Comprehensive Evaluation of Compendial USP<71>, BacT/ALERT Dual-T, and BACTEC FX for the Detection of Product Sterility Testing Contaminants.

Emergence of cell therapy programs in large academic centers has led to an increasing demand for clinical laboratories to assist with product sterility testing. Automated blood culture systems have shown promise as alternatives to the manual USP<71> compendial method, but current published data are limited by small organism test sets, particularly for molds. In 2015, failure of the BACTEC FX system to detect mold contamination in two products prompted us to evaluate three test systems (compendial USP<71>, BACTEC FX, and BacT/ALERT Dual-T) over seven different culture combinations using 118 challenge organisms representative of the NIH cGMP environment. At <96 hours and <144 hours for bacterial and fungal detection, respectively, the compendial USP<71> method significantly outperformed BACTEC FX (84.7% vs 64.4%; p=0.0006), but not the BacT/ALERT at 32.5°C (78.8%; p=0.3116). Extended incubation to 360 hours with terminal visual inspection improved sensitivity, without significant difference between compendial USP<71> and BacT/ALERT (95.7% vs 89.0%, p=0.0860); both systems were better than the BACTEC FX (71.2%; p<0.0001 and p=0.0003, respectively). The BACTEC FX and BacT/ALERT performed equivalently for 30 isolates derived from clinical bloodstream infections, confirming system optimization for clinical organisms rather than environmental contaminants. Paired Sabouraud Dextrose Agar (SDA) plates were always positive for fungi within the acceptable time frame. This study shows that the BACTEC FX is suboptimal for product sterility testing, and provides strong data to support the use of BacT/ALERT at 32.5°C paired with a supplemental SDA plate as an acceptable alternative to compendial USP<71> for product sterility testing.