Caveat: These "tests" are not as exacting as are chemical, analytical tests. All persons involved ought to be properly trained and competent for interpretation. The lab should be officially accredited and involved in an official external competency survey program. The individual test should be quality controlled with a slide containing tissue elements with known positive (graded positives...weak/medium/strong...if possible) and negative control reactions to assure that the test system and reagents all worked properly. It is further optimal if the pathologist selects a tissue block for testing which will include some normal patient tissues which should react positively..."positive internal auto-control" (to assure that the test tissue was properly handled and processed and is capable of positive reactions). Prompt fixation (NBF is considered universally OK with at least 6 hours exposure in fixative & not much more than 48 hours) is vital. Not only should a positive reaction make sense with the control reactions, but it should make morphological sense, also! Finally, the astute clinician should always question an odd or clinically discordant test result!

See March 2008 issue of Archives of Pathology & Laboratory Medicine!

• Other website resources:

  • Dr. Dennis A. Frissman's database of (1)references for every antibody; and, (2)antibody reactions in various lesions...plug your IHC results in and get a differential diagnosis [you will need your own login registration or you can use: username, ebshaw and password, nwmd061ncs]

  • IHC World...........Pathology Outlines

• Nomenclature: Terminology is used very loosely in practice and can include,

  • CD numbers: "Clusters of Differentiation"; these are numbers assigned at the International Leukocyte Workshop (6 workshops as of 12/2001) to various antibodies/clones reacting with the same or similar antigens expressed on hematolymphoid cells (subsequent usage may reveal significant reactions with non-hematolymphoid tissues). Great detail

  • target-antigen designation

  • antibody-producing clone

• Sources we use for doing marker stains on our cases:

  • PAL IHC service @ Lexington Medical Laboratories ("LML"), Inc.... catalogue

  • PhenoPath catalogue

  • Impath physician services

• Unknown primary tumor¹⁰:

• tumor-specific breast-cell marker: BRST-2 (GCDFP-15)...is it primary or metastatic? [met. LMC-02-5503]
  • eccrine tumors can be ER, PR, and GCDFP pos.
  • ER is rarely positive in lung cancer & never positive in colon cancer.
• HER-2: at least when 2-3+, works equally well on Hartmann's-fixed nodes with mets.[LMC-04-2109A3]
• Ki67: use it to help locate the most mitotically active area of tumor for the Elston score (if need be); and it reflects entire % of tumor cells in active-proliferation parts of cell cycle.
• CK7: breast pos. (and so are the intraepidermal benign, clear, Toker cells located singly or as small groups around the nipple duct orifices) & especially in distinguishing intraepidermal Paget's from melanoma (both can be S100 pos.) ²⁰

• CK8:
  • positive cytoplasmic filaments relatively condensed perinuclear ring in lobular and diffuse cytoplasmic positivity in ductal
  • good for (ck7 and pan-ker also) highlighting a highly cellular dispersed invasive tumor pattern in proximity to a lumpectomy surgical margin...possibly more dangerous "close" margin [LMC-01-1915]; remember that it marks both benign and malignant breast epithelium [LMC-01-5927]
  • a micrometastasis marker in nodes from breast cancer surgery but can label plasma cells [LMC-01-7388B&C] (as can EMA) & reticulum cells [LMC-02-4776] but with a more diffuse or granular cytoplasmic pos. different from ca. cells (AE1/AE3 cocktail [our pan-keratin ] has least non-epithelial cell labeling...ck7 just as good/best); pankeratin highlights intracapsular infiltration and transcapsular invasion much superior to H&E [LMC-02-5028].
• E-cadherin: lobular carcinoma is negative; ductal carcinoma is positive; should probably test any relatively small-celled breast cancer with this in the absence of lots of tubule formation [LMC-02-2144;LMC-02-5160]

• ker903(34BE12)(Tavassoli, AJSurg. P 23(9):1048-1058, 1999):
  • ductal hyperplasia (all cells pos....including non-atypia parts of "columnar cell hyperplasia" [LMC-04-5858]) vs. ductal-AH (spotty cells pos. plus myoepithelial cells)[LMC-01-1497] vs. ductal CIS (all cells neg. except myoepithelial cells)...stains "mature epithelials" & ME cells
  • tumor sizing: helps be sure that a benign component is not  counted in the greatest diameter [LMC-01-5927] of a tumor...falsely causing a greater measurement
• laminin or type 4 collagen: stains basement membranes
• S-100: stains myoepithelial (ME) cells AND stains the epithelial cells and nuclei of the acini of microglandular adenosis (MGA) and atypical microglandular adenosis (AMGA)⁷, lesions which lack myoepithelial cells but do have anti-laminin positive intact basement membranes; Paget's can be pos.
• smooth muscle actin (SMA): stains myofibroblasts and ME cells
• CD10: stains ME cells
• CD34 marks myofibroblasts (The Breast J. 7[4]:263-265, 7-8/2001)
• smooth muscle heavy-chain myosin (smm-hc) : labels myoepithelial (ME) cells so as to be excellent help to distinguish invasion (ME neg.)  from CIS (ME pos.)[LMC-01-1384; LMC-02-4316];  in an ordinary adenosis area mixed with invasion (sort out the extent of the invasion); some complex benign stellate adenosis or epitheliosis vs. invasion [S-01-1380; LMC-01-4299]; and invasion vs. a complex papilloma [S-01-3967] or sclerosing papilloma [S-01-9404]
• ER, PR, and HER-2: we have had cases that are HER-2 2+ by IHC & FISH neg. [LMC-01-7037 & 7323]

• to categorize usual tumor types (see notes below table):
  
  

  ---

  Table Of Tumor Types

  ---

• Click To Expand

  Click To Collapse

  caution:
  • gliomas & GFAP: negative GFAP doesn't exclude glioma, especially if vimentin pos.¹⁴
• to predict behavior category:
  • meningioma: 
    1. increased Ki67 rate: increases likelihood to recur (Dr. Peter Burger @ Johns Hopkins [LMC-02-825]; (< about 4% Ki67 positive nuclei in meningioma indicates negative likelihood to recur; about 4-16% positive indicates atypical meningioma and >16% suggests "anaplastic meningioma"...Human Path, 1998.)[LMC-02-4277 5-10%]
    2. PR: 87% of malignant, 57% of atypical, and 36% of benign meningiomas have 0-9% of nuclei PR pos. & only 13% of malignants have 10% or greater positive nuclei¹⁹.
    3. Mitotic index/rate: benign, low (ave. < than 1) per 10 40x hpfs; atypical, ave. 5.15 (+- 0.93); malignant, ave. 17.53 (+- 3.73)¹⁹.
    4. risky histology: 2 or more of: high cellularity, sheeting (absence of lobulation or whorls), prominent nucleoli, and obvious mitotic figures¹⁹. So, worse meningioma outcome when combined PR = 0, mitotic >6, and malignant histology¹⁹.
  • grading glioma/astrocytoma: in a 3-grade system, Ki67 helps chose between I and II....if <3% positive, I; if >7%, II.
• ubiquitin:  labels Alzheimer's disease tangles; EGBs in pilocytic astrocytoma
• peripheral nerve evaluation: S-100 finds them; CD10 stains axonal myelin sheath [LMC-03-2740].

• GI medical issues:
• Helicobacter pylori: the IHC stain is great for identifying the organism, especially when sparsely present (easier than histochem. stains ); and a negative stain is a "must" to be sure no H. p. in any case where EGD is done and results in a biopsy
• LCA or CD3: helps one appreciate lumenal epithelial lymphocytic exocytosis of celiac disease in proximal small bowel (duodenal)...can look H&E normal...[LMC-01-2410; S-01-6591]; and  collagenous or microscopic lymphocytic colitis (diff. dx.) in the colonic biopsies. [S-01-289; S-01-3461; S-01-3499]
• CD10: identifies brush border of absorptive cells:
  1.  so as to note either non-goblet-cell intestinal metaplasia in esophageal biopsies 
  2. or a reduction in celiac disease of normal high ratio of absorptive cells to crypt lining cells [LMC-01-2410]
  3. or reduction or loss of border in disaccharidase deficiency
• CDX2: lung cancer is rarely positive and breast cancer never positive.
• premalignant assessments:
  1. Ki67 : prove a colon polyp (especially if family history of cancer; or is larger size and right colonic location) is not a serrate adenoma by showing that all positive epithelial nuclei are in the deep 66% of crypts [AJSP 23(9):1158-1160, 1999] [S-01-2292; S-01-2127; LMC-01-8025; S-01-14482].
• Barrett's:
1. ck7/ck20: stains to determine if G-E junction focal or short-segment intestinal metaplasia is of the Barrett's type (connoting premalignancy potential) or the gastric type (no or much less neoplastic connotation). Normal pattern=ck20 pos. in surface foveolar cells & ck7 pos. in mucosal glands=gastric-type intestinal metaplasia. A staining reversal occurs in Barrett's-type intestinal metaplasia (Barrett's CK7/20 pattern): ck20 marks only superficial surface & ck7 marks deep & surface^{3, 24} [LMC-01-1860; S-01-5944; S-01-1771]. Have had cases where only partial reversal (& signif. is uncertain) [S-01-11069]
2. p53 is being touted as a Barrett's high-grade dysplasia marker¹⁶ (coupled with c/w histology/cytology), p53 detectibility...nuclear over-expression...is indicative of p53 gene mutation (the normal gene inhibits...brakes...a cell's entry into S-phase).
3. CD10: if you are unsure whether there is a bit of intestinal metaplasia or not in the biopsies, CD10 can tag the brush border of non-goblet absorptive metaplastic cells
• malignancy issues & tumor classification:
  1. pan keratin: to check MALT-like lymphomatous processes and highlight any intra-tumoral epithelials in search of occult/obscure/residual lymphoepithelial lesions.
  2. GIST: (CD117 positive by definition...the product of the c-kit proto-oncogene, but a few said to be negative; 70 % [the myoid GISTs] CD34+; most are desmin neg. [LMC-01-8061], see tumors, below.
  3. unknown primary: COL-1 variant of CEA marks colorectal/intestinal adenocarcinoma [LMC-02-5503], to include small bowel [LMC-04-10307] and some gastrics.
  4. 33% gastric adenoca. can be CD68+ [LMC-03-35].

• Gynecologic:
  • medical issues:
  • tumor identification:
    • ovarian:
    • tubal:
    • uterine:
    • cervical:
    • vulvar:
  • biopsies and CIN  & HPV and early atrophy:
    • Ki67: helps discern early koilocytotic atypia (HPV viropathic change) because more than a few positive nuclei much higher than just basal/parabasal zone (Advances in A. P., 8(2):83-92, 3/01)(Crum, et. al., AJSP 25[7]:884-891, 2001) [S-01-2899; LMC-01-7221; S-01-8116; S-01-8237]; positivity correlates with HPV-type viropathic effect [LMC-01-6187]
    • therefore, if questionable koilocytotic cell change (perinuclear haloes and raisinoid nuclei) and Ki67 positivity not in those nuclei or high in epithelial layer, it may be best to just call minor atypia, if anything [S-01-8117, S-01-8542] ; but, note that,
    • in HPV lesions, Ki67 only stains some of the obviously abnormal and koilocytotic cells positive
    • reactive epithelium, or squamous metaplasia with atypia,  might have "a few" positive nuclei at higher zone...but mostly basal/parabasal [S-01-13041; S-01-13220]
    • reactive/early atrophy has reduced Ki67 positivity, even in basal/parabasal zone [S-01-12475; S-01-13313].

• Hematopathology:
  1. Reed-Sternberg cells: fascin (very R-S cell sensitive; dendritic cells pos.); CD15, CD30
  2. discern obscure follicular pattern in hyperplasia & PTGC: IgD¹⁸
  3. Ki67 proliferation marker: useful in follicular hyperplasia vs. follicular lymphoma...high rate and pattern of follicle polarity favors hyperplasia¹²
  4. hairy-cell leukemia marker: touch prep or marrow smear,  air-dried & then acetone fix prior to IHC with CD11c (LMC-01-8103); monocytes stain pos. (at least with flow cytometry)
  5. Bcl-2: pos. in many types B lymphomas [see write up] and value is in follicular hyperplasia vs. nodular B lymphoma¹²
  6. CD3 pan-T cell: most specific pan-T marker because no CD3 antigen in B cells...but some really immature T neoplasms can be void of CD3 & some can lose their CD3 antigen
  7. CD4 T helper/inducer: these hold the HIV receptor
  8. CD5: pan-T and a B subset (spleen & node mantel zone cells and pleural B lymphs; MCL & CLL/SLL pos. lymphoma) (cutaneous sweat gland sometimes pos.)
  9. CD8 T suppressor/cytotoxic (tumor infiltrating): marker is specific for this cell; labels cells of "large granular lymphocytic leukemia"
  10. CD10 (CALLA): granulocytes, normal follicular-center B cells,  and acute leukemia and follicular lymphoma with pos. cells spilling into the interfollicular zone¹² ; also some early B & T cells²²; and some in MCL  and CML in blast crisis and Burkitt's (polys are positive; some spindly stromal cells are pos.); also stains gut mucosal & renal tubular brush border & peri-axonal sheath.
  11. CD11c: HCL, above 
  12. CD15:
  13. CD20: pan-B cell (the intracellular epitope is L26...clone L26 reagent antibody) but epitope may not have developed yet in less mature, precursor-B neoplasms (see CD79alpha); as many as 20% of myeloma cases are pos. 
  14. CD23: SLL/CLL pos.; B follicular dendritic cells pos.¹²; mature B cells²²
  15. CD 30:
  16. CD56 T natural-killer (NK): best on formalin
  17.  CD68: histiocytic cells; 33% gastric adenoca.
  18. CD79 alpha: is a pan-B but appears earlier and positive in precursor-B tumors; is sort of an immunoglobulin marker & marks plasma cells and 50% myeloma cases positive
  19. CD117: mast cells
  20. CD138: plasma-cell marker (also + with many carcinomas)
  21. cyclin D1: mantel cell lymphoma has pos. cells; we've noted supranuclear "dot" cytoplasmic pos. in ileal epithelium [LMC-03-5237]
  22. ALK-1 protein positive cells in ALCL

• Liver:
1. ubiquitin: labels Mallory's hyalin
2. hunting cholangiocytes (biliary damage...ductopenia): may need to use ck7 and ck19 to hunt residual cholangiocytes; biliary adenoca. CEA, EMA, and keratin pos.⁸
3. tumors, HCC vs. met. adenoca. vs. cholangioca.:
   1. polyclonal CEA: canalicular pattern in 80% hepatomas
   2. Hep Par 1: 95% sensitive for HCC & HCC can be negative & some at least few pos. cells in some gastric, adrenal cortical, & lung cancers.

• Lung:
  • unknown primary met. vs. lung primary: non-small cell non-squamous ca. lung or thyroid TTF-1 positive; neuroend. neg [LMC-02-6548 intermediate neuroend.] for TTF-1.
  • lymphangioleiomyomatosis: HMB-45 + (leiomyosarcoma is HMB-45 -)
  • angiomyolipoma HMB-45 +
  • histo. type of carcinoma (radiation oncology does adjuvant CNS XRT if an adenoca. because much more likely than SCC to have occult CNS mets... remember that bronchogenic ca. may be mixed squamous and adeno): 
    1. small & non-small SCC: ck5 marks SCC (and mesothelioma) & K903 marks SCC; and SCC usually ck7 & LMW-keratin (ck 8, 18, and 19) neg. (non-small, nonsquamous pos. [LMC-02-6548 neuroend.])  and HMK-keratin (ck1, 5, 10, and 14) positive⁴. Small-cell SCCs neuroendocrine neg., TTF-1 neg., & HMW-ker pos.; spindled variant may be scant on keratin and pos. vimentin.¹¹ Cytology: tend to heavier chromatin, less likely large nucleoli, and more likely pyknotic nuclei as cells die; focal, dense squamous cytoplasm.
    2. small & non-small-cell adeno & BAT: most are ck7 pos.; ck20 neg.; 60-75% TTF-1 pos.; villin pos.; LMW-keratin (ck 8, 18, and 19) pos.; also, if CEA+, or CD15+, or LMW-ker+, it favors adeno over squamous. I'd advise using no less than two adeno markers so that a poorly differentiated non-small is SCC only if two or more "adeno" markers are negative.
    3. small-cell neuroendocrine: ck7 neg.; they are typically neuroendocrine pos. (synaptophysin most sens.), HMW-ker neg. and LMW-ker punctate pos., and high % are TTF-1 pos.& may have rare ck7 pos cell.¹¹[a small-bowel met LMC-03-1869]
    4. large-cell neuroendocrine: usually co-express keratin and vimentin 
    5. basaloid: nests of relatively undiff. small cells with lots of necrosis and some peripheral palisading (HMW-ker & LMW-ker pos.¹¹
    6. atypical carcinoid:
    7. carcinoid:
    8. lymphoepithelioma-like carcinoma: large anaplastic cells & most express LMW & HMW ker
    9. giant-cell carcinoma: at least 40% of cells are greater than 40 microns & usually co-express keratin and vimentin¹¹
  • clear-cell carcinoma HMB-45 +
  • ck7 highlights alveolar walls/pneumocytes
  • CD68 shows histiocytic/macrophage component [LMC-02-559]
  • CD34 shows alveolar capillary pattern or absence thereof [LMC-02-559].

• Lymph node micrometastases:
  • carcinomata:
    1. pankeratin (AE1/AE3): best for beast cases, see "breast" above.
    2. benign vs. malignant: it will take skillful comparison of the cytohistology of the primary vs. the node deposit to decide the DDX.
  • melanoma:
    • nodal nevus vs. malignant met.²⁷: capsule & parenchymal septal nevi will have no or very low Ki67 proliferation positivity & melanoma 2-80% of cells positive; & likely HMB45 negative & not have nucleoli²⁸ & can be present in cases which also have mets [L07-5254].
    • S100: global...most sensitive & least specific...will also stain all the numerous intranodal dendritic cells/macrophages (4 types: follicular dendritic, extrafollicular fibroblastic [S100 neg, desmin/sm-actin pos, vimentin pos] histiocytic, and interdigitating ).
    • vimentin: nearly all are pos.¹¹
    • panmelanoma (HMB45, MART1, and tyrosinase cocktail): most likely to be positive for melanoma-specific marking
    • HMB45: 50% of pulmonary primary are pos¹¹; some type of single mononuclear cells [LMC-01-6340] may stain pos. & may not be optimally sensitive¹
    • MART1: highly melanoma specific and no obscuring of nuclear detail

• Oral cavity:
  • S-100: lichen planus lesions sometimes have intraepithelial positive dendritic cells [S-01-3902], while some definite cases [S-01-2388] do not
  • Ki67: increased basal and parabasal positivity in an odontogenic keratocyst [S-01-12766]
• Parathyroid gland:
  • we have found that a cellular adenoma [LMC-02-608] has a Ki67 of about <10% and a carcinoma that subsequently metastasized [LMC-99-2428] was 50-75%.
• Pituitary adenoma/tumors:
  • neuropath lab at Univ. of Va. says that it is one of the few labs in North America focusing IHC & EM in this area 804-924-9175
  • our neurosurgeons don't need marker specification⁹
• Skin:
  • melanoma:
    • S100: though non-specific, nearly all mark with this and it can help prevent a misdiagnosis when working up an atypical fibroxanthoma (AFX)/malignant fibrous histiocytoma (MFH), always be sure that you've checked the melanoma angle with at least this marker [S-02-4382]
    • HMB45:
      • "all" (5-10% false neg.⁴)  melanoma cells pos. (CIS or invasive) vs. spotty pos. with dysplastic nevi vs. neg. with ordinary nevi [S-01-3737]
      • watch out for false pos. monocytes [LMC-01-6340] in lymph nodes¹
      • helpful for an exacting Breslow thickness measure when lots of inflammatory cells and when mixed with residual pre-existing melanocytic nevus [S-01-3737]
    • MART-1(Melan-A): marks benign and malignant (there are false negs.⁴) melanocytes and can help discern upward intra-epidermal scatter [S-01-9340] and it gives a clean view to help discern an accurate Breslow thickness AND to precisely document the early papillary dermal invasion of level II [S-01-9340] or reticular dermal invasion of a Clark level IV [LMC-01-5805; S-01-9973].
    • it may take a cocktail (our panmelanoma) of 3 or more melanoma markers to prove melanoma⁴
  • vascular...CD34: marks vascular endothelium (much better than factor VIII), solitary fibrous tumor, DFSP [LMC-03-6618], and Kaposi's
  • eccrine: calretinin pos.²³
  • mycosis/CTCL (cutaneous T-cell lymphoma): epidermotrophic cells said to be "CD4 helper" pos.

• • mast cells: CD117+ (as are melanocytes)
  • Merkel cell tumor: IHC for neurofilament protein (NFP) pos. in 50% with a "dot" character discernable²⁹; most are CD56 pos.; NE lung met to skin is TTF-1 pos & Merkel neg.; 90% are ck20 pos. with a "dot" character discernable²⁹; & 80% are NSE pos.²⁹; commonly neuropeptide (chromogranin & synaptophysin) pos.
  • carcinoma...possible adnexal
• Soft tissue & vascular:
  • vascular/endothelial: CD34 marks vascular endothelium (much better than factor VIII), solitary fibrous tumor, DFSP [S-01-14379CD34 NEG.], and Kaposi's.
  • AFX/MFH mark with vimentin & alpha-1 ACT [LMC-02-128; S-03-12071] and negative for spindled/sarcomatoid SCC by K903 [LMC-06-8409].
  • peripheral nerve sheath tumors S100 & CD34+.
  • CD-1a marks Langerhans histiocytes (usual histiocytes are negative).
  • CD68: histiocytes⁴.
  • epithelioid angiomyolipoma is HMB-45 + & neg. for S100 & cytokeratin.
  • GIST CD34 + and CD-117 + (see below).
  • IMFT is ALK-1 pos.
  • neurofilament (NFP): neurons, neuroma, neuronal tumors: ganglioneuroma, ganglioneuroblastoma, neuroblastoma, paraganglioma, pheochromocytoma, carcinoid, skin neuroendocrine ca. (Merkel cell), neuroendocrine ca. elsewhere. Presence of S100+ sustentacular cells in an orderly, ordinary pattern tends to favoe a benign paraganglioma or pheochromocytoma (along with smaller size, excellent circumscription, negativity for necrosis, and low mitotic rate).
  • desmin (smooth & skeletal) and myogenin are skeletal muscle marker⁴.
  • smooth muscle actin (SMA) marks smooth muscle⁴.
• Thyroid:
  • unknown primary: non-small cell & non-squamous ca. lung and thyroid are TTF-1 positive (but any definite thyroid cancer primary which is truly TTF-1 negative is a cancer that cannot produce thyroglobulin); medullary ca. TTF-1 pos. & atypical carcinoid of larynx is negative .
  • ck19: marks papillary carcinoma positive (valuable in proving papillary variant of follicular carcinoma) (Human Pathology, Jan. 2001); BUT positive...diffuse reactivity...in only 66% of papillary².
  • ret: nonreactive in benign lesions; pos. in 69% of papillary².
  • calcitonin: marks medullary carcinoma [LMC-01-5521] and thyroid C cells and moderately differentiated neuroendocrine (atypical carcinoid) carcinoma of larynx .
  • HBME-1: positivity may be an indicator of malignancy².
  • 9/138 papillary carcinomas (7 follicular variant & 2 classic) negative for all 3 of ck19, ret, and HBME-1².

• Tumor classification:
  • GIST: the specific marker for this tumor is CD117 (stains the c-kit growth factor receptor) [4/2000] & positivity identifies tumor as from interstitial cell of Cajal and NOT an ordinary sarcoma; GIST is CD34⁺ [LMC-01-8061] & responds to the tyrosine kinase inhibitor STI571 (Gleevec) [6/6/01]; CD117 also stains mast cells and melanocytes [S-02-10146]. Epithelioid GIST may mark poorly for both.
  • neuroendocrine tumors (NET, NEC): need compatible light microscopy nuclei; are NSE pos. but NSE almost too sensitive & poorly specific; NFP & CD56 are sort of pan-neuroendocrine ca. markers & chromogranin & synaptophysin follow. Over 80% of hypercalcemic NET (or NEC) of ovary are WT-1 & infrequent in others, while high % NEC of lung are TTF-1 positive, and most NEC of cervix are HPV positive³¹.
• Urologic:
  • Kidney:
    • medical issues:
    • tumor types:
      1. EMA: RCC positive.
      2. c-kit (CD117) cytoplasmic pos. with papillary and membrane pos. with chromophobe.
      3. EGFR neg. in oncocytoma & pos. in chromophobe²⁵.
      4. CD10 pos. clear cell & papillary & neg. in chromophobe.
      5. IHC & other all types.
      6. urothelial is vimentin neg. (?).
  • Prostate:
    • medical issues:
    • cancer diagnosis:
      1. keratin903: stains basal acinar cells positive; (1) absence of staining in radiated prostate or relative to ASAP (atypical small acinar profiles) lesions PLUS (2) suspicious pattern (3) & cytological features is strong evidence in favor of adenocarcinoma.
      2. P504S¹⁸: stains acinar cancer cells in 92% of cases and also stains hi-grade PIN.
      3. Ki67: a labeling index of 7.1% or higher is related to biochemical failure of therapy, distant metastasis, and cancer specific death ; behaves more like a Gleason 8 [S-04-202].
      4. poorly diff. prostate acinar vs. urothelial: IHC for PSA pos. >95% cases; HMWK & p63 pos. in urothelial & negative in prostate.
      5. neuroendocrine prostate can be TTF-1 positive³⁰.
  • Testis:
    • calretinin: Leydig (and Sertoli ?) cells pos.²³
    • PLAP: seminoma marker (but can see +- in embryonal & neg. in chorio.)
    • vimentin: embryonal & chorio are neg.
    • HCG: trophoblastic positive & not other testicular GCTs.
    • CD30: embryonal carcinoma positive²⁶ & not other testicular GCTs.
    • CD117: seminoma positive²⁶ & not other testicular GCTs.
    • AFP: yolk sac pos. (rare +- embryonal)
  • Urinary bladder & urethra:
    • Ki67 is basal in reactive/hyperplasia & variably upward in dysplasia²¹.
    • reactive vs. CIS: ck20 normally + only for surface "umbrella cells" & all + for CIS; p53, normals neg. & CIS pos .
    • And, using H&E morphology and possibly molecular markers (such as ck7 & ck20 & possibly K903: SCC is ck7/20 negative while TCC is ck7/20 positive; SCC should also be K903 positive), is it SCC or TCC with some SCC features [L06-9298]?

References:

1. Komorowski R, Eval. of Micromet. in SLN of Cutaneous Melanoma, AJSP 25(8):1039-1046
2. Modern Pathology 14:338-342, 2001.
3. Ormsby AH, et. al., @ Cleveland Clinic, Gastroenterology 119:683-690, 9/2000.
4. PhenoPath Clin. Ref. Manual & e-mails
5. Roberts F, et. al., Scotland, AJCP 116:253-262, 8/2001
6. we perform @ LML
7. F. A. Tavassoli, MD, or Daniel Sullivan, MD,  Medical College of Ga., Seminar, 21 April 2001
8. Ackerman's Surgical Pathology [text], 8th Ed., 1996, Juan Rosai
9. members of specialties on our medical staff
10. Applied Immunohistochem. 3(2):101, 1995
11. Dabbs text
12. Chan JKC, syllabus, 111th Semi-annual Seminar (CTTR), 2 Dec. 2001 [we all have a copy]
13. Hoff ER, et. al. HER2/nue Amplification in Breast Cancer, AJCP 117:916-921, June 2002.
14. Burger PC text, SP of the Nervous system and Its Coverings, 4th Ed., 2002 (Dr. Carter's office) 
15. Wu SL, Modern Pathology 15(7):712-717, July 2002.
16. e-mails, letters, communications: GI path. experts
17. AFIP Fascicle, "Tumors of Neuroglia and Choroid Plexus Epithelium.
18. PhenoPath Lab letter, Dr. Allen Gown, 5(3):2, Nov. 2002
19. Hsu DW, et. al., J. Neurosurg., 86:113-120, Jan. 1997.
20. Rosen PP, Rosen's Breast Pathology 2nd Ed., 2001. 
21. Sun W, Applied Immunohistochemistry & Molecular Morphology 10(4):327-331, Dec. 2002 (Dr. Armstrong's journal)
22. Caldwell CW & Lacombe F, Evaluation of Peripheral Blood Lymphocytosis, 125 pages (a flow cytometry book), 2000. [Dr. Armstrong's book]
23. Zymed catalogue, 2002-2003.
24. Goldblum, John R, chair of AP @ Cleveland Clinic, ASCP Teleconference 10/10/03, AV & handout.
25. Shah, Sejal, "Expression of EGFR...", Poster #29, ASCP annual meeting in Las Vegas, 10/2006.
26. Leroy X, et. al., "CD30 and CD117 (c-kit) Used in Combination Are Useful for Distinguishing Emryonal Carcinoma from Seminoma", from France, J. of Histochem. & Cytochem. 50(2):283-5, 2002.
27. Lohmann CM, et. al., @ Memorial SKCC in New York, "Expression of Melanocytic Differentiation Antigens and Ki-67 in Nodal Nevi and Comparison of Ki-67 Expression With Metastatic Melanoma", AJSP 26(10):1351-1357, 2002.
28. Biddle DA, et. al., "Intraparenchymal Nevus Cell Aggregates in Lymph Nodes..." Am. J. Surg. Path. 27(5):673-681, 2003.
    (posted Feb. 2001; latest addition 30 May 2007).
29. McKee, Calonje, & Granter, Pathology of The Skin... Two volumes, 3rd Ed. 2005.
30. True LD, CME course 1 May 2005, 12th Annual SP Seminar, Pittsburgh .
31. Carlson JW, et. al., "Biomarker-assisted...", Histopathology 51:305-312, 2007 [EBS subscribes].
32. Archives of Pathology & Lab. Med., "Special Issue: Diagnostic Immunohistochemistry" 132(3):319-509, March 2008.