until the date of the diagnosis of the lymphoma in the case, were abstracted for cases and controls. All cases and controls were evaluated to assess if the 1987 American College of Rheumatology criteria for rheumatoid arthritis were met, and patients not having rheumatoid arthritis were excluded. The risk of lymphoma was measured as unadjusted and adjusted odds ratios. The study finally consisted of 41 cases and 113 controls.

The table gives exposures linked to an increased risk of lymphoma together with the unadjusted odds ratios. High inflammatory activity was the most prominent risk factor for development of lymphoma, with an odds ratio of 25.8 compared with low inflammatory activity. Inflammatory activity was estimated by a score comprising the whole period of rheumatoid arthritis disease and was based on available data on erythrocyte sedimentation rates, number of swollen and tender joints, and the treating physician's global assessment of disease activity. Other exposures associated with disease severity also entailed an increased odds ratio for lymphoma, such as functional class IV of Steinbrocker (odds ratio 12.9), widespread joint involvement (odds ratio 9.3), and certain extra-articular symptoms.

Disease characteristics associated with increased unadjusted odds ratio for lymphoma

View this table:

Few patients were treated with immunosuppressive drugs, reflecting the standard treatment during the study period. Non-steroidal anti-inflammatory drugs, aspirin, and corticosteroids were in common use, but only a few patients were treated with antimalarials, parenteral gold, D-penicillamine, podophyllotoxin, or sulphasalazine.

We found no association between any specific drug and increased risk of lymphoma. Once drug treatment was adjusted for there was a strong independent association between inflammatory activity and lymphoma.


This study shows a strong association between disease activity in patients with rheumatoid arthritis and risk of developing lymphoma. It strengthens the concept that disease related immune alterations in rheumatoid arthritis also increase the risk of lymphoma, regardless of drug treatment.4 The risk linked to the disease seems larger than risks linked to immunosuppressive treatment seen in other studies.5 Thus our findings provide additional arguments for use of potent immunosuppressive treatment to reduce disease activity, not only to prevent joint damage but possibly also to protect against lymphoma.

Doctors need to be aware of the risk of lymphoma in certain groups of patients with rheumatoid arthritis. In addition, clinical investigators into new drugs for rheumatoid arthritis should take into account the “background” risk of lymphoma in patients with highly active rheumatoid arthritis, who are usually preferred in these trials.


Contributors: EB collected and analysed data, participated in interpreting the data, and wrote the paper. AE initiated and designed the study, supervised data collection, contributed to interpreting the data and writing the paper, and is the study guarantor. PS was responsible for the statistical parts of the study, and helped with analysis, data interpretation, and writing the paper. NF helped with the study design, data interpretation, and writing the paper. LK discussed core ideas and helped in interpreting the data and writing the paper.

Funding: None

Conflict of interest: None.


View Abstract