Impact of clinical sample processing time on the performance and T-cell response of the T-SPOT.TB assay.

OBJECTIVES: T-SPOT.TB, an interferon-gamma release assay (IGRA) for detecting Mycobacterium tuberculosis infection, requires blood processing within 4 hours, limiting clinical flexibility. T-Cell Xtend™ (TCX) extends permissible storage but is costly. This study evaluated delayed processing impacts on T-SPOT.TB consistency in a real-world setting. METHODS: Outpatients undergoing T-SPOT.TB testing were prospectively enrolled. Samples were processed at standard timing (≤4 h), after 24-h delay with TCX, and after 24-h delay without TCX. Consecutive-day fresh samples served as within-subject reproducibility references. RESULTS: Among 443 patients, 282 underwent 24-h delayed testing. Agreement with standard-timing results was 93.3% without TCX (κ=0.859) and 93.0% with TCX (κ=0.860), comparable to fresh-sample reproducibility (95.0%; κ=0.840; all p<0.001). Multivariable analysis identified initial borderline results (OR = 15.21, p<0.001) and 24-h delay (p<0.05) as independent predictors of T-SPOT.TB reversion, regardless of TCX use. Antigen-well spot-forming cell (SFC) counts were significantly reduced after 24-h delay (p<0.001); TCX did not mitigate this decline (p=1.000). Positive-control SFC counts remained stable (p=0.602). CONCLUSIONS: A 24-h processing delay demonstrated high concordance with standard-timing T-SPOT.TB results, supporting clinical feasibility. Nevertheless, borderline results and delayed processing were independently associated with result discordance, warranting cautious clinical interpretation.