Diagnostic accuracy of special AT-rich sequence-binding protein 2 immunohistochemistry for identifying colorectal tumor origin: A systematic review and meta-analysis

Objectives: Special AT-rich sequence-binding protein 2 (SATB2) is an immunohistochemical marker proposed for identifying colorectal origin in primary and metastatic tumors, but reported accuracy varies, and no pooled estimate restricted to verifiable data exists. We performed a systematic review and diagnostic test accuracy meta-analysis of SATB2 immunohistochemistry (IHC) for distinguishing colorectal carcinoma from non-colorectal malignancies. Methods: MEDLINE, Embase, and Web of Science were searched from inception to February 16, 2026, for English- language studies reporting SATB2 IHC against a histopathological reference standard with a non-colorectal comparator. Study selection and data extraction were performed in duplicate, and risk of bias was assessed with quality assessment of diagnostic accuracy studies-2. Studies were pooled only where a complete 2 × 2 table was verifiable against the source publication. Pooled sensitivity and specificity were estimated with a bivariate random- effects model, and positive and negative likelihood ratios (LR+ and LR−) and the diagnostic odds ratio (DOR) with random-effects models on the log scale. The review was registered in International Prospective Register of Systematic Reviews (PROSPERO) (registration number CRD420261396459; retrospective registration). Results: Five studies were included in the qualitative synthesis. Three non-overlapping patient cohorts from two source publications (the resection and biopsy cohorts of Zhang 2018 and Elnady 2022; 274 colorectal and 442 non-colorectal cases) provided verifiable 2 × 2 data for pooling. The pooled sensitivity was 93.9% (95% confidence interval [CI] 90.3–96.3) and the pooled specificity 97.5% (95% CI 95.5–98.6). The pooled LR+ was 37.4 (95% CI 20.9–67.1), the LR− 0.063 (95% CI 0.039–0.101), and the DOR 630 (95% CI 277–1432). Little statistical heterogeneity was detected, though with three cohorts, these estimates have very limited power. Leave-one-out and Zhang-cohort sensitivity analyses did not materially change the estimates. Conclusion: In this limited evidence base, SATB2 IHC was highly accurate for identifying colorectal origin and may be useful as one component of a panel-based diagnostic approach, interpreted alongside morphology, clinical context, and the local staining protocol. The estimates rest on three non-overlapping cohorts from two publications and should be regarded as preliminary rather than establishing performance across all primary, metastatic, biopsy, and carcinoma-of-unknown-primary settings.