5.1 Cytology Specimens for IHC

5.1 Cytology Specimens for IHC

Quick Answer: Cytology IHC is not the same antibody on a smaller sample. Alcohol-fixed smears, air-dried smears, liquid-based monolayers, and alcohol-based cell blocks are different fixatives from 10% neutral buffered formalin (NBF) paraffin tissue. Methanol-based CytoLyt and PreservCyt can weaken or abolish some nuclear and predictive epitopes. Formalin postfixation after alcohol may recover some antigens and fail on others. Predictive markers on cytology are high-risk because scoring systems were built on FFPE cores and resections. CAP 2024 Goldsmith conditional recommendation 9 requires a separate analytic validation for each analyte and each cytology fixation method that is not the original tissue validation set. Good-practice statement 10 recommends a minimum of 10 positive and 10 negative cytology cases if possible, and more if the marker is predictive.

This OpenExamPrep section is independent teaching on cytology specimen handling as it appears under Specimen Handling on the published QIHC topic outline. It does not claim ASCP, CAP, ASCO, or manufacturer approval. Chapter 4 covers routine tissue fixation, processing, and microtomy. Here the job is the cytology stream: smears, liquid-based vials, and cell blocks, and why an FFPE IHC file cannot be copied onto them. Full CAP numeric tables for nonpredictive 10+10 tissue sets, predictive 20+20, lot 1+1, and clone-change revalidation belong in Chapter 12. The cytology separate-validation point is a handling rule and is taught here so it is not skipped.

Why cytology is not small FFPE

A surgical block sits in 10% NBF, is processed through alcohols and xylene, and is embedded in paraffin. Fine-needle aspirates, fluids, brushings, and washings often never see that sequence. They may be smeared into 95% ethanol, air-dried on the bench, collected into methanol-based CytoLyt or PreservCyt, collected into ethanol-based SurePath-type media, or clotted into a cell block that is either formalin-fixed or left in alcohol.

Alcohol is a coagulative dehydrating fixative. It precipitates proteins; it does not build the same methylene bridges that formalin does. Epitopes that survive or even require heat-induced epitope retrieval (HIER) on FFPE may already be available—or already damaged—on an ethanol smear. Protease retrieval that is gentle on a breast core can lift an entire smear off the slide. Predictive cutoffs (PD-L1 tumor proportion, ER percentage or Allred-type scores, HER2 0–3+) were developed on formalin-fixed tissue. Using those cutoffs on a CytoLyt cell block without a cytology-specific validation is a method error, not a close-enough shortcut.

Cellularity is also different. A smear may have three-dimensional clusters, crush, drying rings, mucus, and fewer than 100 tumor cells. A cell block may be mostly blood clot. No polymer detection system repairs a specimen that does not contain evaluable tumor.

Alcohol-fixed smears

Immediate immersion of a wet smear in 95% ethanol (or an ethanol-ether Pap fixative) is the classic Pap-stain preparation. Chromatin is crisp; cells are flattened. For immunocytochemistry:

• Use charged or otherwise adhesive slides. Smears lift during washes and retrieval.
• Do not assume the laboratory's FFPE HIER time is safe. Many ethanol-fixed nuclear antigens still need heat retrieval, often with citrate around pH 6 rather than a long high-pH run that shreds the film.
• Enzyme retrieval frequently destroys alcohol-fixed smears. Treat enzyme-induced epitope retrieval as a last, validated exception, not a default copied from colon biopsies.
• Some membrane and cytoplasmic antigens stain without retrieval on ethanol smears even when the same clone needs HIER on FFPE. That is a protocol difference, not proof the clone is stronger.

If the smear dried before alcohol, you have a mixed air-dried/alcohol specimen. Document it. Mixed fixation is its own method.

Air-dried smears

Air-dried smears are the Romanowsky/Diff-Quik stream: cells look larger, chromatin is coarser, and drying artifact is expected. For IHC they are usually rehydrated and postfixed (methanol, ethanol, or formalin) before antibody. Without postfixation, antigens leak and morphology collapses in aqueous buffers. Air-dried plus later alcohol is not equivalent to immediate wet-fixed alcohol, and air-dried plus later formalin is not equivalent to a formalin cell block. CAP 2024 lists air-dried smears among cytology preparations that need their own validation when they are not the original tissue fixative.

Worked example. A thyroid FNA has one air-dried Diff-Quik slide and one ethanol-fixed Pap slide. The pathologist wants thyroglobulin and calcitonin. You cannot pick the prettier slide and run the surgical-block protocol on both. Each preparation is a different starting matrix. If only one method is validated, stain that method—or validate the other before you report.

Liquid-based cytology

Liquid-based cytology (LBC) collects cells into a vial, then transfers a monolayer to a slide. ThinPrep-type systems use methanol-based CytoLyt (collection, hemolysis, mucolysis) and PreservCyt (preservative vial). SurePath-type systems use ethanol-based preservative. Advantages: less obscuring blood and mucus, leftover vial material for a cell block or molecular tests. For IHC, the vial chemistry is the fixative. A monolayer pulled from PreservCyt has already been methanol-exposed. Running the laboratory's FFPE CK7 protocol on that slide without a PreservCyt validation is the same error as staining an alcohol cell block from an FFPE file.

Keep residual vial fluid if a cell block or extra slides may be needed. Once the vial is discarded, you cannot reconstruct a formalin-fixed block from memory.

Cell blocks: formalin versus alcohol

Cell blocks are how cytology becomes something a histotechnologist can section.

Formalin-fixed cell blocks are the preferred cytology substrate for most predictive IHC when a tissue core is not available, because they share aldehyde cross-links with the assays' original clinical evidence. Alcohol-only blocks are convenient leftovers from Pap vials. Convenience is not concordance.

CytoLyt and PreservCyt antigen effects

CytoLyt and PreservCyt are alcohol-class (methanol) media, not 10% NBF. Cells often see CytoLyt first, then PreservCyt. Published cytology series show that methanol-based media can reduce staining intensity or apparent positivity relative to matched formalin cell blocks. Effects are clone-specific, not a rule that alcohol always kills IHC:

• TTF-1 clone 8G7G3/1 is more vulnerable than some other TTF-1 clones in alcohol cell blocks.
• PD-L1 clones used in lung algorithms, including 22C3 and SP263 in published comparisons, often stain weaker after CytoLyt or PreservCyt than after formalin; concordance at clinical cutoffs can drop.
• Ki-67 / MIB-1: CytoLyt can markedly inhibit MIB-1 labeling; the Ki-67 30-9 clone is less susceptible in published cell-block comparisons. A falsely low MIB-1 index on a CytoLyt block can mis-grade a neuroendocrine or other proliferation-dependent tumor.
• Hormone receptors (ER/PR) have shown alcohol-related loss in some cell-block studies.

Longer alcohol exposure can worsen loss for some antibodies. Do not memorize one complete CytoLyt-sensitive list. Memorize the mechanism: methanol coagulation plus time in the vial can hide or destroy epitopes that an FFPE protocol assumes are cross-linked and retrievable.

Postfixation in formalin

Laboratories often transfer an alcohol-collected pellet into 10% NBF before processing, hoping to convert the specimen into FFPE. Formalin postfixation can partially restore immunoreactivity for some antibodies (certain PD-L1 22C3 laboratory protocols in published cell-line work) and fail to restore others (MIB-1 after CytoLyt; some PR results in alcohol-fixed blocks). Once proteins have been alcohol-denatured, adding formalin does not rewind the specimen to a never-alcoholated core biopsy.

Treat alcohol-then-formalin as a third method, not as honorary NBF-only tissue. CAP 2024 explicitly includes specimens collected in alcohol or alternative fixative media that are postfixed in formalin among cytology types that need separate validation when they differ from the original tissue validation set.

Why predictive markers on cytology are high-risk

Predictive IHC (breast ER/PR/HER2, PD-L1, some ALK IHC, mismatch-repair proteins, Ki-67 used with a numeric cutoff) is high-risk on cytology for stacked reasons:

1. Fixative mismatch as above.
2. Scoring-system mismatch. HER2 0/1+/2+/3+ membrane rules and PD-L1 tumor-proportion or combined positive scores were locked to FFPE morphology. Three-dimensional clusters and stripped nuclei on smears make membrane completeness hard to judge.
3. Minimum tumor-cell counts. Many companion assays expect a stated minimum of viable tumor cells. A hypocellular FNA fails that before the antibody is applied.
4. Controls. An FFPE on-slide control does not prove the cytology slide was retrieved and detected correctly. Cytology validations need cytology-processed positive and negative material.
5. Clinical stakes. A false-negative PD-L1 or ER result can withhold therapy; a false-positive HER2 3+ can send a patient to the wrong drug.

ASCO/CAP breast biomarker guidance still prefers adequately formalin-fixed tissue (cold ischemia of 1 hour or less; formalin fixation 6–72 hours) for ER/PR/HER2. Cytology is a fallback when tissue cannot be obtained, not the first-choice matrix. Independent QIHC study material treats that as a specimen-handling constraint, not as a claim that OpenExamPrep rewrote those breast guidelines.

CAP 2024: separate cytology validation (recommendation 9 and statement 10)

The 2024 CAP analytic-validation guideline (Goldsmith et al.) added cytology-specific statements. Teach them as handling implications:

• Conditional recommendation 9: If IHC is performed on cytologic specimens that are not fixed in the same manner as the tissues used for the initial assay validation, the laboratory should perform separate validations for every new analyte and corresponding fixation method before clinical use.
• Guideline examples include air-dried and/or alcohol-fixed smears, liquid-based preparations, alcohol-fixed cell blocks, and alcohol or alternative-media specimens later postfixed in formalin.
• Good-practice statement 10: A minimum of 10 positive and 10 negative cytology cases is recommended for each such validation, if possible. The medical director should consider increasing the number if predictive markers are being validated. If 10+10 is not feasible (rare antigen, scant positives), document the rationale for fewer cases.

If the original validation used 10% NBF FFPE surgical tissue, a CytoLyt cell block is not covered. If you later add an air-dried smear protocol for the same clone, that is another fixation method and another validation file. CAP commentary notes it is not the intent of statements 9 and 10 to mandate retrospective revalidation of every historical cytology assay that already had documented cytology validation—but missing documentation is not a waiver.

Worked example. CK7 was validated on surgical NBF blocks. A pleural fluid arrives in PreservCyt; the cell block is alcohol-fixed. Reporting CK7 from the surgical file alone does not meet recommendation 9. Build a PreservCyt/alcohol-block set: at least 10 CK7-positive and 10 CK7-negative cytology cases if possible, processed the same way as patient fluids, and file the concordance. If this were PD-L1 instead of CK7, plan a larger set, not a smaller one.

High-yield traps

• Using an FFPE on-slide control as proof that a smear stained correctly.
• Treating CytoLyt and formalin cell blocks as interchangeable because both were labeled cell block.
• Running MIB-1 on CytoLyt material and trusting the index without a CytoLyt-validated Ki-67 clone or method.
• Applying protease retrieval copied from a colon biopsy to an ethanol smear (cells float away).
• Calling a three-cell cluster HER2 3+ with a tissue-only validation.