(30 days)
The Bio-Rad VARIANT™nbs Sickle Cell Program is intended as a qualitative screen for the presence of hemoglobins F, A, S, D, C, and E in eluates of neonatal blood collected on filter paper by high-performance liquid chromatography (HPLC).
The Bio-Rad VARIANT™nbs Sickle Cell Program is intended for Professional Use Only. For In Vitro Diagnostic Use.
The Bio-Rad VARIANT™nbs Sickle Cell Program is for use only with the Bio-Rad VARIANT™nbs Newborn Screening System.
The presence of hemoglobin S (HbS) in a patient blood sample is indicative of sickle cell disease or sickle cell trait. Diagnosis of sickle cell disease prior to the age of four months allows for the administration of a prophylactic treatment with penicillin. Prophylactic treatment with penicillin has shown to decrease morbidity and mortality.
The VARIANTnbs Newborn Screening System uses the principles of high-performance liquid chromatography (HPLC). The VARIANTnbs Sickle Cell Program is based on the chromatographic separation of hemoglobins F, A, S, D, C, and E on a cation exchange cartridge.
The new feature in this submission is the upgrade of the Genetic Data Management (GDM) software. The current software (GDM 2.01) requires Microsoft Windows NT. This product is nearing the end of its lifecycle. GDM 3.0 software is needed to transfer the GDM software to the Microsoft Windows XP Operating System.
Acceptance Criteria and Study Details for Bio-Rad VARIANT™nbs Sickle Cell Program with GDM 3.0
This document outlines the acceptance criteria and the study conducted to demonstrate that the Bio-Rad VARIANT™nbs Sickle Cell Program run on the VARIANT™nbs Newborn Screening System using Genetic Data Management (GDM) 3.0 Software meets these criteria.
1. Table of Acceptance Criteria and Reported Device Performance
The primary acceptance criteria for the new GDM 3.0 software update was to maintain 100% agreement with the predicate device (GDM 2.01) in identifying the presence of specific hemoglobins.
| Acceptance Criteria | Reported Device Performance (GDM 3.0 vs. GDM 2.01) |
|---|---|
| 100% agreement for Hemoglobin | 100% agreement |
| Type FA | 100% agreement |
| 100% agreement for Hemoglobin | 100% agreement |
| Type FAS | 100% agreement |
| 100% agreement for Hemoglobin | 100% agreement |
| Type FAC | 100% agreement |
| 100% agreement for Hemoglobin | 100% agreement |
| Type FAD | 100% agreement |
| 100% agreement for Hemoglobin | 100% agreement |
| Type H | Not explicitly listed as "H" in the provided table, but |
| 100% agreement for all other | the overall "Total" agreement is 100%. |
| specific hemoglobin types | |
| (implied by overall "Total") |
2. Sample Size Used for the Test Set and Data Provenance
- Sample Size: The provided document does not explicitly state the total number of patient samples used in the correlation study. It lists categories of hemoglobin types (FA, FAS, FAC, FAD, and an unspecified "H" type) and indicates 100% agreement for each. While a total count of samples is not given, the implicit message is that all samples tested within these categories showed agreement.
- Data Provenance: The study was conducted "at an external site". The country of origin is not specified but is presumably the United States, given the FDA submission. The study compared results from the predicate device and the new device ran on the same samples on the same day, suggesting it was a prospective comparison of the new software's performance on existing samples, rather than entirely retrospective data analysis.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts
The document describes a "method correlation study" where samples run on the predicate device (GDM 2.01) were repeated on the GDM 3.0 platform. The "ground truth" in this context is established by the results obtained from the predicate device (GDM 2.01). There is no mention of external human experts establishing a separate ground truth for this correlation study, as the comparison is specifically between the two software versions and their agreement.
4. Adjudication Method for the Test Set
No explicit adjudication method is described. The study directly compares the results of the two software versions. Since the goal was to demonstrate 100% agreement, any discrepancy would inherently require investigation. However, with 100% agreement reported, no further adjudication process for conflicting results was needed or mentioned.
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done, and the Effect Size of How Much Human Readers Improve with AI vs. Without AI Assistance
No, a Multi-Reader Multi-Case (MRMC) comparative effectiveness study was not done. This device is an in vitro diagnostic system for qualitative screening of hemoglobins using HPLC and software for data management, not an AI-assisted diagnostic imaging or interpretation tool for human readers. The change is an upgrade to the operating system and features of the data management software.
6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done
The study described is a standalone performance comparison between the GDM 3.0 software and the predicate GDM 2.01 software. It assesses the algorithm's (software's) ability to produce the same qualitative results as the older version when processing the same samples. While human operators are involved in running the HPLC system, the software's output is the focus of the comparison, making it a standalone assessment of the software's correlation with its predecessor.
7. The Type of Ground Truth Used
The ground truth used for this study was the results generated by the predicate device (VARIANT™nbs Sickle Cell Program with GDM 2.01). The objective was to demonstrate that the new GDM 3.0 software produced identical qualitative results to the previously cleared and established predicate device.
8. The Sample Size for the Training Set
The provided summary does not mention a training set in the typical sense for machine learning or AI models. The GDM 3.0 is a software upgrade to an existing HPLC system, primarily focusing on operating system compatibility and usability enhancements, not a new algorithm requiring a training phase from scratch. Therefore, the concept of a "training set" for the algorithm itself is not directly applicable in this context. The development process would have involved internal testing and validation, but this information is not part of the public summary provided.
9. How the Ground Truth for the Training Set Was Established
As explained above, a dedicated "training set" for an AI algorithm is not applicable to this software upgrade. The "ground truth" for ensuring the functionality of the GDM 3.0 software during its development would have been established through rigorous software testing against known sample profiles and expected outputs, confirming its accuracy in processing the chromatographic data according to the established scientific principles of HPLC for hemoglobin separation. This typically involves using reference materials and clinically characterized samples.
§ 864.7415 Abnormal hemoglobin assay.
(a)
Identification. An abnormal hemoglobin assay is a device consisting of the reagents, apparatus, instrumentation, and controls necessary to isolate and identify abnormal genetically determined hemoglobin types.(b)
Classification. Class II (special controls). A control intended for use with an abnormal hemoglobin assay is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to the limitations in § 864.9.