11.1 Staining Troubleshooting
Key Takeaways
- No stain on both the patient and the control is a reagent or protocol miss until proven otherwise; do not report a true negative from a dead run.
- No stain on the patient with a working control is a true biologic negative or a problem in that specimen; a dead control with stain on the patient invalidates the run.
- High background has named mechanisms: endogenous enzyme, endogenous biotin, antibody too concentrated, drying between steps, and retrieval that is too harsh.
- Wrong-compartment color, edge-only chromogen, section wash-off, floaters, and native pigment (melanin, hemosiderin, carbon) are appearance diagnoses distinct from DAB.
- A hook-like picture from antibody excess is dirty, less crisp stain—dilute the primary; do not concentrate it further.
11.1 Staining Troubleshooting
Quick Answer: Read the patient slide and the control together. No stain on both means a reagent or protocol miss. No stain on the patient with a working control is a true negative or a specimen problem. No stain on the control invalidates the run. High background, wrong compartment, edge-only color, wash-off, floaters, and native pigment each have a short list of mechanisms. Operational QC—lots, logs, and validation files—belongs in Chapter 12. This section is what the stained slide looks like and why.
This OpenExamPrep section is independent teaching on IHC stain-appearance troubleshooting covering published QIHC staining topics. It is not an ASCP publication and does not claim Board, CAP, or manufacturer approval.
Start with two slides, not one
Every unexpected IHC result is a paired observation: the patient section and the on-slide or same-run control. Morphology on the hematoxylin counterstain (and the H&E of the same block) tells you whether tissue is present, necrotic, crushed, or the wrong piece. Chromogen tells you whether a reaction product was deposited. Those three views—patient, control, morphology—are the decision table. Do not retiter an antibody because one patient looked blank if you have not asked whether the control is blank too.
Chapter 12 covers lot numbers, temperature logs, and how a laboratory documents a failed run. Here the question is narrower: what does the stain look like, and which mechanism produces that picture?
No stain: patient and control versus patient only versus control only
No stain means expected chromogen is absent in cells that should label, not that the glass is empty. If the section washed off, that is a different row.
| Pattern | What you see | First mechanisms to name | What you do not do |
|---|---|---|---|
| Patient and control negative | Both lack specific chromogen; counterstain may still be present | Missed or expired primary; skipped detection or chromogen; wrong instrument protocol; retrieval omitted or buffer empty; detection kit exhausted | Report the patient as a true negative; repair only the patient slide |
| Patient negative, control positive | Control shows the expected compartment and intensity; patient tissue is present and unstained in the expected cells | True biologic negative; antigen not in this block; necrotic or sampled-out tumor; overfixed or poorly processed patient specimen; wrong tissue submitted | Call the assay failed; change the clone on a single clinical slide without a reason |
| Control negative, patient positive (or both unreadable) | Patient has color; the positive control is blank | Wrong or exhausted control tissue; control dried; control not retrieved; aged control section; mix-up of control and blank | Report the patient's positive or negative result from this run |
Both negative is a run failure until you prove a reagent reached the tissue. Walk the protocol in order: dewax, retrieval, peroxide or biotin block, primary, detection, chromogen, counterstain. A polymer detection that never dispensed, a DAB well filled with water, or a primary vial that still had an unpunctured seal will blank every slide on that program.
Patient only negative is the pattern that tempts a false the-antibody-failed story. If the control is crisp, the antibody and detection worked on that run. Ask whether the patient should have been positive. A colon adenocarcinoma that is CK20-negative can be real. A tonsil that is CD3-negative in T-cell zones is not. If morphology says the antigen should be there, look at that specimen's fixation, processing, and whether you stained the correct block—not at the stock bottle.
Control only negative means you cannot interpret the patient. A stained patient with a dead positive control might be nonspecific chromogen, a floater, or luck. A negative patient with a dead control might be a true negative or the same reagent miss that killed the control. Repeat with a known-reactive control.
Worked scenario: both blank
A pan-cytokeratin run on a skin biopsy and a skin control comes off the stainer with blue nuclei and no brown. The H&E of both blocks shows epithelium. The instrument log shows the primary was not pipetted. This is patient and control negative from a missed primary, not a new retrieval mystery. Restore reagent, repeat both slides, and only then interpret.
Worked scenario: patient blank, control good
The same pan-cytokeratin control is strongly cytoplasmic in epidermis. The patient is a necrotic ulcer base with ghost cells and no intact epithelium on the IHC slide. The blank patient is sampling plus necrosis, not antibody failure. Cut a better block or report that residual epithelium is not present to score.
Weak stain
Weak stain is specific chromogen in the right compartment that is too faint to trust, on tissue that is still on the slide.
| Pattern | Mechanisms |
|---|---|
| Patient and control weak | Under-retrieval; primary too dilute or too short; cold or expired primary; short chromogen development; exhausted DAB or AEC; overly heavy hematoxylin hiding pale brown |
| Patient weak, control strong | Low antigen density in the patient; overfixation of that specimen; decalcification damage; a clone mismatch for that epitope in that tumor |
| Control weak, patient strong | Aged, dried, or poorly chosen control; control from a different processor |
Do not turn up chromogen time as the first fix for a locked predictive assay. Weak nuclear ER after citrate with a strong result after high-pH retrieval is a retrieval story (Chapter 6), not a mountant story.
High background
Background is chromogen or dirty blush outside the expected cell type or compartment. Name a mechanism; do not stop at the word nonspecific.
| Mechanism | Classic picture | Why it happens |
|---|---|---|
| Endogenous enzyme (peroxidase or alkaline phosphatase) | Red cells, neutrophils, and eosinophils labeling on HRP-DAB; brush-border or frozen-tissue blush on AP systems if levamisole was skipped | Tissue enzyme converts substrate without antibody |
| Endogenous biotin | Hepatocytes, renal tubules, and some tumors dirty on ABC/LSAB; cleaner on biotin-free polymer | Avidin binds tissue biotin |
| Titer too concentrated | Diffuse cytoplasmic blush, dirty stroma, loss of a crisp membrane or nuclear edge | Excess primary sticks widely; detection has more to bind |
| Drying between steps | Harsh edges, baked brown, crystalline deposits, especially after retrieval or a dry pad | Dried protein and polymer become sticky films |
| Retrieval too harsh | Frayed section, lost nuclear detail, stromal mud, sometimes lift-off | Over-opened hydrophobic patches bind detection reagents |
A negative reagent control (omit primary, keep detection) splits enzymatic, biotin, and polymer background from primary-antibody background. If the reagent control is dirty, the primary is not the first bottle to dilute. Thick sections and a skipped protein block add uniform muddy stroma; they are supporting actors, not a substitute for the table above.
Wrong localization: nuclear versus expected membrane
Expected compartment is part of the answer, not a style preference.
- Membrane markers such as HER2 and many CD antigens should outline the cell. Nuclear color on a HER2-class assay is not a 3+ membrane score; it is the wrong picture.
- Nuclear markers such as ER, PR, Ki-67, and TTF-1 should sit in nuclei. Cytoplasmic-only ER is not a positive nucleus.
- Diffusion after poor fixation smears cytoplasmic or nuclear proteins into adjacent stroma, a classic problem with some immunoglobulin stains and underfixed cores.
Mechanisms that move signal into the wrong compartment include antibody too concentrated, retrieval too aggressive, drying, a wrong clone, biotin-rich nuclei on avidin-biotin methods, and antigen diffusion from delayed formalin. Compare with the control: if the control is membrane-crisp and the patient is nuclear-only, the patient specimen (fixation, necrosis, crush) is the lead. If both are nuclear when the marker is a membrane target, the assay (clone, titer, retrieval) is the lead.
Edge effect
Edge effect is strong chromogen at the section perimeter with a pale or blank center, or a dark rim around every fragment.
- Drying at the meniscus: reagents evaporate at the edge first.
- Incomplete reagent coverage: a hydrophobic slide or a short dispense leaves the center dry or the edge flooded.
- Fixation rim: an underfixed center versus a well-fixed edge, or the reverse on a large specimen, changes epitope availability.
- Section geometry: folded or thicker edges hold more antibody.
Do not score HER2 or other membrane markers on the dried rim. Read the well-fixed, fully covered center. If the entire fragment is an edge (a tiny biopsy), drying artifact can mimic diffuse positivity. Check the control and whether reagents actually covered the tissue.
Section wash-off
Wash-off is missing tissue, not pale tissue. Glass shows empty spaces without nuclei, fragments in the waste, or a control that vanished while the label is intact.
Causes overlap with microtomy and retrieval: uncharged slides, wet ribbons, aggressive high-pH HIER, fatty or decalcified tissue, thermal shock after retrieval. The troubleshooting move is adhesion and retrieval dose, not a more concentrated primary. A blank area with no nuclei is wash-off. A blank area with blue nuclei and no chromogen is no stain.
Floaters and carryover
A floater is tissue that does not belong to the section: water-bath debris, knife carryover, or a fragment from another case. On IHC it can look like a positive tumor island in negative stroma.
- Morphology will not match the patient's H&E (different epithelium; a shard of unrelated tissue on a skin case).
- The floater may stain perfectly because it is a different specimen that happens to carry the antigen.
- True positivity should fit the architecture of this biopsy.
Do not call a predictive marker on a fragment you cannot match to the H&E. Recut. Flag the water bath if multiple cases show the same extra tissue.
Pigment: melanin, hemosiderin, carbon, and DAB
Brown on an HRP-DAB slide is not automatically immunohistochemistry.
| Pigment | Color and site | How you separate it from DAB |
|---|---|---|
| Melanin | Fine to coarse brown-black in melanocytes, macrophages, melanoma | Present on H&E and on a negative reagent control; Fontana-Masson labels melanin; melanin bleach (permanganate/oxalate class) removes it |
| Hemosiderin | Golden-brown, chunky, in macrophages after hemorrhage | Prussian blue positive; still there without primary |
| Carbon | Jet-black, irregular, lung and nodes | Does not bleach like melanin; unchanged on reagent control |
| Formalin pigment (acid hematin) | Dark, often birefringent, in bloody tissue | Alcoholic picric acid can remove it; not an antibody product |
| DAB | Brown IHC product in the expected compartment | Absent when primary is omitted; not a Fontana synonym |
DAB versus Fontana is a wording trap: Fontana-Masson is a silver method for melanin, not another name for DAB. If the stem says the brown remains on the omit-primary slide, it is pigment, not missed antibody.
AEC red is easier to separate from melanin than DAB brown, which is one reason some laboratories use a red chromogen on pigmented skin or melanoma work-ups—not because red is more specific, but because it is a different color than native melanin.
Hook effect, conceptually
In solution immunoassays, a high-dose hook (prozone) is antigen excess that paradoxically lowers measured signal. Slide IHC is not a one-step sandwich ELISA, but the conceptual cousin shows up in two bench errors.
- Antibody excess: a primary titer far above the validated dilution often produces dirty background and less crisp specific localization (steric crowding, off-target binding). Rarely, specific signal looks paler than a more dilute titer because the picture is muddy rather than because antigen vanished.
- Very high antigen density: crowded epitopes can hinder polymer access so a strongly antigen-positive tissue looks weaker than a moderate control. Uncommon, but a reason not to call a faint dirty stain negative without a dilution series during optimization.
The practical exam move: if raising the primary concentration made the slide dirtier and less readable, you are in titer-too-concentrated / hook-like territory—dilute, do not double again. True absence of antigen with a working control is a clean negative, not a hook.
Worked scenario: dirty membrane marker
HER2 on a breast core shows thick brown in stroma and nuclei, and the control is also muddy. The omit-primary slide is cleaner. Retrieval was an extended high-pH program and the primary was used much more concentrated than the validated dilution. This is titer plus harsh retrieval, not a 3+ membrane result. Repeat at the locked dilution and retrieval; do not score the rim.
Worked scenario: brown melanoma
A Melan-A slide on pigmented melanoma is unreadable: every cell is brown. The negative reagent control is equally brown in the same cells. Fontana would label that pigment; DAB did not create it. Use a red chromogen, a validated bleach-then-IHC sequence, or a marker whose chromogen color is not melanin. Do not call the brown positive Melan-A until pigment is accounted for.
A pan-cytokeratin run shows no brown on the patient skin biopsy and no brown on the skin control. Hematoxylin nuclei are present on both slides, and both blocks contain epithelium on H&E. What is the best first interpretation?
Brown pigment remains in melanoma cells on an omit-primary reagent control that was processed with peroxidase-DAB detection. What is the best interpretation?