Supplement: Orphan-Disease, Rare-Tumor & Cutting-Edge IHC
Addendum to the National Reference Laboratory IHC Menu — Dako Omnis context
Slots into the base document's section numbering. Status flags: T2 = add to the standing reference menu now; LDT = emerging/validated-in-literature, run as in-house validated laboratory-developed test (usually RUO reagent); R = research-only, never for clinical treatment selection.
S0. Why this layer exists — and how to run it
This supplement covers three things the base menu deliberately deferred: (1) fusion/mutation-surrogate IHC that substitutes for molecular testing (SS18-SSX, H3 G34R/V, p65, AFF2, DDIT3, EZHIP), (2) orphan-disease panels that only a national reference center will ever run at volume (neuromuscular, EB antigen mapping, PFIC, amyloid typing, novel MN antigens, prion), and (3) molecular-subtype surrogates entering trials and tumor boards (GATA6/CK17 in PDAC, ASCL1/NEUROD1/POU2F3/YAP1 in SCLC, GPNMB in renal tumors).
Operational rules specific to this layer:
EQA mostly does not exist for these markers (NordiQC/UK NEQAS coverage is sparse). Substitute: genotyped index cases and cell-line FFPE blocks as controls, split-sample exchange with 1–2 peer reference labs, and molecular concordance audits (every surrogate-IHC result vs NGS/FISH for the first 20+ cases, then periodic).
Regulatory framing: most reagents here are RUO. Under EU IVDR logic (Art. 5(5) health-institution exemption) and ISO 15189, each becomes an in-house LDT with a validation dossier (analytical sensitivity/specificity vs genotype, precision, lot-to-lot). In Turkey, align the dossier with TÜRKAK ISO 15189 scope and TİTCK device rules.
Omnis practicalities: most of these are rabbit polyclonals or low-abundance rabbit mAbs → plan TRS High pH HIER + 3-step (linker) detection by default, then de-escalate if signal allows. Batch ultra-low-volume markers (weekly/monthly runs).
Digital pathology hook: QuPath-assisted scoring is already published for GATA6 H-score in PDAC and is the obvious route for Ki-67-like quantitative surrogates — a natural fit for your image-analysis pipeline.
S3-bis. GI, pancreatobiliary & liver (extends base §3)
GATA6
No consensus dx clone; published IHC (COMPASS ancillary work) with pathologist + QuPath-assisted H-score; RNA-ISH alternative
PDAC transcriptional subtype surrogate: GATA6-high = classical (ORR 33% and better mFOLFIRINOX outcomes in COMPASS); GATA6-low = basal-like (ORR 10%, mFFX progression 60% vs 15%)
LDT
CK17 (E3)
Dako/std E3
Dual role: (a) basal-like PDAC marker (with CK5/6+, p63+, GATA6/HNF4α-low); (b) PDAC-vs-reactive panel with IMP3, maspin, S100P
T2
HNF4α
rabbit mAb/polyclonal
Classical-subtype partner; 4-marker panel CK5/6+p63 vs GATA6+HNF4α defines classical/transitional/basal IHC patterns with independent prognostic value
LDT
ATRX + DAXX
ATRX polyclonal/BSB-108 (on menu); DAXX rabbit polyclonal
Loss in ~40% PanNET → ALT phenotype, worse prognosis; distinguishes PanNET from PanNEC (RB/p53 route)
T2
Menin (MEN1)
rabbit mAb/polyclonal
Nuclear loss in MEN1-mutant PanNET; syndromic flag
LDT
BSEP (ABCB11)
rabbit polyclonal
Canalicular loss → PFIC2; orphan pediatric cholestasis service
T2
MDR3 (ABCB4)
P3II-26
Canalicular loss/reduction → PFIC3
T2
Alpha-1-antitrypsin
polyclonal (Dako)
PiZZ globules (PAS-D+ correlate); A1ATD liver
T2
Pitfall note
—
GATA6 also stains many upper-GI/pancreatobiliary lineages — subtype use requires quantitative scoring, not binary read
—
S7-bis. Thoracic (extends base §7)
ASCL1
24B72D11 (BD)
SCLC-A (ASCL1-dominant ≈69% of SCLC); NE-high/DLL3-high
LDT
NEUROD1
EPR20766
SCLC-N (≈17%); NE-high
LDT
POU2F3
rabbit polyclonal (e.g., NBP1-83966)
SCLC-P tuft-cell-like (≈7%); mutually exclusive with ASCL1/NEUROD1; NE-low/DLL3-low — explains "NE-marker-negative SCLC"
LDT
YAP1
63.7
SCLC-Y/inflamed (low-level, mostly combined SCLC; contested as pure subtype); also Hippo-pathway work in mesothelioma; also ST-EPN-YAP1
LDT
NF2/Merlin
rabbit mAb/polyclonal
Loss in mesothelioma (Hippo axis) — adjunct to BAP1/MTAP
R→LDT
Rationale
—
Subtype correlates with DLL3 (tarlatamab), chemo-IO response patterns; expect trial-driven requests. Report as dominant-TF pattern, not single-marker binary
—
S5-bis. Genitourinary (extends base §5)
GPNMB
rabbit mAb (automated assays published)
Sensitive screen for TFE3/TFEB tRCC AND TSC1/2/mTOR-altered tumors (ESC RCC, EVT, LOT, AML/PEComa) — shared MiT-pathway output. Caveats: does not separate those two groups; ~13% equivocal/false-negative vs FISH; confirm with TFE3/TFEB IHC-FISH. TRIM63 RNA-ISH is the emerging alternative
LDT
HOXB13
rabbit mAb (e.g., D7N8O)
Prostate lineage in NKX3.1-dim/PSA-negative metastases
LDT
Prostein (P501S)
10E3
Prostatic lineage (with NKX3.1)
T2
PBRM1
rabbit polyclonal
ccRCC prognostic (with BAP1); research reporting only
R
S6-bis. Gynecologic (extends base §6)
FOXL2
rabbit polyclonal
Sex cord-stromal lineage (adult granulosa); C402G/C134W confirmation stays molecular
T2
BCOR
C-10 (mouse)
High-grade endometrial stromal sarcoma (BCOR-ITD/ZC3H7B::BCOR) — pairs with diffuse cyclin D1; also see sarcoma/CNS entries
LDT
SMARCA4 (BRG1) + SMARCA2 (BRM)
EPNCIR111A (on menu) + BRM polyclonal
SCCOHT: BRG1 loss with BRM co-loss is near-pathognomonic — reflex both in young-female undifferentiated ovarian tumors
T2
Pattern note
—
Mesonephric-like adenocarcinoma: GATA3+/TTF-1+/luminal CD10+ with ER/PR-low, wild-type p53 — panel logic, no new reagent
—
S8-bis. Head & neck, endocrine (extends base §8)
AFF2 (C-terminus)
rabbit anti-AFF2 C-term
Nuclear AFF2 = sensitive & specific surrogate for DEK::AFF2 sinonasal/skull-base carcinoma (papilloma-like, deceptively bland, aggressive); replaces FISH triage
LDT
IDH2 R172 (multi-specific)
MsMab-1 / 11C8B1
IDH2-mutant SNUC (and rare gliomas); imperfect sensitivity — NGS confirms negatives
LDT
NRAS Q61R
SP174
RAS-like thyroid neoplasms; melanoma adjunct
LDT
p27/CDKN1B
SX53G8
MEN4 workup; pituitary/parathyroid context
LDT
NUT scope note
C52B1 (on menu)
Extend NUT IHC beyond NUT carcinoma: NUTM1-rearranged porocarcinoma/adnexal and sarcomas
—
S9-bis. Hematopathology (extends base §9)
BOB1 / OCT2
SP92 (or TG14) / Oct-207
B-cell program integrity: CHL (dim/lost) vs NLPHL/PMBL/LBCL
T2
CD200
e.g., UMAB223
CLL (+) vs MCL (−); HCL (+)
T2
MNDA / IRTA1
253A / rabbit mAb
Marginal-zone vs follicular lymphoma
T2
HGAL (GCET1) / LMO2
MRQ-49 / SP51
Germinal-center program (Hans-plus algorithms)
T2
CXCL13 / ICOS
polyclonal / SP98
TFH phenotype — AITL/nodal TFH lymphoma (with PD-1, CD10, BCL6)
T2
TCF4 (E2-2)
rabbit mAb
BPDCN (with CD123, TCL1)
LDT
TBX21 (T-bet) + GATA3
4B10 + L50-823
PTCL-NOS TBX21 vs GATA3 subtyping (prognostic; entering trials)
LDT
NPM1 (cytoplasmic)
clone 376
Cytoplasmic dislocation = NPM1-mutant AML surrogate on trephines
LDT
CD19
e.g., BT51E
Antigen-escape assessment post-CAR-T/blinatumomab (report presence/loss)
LDT
CD79b / CD22
mAbs
Polatuzumab / inotuzumab target documentation on request
R→LDT
BCMA
mAbs
Myeloma CAR-T/bispecific target — no validated clinical IHC selection assay yet
R
S10-bis. Soft tissue & bone (extends base §10)
SS18-SSX (fusion junction)
E9X9V
Synovial sarcoma: ~100% specific, ~95% sensitive; diffuse strong nuclear
T2
SSX (C-terminus)
E5A2C
~100% sensitive, ~96% specific; run as a pair — concordant staining can replace FISH/NGS in most cases; false negatives in decalcified/poorly fixed small biopsies
T2
CCNB3
rabbit polyclonal
BCOR::CCNB3 sarcoma (with BCOR C-10)
LDT
ETV4
mAb/polyclonal (per Hung et al.)
CIC-rearranged sarcoma (diffuse nuclear; with strong WT1); DUX4 C-terminal Abs remain RUO
LDT
FOSB
5G4
Pseudomyogenic hemangioendothelioma; epithelioid hemangioma
LDT
FOS (c-FOS)
rabbit mAb
FOS-rearranged osteoblastoma/osteoid osteoma vs osteosarcoma
LDT
DDIT3 (CHOP)
e.g., 9C8
Nuclear DDIT3 = FUS/EWSR1::DDIT3 myxoid liposarcoma surrogate
LDT
GLI1
RUO mAb/polyclonal
GLI1-amplified/rearranged mesenchymal tumors ("gastroblastoma-like", plexiform fibromyxoma spectrum)
LDT
SMARCA2 (BRM)
polyclonal
Co-loss with SMARCA4 in thoracic/undifferentiated tumors, SCCOHT
LDT
Pitfall
—
NKX3.1 (EP356) is positive in mesenchymal chondrosarcoma — do not read as prostatic in small-round-cell context
—
S11-bis. Neuropathology (extends base §11)
H3 G34R
RM240 (RevMAb)
Diffuse hemispheric glioma, H3 G34-mutant (~90% of G34 cases); context: OLIG2-negative, ATRX loss, p53+
T2
H3 G34V
RM307 (RevMAb)
The rarer G34V DHG (run both on suspicion; G34M escapes both → sequence)
LDT
p65 (RelA)
e.g., D14E12; nuclear
ZFTA::RELA ST-ependymoma surrogate: ~100% sensitive / ~92% specific for RELA fusion; combine with L1CAM (more sensitive for non-RELA ZFTA partners) ± cyclin D1; double-negative virtually excludes fusion
T2
EZHIP (CXorf67)
rabbit polyclonal (runs on Omnis per RENOCLIP data)
PFA ependymoma: ~93% EZHIP+, remainder H3K27M+ — pairs with H3K27me3 loss; also H3-WT DMG with EZHIP overexpression; germinoma cross-positivity caveat
LDT
BCOR
C-10
CNS tumor with BCOR-ITD (also sarcoma/HG-ESS uses)
LDT
YAP1
63.7
ST-EPN-YAP1 (infant); see also thoracic uses
LDT
MB surrogate panel: GAB1, YAP1, filamin A, OTX2
polyclonal / 63.7 / PM6/317 / rabbit mAb
Provisional medulloblastoma grouping when methylation is unavailable: β-catenin-nuclear=WNT; GAB1/YAP1/filamin A+=SHH; all-neg=group 3/4
LDT
CRX
rabbit polyclonal
Retinoblastoma/pineoblastoma photoreceptor lineage
LDT
PrP (prion)
3F4, KG9, 12F10
CJD surveillance IHC (PrP^Sc deposition patterns). Formic-acid pretreatment, dedicated processing/instrument decontamination, national-surveillance linkage — a defining reference-lab obligation
T2
S12-bis. Dermatopathology (extends base §12)
MCPyV large T
CM2B4
Merkel cell carcinoma: virus-positive vs virus-negative (UV-driven, TTF-1-independent CK20-dot context); etiologic + prognostic
T2
5hmC
rabbit polyclonal
Global loss favors melanoma over nevus; nevoid melanoma adjunct
LDT
MxA
mAb/polyclonal
Type-I interferon signature — dermatomyositis skin (and muscle, see S-NM)
LDT
S13-bis. Nephropathology (extends base §13)
NELL1
rabbit polyclonal
2nd most common MN antigen after PLA2R (~13–16% of PLA2R-negative MN); segmental GBM pattern, IgG1-dominant; associations: malignancy, bucillamine, lipoic acid
T2
EXT1 + EXT2
rabbit polyclonals
~12% of PLA2R-negative MN; autoimmune/membranous-lupus association; bright granular GBM
T2
Sema3B / PCDH7 / NCAM1 / HTRA1
polyclonals
Minor MN antigens (Sema3B pediatric) — batch on request
LDT
IgG subclasses (IgG1–4)
HP600x-series mAbs or sheep polyclonals
MN primary-vs-secondary logic (IgG4-dominant=PLA2R type), PGNMID (monotypic IgG3κ), fibrillary GN
T2
Amyloid typing panel: AA, ATTR, AFib (fibrinogen Aα), ALECT2, κ, λ
mc1 (AA) + polyclonals
IHC pre-typing of amyloid; explicit caveat: IHC mistyping risk is real — laser-microdissection mass spectrometry remains gold standard; DNAJB9 (base menu) covers fibrillary GN
T2
S-NM. Neuromuscular orphan-disease panel (NEW section)
The single largest orphan-disease IHC block a national reference lab should own. Classic clones are Leica/Novocastra heritage; many now validated on FFPE, but keep a frozen-section IF track for dystroglycan and service continuity.
Dystrophin rod / C-term / N-term
DYS1 (Dy4/6D3) / DYS2 (Dy8/6C5) / DYS3 (Dy10/12B2)
DMD (absent) vs BMD (reduced/patchy); all three domains mandatory
α/β/γ/δ-sarcoglycan
Ad1/20A6, βSarc/5B1, 35DAG/21B5, δSarc3/12C1
Sarcoglycanopathies (LGMD R3–R6); secondary reductions cross-panel
Dysferlin
NCL-Hamlet
LGMD R2 / Miyoshi
Merosin (laminin-α2)
Mer3/22B2
MDC1A congenital dystrophy
Emerin
4G5
X-linked Emery-Dreifuss (nuclear rim loss)
Caveolin-3
mAb/polyclonal
LGMD 1C / rippling muscle
Spectrin
RBC2/3D5
Sarcolemmal integrity control for every run
α-dystroglycan
IIH6C4 / VIA4-1
Dystroglycanopathies (glyco-epitope; frozen/WB support)
Utrophin
DRP3/20C5
Compensatory sarcolemmal upregulation in DMD
Fast / slow / neonatal myosin
WB-MHCf / WB-MHCs / WB-MHCn
Fiber typing, grouping, regeneration
MHC class I / II
W6/32 / CR3/43
Sarcolemmal upregulation — inflammatory myopathy screen
C5b-9 (MAC)
aE11
DM capillary deposits; IMNM sarcolemmal deposits
MxA
mAb/polyclonal
Sarcoplasmic MxA = sensitive/specific DM interferon signature
p62 + TDP-43
3/P62-lck + phospho/std
IBM rimmed-vacuole pathology (with COX/SDH histochemistry)
CD56/NCAM (reuse)
123C3
Regenerating fibers
S-EB. Epidermolysis bullosa antigen mapping (NEW section)
IF mapping (frozen preferred) localizes the split and the deficient protein — genotype-guiding orphan service.
Keratin 5 / 14
XM26 / LL002
EB simplex (basal keratinocyte)
Plectin
mAb (e.g., 10F6)
EBS with muscular dystrophy
Integrin α6/β4
mAbs
JEB with pyloric atresia
Laminin-332
GB3
Junctional EB (lamina lucida)
Collagen XVII (BP180)
NC16A-domain mAbs
Junctional EB
Collagen VII
LH7.2
Dystrophic EB (sublamina densa)
Collagen IV
CIV22 (base menu)
Floor/roof reference of the split
S14-bis. Infectious (extends base §14)
Tropheryma whipplei IHC exists but availability is limited — PAS-D morphology + PCR remains the practical reference pathway; list as send-out/LDT-on-demand only.
S18-bis. Pediatric (extends base §18)
ALK IHC (D5F3/5A4, already on menu) gains a neuroblastoma use: ALK-aberrant NB flagging for lorlatinib-era protocols (report intensity/extent; molecular confirms). PHOX2B, INI1, LIN28A already cover the rest.
S-R. Research-only horizon (never for clinical selection)
LAG-3 (17B4 / D2G4O)
Relatlimab context; no CDx requirement
β2-microglobulin / HLA-I (EMR8-5)
Immune-evasion phenotyping
STK11/LKB1
IHC unreliable — molecular only
CLDN6, B7-H3 (CD276), CD70
ADC/CAR-T targets in trials
PLCG2 (SCLC stem-like), TRIM63 RNA-ISH (renal)
Emerging adjuncts
Updated menu arithmetic
GI/pancreatobiliary/liver
+8
Thoracic (SCLC subtyping, NF2)
+5
GU
+4
Gyn (net of cross-listed)
+2
H&N/endocrine
+4
Hematopathology
+13
Soft tissue & bone
+9
Neuropathology
+11
Dermatopathology
+3
Nephropathology (incl. amyloid panel)
+11
Neuromuscular panel
+19
EB mapping (net new)
+6
Research-only shelf
(+5 flagged R)
Supplement total
≈ +95
Combined menu: ~430–460 distinct antibodies — the top of the 300–500 reference-lab envelope, which is exactly where a national center serving all subspecialties should sit. Suggested sequencing: (1) fusion-surrogate sarcoma/CNS block (SS18-SSX pair, H3 G34R/V, p65+L1CAM, BCOR, DDIT3) — highest referral value per antibody; (2) neuromuscular + EB orphan panels — service-defining, low reagent risk (legacy clones); (3) nephropathology antigens + amyloid; (4) subtype-surrogate trio (GATA6/CK17-PDAC, SCLC-TF panel, GPNMB) as trial demand materializes; keep the S-R shelf strictly research-labeled.
Caveats specific to this supplement
Surrogate-IHC ≠ genotype: maintain a standing molecular-concordance log; sequence discordant and antibody-negative-but-suspicious cases (e.g., G34M, variant SS18::SSX, non-RELA ZFTA).
Decalcification and small poorly-fixed biopsies are the dominant false-negative mode for fusion-junction antibodies (documented for E9X9V/E5A2C) — repeat on better material before excluding.
Polyclonal lot drift is the chief analytic risk (EZHIP, NELL1, EXT1/2, CCNB3, GAB1, amyloid panel): re-verify each lot against index-case controls.
GPNMB and other MiT-target markers cannot distinguish translocation-driven from TSC/mTOR-driven tumors — report as pathway-level screen.
Amyloid IHC pre-typing must carry a mandatory LMD-MS recommendation line for therapy-determining cases (ATTR vs AL).
Prion work requires a segregated workflow, formic-acid protocols, and documented decontamination — plan before offering, not after the first referral.
Last updated