MBT Compass HT CA Software; MBT FAST Shuttle US IVD

The MBT Compass HT CA Software and MBT FAST Shuttle US IVD are components of the MALDI Biotyper CA System, a mass spectrometry platform for identifying bacteria and yeast from cultured clinical samples. The system uses matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) to generate protein mass spectra that are compared against a reference library to aid in diagnosis of bacterial and yeast infections, including from blood cultures when used with the optional MBT Sepsityper kit.

The MBT Compass HT CA software adds parallel processing capability to reduce analysis time, user role-based access control, 21 CFR Part 11 compliance features, and an IDealTune auto-tuning function for mass spectrometer maintenance. The MBT FAST Shuttle US IVD is an optional hardware tool that dries samples at controlled temperature (35±1°C) rather than room temperature, reducing drying time. The software also permits declaring low-confidence results as final without mandatory further sample preparation, whereas the predicate required advancing to the next preparation method.

IEC 61326-1, IEC 61326-2-6, IEC 60601-1-2 (electromagnetic compatibility); IEC 61010-1, IEC 61010-2-010, IEC 61010-2-101 (electrical and safety standards); UL and CAN/CSA standards for US and Canadian markets. FDA guidance documents on EMC, device software functions (2023), and cybersecurity in medical devices (2023) were also applied.

Substantial equivalence is established because the device differs from the predicate only in adding two optional components (MBT Compass HT CA software and MBT FAST Shuttle hardware) without changing the fundamental inputs, spectra analysis algorithm, reference library, or diagnostic outputs. Extensive repeatability and reproducibility testing across multiple sites and operators demonstrated that MBT FAST Shuttle drying produces equivalent identification results and log(scores) to room-temperature air-drying. Retrospective validation on 15,270 clinical spectra showed only 0.55% discordance in low-confidence identifications versus extended extraction, with no incorrect identifications. Software verification and validation confirmed the new features do not alter diagnostic functionality. These data establish that the additions maintain safety and effectiveness equivalent to the predicate device.