By: Derrick MoralesPosted on

Why SHBG ELISA standardization is hard

Sex hormone-binding globulin (SHBG) is a liver-derived glycoprotein that binds and transports sex steroids in blood and modulates their distribution between free, albumin-bound, and SHBG-bound pools. Because SHBG levels and binding equilibria influence calculations of free steroid fractions, between-assay bias in SHBG directly propagates into free testosterone/estradiol estimates and study endpoints. Multiple studies show that different SHBG immunoassays yield materially different results, affecting calculated free testosterone and correlations with metabolic traits. CNIB+2PubMed+2

Two structural/biophysical facts amplify the problem:

  1. Most circulating testosterone/estradiol is protein-bound—principally to SHBG and albumin—so any analytical interference with either binding partner can shift apparent equilibria. See concise overviews in the NIH Bookshelf (free > albumin > SHBG fractions; transport roles). CNIB+2CNIB+2

  2. Albumin binds many steroids with low affinity but massive capacity; subtle matrix effects (fatty acids, drugs) can alter albumin’s binding pockets and the speciation of steroids in samples, which in turn can influence apparent immunoreactivity and cross-reactivity profiles. PMC+1

AffiELISA® Human Sex hormone-binding globulin ELISA [ SHBG]

Variability across commercial SHBG ELISA kits: what actually differs?

Capture antibody specificity: monoclonal vs. polyclonal

  • Monoclonal capture antibodies target a single epitope on SHBG. Pros: lot-to-lot consistency and lower off-target binding. Cons: epitope masking by glycosylation variants or ligand occupancy; miss certain isoforms or genetic variants if the epitope is altered. (Genetic variation in SHBG can affect ligand affinity and potentially antigenic presentation.) PubMed

  • Polyclonal capture antibodies recognize multiple epitopes. Pros: broader epitope coverage and higher tolerance to structural microheterogeneity. Cons: greater lot-to-lot drift and risk of low-level cross-reactivity with homologous proteins.

  • Historical ELISA designs for SHBG reported low cross-reactivity with other serum proteins, but performance depends on the exact antibody clone/pool and blocking chemistry. Always verify current kit claims with in-house cross-reactivity panels. PubMed

What to look for in datasheets and validations (and reproduce locally):

  • Epitope mapping or blocking peptide data (monoclonals) and breadth of immunogens (polyclonals).

  • Parallelism across serially diluted serum/CSF/plasma matrices (demonstrates that the sample matrix “behaves” like the calibrator). Guidance on parallelism and other LBA characteristics is detailed in FDA M10 and the FDA Bioanalytical Method Validation documents. U.S. Food and Drug Administration+1

Sandwich format and detection chemistry

  • Two-site (“sandwich”) ELISA (capture + detection) reduces interference compared with competitive formats, but only if the two antibodies recognize non-overlapping epitopes unaffected by steroid occupancy.

  • Biotin/streptavidin systems: evaluate interference from high-dose biotin (common supplement), as recommended by general LBA guidance and good laboratory practice. U.S. Food and Drug Administration+1

Cross-reactivity with albumin-bound steroids: the under-discussed source of bias

Why it matters: In native serum, SHBG is partially occupied by steroids; albumin also carries steroids at high capacity. During sample prep, dilution buffers, surfactants, and organic components can shift the equilibrium between SHBG-bound and albumin-bound fractions. That can:

  • Alter SHBG conformations and epitope exposure (epitope masking/unmasking).

  • Promote dissociation or redistribution of steroids to albumin, changing how the SHBG–steroid complex presents to antibodies.

  • Create apparent cross-reactivity if detection antibodies or blocking agents bind transient steroid-protein complexes.

Key references on distribution/binding that underpin these effects:

  • Overview of testosterone partitioning among free, albumin-bound, and SHBG-bound pools (NIH/NCBI). CNIB

  • Structural/biophysical detail on albumin–testosterone binding (NIH/PMC). PMC

  • Albumin’s broad steroid-binding behavior and buffering capacity (NIH/PMC). PMC

Practical controls to add to your ELISA verification:

  • Spike-recovery with physiologic concentrations of albumin and defined steroid loads (testosterone/estradiol) to mimic native binding states; confirm parallelism vs. calibrator.

  • Test high-fatty-acid sera or add fatty acid supplements to assess allosteric effects on albumin pockets (documented for albumin). PMC

  • Compare results to LC-MS/MS SHBG or steroid panels where possible to decouple immunoreactivity from ligand occupancy artifacts; NIST describes commutability considerations when aligning ligand-binding assays with MS-based results. NIST+1

Image générée

Calibrators and higher-order traceability

Ideal state: Calibrators are traceable to higher-order reference materials and are commutable (behave like native patient samples across methods). The World Health Organization (WHO) provides international standards (IS) and guidance for establishing secondary standards calibrated in IU against IS where appropriate; while there may not be a current WHO IS specifically for SHBG, the principles for secondary standardization and commutability remain directly applicable. Organisation Mondiale de la Santé+1

What labs should require from manufacturers (and verify locally):

  • A clear calibration hierarchy (value assignment procedure, uncertainty) and evidence that kit calibrators are commutable with native serum across at least two orthogonal methods (e.g., different ELISA brands and, if available, an MS-based workflow). See NIST’s definitions/criteria and quantitative approaches to commutability assessment. NIST Publications+2tsapps.nist.gov+2

  • Participation data from recognized standardization efforts (e.g., the CDC Hormone Standardization Programs for steroid hormones—while focused on testosterone/estradiol, they illustrate how ongoing, year-long bias/imprecision assessment improves comparability). Use their documentation as a blueprint for SHBG harmonization initiatives. CDC+2CDC+2

Downstream impact: clinical and epidemiological studies

  • Calculated free testosterone (and related indices such as FAI) depend on SHBG concentration and albumin assumptions; assay bias in SHBG is a first-order driver of error in free-hormone estimates, particularly at high SHBG. Multiple NIH-indexed studies show degraded agreement with equilibrium dialysis when SHBG is elevated or when albumin deviates from fixed assumptions. PubMed+1

  • Population analyses linking SHBG to metabolic syndrome and cardiometabolic risk can be directionally or quantitatively altered by between-assay bias; some correlations appear/disappear depending on assay family. PMC+1

  • Age- and cohort-based trends in SHBG are well documented; if cohorts were measured with different kits or at different times (lot/manufacturer changes), meta-analyses require harmonization layers to avoid spurious conclusions. CNIB

Normalization strategies across laboratories

  1. Common reference panels. Exchange panels of 80–200 native sera spanning physiologic/pathophysiologic SHBG ranges. Include subsets with low albumin, high fatty acids, and known steroid therapy to probe matrix effects. Use robust methods (e.g., Passing–Bablok, Deming) to derive bridging equations between sites/kits. Model bias as a function of concentration and matrix covariates.

  2. Anchor to an orthogonal method. Where feasible, generate an anchor set quantified by a well-characterized LC-MS/MS workflow or by a reference immunoassay with demonstrated commutability across panels (follow NIST criteria for selecting SRMs/EQMs). Use the anchor to perform value transfer to each ELISA. NIST+1

  3. Calibrator value reassignment. If calibrators are non-commutable, re-assign values using pooled human sera with characterization along a calibration hierarchy (WHO secondary standardization principles). Maintain documentation of uncertainties and re-assignment lots. Organisation Mondiale de la Santé+1

  4. Longitudinal QC with external comparators. Run external controls from an independent source each batch; trend bias versus an anchor laboratory. The CDC HoSt framework (even if SHBG isn’t certifiable there) is a model for year-long bias/imprecision surveillance. CDC

  5. Statistical harmonization in analyses. In multi-center datasets, adjust for assay-center as a random effect; apply post-hoc calibration using overlap samples; propagate measurement error into models of free hormones and outcomes (errors-in-variables). (NIH funding announcements for assay validation in multicenter trials outline performance characteristics to document—use these as checklists.) Grants NIH+2Grants NIH+2

Recommendations for assay validation in multi-center SHBG studies

Design controls (pre-analytical):

  • Standardize matrix (serum vs. plasma type), collection tubes, processing time, and freeze–thaw limits across centers.

  • Perform stability studies (short-term, long-term, freeze–thaw) in native serum per FDA guidance; document acceptance criteria and uncertainty budgets. U.S. Food and Drug Administration+1

Analytical validation (per center and cross-center):

  • Accuracy/Trueness & Bias: Compare to anchor method across 60–100 specimens; assess concentration-dependent bias.

  • Precision: Within-run, between-run, between-lot; include reagent lot switch studies.

  • Selectivity/Specificity: Spike common steroids (testosterone, estradiol) and albumin to probe cross-reactivity and epitope masking.

  • Parallelism: Serial dilution of native specimens to test matrix commutability with the calibrator.

  • Reportable range & Linearity: Follow FDA M10 and Bioanalytical Method Validation recommendations for LOD/LOQ, hook effects, and linearity. U.S. Food and Drug Administration+1

Commutability & calibration hierarchy:

  • Demonstrate that calibrators/controls are commutable across platforms; quantify maximum allowable non-commutability bias (MANCB) as recommended by NIST; document uncertainty contributions from calibrator value assignment. tsapps.nist.gov

  • If no higher-order material exists for SHBG, implement a secondary standardization scheme against a de facto reference (well-characterized pooled serum with traceable value assignment) consistent with WHO guidance for secondary standards. Organisation Mondiale de la Santé+1

Ongoing performance monitoring:

  • Inter-site blind replicates (≥5% of total sample load) and EQC rounds each quarter.

  • Change control for kit lots/instrument software; re-bridge to anchor after any change.

  • Archive aliquots for retrospective harmonization if new anchors or commutable materials become available (see NIST program notes on moving toward fresh-frozen, more-commutable matrices). NIST

Image générée

Practical kit-selection checklist (what to ask vendors and verify in-house)

  • Antibody architecture: monoclonal vs. polyclonal; epitope mapping; evidence of tolerance to ligand occupancy.

  • Cross-reactivity panel: tested against albumin-bound steroid conditions; interference from high biotin.

  • Calibrator design: human-matrix based, documented commutability and value assignment; uncertainty statements.

  • Parallelism and matrix effects: internal data on serum/plasma types and lipemic/hemolyzed/icteric specimens.

  • Bridging data: method-comparison vs. at least one external kit and (ideally) an MS anchor.

  • Participation in standardization efforts: any external performance assessments modeled on CDC HoSt programs. CDC

Application angle: implications and mitigation

  • Clinical and epidemiological studies: When SHBG is an exposure, mediator, or covariate (e.g., in metabolic syndrome or aging cohorts), differences between kits can change regression slopes and risk thresholds. Multi-center projects should prespecify assay harmonization steps, include overlap samples, and use hierarchical models with center/assay random effects. PMC

  • Free hormone estimates: Because errors in SHBG and albumin propagate non-linearly into calculated free testosterone, studies with many high-SHBG participants (older men, certain endocrine states) should consider equilibrium dialysis or LC-MS/MS free hormone measurements for a stratified subset to validate calculations. PubMed

Pinned resources (.gov/.edu deep links you can cite in SOPs)

Final word

Reliable SHBG quantification by ELISA is achievable, but it requires discipline around antibody design, matrix-aware interference testing, commutable calibration, and cross-site normalization. Treat calibrators and controls like critical reagents; prove commutability; and pin your network to an orthogonal anchor. With these controls in place, SHBG becomes a far more stable biomarker for translational research and population-scale studies—without compromising interpretability downstream. NIST Publications+2tsapps.nist.gov+2

No related posts.