Radioimmunoassay (RIA) — Quantitative Immunological Detection

RIA remains in active use, particularly in research settings, specialist endocrine laboratories, and for analytes where validated ELISA or chemiluminescence kits are not yet available across all veterinary species. For many routine hormones such as progesterone and cortisol, commercially validated ELISA and automated immunoassay platforms have displaced RIA in most clinical laboratories due to lower regulatory burden. However, RIA is still the reference method against which newer assays are validated, and it retains unique advantages in sensitivity and flexibility for novel analytes.

Iodine-125 (¹²⁵I) is the predominant isotope in modern veterinary RIA. It emits low-energy gamma radiation detectable by standard gamma counters, has a half-life of approximately 60 days (long enough for practical use but short enough to decay to safe levels for waste disposal), and can be conjugated to proteins, peptides, and haptens via established iodination chemistry. Tritium (³H) was used historically for steroids that cannot be easily iodinated without altering their immunoreactivity, but its detection requires liquid scintillation counting, which is more laborious.

Both methods are validated for bovine progesterone and give clinically concordant results when properly performed. ELISA is preferred in most current diagnostic laboratories because it avoids radioactive materials, has longer reagent shelf life, and is amenable to automation. RIA remains preferable when the highest analytical sensitivity is required, when you need to measure a steroid for which no validated ELISA exists in your target species, or when you are validating a new assay and need a gold-standard comparator. Cross-validate whichever method you use against published reference intervals derived with the same technique.

Every assay run should include: (1) a non-specific binding (NSB) control — tracer plus assay buffer but no antibody — to quantify background counts; (2) a zero standard (B0) — tracer plus antibody but no unlabelled antigen — to define maximum binding; (3) at least two or three in-house quality control pools of known concentration spanning the working range of the standard curve; and (4) replicate standards at multiple points on the curve. Intra-assay coefficient of variation should be below 10% and inter-assay CV below 15% for the method to be considered in statistical control.