38 Validation of an Automated Blood Culture Detection System for Sterility Testing of Cellular and Gene Therapy Products

Background Sterility testing of cellular and gene therapy products, such as CAR-T therapy for the treatment of certain hematologic malignancies, is required to ensure cellular product safety prior to infusion in the recipient. The compendial USP <71> sterility testing method (agar-based) is the gold standard for testing stem cell transplant products and is the basis for testing more sophisticated cell therapy products. Automated blood culture systems have emerged as a potential alternative sterility testing method because they offer automated continuous growth detection monitoring, require minimal handling to minimize contamination, and are readily available in most clinical laboratories. Published literature regarding this application is limited. Here, we evaluated the BacT/Alert Virtuo (bioMérieux) automated blood culture detection system and its ability to reliably detect diverse pathogens, ranging from bacteria to mold, in two standard cellular therapy formulations. Methods Six lyophilized organisms (Staphylococcus aureus, Bacillus subtilis, Pseudomonas aeruginosa, Clostridium sporogenes, Candida ­albicans, and Aspergillus brasiliensis; specified in compendial USP <71>) were reconstituted according to the manufacturer specifications and combined with one of two standard cellular therapy product formulations (Plasmalyte-A + 5% HSA and Plasmalyte-A + 5% HSA + 5% DMSO), plus a negative control formulation (saline). Bottles were prepared in sets/pairs (aerobic BacT/Alert FA Plus and anaerobic FN Plus), inoculated with 50 colony forming units (CFUs) per bottle, and incubated on the Virtuo system for a maximum of 14 days. Bottles flagging positive underwent staining (Gram and/or calcofluor) and subculturing to solid agar media for standard-of-care microbial identification using MALDI-TOF MS (bacteria and yeast) or macroscopic and microscopic identification (mold). Results The automated detection system correctly flagged all contrived culture sets (18/18) within 72 hours. There was one set each of B. subtilis and P. aeruginosa which only the aerobic bottle flagged positive, although both organisms are known to preferentially grow aerobically. Of note, Ralstonia was unexpectedly co-detected in one set of C. sporogenes, C. albicans, and saline-only negative control. Additional testing confirmed the saline control was contaminated with Ralstonia. Lastly, bottles inoculated with mold correctly flagged positive on the Virtuo but staining and agar growth remained negative. These no-organism-seen (NOS) bottles were reloaded on the Virtuo and flagged positive again <24 hours, from which, septate hyphae were observed by calcofluor staining and mold was recovered and identified. Conclusion The Virtuo system with FA Plus and FN Plus bottles can successfully detect the presence of organisms in two standard cellular therapy formulations. Importantly, although mold contamination was adequately detected by the Virtuo, mold recovery and identification may require subculture optimization. In summary, our findings support that automated blood culture detection systems are a reliable alternative method of sterility testing for cellular and gene therapy products.