Open Access

Simone Fischer, Christa Meisinger, Dennis Freuer

• Objectives Among the over 80 different autoimmune diseases, psoriasis (PsO), rheumatoid arthritis (RA), and ankylosing spondylitis (AS) are common representatives. Previous studies indicated a potential link with cancer risk, but suffered often from low statistical power. Thus, we aimed to synthesize the evidence and quantify the association to different female-specific cancer sites. Methods The systematic review was performed according to PRISMA guidelines. A search string was developed for the databases PubMed, Web of Science, Cochrane Library and Embase. Results were screened independently by two investigators and the risk of bias was assessed using the ROBINS-E tool. Meta-analyses were performed using inverse variance weighted random-effects models. Statistical between-study heterogeneity was quantified by calculating Cochran's Q, , and Higgins' I2 statistics. Sources of heterogeneity were analyzed and adjusted for within an intensive bias assessment in the form ofObjectives Among the over 80 different autoimmune diseases, psoriasis (PsO), rheumatoid arthritis (RA), and ankylosing spondylitis (AS) are common representatives. Previous studies indicated a potential link with cancer risk, but suffered often from low statistical power. Thus, we aimed to synthesize the evidence and quantify the association to different female-specific cancer sites. Methods The systematic review was performed according to PRISMA guidelines. A search string was developed for the databases PubMed, Web of Science, Cochrane Library and Embase. Results were screened independently by two investigators and the risk of bias was assessed using the ROBINS-E tool. Meta-analyses were performed using inverse variance weighted random-effects models. Statistical between-study heterogeneity was quantified by calculating Cochran's Q, , and Higgins' I2 statistics. Sources of heterogeneity were analyzed and adjusted for within an intensive bias assessment in the form of meta-regression, outlier, influential, and subgroup analyses. A range of methods were used to test and adjust for publication bias. Results Of 10,096 records that were originally identified by the search strategy, 45 were included in the meta-analyses. RA was inversely associated with both breast and uterine cancer occurrence, while PsO was associated with a higher breast cancer risk. Outlier-adjusted estimates confirmed these findings. Bias assessment revealed differences in geographic regions, particularly in RA patients, with higher estimates among Asian studies. An additional analysis revealed no association between psoriatic arthritis and breast cancer. Conclusions RA seems to reduce the risk of breast and uterine cancers, while PsO appears to increase breast cancer risk. Further large studies are required to investigate potential therapy-effects and detailed biological mechanisms.…