Purpose: Test the hypothesis that puerperal mastitis may alter immunity related to the mucin (MUC) family of glycoproteins and lower risk of ovarian cancer.

Methods: In two case-control studies conducted in New England between 1998 and 2008, we examined the association between self-reported mastitis and ovarian cancer in 1,483 women with epithelial ovarian cancer and 1,578 controls. IgG1 antibodies against (MUC1) CA15.3 and (MUC16) CA125 were measured using electrochemiluminescence assays in a subset of controls (n = 200). Preoperative CA125 was recorded in 649 cases. The association between ovarian cancer and mastitis was assessed using unconditional logistic regression to calculate adjusted odds ratios, OR, and 95 % confidence intervals (CI). Associations between mastitis and anti-CA15.3 and anti-CA125 antibodies and preoperative CA125 levels were evaluated using adjusted linear regression models.

Results: Prior mastitis was associated with a significantly lower risk of ovarian cancer: OR (and 95 % CI) of 0.67 (0.48, 0.94) adjusted for parity, breastfeeding, and other potential confounders. The association was strongest with 2 or more episodes of mastitis, and risk declined progressively with increasing number of children and episodes of mastitis. Among controls, prior mastitis was associated with significantly higher anti-CA15.3 and anti-CA125 antibody levels and, among cases, with significantly lower preoperative CA125 levels.

Conclusion: Puerperal mastitis may produce long-lasting anti-mucin antibodies that may lower the risk of ovarian cancer, plausibly through enhanced immune surveillance. Studying immune reactions related to MUC1 and MUC16 in the 10-20 % of breastfeeding women who develop mastitis may suggest ways to duplicate its effects through vaccines based on both antigens.