NMPA Reviewer Q&A 2021/9/6


2021-09-07

1.If the registrant claims that a new clinical function of a medical device with software is a minor change, is it possible to ignore such change in the release version?

Definitely not. Changes in clinical functions shall be major updates of software. If the naming rules for software version number specified by the manufacturer cannot clearly distinguish major and minor software updates, software release version number shall follow its higher risk. The software release version in China shall reflect such changes.

For example, the naming rule of software version number developed by the manufacturer is X.Y.Z, of which X represents major update, Y represents minor update and Z represents correction. If the change of clinical function is reflected in the Y field, the software release version number shall be determined as X.Y.

2.What should be paid attention to in the clinical trials of in vitro diagnostic reagent?

In clinical trials, when clinical samples of the type specified in the instructions for use (IFU) are tested, attention shall be paid to the conformity of technical contents such as sample collection, sample storage conditions, sample storage time, sample processing methods, etc. The claims in the test reagent IFU shall be supported by preclinical studies, and attention shall be paid to the requirements in the IFU of test reagent, reference reagent and recheck reagent.

For example, clinical trials of nucleic acid detection reagents should note that:

Whether the samples can be reused for multiple times during the evaluation of analytical performances of in vitro diagnostic reagents?

The analytical performance of in vitro diagnostic reagents includes multiple contents such as accuracy, precision, limit of detection and specificity. It is necessary to include the samples from different sources, types, concentrations and other characteristics according to the specific requirements for various performances, so as to fully evaluate various performances of product. Therefore, the re-use of samples shall be avoided as far as possible in the evaluation process of analytical performances.