IHC biomarkers bridge pathology and therapeutics. Predictive markers identify patients likely to respond to specific treatments — they select therapy. Prognostic markers correlate with patient outcome (survival, recurrence) regardless of treatment — they inform prognosis. Many markers serve both roles.
Hormone Receptors in Breast Cancer
Estrogen receptor (ER) and progesterone receptor (PR) are the most important predictive markers in breast cancer. ER-positive tumors respond to endocrine therapy (tamoxifen, aromatase inhibitors). ER is scored by the Allred score or H-score, which combine intensity (0-3) and percentage of positive nuclei. ASCO/CAP guidelines define ER positivity as ≥1% nuclear staining in invasive tumor cells. PR adds independent prognostic information and predicts response to combined endocrine therapy.
HER2 (human epidermal growth factor receptor 2) is both prognostic (HER2 amplification correlates with aggressive behavior) and predictive (HER2-targeted therapies — trastuzumab, pertuzumab, T-DM1). HER2 IHC is scored 0-3+: 0/1+ is negative, 2+ is equivocal (requires FISH confirmation), 3+ is positive (uniform, intense membranous staining in >10% of cells). ASCO/CAP 2018 guidelines require that 3+ cases show complete, circumferential membranous staining at high intensity.
PD-L1 in Multiple Cancers
Programmed death-ligand 1 (PD-L1) expression predicts response to immune checkpoint inhibitors (pembrolizumab, nivolumab, atezolizumab). PD-L1 IHC uses different companion diagnostic assays for different drugs and tumor types: 22C3 assay (Dako) for pembrolizumab in lung cancer; 28-8 for nivolumab; SP142 for atezolizumab in triple-negative breast cancer and urothelial carcinoma; SP263 for durvalumab. Scoring varies by tumor type: tumor proportion score (TPS, % of positive tumor cells) for lung cancer, combined positive score (CPS, includes immune cells) for gastric and cervical cancer, and tumor area positivity for urothelial carcinoma.
Mismatch Repair Proteins
MLH1, MSH2, MSH6, and PMS2 IHC detects deficient mismatch repair (dMMR), which predicts response to immune checkpoint inhibitors regardless of tumor type (pembrolizumab is FDA-approved for all dMMR solid tumors). Loss of nuclear expression of one or more MMR proteins indicates dMMR. MLH1 loss is often due to promoter methylation (sporadic), while MSH2 loss is more frequently germline (Lynch syndrome). Combined with microsatellite instability testing by PCR, dMMR IHC is the standard screening method for Lynch syndrome in colorectal and endometrial cancers.
Ki-67 Proliferation Index
Ki-67 (MIB-1 clone) marks all active phases of the cell cycle (G1, S, G2, M) but not quiescent (G0) cells. The Ki-67 index is reported as the percentage of positive tumor nuclei in the highest proliferation area (hot spot). It is prognostic in breast cancer (distinguishing luminal A from luminal B subtypes), neuroendocrine tumors (grading), and lymphomas. Standardization of counting methods (manual vs. digital image analysis, hot spot vs. average counting) remains challenging.
Additional Predictive Markers
ALK IHC identifies ALK-rearranged non-small cell lung cancer, predicting response to ALK inhibitors (crizotinib, alectinib). ROS1 IHC screens for ROS1 rearrangements. BRAF V600E mutation-specific IHC (VE1 clone) detects the most common BRAF mutation in melanoma, colorectal cancer, and papillary thyroid cancer. C-KIT (CD117) expression in GIST predicts response to imatinib mesylate. EGFR IHC is not recommended for predicting EGFR mutation status in lung cancer — mutation testing by molecular methods is required.
Quality and Standardization
Predictive IHC requires rigorous quality assurance. Each assay must be validated against a gold-standard method (FISH, sequencing) with defined sensitivity and specificity. External quality assessment (EQA) participation is mandatory for laboratories performing predictive IHC. Turnaround time targets are critical — predictive marker results are needed within 7-10 working days to avoid delaying treatment decisions.
