Biomarker Guide · Path-iQ Global Pathology Review
A companion diagnostic (CDx) is an in vitro diagnostic test that provides information essential for the safe and effective use of a corresponding therapeutic product. The pathologist is the critical gatekeeper of CDx testing — responsible for selecting appropriate tissue, ensuring assay quality, interpreting results, and communicating findings that directly determine which molecularly targeted therapy a patient receives. As of 2026, the FDA has approved over 50 companion diagnostic tests across more than 20 oncology indications.
| Biomarker | Test Method | Approved CDx Platform | Targeted Therapy | Expected Prevalence |
|---|---|---|---|---|
| EGFR exon 19 del / L858R | PCR / NGS | cobas EGFR v2; therascreen EGFR; FoundationOne CDx | Osimertinib (1L), erlotinib, gefitinib, afatinib, dacomitinib | 10–15% (Western); 30–40% (East Asian) |
| EGFR T790M (resistance) | Plasma ctDNA or tissue NGS | cobas EGFR v2 (plasma) | Osimertinib (2L after 1G EGFR-TKI) | ~50% of EGFR-mut post-1G TKI resistance |
| EGFR exon 20 insertion | NGS | FoundationOne CDx; OncomineDx Target | Amivantamab; mobocertinib | ~2–3% of NSCLC |
| ALK rearrangement | FISH / IHC (D5F3) / NGS | VENTANA ALK (D5F3); Vysis ALK Break Apart FISH; FoundationOne CDx | Crizotinib, alectinib, brigatinib, lorlatinib, ceritinib | ~3–5% |
| ROS1 rearrangement | FISH / IHC / NGS | Abbott Vysis ROS1 FISH; FoundationOne CDx | Crizotinib, entrectinib, lorlatinib | ~1–2% |
| BRAF V600E | PCR / IHC (VE1) / NGS | cobas BRAF V600; FoundationOne CDx | Dabrafenib + trametinib | ~2–4% |
| MET exon 14 skipping | RNA-based NGS or DNA NGS | FoundationOne CDx; OncomineDx | Tepotinib, capmatinib | ~3–4% |
| RET rearrangement | FISH / NGS (RNA preferred) | FoundationOne CDx | Selpercatinib, pralsetinib | ~1–2% |
| NTRK1/2/3 fusion | IHC (pan-TRK); confirmed by NGS | FoundationOne CDx; OncomineDx | Larotrectinib, entrectinib (tumour-agnostic) | <1% |
| PD-L1 expression (TPS) | IHC | 22C3 (Dako/Agilent) — pembrolizumab CDx; 28-8; SP142; SP263 | Pembrolizumab (TPS ≥1%; ≥50% 1L); atezolizumab (TC/IC ≥50%) | TPS ≥50%: ~30% of NSCLC |
| KRAS G12C | PCR / NGS | therascreen KRAS RGQ; FoundationOne CDx | Sotorasib, adagrasib | ~13% of NSCLC; most common KRAS mutation |
| Biomarker | Test Method | CDx Platform | Therapy | Prevalence |
|---|---|---|---|---|
| HER2 IHC 3+ or ISH amplified | IHC + FISH/CISH/DISH | PATHWAY anti-HER2 (4B5); BOND Oracle HER2; HER2 IQFISH pharmDx | Trastuzumab, pertuzumab, T-DM1, T-DXd (HER2-positive) | ~15–20% of breast cancers |
| HER2-low (IHC 1+ or 2+/ISH-) | IHC ± FISH | PATHWAY anti-HER2 (4B5); SP3 antibody | Trastuzumab deruxtecan (T-DXd) | ~45–55% of HR+ breast cancers |
| PIK3CA mutation (HR+/HER2-) | NGS or therascreen PIK3CA | therascreen PIK3CA RGQ PCR Kit (tissue + plasma) | Alpelisib + fulvestrant | ~40% of HR+/HER2- breast cancers |
| ESR1 mutation (metastatic) | Plasma ctDNA NGS | Guardant360 CDx | Elacestrant (1st oral SERD) | ~25–40% of HR+ mBC after AI therapy |
| BRCA1/2 germline (HER2-) | Germline sequencing | BRACAnalysis CDx (Myriad) | Olaparib, talazoparib (PARP inhibitors) | ~5–7% of unselected breast cancer |
| PD-L1 (SP142, CPS) — TNBC | IHC | VENTANA PD-L1 (SP142) — IC ≥1% | Atezolizumab + nab-paclitaxel (1L TNBC) | IC≥1%: ~40% of TNBC |
| NTRK1/2/3 fusion | NGS (tumour-agnostic) | FoundationOne CDx | Larotrectinib, entrectinib | <1% |
| Biomarker | Method | CDx Platform | Therapy / Exclusion | Prevalence |
|---|---|---|---|---|
| KRAS / NRAS (RAS) wildtype | PCR / NGS | therascreen KRAS RGQ; FoundationOne CDx | Cetuximab, panitumumab (RAS mut = contraindication) | RAS wildtype: ~50% of mCRC |
| BRAF V600E | PCR / IHC / NGS | cobas BRAF V600; FoundationOne CDx | Encorafenib + cetuximab (BRAF-mut mCRC); also distinguishes sporadic vs Lynch MSI-H | ~8–12% of CRC |
| MSI-H / dMMR | IHC (MMR panel) or PCR (MSI) | Dako MMR IHC + MSI PCR; FoundationOne CDx | Pembrolizumab (1L mCRC); nivolumab ± ipilimumab | ~4–5% of mCRC (higher in stage II: 15–20%) |
| HER2 amplification | IHC + ISH / NGS | FoundationOne CDx; HERACLES-RESCUE criteria | Trastuzumab deruxtecan; tucatinib + trastuzumab | ~2–3% of mCRC; enriched in RAS wildtype |
| NTRK1/2/3 fusion | NGS | FoundationOne CDx | Larotrectinib, entrectinib | <1% |
| Tumour Type | Biomarker | Therapy | Platform |
|---|---|---|---|
| Melanoma | BRAF V600E/K | Vemurafenib + cobimetinib; dabrafenib + trametinib; encorafenib + binimetinib | cobas 4800 BRAF V600; THxID BRAF |
| Gastric / GEJ adenocarcinoma | HER2 IHC/ISH | Trastuzumab + chemotherapy; trastuzumab deruxtecan (HER2+ 2L) | PATHWAY HER2 (4B5); BOND-3 Oracle |
| Gastric / GEJ adenocarcinoma | PD-L1 CPS (22C3) | Pembrolizumab (CPS ≥1 for 1L chemo + pembro); nivolumab (CPS ≥5) | Dako PD-L1 IHC 22C3 |
| Ovarian cancer | BRCA1/2 germline + somatic | Olaparib, niraparib, rucaparib (PARP inhibitors; HRD testing also used) | BRACAnalysis CDx; myChoice CDx (HRD) |
| Prostate cancer (mCRPC) | BRCA1/2 / HRR mutations | Olaparib (BRCAm); rucaparib; niraparib | FoundationOne CDx; BRACAnalysis CDx |
| Bladder / urothelial | FGFR2/3 mutation or fusion | Erdafitinib | therascreen FGFR RGQ RT-PCR |
| Cholangiocarcinoma | FGFR2 fusion / rearrangement | Pemigatinib, futibatinib, infigratinib | FoundationOne CDx |
| Any solid tumour (tumour-agnostic) | MSI-H / dMMR | Pembrolizumab (tumour-agnostic approval 2017) | MMR IHC; MSI PCR; FoundationOne CDx |
| Any solid tumour (tumour-agnostic) | TMB-H (≥10 mut/Mb) | Pembrolizumab (TMB-H tumour-agnostic) | FoundationOne CDx |
| Any solid tumour (tumour-agnostic) | NTRK1/2/3 fusion | Larotrectinib, entrectinib | FoundationOne CDx; OncomineDx |
| AML | FLT3-ITD / TKD | Midostaurin, gilteritinib (FLT3-mut AML) | LeukoStrat CDx FLT3 Mutation Assay |
| NSCLC / thyroid | RET fusion / mutation | Selpercatinib, pralsetinib | FoundationOne CDx |
PD-L1 immunohistochemistry is simultaneously one of the most widely used and most technically inconsistent biomarker tests in oncology. Four major PD-L1 assays — 22C3 (Dako), 28-8 (Dako), SP142 (Ventana), and SP263 (Ventana) — are FDA-approved as CDx or complementary diagnostics for different immunotherapy agents on different staining platforms. They are not interchangeable.
| Clone | Platform | Approved Indication | Scoring System | Key Threshold |
|---|---|---|---|---|
| 22C3 | Dako Autostainer Link 48 | Pembrolizumab — NSCLC, cervical, head & neck, gastric, urothelial, TNBC, endometrial, TMB-H | TPS (tumour proportion score) for NSCLC; CPS (combined positive score) for other tumour types | TPS ≥50% = 1L pembrolizumab monotherapy NSCLC; CPS ≥1 for many indications |
| 28-8 | Dako Autostainer Link 48 | Nivolumab — NSCLC, melanoma, HNSCC, gastric, urothelial, MSI-H CRC | TPS (tumour cells only) | Variable by indication; ≥1% to ≥5% for different nivolumab indications |
| SP142 | VENTANA BenchMark ULTRA | Atezolizumab — NSCLC (TC/IC); TNBC (IC ≥1%) | TC% (tumour cell) + IC% (immune cell); scored separately | TC ≥50% OR IC ≥10% for NSCLC; IC ≥1% for TNBC + nab-paclitaxel |
| SP263 | VENTANA BenchMark ULTRA | Durvalumab — NSCLC; nivolumab (some indications in ex-US) | TPS (tumour cells only) | TC ≥25% for durvalumab in NSCLC (adjuvant/unresectable) |
The Blueprint Comparability Study (phases 1 and 2) demonstrated that 22C3, 28-8, and SP263 are approximately comparable for tumour cell scoring (TC), but SP142 scores consistently lower — particularly for TC. Pathologists must use the analytically validated clone linked to the specific therapeutic agent being considered. Cross-platform or cross-clone reporting introduces unacceptable risk of misclassification.
The most common reason for CDx testing failure is inadequate tissue — either insufficient tumour cellularity in the submitted block, or exhaustion of tissue through sequential IHC runs before molecular testing is requested. Best practice principles for tissue stewardship: