Sources of Error and False Results in Serological Testing

The physical and chemical state of forensic samples at the time of testing is the single greatest source of false negative results in serological analysis. Unlike clinical samples collected under controlled conditions, forensic stains are deposited on surfaces with unpredictable absorbency, exposed to environmental variables for unknown periods, and often of uncertain age when submitted to the laboratory.

Protein degradation is the primary mechanism. Serological targets, including blood group antigens, seminal plasma proteins, and salivary amylase, are all proteins or glycoproteins. Denaturation destroys the three-dimensional epitope structure that antibodies recognise, even if the primary amino-acid sequence remains intact. Elevated temperature accelerates denaturation: a bloodstain on dark clothing left in a car on a hot day may show markedly reduced immunoreactivity within hours.

Ultraviolet exposure cleaves peptide bonds and generates oxidative damage. Microbial proteases from bacterial growth on moist stains digest protein epitopes directly. The practical consequence is that a negative serological result from an aged or environmentally compromised sample cannot be interpreted as confirming the absence of the fluid: it may simply reflect destruction of the target.

Inhibitors are substances that are present in the sample or its substrate and interfere with the assay chemistry without necessarily destroying the analyte. Haem and haematin, released by red cell lysis, inhibit many enzyme-linked steps in ELISA by competing with the horseradish peroxidase reporter enzyme.

Melanin from dark fabric dyes binds antibody non-specifically and reduces effective antibody concentration. Indigo dye from denim is a well-documented inhibitor of both PCR and immunoassays. Tannic acid from leather substrates precipitates proteins. Bile salts, relevant in samples from clothing soaked in gastrointestinal fluid, disrupt hydrophobic interactions that stabilise antibody-antigen binding. The standard approach to inhibitor detection is to run the sample extract at multiple dilutions: if the result becomes positive at higher dilution, the concentrated sample contains an inhibitor.

The hook effect is an assay-design artefact specific to sandwich immunoassay formats, including two-site ELISA and lateral-flow immunoassays. In a correctly functioning sandwich assay, the capture antibody immobilised on the solid phase binds one epitope of the target antigen, and the detection antibody (conjugated to a reporter) binds a second, non-overlapping epitope. The signal is proportional to the amount of antigen captured in this three-component complex.

When antigen concentration greatly exceeds the binding capacity of both antibodies, each antigen molecule occupies either a capture site or a detection antibody molecule but not both simultaneously. Fewer complete sandwiches form. Signal generation falls.

At extreme antigen excess, the signal can drop below the assay's cut-off, and a strongly positive sample reads as negative. This is the hook. The name comes from the shape of the dose-response curve when plotted across a wide concentration range: signal rises with increasing antigen, reaches a maximum, then falls back, producing a curve that resembles a hook at the high-concentration end.

In forensic serology, the hook effect is most relevant in tests for prostate-specific antigen (PSA) used to identify seminal fluid, and in tests for human chorionic gonadotropin (hCG) used to detect foetal or pregnancy-related material.

A concentrated undiluted extract from a semen stain may trigger the hook effect in a PSA lateral-flow test. The detection protocol should include parallel testing of an undiluted and a diluted extract. If the diluted extract produces a stronger positive than the undiluted one, the hook effect is confirmed.

Every antibody in forensic use was raised against a defined antigen, but no antibody is perfectly monospecific. Cross-reactivity arises when a molecule other than the intended target shares enough structural features with the immunising antigen that the antibody binds it with measurable affinity. The signal produced may be indistinguishable from a true positive unless orthogonal methods are applied.

In species identification using precipitin tests, antisera raised against human serum albumin or human IgG can cross-react with serum proteins from non-human primates. The degree of cross-reactivity reflects phylogenetic distance: chimpanzee and gorilla proteins share more than 98% sequence identity with their human counterparts, and a precipitin test using a poorly characterised anti-human antiserum may give a visible precipitin line with primate blood.

Cases involving wildlife crime, where the question is whether a stain is human or animal, require antisera with documented cross-reactivity profiles across the relevant species, not just human positive and blank negative controls.

In body-fluid identification, lateral-flow assays for salivary amylase (alpha-amylase) have documented cross-reactivity with pancreatic amylase. A swab from a surface contaminated with vomit or faecal material may test positive in an assay designed to detect saliva, because both salivary and pancreatic amylase share the same enzyme but are encoded by different genes (AMY1 vs AMY2).

The ABAcard HemaTrace, widely used for human bloodstain confirmation, shows cross-reactivity with ferret and some primate blood. Manufacturers' cross-reactivity tables are a mandatory starting point for any new forensic application of a lateral-flow device. That profile is why confirmatory immunological testing for human blood is reported as species-specific rather than human-exclusive, since the anti-human haemoglobin antibody immobilised on the HemaTrace membrane also binds haemoglobin from a small set of other species.

Every quantitative or semi-quantitative immunoassay requires a threshold, the cut-off, below which a result is reported as negative and above which it is reported as positive. The placement of this cut-off is a deliberate design decision that determines the assay's sensitivity and specificity, and therefore its false negative and false positive rates.

There is no cut-off that simultaneously minimises both. Lowering the cut-off increases sensitivity (fewer missed true positives) but increases the false positive rate. Raising it increases specificity but increases the false negative rate.

In clinical diagnostics, cut-offs are set by receiver operating characteristic (ROC) analysis across reference populations of known-positive and known-negative samples. Forensic applications use the same mathematical approach but face a different population: the reference samples must include degraded, inhibited, and contaminated specimens that reflect casework conditions, not fresh clinical specimens.

A cut-off validated on clean laboratory reference samples will have a different sensitivity in degraded forensic samples. This is why validation studies for forensic serological methods must include samples prepared to mimic realistic crime scene conditions. Those studies are what establish the assay's analytical sensitivity and specificity in the first place, using blank replicates to fix the limit of detection and challenge panels to document cross-reactivity.

Non-specific binding is the primary driver of background signal in ELISA. Proteins in the sample adsorb to the polystyrene microplate surface or to the detection antibody by hydrophobic and electrostatic interactions that are not antigen-antibody specific. Blocking steps using bovine serum albumin (BSA), casein, or non-fat dried milk are designed to saturate non-specific binding sites before the assay antibody is added.

Haem compounds from blood samples are particularly problematic: they bind non-specifically and carry intrinsic peroxidase activity that generates a coloured product in the TMB substrate step even without any secondary antibody present. This haem-driven background must be controlled by including a no-antibody blank well for each sample, or by using alternative substrates less susceptible to pseudoperoxidase interference.

Human factors account for a substantial proportion of forensic serological errors. Studies of laboratory quality failures across forensic disciplines consistently find that pipetting error, incorrect reagent preparation, mislabelling, and failure to follow written protocols are more common sources of incorrect results than assay chemistry failures. The serological laboratory must treat operator error as a system design problem, not an individual failure, and build controls that detect procedural deviations before results are reported.

Standard controls placed in every assay run serve as the primary detection mechanism. A positive control sample of known concentration confirms that the assay chemistry is functioning. A negative control confirms the absence of contamination and non-specific signal at the background level. A reagent blank (buffer only, no sample) distinguishes substrate autofluorescence or pseudoperoxidase signal from sample-derived signal. If any control produces an out-of-range result, the entire run is invalidated and must be repeated before casework results are released.

Contamination is a specific class of procedural failure that introduces foreign biological material into the sample or introduces sample material into a negative control. In serological testing, the most consequential form is secondary transfer of biological material between evidence items during handling, which can generate genuine but misleading biological signals on a substrate that was not present at the crime scene.

Secondary transfer is documented in literature on blood pattern analysis and touch-DNA studies. The serological implication is that a confirmed positive result on an exhibit does not prove that biological material was deposited at the crime scene: it proves only that it was present at the time of testing.

Documentation and chain of custody records provide the evidentiary framework for distinguishing pre-analytical contamination (which occurred before the sample reached the laboratory) from intra-laboratory contamination (introduced during processing). The two have different legal consequences and different remedies. Intra-laboratory contamination is the laboratory's responsibility; pre-analytical contamination is an investigative and collection issue. Both must be documented clearly in the case report.

No single serological test is sufficient for a forensic conclusion. Every major forensic serology standard, from the SWGMAT Body Fluid Identification guidelines in the US to the ENFSI body fluid best practice manual and the standards used under India's NABL scheme, requires that a presumptive positive result from a screening assay be confirmed by at least one additional test using a different analytical principle before a conclusion is reported as confirmed.

This two-tier confirmatory hierarchy catches errors that arise within any single assay format, because the mechanisms of false results in different assay types are largely independent.

For bloodstain identification, a typical hierarchy is: catalytic colour test (Kastle-Meyer or leuco-malachite green) as a presumptive screen, followed by lateral-flow HemaTrace or ELISA for human haemoglobin as confirmation.

Kastle-Meyer is inhibited by strong oxidising agents and can be false-positive from plant peroxidases. HemaTrace is specific to human/primate haemoglobin and targets a different protein property. A result positive on both, from a correctly controlled run, is substantially more reliable than either alone. For semen, the microscopic identification of spermatozoa or seminal vesicle cells provides confirmation by a completely different analytical principle (morphology) independent of immunochemistry.

External proficiency testing is the systematic check on whether a laboratory's error rates in practice match its validated performance specifications. Programmes run by the Collaborative Testing Services (CTS), the American Society of Crime Laboratory Directors (ASCLD), and equivalent bodies in the UK and EU send known samples with expected results to participating laboratories.

The laboratory tests them blind and returns its findings. Discordant results trigger a root-cause investigation: the laboratory must identify whether the failure arose from a chemistry, equipment, or personnel factor and document corrective action. Persistent discordant results in proficiency testing are a regulatory trigger for accreditation review.