Background/Purpose: The risk of cancer compared with general population has been elevated in several connective tissue diseases. However, the standardized incidence ratios (SIRs) for overall cancer in patients with systemic sclerosis (SSc) were inconsistent. Moreover, most of existing studies were limited in size and based on hospital case series, with attendant selection and referral biases. We therefore aimed to examine, using meta-analysis, cancer risk in patients with SSc deriving from population-based cohort studies, as compared with the expected risks in age- matched background populations.

Methods: We searched five different databases (MEDLINE, Scopus, CINAHL, Web of Science and Cochrane Collaboration databases), reference lists of retrieved studies and review articles from January 1966 until May 2012. Population-based cohort studies relevant for determining cancer risk in patients with SSc were included. All papers fulfilling the strict inclusion criteria were scrutinized for data on population size, time of follow-up and observed to expected cancer rates (standardized incidence ratio (SIR)). Two investigators independently evaluated the quality of the studies by using a scoring system that was created on the basis of a recently used system designed with reference to MOOSE, QUAST and STROBE. Data syntheses were based on random effects model.

Results: Seven articles were included. The pooled SIR for overall cancer was 1.50 (95% CI: 1.23-1.83). The pooled SIR of 1.85 (95% CI: 1.49-2.31) for men was significantly higher than that of 1.33 (95% CI: 1.18-1.49) for women (p =0.009) and stratification on sex eliminated heterogeneity. Stratification based on type of SSc did not produce statistically significant differences in the pooled SIR between limited SSc and diffuse SSc (p = 0.98). The significant increased risk of cancer of the lung (SIR: 3.18; 95% CI: 2.09-4.85), the liver (SIR: 4.36; 95% CI: 2.00-9.51), the hematological system (SIR: 3.55; 95% CI: 1.70-7.43), non-Hodgkin lymphoma (SIR: 2.71; 95% CI: 1.43-5.14), leukemia (SIR: 2.75; 95% CI: 1.32-5.73) and the bladder (SIR: 2.00; 95% CI: 1.06-3.77) was observed. The pooled SIR for non-melanoma skin cancer was significantly increased only in men, while the pooled SIR for bladder cancer was significantly increased only in women. The appearance of funnel plots was symmetrical and Egger’s test results were not significant (p = 0.60).

Conclusion: In conclusion, SSc are associated with an increased risk of cancer, particularly lung, liver, hematological and bladder cancer. Men with SSc are at higher cancer risk than women.