Scleroderma and malignancy: an epidemiological study.

OBJECTIVES Although case reports and some patient series suggest an increased risk of cancer among patients with scleroderma, there are no population based studies to support this association. A population based follow up study was therefore carried out of 233 patients with scleroderma from the six-county Uppsala health care region of Sweden for the time period 1955-84. METHODS Using the inpatient registry for the Uppsala health care region, all patients with scleroderma were identified. Their unique identification codes were then used to perform a record linkage with the National Cancer Registry. Expected cancer rates were determined using the age and gender specific rates for the Uppsala health care region. RESULTS The standardised incidence ratio (SIR) for all cancers among these patients was significantly increased (SIR = 2.4; 95% CI = 1.5 to 3.6). The SIRs for lung cancer (SIR = 7.8; 95% CI = 2.5 to 18.2) and non-Hodgkin's lymphoma (SIR = 9.6; 95% CI = 1.1 to 34.5) were also significantly increased. Excluding patients who were diagnosed with cancer within a year of their scleroderma diagnosis resulted in similar findings, though the SIR for non-Hodgkin's lymphoma was no longer statistically significant. CONCLUSIONS Larger population based investigations of cancer risk among patients with scleroderma are needed to confirm these initial findings and to evaluate in greater detail possible cancer risk among these patients.