9.1 Morphology, Anatomy, and Basic Pathology

9.1 Morphology, Anatomy, and Basic Pathology

Quick Answer: IHC is a colored stain, not a diagnosis. Read the H&E first so you know which cell you are scoring. True signal sits in a biologically expected compartment—nuclear, cytoplasmic, membranous, Golgi, canalicular, or dot-like—and should appear in internal controls such as residual normal epithelium. Use carcinoma versus sarcoma versus lymphoma only as a first-line panel frame. Edge burn, pigment, crush, and floaters are not antigen.

This OpenExamPrep section is independent teaching on morphology, expected IHC localization, and basic pathology frames covering published QIHC staining topic areas under tissues (morphology, staining patterns, microorganisms, and basic pathology). It is not an ASCP publication and does not claim Board, CAP, or manufacturer approval. It is staining-pattern teaching, not a separate diagnostic-panel domain.

Why IHC is interpreted with H&E

A brown or red precipitate does not name a tumor. The hematoxylin and eosin (H&E) section tells you whether the cells of interest form glands, squamous pearls, spindle fascicles, or a discohesive round-cell sheet. It also shows whether residual normal structures remain—bronchiolar epithelium, prostatic glands, colonic crypts, breast ducts, thyroid follicles—that can serve as internal controls.

A practical sequence on the bench and at the microscope:

1. Confirm the block and level match the request (correct specimen, not a recut of the wrong cassette).
2. Read the H&E for architecture, cell type, necrosis, pigment, crush, and cautery.
3. Predict the compartment the antibody should occupy if the antigen is present in that cell type.
4. Confirm the external control stained as expected for this run.
5. Look for internal-control staining in residual normal tissue on the patient slide.
6. Only then interpret the tumor cells—or the suspected microorganisms in Section 9.4.

If the H&E shows crush or cautery, nuclear transcription-factor stains often fail or look smudged in that zone even when the clone and the rest of the run are fine. If the H&E already shows anthracotic pigment, hemosiderin, or melanin, a DAB chromogen can be indistinguishable from that pigment unless you use a contrasting red chromogen, a laboratory-validated bleach step, or a careful side-by-side with the H&E.

IHC never replaces anatomy. CK7 in a gland that is clearly a residual bronchiole is not metastatic carcinoma. TTF-1 in residual pneumocytes at the edge of a lung scar is not proof that a spindle plaque is adenocarcinoma. Nuclear p63 in basal cells of a prostatic gland argues the gland still has a basal layer; it is not a tumor-positive stain.

Expected subcellular localization

Antibodies are validated for a compartment. Signal in the wrong place is either a documented biologic variant or artifact. Do not promote cytoplasmic blush to a nuclear-positive result because the tumor type “should” have been positive.

Nuclear stains are read in nuclei. A TTF-1 that only dirties cytoplasm is not a positive nuclear result. Membrane stains are read on membranes. Cytoplasmic cytokeratin in a carcinoma is expected; cytokeratin confined to a crushed edge with a clean center is suspicious for artifact.

Dot-like keratin in Merkel cell carcinoma is a true pattern: perinuclear intermediate-filament balls. Do not dismiss it as artifact because it is not diffuse. Conversely, a few DAB specks over neutrophils, over glass, or over a fold are not Merkel dots.

Internal controls, especially residual normal epithelium

External on-slide or separate controls prove the run worked. Internal controls prove the tissue on that slide can still show the antigen after fixation, processing, and retrieval.

Useful internal epithelial examples:

• Residual squamous epithelium should label with high-molecular-weight keratins / CK5/6, and basal-to-mid-layer nuclei should label with p40/p63, depending on clone and site.
• Residual lung pneumocytes and bronchiolar epithelium often provide nuclear TTF-1 and granular cytoplasmic napsin A in type II pneumocytes and some bronchiolar cells (clone-dependent patterns).
• Residual colonic epithelium is a built-in CK20 and CDX2 check.
• Residual prostatic basal cells are why PIN-4-type cocktails work: p63 and high-molecular-weight cytokeratin should mark basal cells if benign architecture remains (Section 9.3).
• Residual thyroid follicles should show thyroglobulin in colloid/epithelium and nuclear TTF-1/PAX8 in follicular cells.

If the tumor is negative and the residual normal structure that should be positive is also negative, treat the slide as a failed stain (retrieval, drying, omitted primary, wrong antibody, over-fixation of that block)—not as a biologically negative tumor. If the tumor is negative and the internal control is positive, a true negative is more believable.

Absence of residual normal tissue (a tiny crushed biopsy or a pure tumor block) means you lean harder on the external control and on morphology. You cannot invent an internal control that is not on the slide.

Carcinoma versus sarcoma versus lymphoma as a panel frame

QIHC work is staining and preliminary pattern reading, not issuing the surgical pathology diagnosis. A simple lineage frame still decides which first antibodies are even rational.

This frame prevents wasting tissue on SATB2 when the H&E is obviously a lymphoma, and it prevents calling a spindle plaque a carcinoma because one cell took up dirty DAB. Vimentin is too widely expressed to prove sarcoma by itself. CD45 supports a hematolymphoid lineage when it labels the tumor cells, not the tumor-infiltrating lymphocytes you accidentally scored.

Mesothelioma, germ-cell tumors, and melanoma sit in the gaps of a three-bucket frame. The technologist’s job is not to issue those names from one cocktail. The job is to recognize that expected staining patterns follow cell type, and that a pan-keratin cocktail is a lineage screen, not a magic tumor name.

Artifact versus true stain

True stain occupies the expected compartment in the expected cell, appears in internal control tissue when that tissue is present, survives comparison with the negative reagent control (omit primary or isotype), and matches cell borders or nuclei rather than glass, folds, or cautery rims.

Common mimics:

• Edge artifact / penumbra: strong stain at the tissue margin that fades centrally, often from drying, free antibody, or retrieval damage at the edge. Do not score only the burnt rim.
• Section folds and chatter: trapped chromogen looks positive.
• Crush and cautery: nuclear stains smear; membranes disappear.
• Pigment: anthracosis (black, chunky, in macrophages), hemosiderin (golden-brown, granular, Prussian-blue positive on a separate stain), melanin (fine brown, bleach-sensitive), formalin-heme pigment (often in blood-rich tissue). DAB is also brown. If the H&E already shows pigment in the same cells, do not call it IHC signal.
• Floaters and knife carryover: a fragment of control or another case. The floater’s morphology often does not match the H&E tumor.
• Dirty background: collagen, necrosis, and serum can bind polymer. Background does not respect compartments.
• Endogenous pigments versus organisms: carbon and hemosiderin are not Helicobacter; melanin is not a fungus (Section 9.4).

Worked example

A lung wedge has keratin-positive spindle cells at the pleural surface. The H&E shows entrapped hyperplastic pneumocytes and a separate fibrous plaque. TTF-1 labels cuboidal cells lining clefts (residual epithelium) and does not label the spindle cells. The keratin in the spindles is weak, cytoplasmic, and strongest at the cautery edge. The better reading is entrapped epithelium plus edge artifact, not sarcomatoid carcinoma proven on a single dirty AE1/AE3. A true epithelial malignancy should show keratin in the neoplastic population in a pattern that matches the H&E tumor, with internal controls behaving as expected.

Morphology first, compartment second, internal control third, tumor last.