Rouleau · Endocrine connections 2019 · Retrospective diagnostic method-comparison study · n=435 serum samples

Discordance between testosterone measurement methods in castrated prostate cancer patients.

Cited 9 times in the scientific literature.

Level 3 - non-randomized controlled study

Retrospective diagnostic method-comparison study comparing an index immunoassay against a gold-standard reference (LC-MS/MS).

PubMed 30673630 · doi:10.1530/EC-18-0476 · record verified 2026-08-29

What was done

A retrospective laboratory comparison evaluated 435 serum samples collected from castrated prostate cancer patients between April and September 2017. Total testosterone in identical serum samples was quantified using both liquid chromatography-tandem mass spectrometry (LC-MS/MS, reference standard) and an electrochemiluminescent immunoassay to evaluate immunoassay accuracy around the clinical castrate threshold (0.7 nM).

What was found

Mean testosterone concentrations were significantly higher via immunoassay than LC-MS/MS (0.672 ± 0.359 nM vs. 0.461 ± 0.541 nM; P < 0.0001). The apparent testosterone breakthrough rate (>0.7 nM) was 22.1% with immunoassay versus 13.1% with mass spectrometry (P < 0.0001). Among samples measured at ≥0.7 nM by immunoassay, 50% were <0.7 nM by LC-MS/MS. Conversely, 2.95% of samples measuring <0.7 nM by immunoassay were quantified at ≥0.7 nM by LC-MS/MS.

Why it matters

Standard immunoassays frequently overestimate testosterone in the castrate range, creating a substantial false-positive rate for castration failure that could trigger inappropriate changes in prostate cancer therapy without mass spectrometry confirmation.

Limits

The study was retrospective, evaluated individual serum samples rather than longitudinal patient courses, and included no clinical follow-up or oncologic outcome tracking. Patient demographics, specific androgen deprivation regimens, and testing blinding were not reported in the abstract.

Cited by