COMPARATIVE DIAGNOSTIC PERFORMANCE OF ALK IMMUNOHISTOCHEMISTRY AND FLUORESCENCE IN SITU HYBRIDIZATION IN NON-SMALL CELL LUNG CANCER: A SYSTEMATIC REVIEW AND META-ANALYSIS

Abstract

Background: Testing for ALK rearrangement is essential in non-small cell lung cancer (NSCLC). Immunohistochemistry (IHC) and fluorescence in situ hybridization (FISH) identify related but distinct molecular features, and discordant findings may affect access to targeted treatment. Objective: To quantify the diagnostic sensitivity, specificity and agreement of ALK IHC relative to FISH and examine assay-related and methodological causes of variation. Methods: Six original paired-testing cohorts comprising 1,868 evaluable specimens were synthesized. Study-level 2 × 2 tables were analyzed using separate logit-scale random-effects models with Paule–Mandel variance estimation. Study selection was presented using the PRISMA 2020 framework, with diagnostic reporting informed by PRISMA-DTA. Study-specific assay thresholds, verification procedures and applicability were evaluated; leave-one-study-out analyses examined stability. Results: Among 1,868 specimens, 298 were positive by both assays, 1,550 were negative by both, 13 were IHC positive/FISH-negative and seven were IHC-negative/FISH-positive. Crude agreement was 98.9% (Cohen’s κ = 0.961). Pooled sensitivity was 96.2% (95% CI 91.7–98.3%; I² = 20.0%), and pooled specificity was 98.7% (95% CI 96.8 99.4%; I² = 54.9%). Sensitivity ranged from 90.9% to 100.0% and specificity from 96.6% to 100.0%. In the largest fully paired cohort (n = 1,031), D5F3 and 5A4 each detected 30/33 FISH-positive tumors, but their specificity differed (99.8% versus 98.3%). Conclusion: Validated ALK IHC shows high concordance with FISH, but predictive value depends on prevalence, staining platform and case selection. FISH is an operational comparator rather than an error-free reference for actionable ALK fusion expression. Borderline or discordant findings warrant contextual pathological review and orthogonal molecular evaluation where clinically indicated.