Pooling sputum samples for the Xpert MTB/RIF Ultra assay: A sensitive and effective screening strategy

Area of Science:

• Microbiology
• Infectious Diseases
• Molecular Diagnostics

Background:

• The standard Xpert MTB/RIF assay has limitations in sensitivity for diagnosing paucibacillary tuberculosis.
• Pooled testing strategies are being explored to improve efficiency and cost-effectiveness in tuberculosis diagnostics.

Purpose of the Study:

• To evaluate the sensitivity and effectiveness of pooled Xpert MTB/RIF Ultra (Ultra) testing for mass tuberculosis screening.
• To determine optimal pool sizes for Ultra pooled testing across various tuberculosis prevalence rates.

Main Methods:

• Frozen sputum samples, previously tested by individual Xpert assays, were pooled in groups of 4, 8, and 16.
• Pooled samples were tested using the Xpert MTB/RIF Ultra assay, with each pool containing a single positive sample at varying bacterial loads.
• Simulations of TB screening scenarios were conducted at prevalence ranges of 0.2-1.0% to calculate cartridge efficiency.

Main Results:

• The overall sensitivity of Ultra pooled testing was high, reported at 88.9% (95% CI: 75.9-96.3).
• Sensitivity was higher when positive samples had greater mycobacterial loads.
• A pool size of 8 demonstrated at least 80% cartridge savings, even at the highest simulated prevalence.

Conclusions:

• Sputum pooling with Xpert MTB/RIF Ultra is a sensitive and effective strategy for tuberculosis screening.
• Despite its effectiveness, widespread screening in resource-limited settings may necessitate the development of novel, lower-cost, high-throughput tools, potentially utilizing non-sputum specimens.