ABSTRACT Background Muir‐Torre syndrome (MTS) is a hereditary tumor predisposition syndrome associated with sebaceous neoplasms. Immunohistochemistry (IHC) for loss of mismatch repair (MMR) proteins in these tumors is used as a screening test, but its diagnostic accuracy has not been rigorously assessed. Methods We conducted a meta‐analysis of 25 studies involving 692 patients who underwent IHC testing for MMR protein loss in sebaceous neoplasms. Results The pooled sensitivity was 0.84 (95% CI: 0.75–0.90) and specificity was 0.46 (95% CI: 0.28–0.66), with significant inter‐study heterogeneity in specificity ( I 2 , 77%). Restricting the meta‐analysis to more rigorous studies with exposure‐based designs and germline mutation as the reference standard yielded higher sensitivity (0.91; 95% CI: 0.83–0.96) and lower specificity (0.14; 95% CI: 0.06–0.27). Hypothetically restricting testing to patients under 60 years or tumors outside the head/neck locations increased specificity (0.87 and 0.88, respectively) but reduced sensitivity (0.60 and 0.37, respectively). A two‐antibody panel (MSH6 and PMS2) performed equivalently to a four‐antibody panel (MLH1, MSH2, MSH6, and PMS2). Conclusions IHC testing can discriminate between sporadic and MTS‐associated sebaceous neoplasms, but diagnostic utility is limited by low specificity. Most MMR‐deficient cases are not due to MTS.