Formalin fixation

In brief

What formalin is

Formaldehyde is the smallest aldehyde: one carbon atom double-bonded to an oxygen atom and carrying two hydrogens, written H₂C=O, with a molecular weight of 30 ^1,2. At room temperature it is a gas, and what laboratories buy is that gas bubbled through water to saturation, about 37% formaldehyde ². "Formalin" is the name of the solution, not of the molecule; it began as a trade name: in the 1890s the industrial precursor of Schering AG patented the aqueous solution of formaldehyde gas under that name, as recounted in the introduction of a 2013 fixation study ⁵. When a pathologist says 10% formalin, she means that 37% stock diluted one part in ten, so the working fixative is about 3.7 to 4% formaldehyde, which is about 1.3 molar ^1,2.

In water, almost none of it stays as formaldehyde. Each molecule takes up a molecule of water and becomes methylene glycol, HO–CH₂–OH, a hydrate that does not cross-link on its own but turns back into formaldehyde continuously, and the balance lies overwhelmingly on the hydrate's side: direct nuclear magnetic resonance measurements find only traces of molecular formaldehyde in dilute solutions, with methylene glycol and a little of its dimer making up the rest ^3,4. Infrared and Raman spectra of commercial formaldehyde solutions show no detectable carbonyl formaldehyde, the reactive form with the carbon–oxygen double bond ¹. Only a tiny fraction of the molecules in a bottle of formalin are reactive at any moment, and more is released only as tissue uses the reactive fraction up ^1,5. That one fact explains most of this page: formalin enters tissue quickly, as methylene glycol, and fixes it slowly, as formaldehyde ^1,2. The same equilibrium underlies a regulatory quarrel over hair-smoothing products: the Cosmetic Ingredient Review panel concluded in 2013 that the two forms interconvert continuously and that heat shifts the balance toward formaldehyde ³¹, and a consultancy answered in 2014 that methylene glycol is a distinct chemical with no toxicity of its own beyond the formaldehyde in equilibrium with it ³².

Two other ingredients matter. Methanol, about 10% of commercial stock and about 1% of the diluted fixative, is there to stop methylene glycol molecules from linking into chains, and because it is an alcohol it gives commercial formalin an initial alcohol-fixation phase before the cross-linking begins ¹; another review puts the methanol in stock at 10 to 15% ². On standing, formaldehyde oxidizes to formic acid, which is why plain formalin turns acid, to about pH 4 at room temperature ^1,2. Paraformaldehyde is the solid polymer ^1,33. For critical work, laboratories make formaldehyde from it by heating it in water, or simply letting it stand, until it depolymerizes; the solution made this way is purer than commercial stock and gathered no significant acid over 8 days of storage ³³.

Ten percent neutral buffered formalin, abbreviated NBF, is the same 4% formaldehyde made up in phosphate buffer near pH 7; one carefully described protocol uses 100 mM phosphate adjusted to pH 6.8 to 7.2 ⁵. The buffer is not a refinement. At pH 4 many of the amino groups formaldehyde reacts with carry a positive charge and are unavailable, and the formaldehyde bound by collagen rises from 0.05 mmol per gram at pH 4 to 0.4 mmol per gram at pH 7 to 8 ¹. Acid formalin also reacts with the heme of hemoglobin to form acid formalin hematin, a dark brown crystalline pigment that lies beside red cells and can be mistaken for microorganisms; buffered formalin prevents it ¹¹.

The history, and what was reversed

Formaldehyde was first made in 1859 ², in a line of chemistry that histories of the reagent trace through Justus von Liebig, Alexander Butlerow and August Wilhelm Hofmann ⁴⁰. Its arrival in the microscope room was an accident: in 1893 the German Ferdinand Blum found that it fixed tissue, and formalin rapidly became popular for hardening and preserving whole human and animal specimens ^14,15,40. It displaced ethyl alcohol, which Dutch scientists had used since the eighteenth century, and the reason the record gives for the switch is that formalin was cheaper ¹⁵. Diagnostic microscopy then grew up around formalin fixation and paraffin embedding together ¹⁵.

Most of what followed was a slow discovery of the price. Formalin's damage to tissue structure was argued over for decades after Blum ¹⁴. Immunohistochemistry, the staining of one chosen molecule with an antibody, made the damage visible, because many antigens that stain in frozen tissue seemed concealed or destroyed after formalin ⁴. When DNA analysis reached archival blocks, formalin turned out to write false mutations into the record ¹³. Each time, the profession kept formalin and engineered around it, with antigen retrieval, enzymes that clean up damaged DNA, and fixation windows written into guidelines ^9,34,36. The timeline marks every belief that did not survive as a reversal.

How it fixes: the chemistry of a cross-link

Fixation means stopping the tissue's own destructive enzymes and hardening its structures so they survive dehydration in alcohol, clearing and wax ^1,5. Formaldehyde does it in steps. First it adds to a nitrogen atom, chiefly the free amino group at the end of the side chain of the amino acid lysine, forming a methylol, a –CH₂OH tag on the protein ^1,2,14. With a primary amine it can also form a Schiff base, a carbon double-bonded to a nitrogen ¹. A methylol can then react with a second group nearby to make a methylene bridge, a single –CH₂– carbon holding two parts of a protein, or two proteins, together ^1,6.

Mass spectrometry, which weighs molecular fragments precisely enough to tell what has been added to them, mapped the reactions in model peptides treated with excess formaldehyde ⁶. Formaldehyde reacted with the free amino group at the start of the chain and with the side chains of arginine, cysteine, histidine and lysine ⁶. When the formaldehyde was offered a second molecule to link, the amino acid glycine, the resulting adduct attached rapidly to the chain's free amino end, to arginine and to tyrosine, and less readily to asparagine, glutamine, histidine and tryptophan ⁶. In a whole small protein, insulin, arginine, tyrosine and lysine were the most reactive and seven bridges formed inside the molecule, while eight residues that could in theory have reacted did not, probably because they are intrinsically sluggish, buried in the folded protein, or both ⁷. One result corrects the textbook sketch: formaldehyde did not link two primary amino groups on separate molecules to each other ⁶.

The first reactions are quick and reversible; the stable network takes much longer ¹. A review of the chemistry places initial cross-linking at 24 to 48 hours after formaldehyde arrives and the formation of stable, essentially permanent bridges at about 30 days ¹. In rabbit liver, binding of radiolabeled formaldehyde was half complete at about 100 minutes and reached a plateau at about 24 hours, faster at 37 °C than 25 °C and faster at pH 7 than pH 4 ⁸. What this preserves is remarkably fine: in ribonuclease A, an enzyme used as a model, formaldehyde left the protein's folded shape intact, and heating later recovered its activity ².

How reversible is it, then? The evidence points both ways because it describes different bonds. Washing removes loosely bound formaldehyde within hours, but the rest resists weeks of washing and high concentrations of urea, and some leaves only after years in water ¹⁴; in the rabbit-liver experiment, 26 days of rinsing left 10 to 20% of the label in place ⁸. Antigen retrieval exploits the weak bonds. Heating a section in solution to about 100 °C hydrolyzes Schiff bases while the stable methylene bridges remain, and the protein unfolds enough to show an antibody its target ⁴. When the method was introduced in 1991, 39 of 52 antibodies stained better, 9 were unchanged and 4 were worse ³⁴; some epitopes still refuse to come back ².

Getting it in: penetration, time, heat and volume

Formalin has to reach a cell before it can fix it, and it travels by diffusion, the slow random spread of molecules from where they are concentrated to where they are not, which physicists describe with Fick's laws ¹. In 1941 Peter Medawar, the zoologist who later won a Nobel Prize for work on immune tolerance, summarized the result for fixatives as d = K√t: the depth reached, d, grows with the square root of the time, t, multiplied by a coefficient, K, that differs from one fixative to another (Medawar PB. The rate of penetration of fixatives. J R Microsc Soc 1941;61:46-57, as cited in ²). The review that quotes the formula gives no value of K for formaldehyde, only Dempster's rank order, acetic acid fastest and formaldehyde next ², so the formula and the clinical rule of thumb are different kinds of statement: one a law of diffusion, the other a crude average. The rule of thumb pathology teaches is that formalin penetrates about 1 mm an hour, and the fixation time usually recommended for neutral buffered formalin is 24 hours ².

The rule misleads in three ways. First, the square root: on Medawar's own relation, a front that has gone 1 mm in the first hour has gone about 2 mm at 4 hours and about 5 mm at 24 hours, not 24 mm; that is arithmetic on the formula, not a measurement ². Second, K is not one number; the literature reports many different values ¹, and when penetration into whole human spleens was measured over 1 to 25 days for three fixatives, formaldehyde among them, the fixatives moved more slowly than earlier experiments in gels and animal tissue had suggested ⁵¹. Third, and most important, penetration is not fixation: methylene glycol arrives well before formaldehyde has bound, which is why a large specimen often stains differently at its surface and in its middle ^1,2.

Heat speeds the chemistry and cold slows it ^5,8. Warming shifts the equilibrium toward reactive formaldehyde, about threefold between 20 °C and 45 °C, and speeds cross-linking itself ⁵; tissue sections reach binding equilibrium in about 24 hours at 25 °C and in under 18 hours at 37 °C ². But heating a specimen before the fixative has reached its center damages the center: 4 mm slices of tonsil put straight into 40 °C formalin were well fixed at the edge and poorly fixed inside ⁵. The answer one industry group devised was to soak tissue first in 4 °C formalin, where cross-linking almost stops while the fixative diffuses in, and then to warm it; 2 hours cold and 2 hours at 45 °C gave morphology indistinguishable from 24 hours at room temperature ⁵.

Volume and thickness are the other levers. A review of the chemistry advises 25 parts fixative to 1 part tissue and never fewer than 10 ¹; an artifact catalog gives 20 parts fixative and a slice no thicker than 6 mm ¹¹. Tissue stuck to the container wall, pieces stuck together, too little fixative and too thick a slice all leave regions where autolysis, the tissue digesting itself, continues before formalin arrives ¹¹. And the clock starts before the formalin does. Phosphorylated proteins are reported to begin degrading within 30 minutes of the tissue losing its blood supply, and in 93 breast cancers with documented time to fixation several lost their staining within 1 to 2 hours ⁵²; in colorectal cancer, the one hour of cold ischemia that guidelines permit was highly damaging to the phosphorylated form of EGFR, a growth-signal receptor; phosphorylation is the phosphate tag that switches such proteins on, and it is lost quickly once blood supply stops ¹⁰.

When it runs short, and when it runs away

Under-fixation is the failure the profession long underrated ²⁷. In 24 estrogen-receptor-positive breast cancers fixed for set times, the mean receptor score on a 0-to-7 scale was 2.46 after 3 hours, 5.75 after 6 hours and 6.70 after 8 hours, and needle biopsies whose results disagreed with the later excision had been fixed for a mean of 1.2 hours against 6.3 hours for those that agreed ²⁶. In prostate tissue, same-day processing produced the weakest staining for the protein p27, a brake on cell division whose loss is used as a marker of aggressive cancer, with a visible gradient from a well-stained rim to a weak center, and the authors concluded that for that marker under-fixation was much more of a problem than the widely feared over-fixation ²⁷. Newer studies outside the breast cut both ways. In 30 colorectal cancers stained 600 times for mismatch-repair proteins, fixation under 20 hours produced 47.8% of the failed stains and fixation over 90 hours 28.1%, with cold (4 °C) fixation failing least ⁵⁴; in a Japanese survey of gastric-cancer HER2 testing, longer fixation of surgical specimens went with more 0 and 1+ scores and fewer 3+ ⁵⁵; yet in 941 endometrial cancers, neither prolonged cold ischemia nor time in fixative outside 6 to 72 hours changed estrogen, progesterone, HER2 or mismatch-repair staining, and the authors asked whether breast rules should be applied to hysterectomies at all ⁵⁶.

Over-fixation is real but slower ^28,29. In a 2026 study of fifty cores from a single mastectomy, receptor and Ki67 staining (Ki67 is a protein present only in dividing cells, used to gauge how fast a tumor grows) was stable from 0.5 to 48 hours and declined significantly beyond 72 hours, while unbuffered formalin lowered the proportion of estrogen-receptor-positive cells only slightly, from 96.89% to 94.32%, without changing the clinical score, although in the same experiment every marker fell once cold ischemia passed 2 hours ²⁸; for DNA the buffer matters more than for stains, since in animal tissues unbuffered formalin fragmented DNA and lowered PCR efficiency compared with buffered formalin ⁵⁷. In animal organs fixed for three to five weeks, an extra day in alcohol, xylene, wax or heat did no harm, but organs fixed for a year lost or weakened some immunostains ²⁹. In human brains fixed for 1 to 20 years, staining for ferritin, vimentin and collagen IV and a silver stain faded with time, while other markers did not ⁵⁸.

These are not laboratory curiosities. The ASCO/CAP guidelines were written because about 20% of HER2 testing and up to 20% of estrogen- and progesterone-receptor testing worldwide were estimated to be inaccurate, which the ER/PgR guideline blamed on variation in handling before the test, in positivity thresholds and in interpretation criteria ^42,44. A team that studied the chemistry pointed out that even the 2007 window of 6 to 48 hours allowed an eightfold variation in fixation time and was never meant to be optimal for every assay ⁵. The stakes have risen. After a phase 3 trial, DESTINY-Breast04, showed that trastuzumab deruxtecan, an antibody carrying a chemotherapy payload, helps patients whose tumors score only 1+ or 2+ without gene amplification ⁵⁹, the 2023 ASCO/CAP update affirmed its 2018 rules, declined to create new 'HER2-low' or 'ultralow' result categories, and urged laboratories to separate a score of 0 from 1+ now that the line decides eligibility for the drug ⁴⁸; a second trial has since led expert groups to define an 'ultralow' category of faint staining within score 0 ⁵⁹. That line is exactly where fixation bites: when 1,107 paired breast-cancer samples were rescored under the three-tier scheme, agreement fell to 69.7%, driven by shifts between 0 and 1+, and specimens referred from outside laboratories agreed less often, which the authors ascribed to pre-analytical factors such as fixation ⁶⁰. In a 2026 series of 956 breast cancers, 1.1% were negative by immunohistochemistry but amplified by FISH (fluorescence in situ hybridization, which counts copies of the gene with glowing probes), more often in biopsies from metastatic sites where ischemia and fixation times had not been documented ⁶¹.

For nucleic acids the same two failures exist, measured more recently. In liver-cancer tissue fixed for 1 to 240 hours, total RNA and long RNA fragments fell as fixation lengthened and sequencing reads of control genes fell with them, so the authors advised keeping within the 6-to-72-hour window for RNA as well ⁶². In pig liver, DNA and RNA both deteriorated as fixation went from 1 to 3 to 7 days; 20% formalin damaged DNA more than 10% in small pieces; and blocks under a year old gave the best DNA ⁶³. In veterinary tissues, unbuffered formalin and prolonged fixation each advanced DNA fragmentation and lowered PCR efficiency, and storing fixed tissue in 70% ethanol prevented the loss of immunostaining that long fixation causes ⁵⁷. One multi-laboratory study of a cell line fixed for four different times found that the fixation-specific sequencing errors concentrated at the surface of the block, like an edge effect, while the inner tissue showed no quality loss related to fixation time ⁶⁴.

For DNA the trade-off runs differently. A systematic review of the forensic literature lists the fixation process among the factors that cannot be controlled once a block exists, which is why its advice concentrates on the later steps, extraction and amplification ²². A 2011 review found no reported influence of paraffin-block storage time on immunostaining ⁵³, but a 2026 study that measured it found PD-L1 staining nearly halved after five years in the block and reduced by about 60% within two months on cut slides, and HER2 antigen on slides decayed over years at room temperature ^65,66; DNA in stored blocks keeps accumulating damage too: sequencing quality fell with block age, most of all after 5 years, and even 2 to 4 years of storage added false variants ³⁰.

What it does to DNA and RNA

The forensic review puts it without euphemism: preserving tissue in formalin is known to damage DNA through cross-linking, and the result is severely fragmented DNA of variable yield, which limits what can be done with it ²². Formaldehyde reacts directly with the bases and ties DNA to the histone proteins wrapped around it ². The fragments are short: PCR from formalin-fixed tissue is usually assumed to give products of about 200 base pairs on average ², a mastectomy series obtained 300 to 400 base-pair products ¹², and in whole-exome sequencing, which reads all the protein-coding parts of the genome, more than 20% of the fragments from fixed tissue were so short that the two reads taken from either end overlapped, sequencing the same bases twice ³⁵.

The fragments also carry false letters. Cytosine, one of DNA's four bases, can lose an amino group, a reaction called deamination, and become uracil, which copying enzymes read as thymine; so a C becomes a T, and on the opposite strand a G becomes an A ^21,36. In 1999 a controlled comparison found up to one such false mutation per 500 bases in fixed material, 27 of 28 of them C to T or G to A, and more artifacts the fewer cells were used ¹³. An enzyme, uracil-DNA glycosylase, removes the uracil before copying and sharply reduces these artifacts without erasing true mutations in the cancer genes KRAS or EGFR ³⁶; but at CpG sites, where cytosine carries a methyl group, deamination makes thymine directly, and the enzyme cannot remove it ⁶⁷. In four patients' matched exomes, artifactual C>T and G>A changes ran at 200 to 1,200 per million bases in fixed tissue and were absent from frozen tissue ³⁵.

At the scale of whole genomes, formalin leaves a recognizable fingerprint, a mutational signature closely resembling one seen in a DNA-repair deficiency, and an algorithm called FFPEsig can correct a tumor's mutation catalog for it ⁵⁰. Larger comparisons followed. In England's 100,000 Genomes Project, 578 formalin-fixed cancer genomes set against 11,014 fresh-frozen ones showed three artifact signatures, one known and two new, yet still yielded the clinically actionable variants, and the authors built a score that grades how damaged each sample is ⁶⁸. In 56 matched pairs, fixation enriched false calls about twentyfold across every class of mutation and blurred a marker of DNA-repair deficiency, which a machine-learning filter trained to recognize formalin's errors largely restored ⁶⁹. In 50 pairs of cryopreserved and fixed tumors, fixed tissue reported a tumor mutation burden, the count of mutations per million bases that is used to choose patients for immunotherapy, of 13.7 against 6.4, and only 43.5% of the variants called in one matched the other ⁷⁰. The damage also continues after fixation ³⁰. In 281 FFPE tissues aged up to 17 years, sequencing quality declined with storage, and paired samples stored 2 to 4 years gained 191 false-positive variants against 154 true ones, all above 5% of reads; in four samples that were low at baseline, storage raised the measured tumor mutation burden artificially, and pretreatment with a DNA repair enzyme largely restored all of this ³⁰.

The counterweight deserves equal weight. Archival blocks are now regarded as a treasure for research rather than a dead end ²⁰, and more than a hundred published studies describe ways to get usable DNA out of them ²². About 7,000 genes were read from punches 100 micrometers across and 10 micrometers thick, a few dozen cells, from mouse liver blocks two years old, and tumor microenvironments were characterized in human lung cancer blocks six years old ⁷¹. A lung block from a child who died in 1912 yielded metagenomic reads suggesting Rickettsia felis and a Mucor fungus, although the authors could not exclude contamination ⁷². Tissue from different sources can also differ in molecular quality simply because it was fixed and processed differently ⁷³. The archive is damaged; it is not unreadable ^22,71.

RNA fares as badly and is measured less often. Longer fixation lowered total RNA and long fragments and cut sequencing reads of control genes in liver cancer fixed for 1 to 240 hours ⁶²; in post-mortem brain, messenger RNA detectability changed with fixation time while background rose ⁷⁴; and in spinal cord and hypothalamus fixed for weeks to years, the usual RNA quality metrics did not change, yet the probes used for transcriptome sequencing ligated poorly and expression of nervous-system genes read artificially low, so a block can pass its quality checks and still mislead ⁷⁵.

For anyone hoping to sequence a parasite out of a block, the consequences are practical. Assays must target short stretches of DNA, because long targets fail on fragmented templates ^2,12. Low-level findings must be confirmed by an independent method or a second amplification, because formalin artifacts look like real variants ^13,21. And older blocks give worse answers, so a negative result from an old archive means less than a negative from a new one ^24,30.

How fixation is checked, and what is done when it fails

Fixation itself is almost never measured ⁷⁶. A 2021 paper begins with the plain statement that there are currently no analytical methods of detecting fixation, and that as a result fixation in clinical practice is highly variable and a persistent source of error ⁷⁶. What is measured is time ⁹. For breast cancer, the guidelines require that the time to fixation and the time in fixative be recorded and taken into account, with a disclaimer on the report when handling requirements were not met ⁹. The research alternative is acoustic: the same industry group built an instrument that listens to formalin diffusing into a core, and in 87 tonsil samples fixation timed by it, averaging 4.2 hours in cold formalin, matched 24-hour fixation for FOXP3, a protein marking regulatory immune cells that is sensitive to under-fixation, while deliberately short fixation suppressed it ⁷⁶. The same group later showed that formaldehyde cross-linking changes the infrared spectrum of a section, most of all in the band that reports protein folding, and trained a model that predicted the fixation time of unknown blocks to within 1.4 hours on average, validated against bcl-2, FOXP3 and Ki-67 staining ⁷⁷; neither method is in clinical use.

Everything else is indirect. Immunostains carry internal controls, normal cells on the same slide that should stain, and a weak center beside a strong rim betrays under-fixation ^27,44. A 2011 review identified 62 variables between surgery and staining, of which only 27 had been studied and 15 shown to change results ⁵³. For DNA, laboratories check how long a fragment can be amplified and watch sequencing quality metrics, which decline with block age ^12,30.

Several repairs exist, some routine and some still new. Heat-induced antigen retrieval recovers many masked epitopes ^20,34. It is not free: heating fragments the DNA in the section and lowers its yield, although short targets still amplify from immunostained slides ⁷⁸, and the high-pH buffer most laboratories use breaks open nuclei in thyroid, lung and kidney so that nuclear antigens stain weakly or not at all unless a citrate buffer is used instead ⁷⁹. Uracil-DNA glycosylase removes most deamination artifacts before sequencing ^36,67, a DNA repair enzyme restores much of the damage of long storage ³⁰, and FFPEsig corrects mutation catalogs computationally ⁵⁰. At the bench, injecting formalin into large specimens improves staining ²⁷, and chilling a mastectomy for 20 minutes so that it can be sliced at 4 mm intervals, to help fixation, did not harm receptor or DNA testing ¹².

What formalin does to parasites

In the stool laboratory, formalin is the parasite hunter's ally ^37,80. Stool for microscopy is preserved in formalin and then concentrated by sedimentation with a solvent, so that eggs, cysts and larvae collect in a sediment that can be examined ^37,41. The solvent was originally diethyl ether; in 1979 ethyl acetate was shown to concentrate organisms as well or better, with no distortion of their shape, while being less flammable and less hazardous ³⁷, and a 1981 comparison in 62 stools found the same ⁴¹. A 2026 meta-analysis of 40 studies comparing this formalin-ether or formalin-ethyl acetate method, known after Ritchie, with the Kato-Katz smear found markedly higher odds of detecting Strongyloides stercoralis by concentration (odds ratio 5.02, 95% confidence interval 1.75 to 14.41), while the smear did better for Schistosoma mansoni (odds ratio 0.59, 0.39 to 0.90) ⁸¹. Formalin preservation itself costs little sensitivity: in 574 formalin-fixed stools from Laos tested four ways and analyzed with a model that needs no gold standard, the concentration method found an estimated 72.7% of Ascaris lumbricoides infections (credible interval 68.9 to 76.6%), figures similar to those reported for fresh stool ⁸².

For the parasite most firmly tied to cancer the method is central. In 455 stools from northeastern Thailand, where infection with the liver fluke Opisthorchis viverrini is closely associated with cholangiocarcinoma, a bile-duct cancer, formalin-ethyl acetate concentration found fluke eggs in 16.48% of specimens against 10.55% for an automated analyzer and 8.57% for the Kato-Katz smear ⁸⁰. The study reported a sensitivity of 98.67% and a specificity of 97.63% for concentration, but the abstract does not state its reference standard, and its figure of 100% sensitivity for the smear, which found half as many positives, is hard to reconcile ⁸⁰. Across 40 studies, however, the concentration method found Opisthorchis viverrini no more often than the Kato-Katz smear (odds ratio 1.09, 95% confidence interval 0.61 to 1.92), so the Thai advantage is one study's result, not a rule ⁸¹. What is clear is that formalin does not distort eggs; it keeps them readable ^37,41.

It is often said that formalin kills parasites; for thick-shelled worm eggs that is a half-truth, and the evidence behind it is narrower than the saying ⁸³. Eggs of the pig roundworm Ascaris suum are resistant enough that dilute formalin, 0.5 to 5%, is among the solutions existing test methods use to incubate them so that they develop, and in a 2024 test of the reagents used in sanitation assays, 10% formalin, like every other extraction solution, had minimal effect on their viability over the brief, minutes-long exposures of a laboratory recovery protocol ⁸³. Whether eggs survive the days spent in a fixative or the weeks in a preserved stool specimen was not tested, so the right conclusion is narrow: a brief wash in formalin does not make such eggs harmless, and a formalin-preserved egg is not necessarily a dead one ⁸³.

In tissue sections, eggs and worms generally remain recognizable after formalin; what loses them is mostly sampling and unfamiliarity ^38,84. In 228 Tanzanian women, Schistosoma haematobium eggs were found in the cervix in 49% by examining a squashed whole biopsy, in 18% by routine histologic sections and in 3% by cytology, the examination of loose cells scraped from the surface, and histology succeeded mainly where eggs were dense ³⁸. Two lung nodules that looked like cancer were pulmonary dirofilariasis, in which dead and dying Dirofilaria immitis worms block branches of the lung arteries and provoke a granulomatous reaction, a walled-off knot of immune cells ⁸⁵. Pinworm remnants and the tissue changes around them were still describable in appendectomy blocks stored for 12 to 22 years ²³. The recognition problem is real: parasitic cysts are rarely seen by pathologists and carry many morphologic pitfalls ⁸⁴, and formalin's own pigment can imitate microorganisms ¹¹. Sizes are also slightly off, since fixation and processing shrank kidney tumors by a mean of 11.4%; whether eggs shrink by the same amount has not been measured ⁸⁶.

Parasite DNA fares worse than parasite shape, but not hopelessly ^23,39. Of five ways of keeping Cryptosporidium-positive stool for a month, freezing preserved PCR detection best (80%) and 10% formalin worst (0%) ³⁹. In tissue blocks, pinworm DNA was amplified, using the mitochondrial gene cox1, from 12 of 17 appendectomy blocks stored 12 to 22 years ²³, a tapeworm larva in a lung was identified to species from an FFPE biopsy ⁸⁷, and a validated Echinococcus PCR aimed at 12S rDNA was positive in 9 of 15 viable lesions, all 6 from blocks under 6 years old and, by arithmetic from the published counts, only 3 of the 9 older ones, leading its authors to recommend blocks under 5 years ²⁴. Yields elsewhere are of the same order: in 35 hydatid-cyst blocks from northern Iran, two mitochondrial genes amplified from 16 (45.7%) and 12 (34.3%) ⁸⁸, and a multiplex real-time PCR designed around short mitochondrial targets, validated on 81 fresh and fixed echinococcosis tissues with 100% specificity, is the kind of assay the fragments allow ⁸⁹. Metagenomic sequencing of a fixed hip-cyst biopsy named Spirometra mansoni when treatment for the suspected organism had failed ⁹⁰, and infrared microscopy, which reads chemistry rather than DNA, told Echinococcus elements from human tissue in sections from 11 patients ⁹¹. For abdominal angiostrongyliasis, PCR on FFPE bowel had 55% sensitivity and 100% specificity and was positive in 4 of 20 cases where no worm had been seen ⁹². In routine metagenomic sequencing of 623 FFPE samples, parasites accounted for 9 of 229 positive results, and 9.6% of samples were uninterpretable ²⁵.

Alternatives, and the case against formaldehyde

Alcohol-based fixatives work by coagulation, precipitating proteins rather than cross-linking them, so they leave immunostaining uncompromised and DNA and RNA undamaged ¹⁵. A Leiden laboratory used one such fixative for skin biopsies from 1987 to 2001 and another since 2004 for more than 40,000 skin biopsies and more than 100,000 cervical samples, and predicted in 2008 that formalin might eventually be legislated out of diagnostic pathology once health authorities realized that it invalidates expensive tests ¹⁵. The prediction came from the laboratory that uses those fixatives, and it has half come true: in 2016 the European Union proposed a ban on formalin and granted diagnostic pathology only a temporary exemption, which has driven the search for substitutes in Europe ⁴⁶; neutral buffered formalin nonetheless remains the fixative of choice in most laboratories ². Commercial "molecular fixatives" built on alcohol or acetone preserve nucleic acids better than formalin, but they shrink tissue and alter red cells, and across published comparisons few outperform formalin for immunostaining: in one study, formalin gave good staining with 84% of 85 diagnostic antibodies against 66% for one alternative ². Their cost and the need to revalidate every antibody have kept them out of most laboratories ².

PAXgene Tissue, developed within SPIDIA, a European Union project on standardizing the handling of specimens before testing, is described by its maker as a non-cross-linking mixture of alcohols, an acid and a soluble organic compound ⁹³. In 70 normal tissues split between it and formalin, morphology was adequate for diagnosis in most cases, some antigens stained without retrieval and others with standard formalin protocols, and in situ hybridization, which locates specific DNA or RNA in the section with labeled probes, and special stains performed comparably ⁹³. The same paper names the obstacles frankly: pathologists are trained to read formalin's artifacts, formalin is cheap, and tissue fixed otherwise does not look quite the same ⁹³.

Glyoxal, a two-carbon dialdehyde, was called the leading alternative to formaldehyde as a histological fixative in 2007 by a reviewer writing from a private consultancy ⁹⁴, and the claim has since been tested in clinics. It reacts faster than formaldehyde and poses a lower inhalation risk, forms small adducts that leave most protein groups available to antibodies, impairs mainly arginine, and gives hematoxylin and eosin slides, the routine purple-and-pink stain, of unusual clarity; silver stains for Helicobacter pylori, the stomach ulcer bacterium, however, work poorly after it ⁹⁴. In cells imaged by fluorescence microscopy, glyoxal acted faster than paraformaldehyde, cross-linked proteins more effectively and gave better staining for most targets ⁹⁵. The commercial glyoxal fixatives of the 2000s were acidic, which a Turin group blamed for their poor record; it removed the acid with an ion-exchange resin and reported, in 30 colon and stomach cancers, morphology and immunostaining comparable to formalin and longer DNA fragments ⁹⁶. An open-label multicenter non-inferiority trial of this glyoxal acid-free fixative, published online in 2023 with co-authors employed by its maker, ADDAX Biosciences, found diagnostic quality non-inferior to formalin in every organ tested ⁴⁶. Independent results are less kind: in Tennessee, formalin was slightly superior to glyoxal across 34 antibodies, though all were judged usable ⁹⁷, and 2% acid-free glyoxal produced deleterious changes in morphology and staining in a second study from the same laboratory ⁹⁸; a multi-institutional veterinary study found better DNA yields but lost mast-cell granules and concluded that the product is not yet economically competitive ⁹⁹. A comparison of more than 200 specimens in home-made and commercial substitutes concluded that a formaldehyde-free laboratory is possible, noting that a medium-to-large laboratory may use more than 3,500 liters of formalin a year ¹⁰⁰; its first author is affiliated with a company, EuroClone ¹⁰⁰.

The other case against formaldehyde, besides what it does to molecular tests, is the safety of the people who use it ^93,100. In June 2004 an IARC working group concluded that formaldehyde is carcinogenic to humans, the category IARC calls Group 1, with sufficient evidence that it causes nasopharyngeal cancer, evidence for leukemia it called "strong but not sufficient", and limited evidence for cancer of the nose and sinuses ¹⁶. Five years later a second working group upgraded the leukemia evidence to sufficient ^17,18, a conclusion the European committees that set workplace limits have not accepted ¹⁹. The body makes formaldehyde itself, as a by-product of one-carbon metabolism, and relies on an enzyme and on DNA-repair pathways to detoxify it; cells that cannot detoxify it, or cannot repair DNA cross-links, die when given folate derivatives, the vitamin whose breakdown releases it, which is one reason the leukemia question is argued about at all ¹⁰¹. Among funeral-industry workers who died between 1960 and 1986, deaths from myeloid leukemia rose with years of embalming and with peak exposure; compared with fewer than 500 lifetime embalmings, the odds were 3.9 times higher after more than 34 years of embalming (95% confidence interval 1.2 to 12.5) and 3.1 times higher above a cumulative exposure of 9,253 parts per million-hours (1.0 to 9.6) ¹⁰². That is an association built from exposures reconstructed by interviewing relatives and coworkers, not proof of cause ¹⁰². A 2020 meta-analysis of 31 studies found no overall increase in lung cancer (pooled risk 1.04, 0.97 to 1.12), with a small rise in higher-quality studies (1.13, 1.08 to 1.19) ¹⁰³. A 2025 Italian case-control study of 867 lymphomas found no link with lymphoma overall but a doubled risk of multiple myeloma with occupational exposure (odds ratio 2.0, 95% confidence interval 1.19 to 3.31) ¹⁰⁴; a 2026 meta-analysis of 22 studies of exposed workers found changes in white-cell counts but no dose-response, and argued that inhaled formaldehyde does not travel beyond the tissues it first touches ¹⁰⁵. The Global Burden of Disease study attributed 592 nasopharyngeal-cancer deaths worldwide in 2021 to occupational formaldehyde ¹⁰⁶. In pathology laboratories the exposures are not trivial: in four Tehran hospitals 91% of measured exposures exceeded the US NIOSH recommended limit of 0.016 parts per million, highest among technicians ¹⁰⁷; personal samples in one Iranian university's hospital laboratories averaged 0.195 ppm ¹⁰⁸; and anatomy technicians in an Australian university averaged 1.1 ppm over a shift as an 8-hour time-weighted average, with their upper exposures above that country's limit ¹⁰⁹.

What was believed and is not

Formalin has been in use since 1893, long enough for its lore to outlive the evidence ¹⁴. Several things still taught about it are rules of thumb that were never measured, or findings that later work overturned or narrowed: how fast it penetrates, whether too long is worse than too short, whether it destroys antigens, how faithfully a fixed block can be sequenced, how stable that block is in storage, whether a stored block or a cut slide keeps its stains, and how firmly formaldehyde is tied to leukemia ^{2,13,17,27,30,34,65}.

One belief on the list is this season's own: that formalin is part of why parasites in tumors might go unseen. The evidence supports a narrower and more useful version of it, set out in the last card, and it is offered in the same spirit as the others, as a correction that sharpens the question instead of closing it ^24,38.

What is still unknown

The searches behind this page found no study measuring how formalin fixation affects the recovery of parasite DNA from human tumor tissue specifically; the nearest evidence comes from non-tumor lesions in which the parasite was already known to be present, in liver, appendix and bowel, and it shows feasibility and, for one tapeworm, steep losses with block age ^23,24,92. Nor did they find a study that fixed known helminth material, processed it to paraffin and scored how often blinded pathologists recognized it, so the claim that formalin preserves helminth morphology rests on convergent but indirect evidence ^38,85.

Fixation itself still cannot be measured in routine clinical practice, although two research methods, one acoustic and one infrared, now exist ^76,77, and the time windows in use were validated for a few breast-cancer receptors ^9,26 and have since been found not to matter for four markers in endometrial cancer ⁵⁶. Whether those windows suit antibodies against parasite antigens, or parasite DNA, has not been studied; of 62 recognized pre-analytical variables, only 27 had been examined for their effect even on ordinary immunostaining by 2011 ⁵³. How much eggs and larvae shrink in processing, and whether the 11.4% shrinkage measured in kidney tumors applies to them, is unknown ⁸⁶.

Two questions sit at the edge of current methods. Metagenomic sequencing of FFPE tissue works in routine practice, but about one sample in ten fails quality control or raises suspicion of contamination ²⁵, and even a striking result from an old block, such as reads of Rickettsia and Mucor in a lung from 1912, cannot always be told apart from contamination ⁷². And the alternatives to formalin have been compared for human antigens, special stains and nucleic acids, but not for how well they preserve worms and eggs, although at least one, glyoxal, is known to spoil a silver stain used for bacteria ^93,94.

How it connects to parasites and cancer

Formalin is the gate through which most tissue diagnoses of cancer pass, and FFPE blocks are usually the only material left for later molecular work ^21,22. It is a gate the profession knows to be leaky even for the molecules it is trying to find: up to 20% of receptor results and about 20% of HER2 results were judged potentially inaccurate when the guidelines were written, from pre-analytic variation, positivity thresholds and interpretation criteria ^42,44. For an organism nobody is looking for, no comparable error rate has been published; the most recent review of parasitic lesions in pathology offers expert warnings about pitfalls, not numbers ⁸⁴.

Each piece of the connection has its own strength. That formalin fragments DNA and writes false mutations into it is established beyond reasonable doubt, by controlled human-tissue studies over more than twenty-five years ^13,30,35,50. That parasite DNA can nonetheless be recovered from blocks with targeted assays is well supported in specific diseases, with sensitivity from 55% to 100% depending on the organism, the assay and the age of the block ^23,24,92. That formalin preserves the shape of eggs and worms in tissue is supported by convergent evidence but has never been measured directly ^38,85. That parasites have been systematically missed inside human tumors because of formalin is not supported by any study; the strongest proof that parasite-derived cells can hide in a tumor-like lesion is a single case, recognized because someone ran a eukaryote-targeted PCR ⁴⁵.

So the honest form of the season's question about formalin is narrower than it first sounds. On the evidence above, formalin is unlikely to be why a parasite in a tumor would be missed under the microscope; sampling and recognition are the larger filters ^38,84. Formalin is the reason that proving, after the fact, that a parasite was there is difficult, and the reason that a negative molecular result from an old block means little ^24,30. The tools to ask properly exist: metagenomic sequencing of FFPE tissue in routine practice, short-target assays, repair enzymes, blocks less than five years old and confirmation by an independent method ^21,24,25,30. And the same chemical is, in the stool laboratory, how the parasite most firmly linked to cancer is counted: in one large Thai study, formalin-ethyl acetate concentration found more Opisthorchis viverrini infections than either method it was compared with ⁸⁰, although across 40 studies it and the Kato-Katz smear found that fluke equally often ⁸¹.

The tests, one by one

• Recorded cold-ischemia time and time in fixativeThe only routine 'measurement' of fixation: the clock times written on the requisition, which the ASCO/CAP guidelines (American Society of Clinical Oncology and College of American Pathologists) require for breast cancer ⁹.Required for breast-cancer biomarker testing ⁹As good as the record. A 2021 study states that no analytical method of detecting fixation is in clinical use ⁷⁶.Thickness, volume, temperature and how far formaldehyde actually got; when the times are missing, discordant results become more common ⁶¹. When a manual immunostaining service opened in southern Ethiopia, 68% of 125 specimens arrived without a cold-ischemia time and 60% without a fixation time; where recorded, fixation ran a median of 72 hours and up to 192 ¹¹⁰.
• Internal controls and staining gradients on immunohistochemistryNormal cells on the same slide that should stain, and a rim-to-center gradient that betrays under-fixation ^27,44.Expected on every clinical receptor stain ⁴⁴Good at flagging gross failure; the ER/PgR (estrogen- and progesterone-receptor) guideline still estimated that up to 20% of receptor results worldwide may be inaccurate ⁴⁴.Subtle loss near a cutoff, such as HER2 0 versus 1+ (HER2 is a growth-factor receptor scored 0 to 3+ on the slide), which now decides eligibility for trastuzumab deruxtecan, an antibody that carries a chemotherapy drug to HER2-bearing cells ⁴⁸.
• DNA quality checks before sequencingAmplifiable fragment length and sequencing quality metrics measured on DNA extracted from the block ^12,30.Routine before clinical next-generation sequencing ³⁰Reliable for predicting failure: quality metrics fall with block age, especially beyond 5 years ³⁰.Artifacts that pass quality control; 2 to 4 years of storage added 191 false variants against 154 true ones above 5% allele frequency, that is, present in more than 1 in 20 of the DNA reads at that position ³⁰.
• Acoustic time-of-flight monitoringResearch instrument that listens to formalin diffusing into a tissue core in real time ⁷⁶.Research only; developed by Roche Tissue Diagnostics scientists ⁷⁶In 87 tonsil samples, dynamically timed fixation matched 24-hour fixation for staining of FOXP3, a protein that marks regulatory immune cells ⁷⁶.Tested on tonsil cores with two markers; not validated for other tissues or for DNA ⁷⁶; an infrared method from the same company predicts fixation time to within 1.4 hours but is likewise research-only ⁷⁷.
• Formalin-ethyl acetate stool concentrationStool is preserved in formalin and then concentrated by sedimentation with ethyl acetate, so that eggs, cysts and larvae collect in a sediment for the microscope ^37,41.Routine concentration method in stool parasitology, compared in the literature with the Kato-Katz smear ^80,81For the carcinogenic liver fluke in 455 Thai stools it found the most egg-positive specimens of three methods, with a reported sensitivity of 98.67% and specificity of 97.63%; the abstract does not state the reference standard ⁸⁰.Formalin-preserved stool is a poor source of DNA, and the Kato-Katz smear finds Schistosoma mansoni more often ^39,81.

Where it connects

Sources

110 sources, numbered as they are cited. Every one was checked against PubMed or its publisher before it was cited here; the note under each says what it shows and what it does not.

1. 1
   Thavarajah R, Mudimbaimannar VK, Elizabeth J, et al. Chemical and physical basics of routine formaldehyde fixation.J Oral Maxillofac Pathol · 2012 · 16(3):400-5doi:10.4103/0973-029X.102496 · PMID 23248474

   A narrative review of formalin chemistry and physics (methylene glycol, pH, osmolarity, volume, timing); its numbers are compiled from earlier work and appear only in the full text, not the abstract.

2. 2
   Howat WJ, Wilson BA. Tissue fixation and the effect of molecular fixatives on downstream staining procedures.Methods · 2014 · 70(1):12-9doi:10.1016/j.ymeth.2014.01.022 · PMID 24561827

   A review giving formalin's composition, Medawar's formula, the 1 mm per hour rule of thumb and a literature comparison of alternative fixatives; it reports others' measurements rather than new ones.

3. 3
   Rivlin M, Eliav U, Navon G. NMR studies of the equilibria and reaction rates in aqueous solutions of formaldehyde.J Phys Chem B · 2015 · 119(12):4479-87doi:10.1021/jp513020y · PMID 25742498

   Direct 1H and 13C NMR detection of molecular formaldehyde in dilute solution across pH 1.8 to 7.8 and 5 to 60 °C; physical chemistry, not tissue.

4. 4
   Shiurba RA, Spooner ET, Ishiguro K, et al. Immunocytochemistry of formalin-fixed human brain tissues: microwave irradiation of free-floating sections.Brain Res Brain Res Protoc · 1998 · 2(2):109-19doi:10.1016/s1385-299x(97)00029-9 · PMID 9473616

   Explains the methylene glycol equilibrium and why heating restores antigens (Schiff bases hydrolyze, methylene bridges survive); a protocol paper on brain tissue, not a measured comparison.

5. 5
   Chafin D, Theiss A, Roberts E, et al. Rapid two-temperature formalin fixation.PLoS One · 2013 · 8(1):e54138doi:10.1371/journal.pone.0054138 · PMID 23349806

   Shows that cross-linking nearly stops at 4 °C and that 2 hours cold then 2 hours at 45 °C matched 24-hour fixation in 38 tissue samples; the first author was at Ventana Medical Systems, which sells staining instruments and reagents, and the work ran on that company's platform. Its introduction states, as read in the full text, that the industrial precursor to Schering AG patented the aqueous solution of formaldehyde gas under the name 'formalin' in the 1890s.

6. 6
   Metz B, Kersten GFA, Hoogerhout P, et al. Identification of formaldehyde-induced modifications in proteins: reactions with model peptides.J Biol Chem · 2004 · 279(8):6235-43doi:10.1074/jbc.M310752200 · PMID 14638685

   Mass-spectrometry map of which amino acids formaldehyde modifies and how (methylols, Schiff bases, methylene bridges); done on short model peptides in solution, not on fixed tissue.

7. 7
   Metz B, Kersten GFA, Baart GJE, et al. Identification of formaldehyde-induced modifications in proteins: reactions with insulin.Bioconjug Chem · 2006 · 17(3):815-22doi:10.1021/bc050340f · PMID 16704222

   Extends the peptide map to a whole small protein and shows that folding hides some reactive residues; a vaccine-chemistry study of one protein, not a tissue study.

8. 8
   Helander KG. Kinetic studies of formaldehyde binding in tissue.Biotech Histochem · 1994 · 69(3):177-9doi:10.3109/10520299409106282 · PMID 8068812

   Radiolabeled formaldehyde in rabbit liver: half-maximal binding at about 100 minutes, plateau at about 24 hours, and slow loss on rinsing; one animal tissue under laboratory conditions.

9. 9
   Wolff AC, Hammond MEH, Hicks DG, et al. Recommendations for human epidermal growth factor receptor 2 testing in breast cancer: American Society of Clinical Oncology/College of American Pathologists clinical practice guideline update.Arch Pathol Lab Med · 2014 · 138(2):241-56doi:10.5858/arpa.2013-0953-SA · PMID 24099077

   Read in full text: time in fixative 'increased from 6 to 48 hours to 6 to 72 hours' to match the ER/PgR guideline, and both cold ischemia and fixation time must be recorded; most committee members disclosed no relevant commercial ties. Published online in October 2013 and co-published in J Clin Oncol 2013;31(31):3997-4013.

10. 10
    Theiss AP, Chafin D, Bauer DR, et al. Immunohistochemistry of colorectal cancer biomarker phosphorylation requires controlled tissue fixation.PLoS One · 2014 · 9(11):e113608doi:10.1371/journal.pone.0113608 · PMID 25409462

    In 23 colorectal cancers, a one-hour cold ischemia interval damaged phosphorylated EGFR and a 4-hour two-temperature protocol stained some phospho-markers better; four of five authors were Ventana employees.

11. 11
    Bindhu P, Krishnapillai R, Thomas P, et al. Facts in artifacts.J Oral Maxillofac Pathol · 2013 · 17(3):397-401doi:10.4103/0973-029X.125206 · PMID 24574659

    A catalog of processing artifacts, including the 20-to-1 volume rule and acid formalin hematin pigment that can mimic microorganisms; a descriptive review without measurements.

12. 12
    Kumarapeli AR, Bellamy W, Olgaard E, et al. Short-Duration Rapid Chilling of Mastectomy Specimens Does Not Interfere With Breast Cancer Biomarker and Molecular Testing.Arch Pathol Lab Med · 2019 · 143(1):92-98doi:10.5858/arpa.2017-0377-OA · PMID 29932859

    Chilling mastectomies to slice them at 4 mm did not harm receptor testing, and FFPE DNA gave 300 to 400 base-pair PCR products; one institution's retrospective review.

13. 13
    Williams C, Pontén F, Moberg C, et al. A high frequency of sequence alterations is due to formalin fixation of archival specimens.Am J Pathol · 1999 · 155(5):1467-71doi:10.1016/S0002-9440(10)65461-2 · PMID 10550302

    Controlled comparison of frozen and formalin-fixed material: up to one false mutation per 500 bases in fixed tissue, almost all C to T or G to A; it showed the problem, not its chemical cause.

14. 14
    Puchtler H, Meloan SN. On the chemistry of formaldehyde fixation and its effects on immunohistochemical reactions.Histochemistry · 1985 · 82(3):201-4doi:10.1007/BF00501395 · PMID 3997553

    States that Blum discovered formalin fixation by accident in 1893 and argues from chemical data that firmly bound formaldehyde is not removed by washing; a chemical argument, not a new experiment.

15. 15
    Boon ME, Kok LP. Theory and practice of combining coagulant fixation and microwave histoprocessing.Biotech Histochem · 2008 · 83(6):261-77doi:10.1080/10520290802553476 · PMID 19031284

    History of formalin replacing alcohol and the Leiden experience with alcohol-based fixatives; the authors evaluate fixatives developed and used in their own laboratory, so it is an advocate's account.

16. 16
    Cogliano VJ, Grosse Y, Baan RA, et al. Meeting report: summary of IARC monographs on formaldehyde, 2-butoxyethanol, and 1-tert-butoxy-2-propanol.Environ Health Perspect · 2005 · 113(9):1205-8doi:10.1289/ehp.7542 · PMID 16140628

    The June 2004 IARC working group's conclusion that formaldehyde is carcinogenic to humans (sufficient for nasopharyngeal cancer, 'strong but not sufficient' for leukemia); a summary of Volume 88, not the volume itself.

17. 17
    Baan R, Grosse Y, Straif K, et al. A review of human carcinogens--Part F: chemical agents and related occupations.Lancet Oncol · 2009 · 10(12):1143-4doi:10.1016/s1470-2045(09)70358-4 · PMID 19998521

    Meeting report of the October 2009 IARC working group that produced Monograph Volume 100F; it has no abstract, so the leukemia conclusion is quoted here through [66] and [68].

18. 18
    Lan Q, Smith MT, Tang X, et al. Chromosome-wide aneuploidy study of cultured circulating myeloid progenitor cells from workers occupationally exposed to formaldehyde.Carcinogenesis · 2015 · 36(1):160-7doi:10.1093/carcin/bgu229 · PMID 25391402

    States that IARC concluded there is sufficient evidence that formaldehyde causes leukemia, particularly myeloid leukemia; the study itself looks for leukemia-like chromosome changes in 29 exposed Chinese workers against 23 controls, a test of plausibility, not of risk.

19. 19
    Mundt KA, Thompson WJ. Formaldehyde and Myeloid Leukemia: Diverging Tracks of Human Health Science and Regulation.Risk Anal · 2026 · 46(3):e70198doi:10.1111/risa.70198 · PMID 41731959

    Reviews the conflicting hazard assessments since 2009 and reports that the EU SCOEL and ECHA concluded formaldehyde is unlikely to cause myeloid leukemia; it argues the skeptical side, its second author is an independent consultant, and the first author's earlier reviews of the same question were written at Ramboll Environ, an industry consultancy, which should be weighed.

20. 20
    Shi SR, Shi Y, Taylor CR. Antigen retrieval immunohistochemistry: review and future prospects in research and diagnosis over two decades.J Histochem Cytochem · 2011 · 59(1):13-32doi:10.1369/jhc.2010.957191 · PMID 21339172

    Twenty-year review by the method's originators of how antigen retrieval spread across pathology; a review by interested parties, not an independent audit.

21. 21
    Do H, Dobrovic A. Sequence artifacts in DNA from formalin-fixed tissues: causes and strategies for minimization.Clin Chem · 2015 · 61(1):64-71doi:10.1373/clinchem.2014.223040 · PMID 25421801

    A mini-review of the lesions formalin leaves in DNA (fragmentation, deaminated bases, abasic sites) and how to guard against them; a review, not new data.

22. 22
    Reid KM, Maistry S, Ramesar R, et al. A review of the optimisation of the use of formalin fixed paraffin embedded tissue for molecular analysis in a forensic post-mortem setting.Forensic Sci Int · 2017 · 280:181-187doi:10.1016/j.forsciint.2017.09.020 · PMID 29078160

    Systematic review of 111 articles on getting usable DNA from FFPE tissue, stating plainly that formalin yields severely fragmented DNA; forensic setting, not parasite detection.

23. 23
    Haghshenas M, Koosha M, Latifi A, et al. Detection of Enterobius vermicularis in archived formalin-fixed paraffin-embedded (FFPE) appendectomy blocks: It's potential to compare genetic variations based on mitochondrial DNA (cox1) gene.PLoS One · 2023 · 18(2):e0281622doi:10.1371/journal.pone.0281622 · PMID 36758053

    Pinworm DNA amplified from 12 of 17 appendectomy blocks stored 12 to 22 years; blocks were chosen because worms were already seen, so it shows feasibility, not sensitivity.

24. 24
    Grimm J, Krickl J, Beck A, et al. Establishing and evaluation of a polymerase chain reaction for the detection of Echinococcus multilocularis in human tissue.PLoS Negl Trop Dis · 2021 · 15(2):e0009155doi:10.1371/journal.pntd.0009155 · PMID 33630840

    A validated tapeworm PCR on FFPE tissue: all 6 lesions in blocks under 6 years old were positive, but only 3 of 9 older blocks were; liver disease, not tumors.

25. 25
    Hamelin B, Hosch S, Neidhöfer C, et al. Unbiased DNA Pathogen Detection in Tissues: Real-World Experience With Metagenomic Sequencing in Pathology.Lab Invest · 2026 · 106(1):104254doi:10.1016/j.labinv.2025.104254 · PMID 41167475

    Routine metagenomic sequencing of 623 FFPE samples in a Swiss pathology institute found a plausible pathogen in 36.8%, parasites in 9 cases; infectious-pathology cases, not a tumor survey. Published online 29 October 2025.

26. 26
    Goldstein NS, Ferkowicz M, Odish E, et al. Minimum formalin fixation time for consistent estrogen receptor immunohistochemical staining of invasive breast carcinoma.Am J Clin Pathol · 2003 · 120(1):86-92doi:10.1309/QPHD-RB00-QXGM-UQ9N · PMID 12866377

    Timed fixation of 24 breast cancers showing estrogen-receptor staining is unreliable under 6 to 8 hours; one laboratory, one marker.

27. 27
    De Marzo AM, Fedor HH, Gage WR, et al. Inadequate formalin fixation decreases reliability of p27 immunohistochemical staining: probing optimal fixation time using high-density tissue microarrays.Hum Pathol · 2002 · 33(7):756-60doi:10.1053/hupa.2002.126187 · PMID 12196928

    Prostate tissue fixed 0 to 8 days: same-day processing gave the weakest staining and 8 days was still excellent; one antigen, so the conclusion cannot be generalized to every marker.

28. 28
    Ndengue CP, Atangana PJA, Ateba GR, et al. Pre-analytical variables affecting breast cancer biomarker expression: A controlled single-specimen study of fixation duration, cold ischemia time, and fixative preparation in a low-resource setting.PLoS One · 2026 · 21(6):e0343185doi:10.1371/journal.pone.0343185 · PMID 42247401

    Fifty cores from one mastectomy: receptor staining stable from 0.5 to 48 hours, falling beyond 72 hours, and only slightly worse in unbuffered formalin; a single patient, so it is a proof of concept.

29. 29
    Honeywell R, Donner M, Criswell S. Thorough formalin fixation confers a protective effect for histochemical and immunohistochemical testing on tissues exposed to prolonged time in ethanol, xylene, heat, or paraffin.Histochem Cell Biol · 2026 · 164(1)doi:10.1007/s00418-026-02545-y · PMID 42825935

    Well-fixed animal tissue tolerated an extra day in processing reagents, but organs fixed for a year lost some immunoreactivity; animal organs, not human surgical specimens.

30. 30
    Wu AA, Haley L, Pallavajjala A, et al. Long-Term Storage of Formalin-Fixed Paraffin-Embedded Tissues Negatively Impacts Next-Generation Sequencing: Successful Restoration with a DNA Repair Enzyme.Clin Chem · 2026 · 72(10):1067-1078doi:10.1093/clinchem/hvag066 · PMID 42322401

    281 FFPE tissues aged 0 to 17 years: sequencing quality fell with storage and 2 to 4 years added more false variants than true ones; repair-enzyme rescue was tested on 10 paired samples.

31. 31
    Boyer IJ, Heldreth B, Bergfeld WF, et al. Amended safety assessment of formaldehyde and methylene glycol as used in cosmetics.Int J Toxicol · 2013 · 32(6 Suppl):5S-32Sdoi:10.1177/1091581813511831 · PMID 24335968

    Cosmetic Ingredient Review panel; concludes that the two interconvert continuously and that hair-smoothing products are unsafe at present use.

32. 32
    Golden R, Valentini M. Formaldehyde and methylene glycol equivalence: critical assessment of chemical and toxicological aspects.Regul Toxicol Pharmacol · 2014 · 69(2):178-86doi:10.1016/j.yrtph.2014.03.007 · PMID 24709515

    Written at ToxLogic, a consultancy; argues against treating methylene glycol as formaldehyde for regulatory purposes.

33. 33
    Helander KG. Formaldehyde prepared from paraformaldehyde is stable.Biotech Histochem · 2000 · 75(1):19-22doi:10.3109/10520290009047980 · PMID 10810978

    Shows that formaldehyde made from paraformaldehyde is purer than commercial stock and stays stable for days; a small chemistry study, not a comparison of diagnostic outcomes.

34. 34
    Shi SR, Key ME, Kalra KL. Antigen retrieval in formalin-fixed, paraffin-embedded tissues: an enhancement method for immunohistochemical staining based on microwave oven heating of tissue sections.J Histochem Cytochem · 1991 · 39(6):741-8doi:10.1177/39.6.1709656 · PMID 1709656

    The paper that introduced heat-induced antigen retrieval (39 of 52 antibodies improved); the first author worked at BioGenex Laboratories, a commercial reagent company.

35. 35
    Oh E, Choi YL, Kwon MJ, et al. Comparison of Accuracy of Whole-Exome Sequencing with Formalin-Fixed Paraffin-Embedded and Fresh Frozen Tissue Samples.PLoS One · 2015 · 10(12):e0144162doi:10.1371/journal.pone.0144162 · PMID 26641479

    Matched frozen and FFPE exomes from 4 patients: short inserts and 200 to 1,200 artifactual C>T/G>A changes per million bases in FFPE only; four patients, so the rate is indicative.

36. 36
    Do H, Dobrovic A. Dramatic reduction of sequence artefacts from DNA isolated from formalin-fixed cancer biopsies by treatment with uracil- DNA glycosylase.Oncotarget · 2012 · 3(5):546-58doi:10.18632/oncotarget.503 · PMID 22643842

    Identifies uracil from cytosine deamination as a major source of formalin sequencing artifacts and removes much of it enzymatically; tested on a handful of cancer genes.

37. 37
    Young KH, Bullock SL, Melvin DM, et al. Ethyl acetate as a substitute for diethyl ether in the formalin-ether sedimentation technique.J Clin Microbiol · 1979 · 10(6):852-3doi:10.1128/jcm.10.6.852-853.1979 · PMID 574877

    The study that let stool laboratories replace flammable ether with ethyl acetate without losing organisms or distorting their shapes; a short comparative report.

38. 38
    Poggensee G, Sahebali S, Van Marck E, et al. Diagnosis of genital cervical schistosomiasis: comparison of cytological, histopathological and parasitological examination.Am J Trop Med Hyg · 2001 · 65(3):233-6doi:10.4269/ajtmh.2001.65.233 · PMID 11561710

    Same 228 Tanzanian women tested three ways: eggs found in 49% by compressed biopsy, 18% by histologic sections and 3% by cytology; it measures sampling and method, not formalin chemistry directly.

39. 39
    Abdelsalam IM, Sarhan RM, Hanafy MA. The Impact of Different Copro-preservation Conditions on Molecular Detection of Cryptosporidium Species.Iran J Parasitol · 2017 · 12(2):274-283PMID 28761489

    Ten Cryptosporidium-positive stools kept five ways for a month: freezing preserved PCR detection best and 10% formalin worst (0%); a small sample, one gene target.

40. 40
    Musiał A, Gryglewski RW, Kielczewski S, et al. Formalin use in anatomical and histological science in the 19th and 20th centuries.Folia Med Cracov · 2016 · 56(3):31-40PMID 28275269

    A history of formalin from primary texts, naming the chemists and anatomists involved; a historical review, with no laboratory data.

41. 41
    Erdman DD. Clinical comparison of ethyl acetate and diethyl ether in the formalin-ether sedimentation technique.J Clin Microbiol · 1981 · 14(5):483-5doi:10.1128/jcm.14.5.483-485.1981 · PMID 7309845

    62 fresh stools concentrated both ways: comparable recovery of eggs, cysts and larvae, with practical cautions; one laboratory.

42. 42
    Wolff AC, Hammond MEH, Schwartz JN, et al. American Society of Clinical Oncology/College of American Pathologists guideline recommendations for human epidermal growth factor receptor 2 testing in breast cancer.J Clin Oncol · 2007 · 25(1):118-45doi:10.1200/JCO.2006.09.2775 · PMID 17159189

    The first ASCO/CAP HER2 guideline, estimating that about 20% of HER2 testing may be inaccurate; its fixation window (6 to 48 hours) is quoted here through a later paper, not from this abstract.

43. 43
    Möhner M, Wendt A. A diagnostic bias might be a much simpler explanation for the apparently elevated risk for nasopharyngeal cancer with respect to formaldehyde.J Occup Med Toxicol · 2016 · 11:54doi:10.1186/s12995-016-0143-4 · PMID 27980603

    Editorial from the German Federal Institute for Occupational Safety and Health arguing that diagnostic bias may explain the nasopharyngeal-cancer excess in the US industrial cohort; it also records the 2009 IARC conclusion on leukemia. A dissent, not new data.

44. 44
    Hammond MEH, Hayes DF, Dowsett M, et al. American Society of Clinical Oncology/College Of American Pathologists guideline recommendations for immunohistochemical testing of estrogen and progesterone receptors in breast cancer.J Clin Oncol · 2010 · 28(16):2784-95doi:10.1200/JCO.2009.25.6529 · PMID 20404251

    The ER/PgR guideline stating that up to 20% of receptor results worldwide may be wrong, mostly from pre-analytic variation; panel disclosures are in the full text and were not reviewed here.

45. 45
    Muehlenbachs A, Bhatnagar J, Agudelo CA, et al. Malignant Transformation of Hymenolepis nana in a Human Host.N Engl J Med · 2015 · 373(19):1845-52doi:10.1056/NEJMoa1505892 · PMID 26535513

    A single case in which cancer-like cells in a man with HIV were tapeworm cells, recognized only by eukaryote-targeted PCR; one patient, profound immunosuppression.

46. 46
    Ryska A, Sapino A, Landolfi S, et al. Glyoxal acid-free (GAF) histological fixative is a suitable alternative to formalin: results from an open-label comparative non-inferiority study.Virchows Arch · 2024 · 485(2):213-222doi:10.1007/s00428-023-03692-6 · PMID 37996705

    Multicenter non-inferiority trial scored with four binary questions, published online in 2023; co-authors from ADDAX Biosciences and 1MED; the abstract records that the EU proposed a formalin ban in 2016 with a temporary exemption for diagnostics.

47. 47
    Wolff AC, Hammond MEH, Allison KH, et al. Human Epidermal Growth Factor Receptor 2 Testing in Breast Cancer: American Society of Clinical Oncology/College of American Pathologists Clinical Practice Guideline Focused Update.J Clin Oncol · 2018 · 36(20):2105-2122doi:10.1200/JCO.2018.77.8738 · PMID 29846122

    Focused update on interpretation and unusual in situ hybridization patterns; the abstract does not restate the fixation window, which is carried over from 2013.

48. 48
    Wolff AC, Somerfield MR, Dowsett M, et al. Human Epidermal Growth Factor Receptor 2 Testing in Breast Cancer: ASCO-College of American Pathologists Guideline Update.J Clin Oncol · 2023 · 41(22):3867-3872doi:10.1200/JCO.22.02864 · PMID 37284804

    The current HER2 update: affirms the 2018 recommendations and stresses separating IHC 0 from 1+ now that trastuzumab deruxtecan is approved for low HER2; its abstract does not restate the fixation window.

49. 49
    Allison KH, Hammond MEH, Dowsett M, et al. Estrogen and Progesterone Receptor Testing in Breast Cancer: ASCO/CAP Guideline Update.J Clin Oncol · 2020 · 38(12):1346-1366doi:10.1200/JCO.19.02309 · PMID 31928404

    Current ER/PgR guideline, creating the ER Low Positive category and requiring that the status of controls be reported for 0 to 10% staining; the abstract does not restate the 6-to-72-hour window.

50. 50
    Guo Q, Lakatos E, Bakir IA, et al. The mutational signatures of formalin fixation on the human genome.Nat Commun · 2022 · 13(1):4487doi:10.1038/s41467-022-32041-5 · PMID 36068219

    Derives the genome-wide mutational signature of formalin and a computational correction (FFPEsig); corrects the statistics of a mutation catalog, not individual calls.

51. 51
    Start RD, Layton CM, Cross SS, et al. Reassessment of the rate of fixative diffusion.J Clin Pathol · 1992 · 45(12):1120-1doi:10.1136/jcp.45.12.1120 · PMID 1479044

    Measured fixative penetration into whole human spleens over 1 to 25 days and found it slower than earlier gel and animal studies; the abstract gives no single rate.

52. 52
    Vassilakopoulou M, Parisi F, Siddiqui S, et al. Preanalytical variables and phosphoepitope expression in FFPE tissue: quantitative epitope assessment after variable cold ischemic time.Lab Invest · 2015 · 95(3):334-41doi:10.1038/labinvest.2014.139 · PMID 25418580

    93 breast cancers with documented time to fixation: several labile phosphoproteins lost antigenicity within 1 to 2 hours of cold ischemia; the 30-minute figure is quoted in its introduction from earlier work, not measured here; research epitopes, not routine diagnostic markers.

53. 53
    Engel KB, Moore HM. Effects of preanalytical variables on the detection of proteins by immunohistochemistry in formalin-fixed, paraffin-embedded tissue.Arch Pathol Lab Med · 2011 · 135(5):537-43doi:10.5858/2010-0702-RAIR.1 · PMID 21526952

    A National Cancer Institute review of 39 studies: 62 pre-analytical variables identified, 27 studied, 15 shown to affect staining; it summarizes, it does not test.

54. 54
    Grillo F, Ali M, Paudice M, et al. Impact of formalin fixation on mismatch repair protein evaluation by immunohistochemistry.Virchows Arch · 2023 · 483(5):677-685doi:10.1007/s00428-023-03661-z · PMID 37773452

    Leftover tissue from 30 colorectal resections fixed five ways and stained 600 times; the first study designed to test fixation on mismatch-repair stains; one center.

55. 55
    Abe H, Kuwata T, Kushima R, Ushiku T. Nationwide survey on HER2 and PD-L1 testing practices in gastric cancer across Japan.Gastric Cancer · 2025 · 28(2):294-300doi:10.1007/s10120-024-01571-w · PMID 39656340

    Questionnaire answered by 155 institutions; the fixation effect is an association in survey data, and the authors also blame differences in scoring between pathologists.

56. 56
    Haight PJ, Lammers S, Kistenfeger Q, et al. Cold ischemia time and formalin fixation time in endometrial cancer: Should breast cancer guidelines for preanalytical variables be applied to hysterectomy specimens?Gynecol Oncol · 2024 · 191:194-200doi:10.1016/j.ygyno.2024.10.015 · PMID 39442372

    941 patients, median cold ischemia 33 minutes; a negative result for four markers in one tumor type.

57. 57
    Chambers JK, Watanabe K, Michishita M, et al. Standardization of tissue handling in veterinary pathology: How fixation affects morphological and molecular examination results.Vet Pathol · 2025 · 62(6):949-962doi:10.1177/03009858251349126 · PMID 40574335

    Animal tissues; unbuffered and prolonged fixation each fragmented DNA and lowered PCR efficiency, and coating blocks with paraffin prevented DNA fragmentation over time; veterinary material.

58. 58
    Frigon EM, Ma W, Tremblay C, et al. Effects of prolonged fixation on vascular biomarkers in postmortem human brains.J Cereb Blood Flow Metab · 2026 · 271678X261461536doi:10.1177/0271678X261461536 · PMID 42257420

    Human brains fixed 1 to 20 years: some stains (ferritin, vimentin, collagen IV, a silver stain) faded with time and others did not; postmortem brain, not surgical tumor tissue.

59. 59
    Arora N, Bajpai J, Bal A, et al. Challenges & recommendations for identification of human epidermal growth factor receptor-2 (HER2)-low metastatic breast cancer in India: Expert opinion statement.Indian J Med Res · 2025 · 162(3):279-292doi:10.25259/IJMR_2009_2024 · PMID 41296586

    Expert statement that records the DESTINY-Breast04 and DESTINY-Breast06 trials, defines HER2-ultralow as a score of 0 with some membrane staining and lists fixation among the variables that decide low scores; opinion, not data.

60. 60
    Juneja A, Sahay A, Padwale P, et al. Concordance of conventional two-tier HER2 classification versus three-tier classification (including HER2-low status) in invasive breast carcinoma: a retrospective real-world analysis.J Clin Pathol · 2026 · 79(8):526-533doi:10.1136/jcp-2026-210710 · PMID 42191357

    1,107 paired samples from one Indian cancer center; concordance 69.7% (kappa 0.53) under the three-tier scheme, and referral specimens worse (odds ratio 1.72).

61. 61
    Wilcock D, Jedrzkiewicz J, Cleary AS, et al. Breast Cancers With Discordant Human Epidermal Growth Factor Receptor 2 Fluorescence In Situ Hybridization-Positive (Group 1) and Immunohistochemistry-Negative (0/1+) Results Following American Society of Clinical Oncology/College of American Pathologists Guideline.Arch Pathol Lab Med · 2026 · 150(7):503-512doi:10.5858/arpa.2025-0490-OA · PMID 41819599

    In 956 breast cancers tested both ways, 1.1% were IHC-negative but FISH-positive, more often from metastatic sites with undocumented fixation; one institution, and fixation is one candidate cause among several.

62. 62
    Amemiya K, Hirotsu Y, Nagakubo Y, et al. Influence of formalin fixation duration on RNA quality and quantity from formalin-fixed paraffin-embedded hepatocellular carcinoma tissues.Pathol Int · 2023 · 73(12):593-600doi:10.1111/pin.13385 · PMID 37933792

    Timed fixation, 1 to 240 hours, of hepatocellular carcinoma; blocks stored 500 days also lost RNA yield; one tumor type and one sequencing platform.

63. 63
    Kunita A, Shinozaki-Ushiku A, Nagumo S, Ushiku T. Preanalytical Determinants of DNA and RNA Quality in FFPE Tissues: Practical Recommendations for Molecular Testing.Pathol Int · 2026 · 76(5):e70118doi:10.1111/pin.70118 · PMID 42116622

    Pig liver trimmed to 2, 10 or 50 mm and fixed 1, 3 or 7 days in 10% or 20% formalin, plus clinical blocks; recommends 10% formalin for 24 hours or less and blocks under one year old.

64. 64
    Zhang Y, Blomquist TM, Kusko R, et al. Deep oncopanel sequencing reveals within block position-dependent quality degradation in FFPE processed samples.Genome Biol · 2022 · 23(1):141doi:10.1186/s13059-022-02709-8 · PMID 35768876

    A characterized diploid cell line fixed for four durations and sequenced by four cancer panels in several laboratories; a cell block, not tissue.

65. 65
    Gambella A, Mastracci L, Trambaiolo C, et al. Preanalytical Determinants of Patient Eligibility for Immune Checkpoint Inhibitors Revealed by Quantitative Programmed Cell Death Ligand-1 (PD-L1) Analysis.Lab Invest · 2026 · 106(8):106138doi:10.1016/j.labinv.2026.106138 · PMID 42176954

    Tonsil used as a standard tissue; cold ischemia above 30 hours cut PD-L1 staining by more than 52%, block aging by 49.2% at 60 months and slide aging by about 60% in 60 days; cold formalin slowed but did not stop the loss.

66. 66
    Scott L, Benanto J, Robbins CJ, et al. Temperature-Dependent Long-term and Short-term Stability of Human Epidermal Growth Factor Receptor 2 Antigenicity on Pathology Slides.Arch Pathol Lab Med · 2026 · 150(9):707-712doi:10.5858/arpa.2025-0447-OA · PMID 41997586

    Surplus microarray slides stored for years at room temperature or at -80 °C compared with freshly cut slides; loss over years at room temperature, negligible over two months in either condition.

67. 67
    Do H, Wong SQ, Li J, et al. Reducing sequence artifacts in amplicon-based massively parallel sequencing of formalin-fixed paraffin-embedded DNA by enzymatic depletion of uracil-containing templates.Clin Chem · 2013 · 59(9):1376-83doi:10.1373/clinchem.2012.202390 · PMID 23649127

    Carries the uracil fix into next-generation sequencing and shows its limit: artifacts at methylated CpG sites are resistant; lung cancer samples on one amplicon panel.

68. 68
    Basyuni S, Heskin L, Degasperi A, et al. Large-scale analysis of whole genome sequencing data from formalin-fixed paraffin-embedded cancer specimens demonstrates preservation of clinical utility.Nat Commun · 2024 · 15(1):7731doi:10.1038/s41467-024-51577-2 · PMID 39231944

    National program data; the approach characterizes artifacts rather than discarding samples, and fresh-frozen tissue remains the authors' gold standard.

69. 69
    Domenico D, Gundem G, Levine MF, et al. Enabling whole genome sequencing analysis from FFPE specimens in clinical oncology.Nat Commun · 2025 · 16(1):10649doi:10.1038/s41467-025-65654-7 · PMID 41309586

    56 fresh-frozen and FFPE pairs; the FFPErase filter reached 99% sensitivity against FDA-approved panel tests; one cancer center.

70. 70
    Dixon K, Lee JH, Miller R, et al. Cryopreserved Tissue Biospecimens Offer Superior Quality for Whole-Genome Sequencing of Various Cancers Compared to Paired Formalin-Fixed Paraffin-Embedded Tissues.Int J Mol Sci · 2025 · 26(22):11038doi:10.3390/ijms262211038 · PMID 41303519

    The first author works for Specicare, a company that sells cryopreservation, and several co-authors for a sequencing company; the direction of the result matches independent work, but the magnitude should be read with that in mind.

71. 71
    Matsunaga H, Arikawa K, Yamazaki M, et al. Reproducible and sensitive micro-tissue RNA sequencing from formalin-fixed paraffin-embedded tissues for spatial gene expression analysis.Sci Rep · 2022 · 12(1):19511doi:10.1038/s41598-022-23651-6 · PMID 36376423

    About 7,000 genes read from 100-micrometer punches of 2-year-old mouse blocks and 6-year-old human lung cancer blocks; one author lists a company affiliation (Frontier Biosystems).

72. 72
    Xiao Y, Sheng ZM, Taubenberger JK. Precursors to the 1918 pandemic: Finding Rickettsia felis and Mucor co-infection associated with a death in 1912.Int J Paleopathol · 2026 · 52:69-75doi:10.1016/j.ijpp.2026.01.001 · PMID 41529435

    Metagenomic reads from a 1912 FFPE lung block suggested Rickettsia felis and Mucor; the authors state that contamination cannot be ruled out and no tissue remained for confirmation.

73. 73
    Srinivasan M, Sedmak D, Jewell S. Effect of fixatives and tissue processing on the content and integrity of nucleic acids.Am J Pathol · 2002 · 161(6):1961-71doi:10.1016/S0002-9440(10)64472-0 · PMID 12466110

    A review of how procurement, fixation and processing change the molecular quality of tissue; it frames the problem, without new data.

74. 74
    Hurler CA, Liebscher S, Arzberger T, Jäkel S. Impact of fixation duration on messenger RNA detectability in human formalin-fixed paraffin-embedded brain tissue.Brain Commun · 2024 · 6(6):fcae430doi:10.1093/braincomms/fcae430 · PMID 39659968

    RNAscope in post-mortem brain; descriptive, with no single threshold.

75. 75
    Howe S, McCann H, Xiong Z, et al. Long-term fixation impact on archived human nervous tissues for sequencing-based transcriptomics.Brain Commun · 2025 · 7(6):fcaf428doi:10.1093/braincomms/fcaf428 · PMID 41262414

    Sydney Brain Bank tissue; the failure was in probe ligation, not in the RNA integrity number or DV200.

76. 76
    Bauer DR, Leibold T, Chafin DR. Making a science out of preanalytics: An analytical method to determine optimal tissue fixation in real-time.PLoS One · 2021 · 16(10):e0258495doi:10.1371/journal.pone.0258495 · PMID 34648597

    Acoustic monitoring of formalin diffusion in 87 tonsil samples, opening with the statement that no analytical method of detecting fixation exists in practice; two authors were Roche Tissue Diagnostics employees.

77. 77
    Bauer DR, Chafin DR. Assessing Tissue Fixation Time and Quality with Label-free Mid Infrared Spectroscopy and Machine Learning.Biopreserv Biobank · 2023 · 21(2):208-216doi:10.1089/bio.2022.0108 · PMID 36516138

    Both authors at Roche Diagnostics Solutions (Ventana); more than 100 samples; a method paper.

78. 78
    Johann DJ, Shin IJ, Roberge A, et al. Effect of Antigen Retrieval on Genomic DNA From Immunodissected Samples.J Histochem Cytochem · 2022 · 70(9):643-658doi:10.1369/00221554221124163 · PMID 36129255

    Clinical specimens and xenografts; heat retrieval gave the best staining but extensive DNA fragmentation, and mutations were still detectable by droplet PCR and targeted sequencing.

79. 79
    Wang H, Luo X, Qiang X, Dai X. Heat-induced antigen retrieval by using Tris-EDTA solution destroys nuclear structure in certain tissues.Biochem Biophys Rep · 2026 · 45:102443doi:10.1016/j.bbrep.2026.102443 · PMID 41568059

    Mouse and human tissues; nuclear leakage in thyroid, lung and kidney with Tris-EDTA but not citrate; one hospital laboratory.

80. 80
    Boonsuya A, Arunsan P, Pechdee P, et al. Detection of the carcinogenic liver fluke, Opisthorchis viverrini: comparison of two coprological methods versus the automatic feces analyzer.Trop Biomed · 2024 · 41(3):264-270doi:10.47665/tb.41.3.005 · PMID 39548779

    455 stools from northeastern Thailand: formalin-ethyl acetate concentration found the most liver-fluke eggs; it compares methods against each other, not against a molecular reference.

81. 81
    Jongthawin J, Mahittikorn A, Wangdi K, et al. Comparison of Ritchie and Kato-Katz methods for the detection of intestinal helminths in humans: a systematic review and meta-analysis.Parasit Vectors · 2026 · 19(1)doi:10.1186/s13071-026-07437-7 · PMID 42141455

    Meta-analysis of 40 studies comparing formalin-ether concentration with the Kato-Katz smear; heterogeneity was very high, so the pooled odds ratios are rough guides.

82. 82
    Larkins A, Keokhamphavanh B, Knight B, et al. Diagnostic sensitivity of formalin-fixed faecal microscopy for the detection of soil-transmitted helminths.Trans R Soc Trop Med Hyg · 2025 · 119(6):596-605doi:10.1093/trstmh/traf011 · PMID 39921402

    Bayesian latent-class estimates on formalin-fixed samples only, so the comparison with fresh stool is against earlier studies, not a paired test.

83. 83
    Naidoo D, Archer CE. Effects of reagents and chemicals used in existing sanitation helminth test methods on the viability of Ascaris suum eggs.J Parasitol · 2024 · 110(6):709-716doi:10.1645/22-58 · PMID 39701159

    Found that 10% formalin on its own, like every other extraction solution, had minimal effect on the viability of pig roundworm eggs; the abstract gives no exposure time for formalin alone and up to 15 or 30 minutes for the combined and flotation steps, so the exposures were minutes, not days; sanitation methods and Ascaris suum, not human tissue.

84. 84
    Trecourt A, Radobonirina M. [Infectious cysts and pseudocysts: When parasites want to mimic tumors!].Ann Pathol · 2025 · 45(6):513-522doi:10.1016/j.annpat.2025.04.003 · PMID 40328550

    A French review stating that parasitic cysts are rarely seen by pathologists and carry many morphologic pitfalls; expert opinion with no pooled numbers.

85. 85
    Jacob S, Parameswaran A, Santosham R, et al. Human pulmonary dirofilariasis masquerading as a mass.Asian Cardiovasc Thorac Ann · 2016 · 24(7):722-5doi:10.1177/0218492316658569 · PMID 27432058

    Two lung nodules suspected to be cancer were dog heartworm; two cases, showing worms recognized after excision rather than missed.

86. 86
    Tran T, Sundaram CP, Bahler CD, et al. Correcting the Shrinkage Effects of Formalin Fixation and Tissue Processing for Renal Tumors: toward Standardization of Pathological Reporting of Tumor Size.J Cancer · 2015 · 6(8):759-66doi:10.7150/jca.12094 · PMID 26185538

    34 kidney tumors shrank a mean 11.4% from fresh to slide; measured in kidney tumors, so its application to parasite eggs is an assumption.

87. 87
    Koonmee S, Intapan PM, Yamasaki H, et al. Molecular identification of a causative parasite species using formalin-fixed paraffin embedded (FFPE) tissues of a complicated human pulmonary sparganosis case without decisive clinical diagnosis.Parasitol Int · 2011 · 60(4):460-4doi:10.1016/j.parint.2011.07.018 · PMID 21835263

    A tapeworm larva identified to species from FFPE lung biopsy DNA; a single case report.

88. 88
    Gholami Koohestan M, Saberi R, Daryani A, et al. Identification and genotyping of Echinococcus granulosus from human clinical samples in Guilan province, north of Iran.Parasite Epidemiol Control · 2024 · 25:e00353doi:10.1016/j.parepi.2024.e00353 · PMID 38699631

    35 FFPE hydatid cysts; block ages are not given in the abstract; all sequences were E. granulosus sensu stricto.

89. 89
    Knapp J, Lallemand S, Monnien F, et al. Real-time multiplex PCR for human echinococcosis and differential diagnosis.Parasite · 2023 · 30:3doi:10.1051/parasite/2023003 · PMID 36700708

    French national reference center; 81 fresh and FFPE tissues; 100% specificity for all targets, with companion assays for Taenia solium and Toxocara.

90. 90
    Hu D, Jin W, Ding H, et al. Spirometra mansoni sparganosis identified by metagenomic next-generation sequencing: a case report.Int J Infect Dis · 2023 · 128:128-131doi:10.1016/j.ijid.2022.12.038 · PMID 36592686

    One patient; confirmed by cox1 PCR and phylogeny; one co-author is at a commercial diagnostics company.

91. 91
    Brunner A, Unterberger SH, Auer H, et al. Suitability of Fourier transform infrared microscopy for the diagnosis of cystic echinococcosis in human tissue sections.J Biophotonics · 2024 · 17(6):e202300513doi:10.1002/jbio.202300513 · PMID 38531615

    11 confirmed patients, archived blocks; a feasibility study with no blinded accuracy figures.

92. 92
    Rodriguez R, da Silva ACA, Müller CA, et al. PCR for the diagnosis of abdominal angiostrongyliasis in formalin-fixed paraffin-embedded human tissue.PLoS One · 2014 · 9(4):e93658doi:10.1371/journal.pone.0093658 · PMID 24705328

    Worm PCR on FFPE bowel: 55% sensitivity, 100% specificity, and positive in 4 of 20 cases where no worm was seen; small groups.

93. 93
    Kap M, Smedts F, Oosterhuis W, et al. Histological assessment of PAXgene tissue fixation and stabilization reagents.PLoS One · 2011 · 6(11):e27704doi:10.1371/journal.pone.0027704 · PMID 22110732

    70 split normal tissues: PAXgene morphology adequate for diagnosis in most; work from the EU SPIDIA project testing the commercial PreAnalytiX product, whose composition is given only as the maker describes it.

94. 94
    Dapson RW. Glyoxal fixation: how it works and why it only occasionally needs antigen retrieval.Biotech Histochem · 2007 · 82(3):161-6doi:10.1080/10520290701488113 · PMID 17987441

    Describes glyoxal chemistry and notes that silver stains for Helicobacter pylori fail after it; the author's affiliation is a private consultancy, and the paper is a review.

95. 95
    Richter KN, Revelo NH, Seitz KJ, et al. Glyoxal as an alternative fixative to formaldehyde in immunostaining and super-resolution microscopy.EMBO J · 2018 · 37(1):139-159doi:10.15252/embj.201695709 · PMID 29146773

    Glyoxal fixed cultured cells faster and stained better than paraformaldehyde for most targets; research microscopy, not diagnostic pathology.

96. 96
    Bussolati G, Annaratone L, Berrino E, et al. Acid-free glyoxal as a substitute of formalin for structural and molecular preservation in tissue samples.PLoS One · 2017 · 12(8):e0182965doi:10.1371/journal.pone.0182965 · PMID 28796828

    30 colon and stomach cancers fixed in parallel; from the Turin group that devised the fixative, which was later commercialized by ADDAX Biosciences.

97. 97
    Criswell SL, Altman S, Peeler C, et al. Glyoxal fixation: an immunohistochemistry assay evaluation.J Histotechnol · 2022 · 45(2):77-84doi:10.1080/01478885.2021.1996125 · PMID 34878359

    34 antibodies; formalin slightly superior but all glyoxal fixatives judged acceptable; one academic laboratory in Tennessee.

98. 98
    Peeler C, Pitzer CR, Paez HG, Criswell S. Histochemical and morphological evaluation of a glyoxal acid-free fixative.Biotech Histochem · 2024 · 99(1):49-58doi:10.1080/10520295.2023.2300797 · PMID 38164087

    Formalin and an acidic commercial glyoxal behaved alike while 2% acid-free glyoxal was deleterious; independent of the maker.

99. 99
    Zappulli V, Moccia V, Torrigiani F, et al. Non-toxic acid-free glyoxal fixative for veterinary gross specimen preservation, histopathology, immunohistochemistry, and molecular analysis.Vet Pathol · 2026 · 63(2):282-295doi:10.1177/03009858251372572 · PMID 41059728

    59 biopsies and 21 necropsies from several animal species; one co-author is at ADDAX Biosciences.

100. 100
     Zanini C, Gerbaudo E, Ercole E, et al. Evaluation of two commercial and three home-made fixatives for the substitution of formalin: a formaldehyde-free laboratory is possible.Environ Health · 2012 · 11:59doi:10.1186/1476-069X-11-59 · PMID 22947094

     More than 200 specimens fixed in parallel in formalin substitutes; the first author is affiliated with EuroClone S.p.A., a company, which should be weighed against its advocacy.

101. 101
     Burgos-Barragan G, Wit N, Meiser J, et al. Mammals divert endogenous genotoxic formaldehyde into one-carbon metabolism.Nature · 2017 · 548(7669):549-554doi:10.1038/nature23481 · PMID 28813411

     Shows that formaldehyde is generated from the breakdown of folate inside cells and that human, mouse and chicken cells unable to detoxify it, or unable to repair DNA cross-links, die when given folate derivatives; cell-culture work, not an exposure study.

102. 102
     Hauptmann M, Stewart PA, Lubin JH, et al. Mortality from lymphohematopoietic malignancies and brain cancer among embalmers exposed to formaldehyde.J Natl Cancer Inst · 2009 · 101(24):1696-708doi:10.1093/jnci/djp416 · PMID 19933446

     Case-control study of funeral workers: myeloid leukemia deaths rose with years of embalming and peak exposure; exposures reconstructed from interviews with relatives and coworkers, so association, not proof.

103. 103
     Kwak K, Paek D, Park JT. Occupational exposure to formaldehyde and risk of lung cancer: A systematic review and meta-analysis.Am J Ind Med · 2020 · 63(4):312-327doi:10.1002/ajim.23093 · PMID 32003024

     Pooled 31 studies: no overall rise in lung cancer with occupational formaldehyde, a small rise in higher-quality studies; observational data.

104. 104
     Cocco P, Meloni F, Trobbiani C, et al. Occupational exposure to formaldehyde and risk of lymphoma subtypes: results of a multicentre Italian case-control study.Environ Health · 2025 · 24(1):82doi:10.1186/s12940-025-01232-0 · PMID 41146278

     867 lymphoma cases and 774 controls with expert-assessed exposure; a myeloma signal only, with trends across exposure metrics, and no association with lymphoma overall.

105. 105
     Pourbabaki R, Rahimian F, Ghaffari ME, et al. Hematological changes in formaldehyde-exposed workers: A critical review and meta-analysis.Toxicol Ind Health · 2026 · 42(10):424-439doi:10.1177/07482337261472039 · PMID 42538616

     22 studies, 10 pooled; the authors stress poor exposure assessment, find no dose-response, and conclude that inhaled formaldehyde stays at the portal of entry, the position also held by industry-linked reviewers.

106. 106
     Li J, Lai Y, Tang A, Yi X. A global analysis of nasopharynx cancer burden attributable to occupational formaldehyde exposure.Front Oncol · 2025 · 15:1660669doi:10.3389/fonc.2025.1660669 · PMID 41079100

     Modeled estimates from the Global Burden of Disease 2021 study (592 deaths, 95% uncertainty interval 401 to 857); a model, not a count.

107. 107
     Foroughi P, Golbabaei F, Sadeghi-Yarandi M, et al. Occupational exposure, carcinogenic and non-carcinogenic risk assessment of formaldehyde in the pathology labs of hospitals in Iran.Sci Rep · 2024 · 14(1):12006doi:10.1038/s41598-024-62133-9 · PMID 38796506

     Four Tehran hospitals; 91.23% of exposures above the US NIOSH limit of 0.016 ppm, highest among laboratory technicians; risk estimates use the US EPA method.

108. 108
     Kangarlou MB, Fatemi F, Dehdashti A, et al. Occupational health risk assessment of airborne formaldehyde in medical laboratories.Environ Sci Pollut Res Int · 2023 · 30(17):50392-50401doi:10.1007/s11356-023-25523-6 · PMID 36795204

     30 employees of one Iranian university's hospital laboratories; personal exposure 0.0156 to 0.594 ppm, mean 0.195 ppm; one institution.

109. 109
     McPhail LA, Whitelaw JL, Apthorpe L. Evaluation and control of formaldehyde exposures in a modern university anatomy laboratory.J Occup Environ Hyg · 2026 · 23(7):473-484doi:10.1080/15459624.2026.2657305 · PMID 42214085

     14 personal samples from three technicians; small, but with measured 8-hour averages (geometric mean 1.1 ppm) whose 95th percentile exceeded the Australian workplace limit.

110. 110
     Taylor L, Nelson RA, Geremew TT, et al. Addressing Global Disparities in Breast Cancer by Transforming Surgical Pathology in Ethiopia: Early Implementation of Immunohistochemistry.JCO Glob Oncol · 2025 · 11:e2500178doi:10.1200/GO-25-00178 · PMID 41397198

     125 specimens from 21 facilities, 2021 to 2024; an implementation report, not an accuracy study.