Diagnostic accuracy of a single-reaction, pan-lineage Lassa virus RT-PCR at two Nigerian laboratories: a retrospective and prospective cohort study

Abstract

Background

Lassa virus (LASV) intra- and inter-lineage nucleotide diversity has made it challenging to develop a single-reaction RT-PCR assay that can detect all viral variants with equal sensitivity. We evaluated PANDAA ^® LASV, a single-reaction pan-lineage RT-PCR assay, at two Nigerian Lassa fever laboratories against the RealStar Lassa Virus RT-PCR Kit 2.0, the current diagnostic standard in Nigeria, which requires two separate reactions per patient sample to mitigate against false negative results.

Methodology and Principal Findings

We conducted a diagnostic accuracy study using retrospectively ( n = 200) and prospectively ( n = 100) plasma samples that had been collected from patients meeting the standard case definition of suspected Lassa fever for testing by real-time PCR at Abubakar Tafawa Balewa University Teaching Hospital in Bauchi and Federal Medical Centre in Owo. PANDAA ^® LASV sensitivity was 98.6% (138/140; 95% CI 94.9–99.8%), specificity 99.4% (155/156; 96.5–100%), and accuracy 99.0% (293/296; 97.1–99.8%). The positive likelihood ratio was 153.8 (21.8–1084.9) and the negative likelihood ratio 0.01 (0.00–0.06). PANDAA Ct values correlated significantly with the reference assay L gene target (Spearman’s ρ = 0.80; p < 0.0001) and GPC target ( ρ = 0.69; p < 0.0001). Three confirmed discordant results occurred at high Ct values near assay detection limits. Performance was consistent across sites with overlapping confidence intervals for sensitivity and specificity and no difference in median Ct values for any assay target between laboratories. Ct values were significantly lower in patients ≥ 45 years ( p = 0.008).

Conclusions and Significance

PANDAA ^® LASV demonstrated high diagnostic accuracy to detect Lassa virus RNA compared to the dual-reaction reference assay. Reducing the operational burden from two PCR reactions per sample to a single PANDAA ^® LASV reaction while maintaining equivalent sensitivity and specificity could have a direct impact on labour and reagent costs, and turnaround time. Clinical validation against additional LASV lineages is an immediate priority.

This Zenodo record is a permanently preserved version of a PREreview. You can view the complete PREreview at https://prereview.org/reviews/24095257.

This study evaluates PANDAA LASV, a single-reaction RT-PCR assay for detecting Lassa virus, using 296 clinical samples from two Nigerian laboratories. The assay showed 98.6% sensitivity and 99.4% specificity compared with RealStar RT-PCR (the current standard). Only three samples remained discordant after repeat testing; performance was similar across both sites.

I think this is a meaningful contribution given Lassa virus has substantial genetic variation, making it difficult to develop a single assay that detects different lineages. The study provides promising clinical evidence that a single-reaction test could perform comparably to the current two-reaction method, while potentially reducing laboratory costs and workload.

Major issues

Although PANDAA is designed to detect all seven Lassa virus lineages, the clinical samples mainly represented lineages II and III. The authors should be more cautious about generalizing their findings to other lineages without additional clinical evidence.
Initially, 18 samples had conflicting results between the two assays, but only three remained discordant after repeat testing. Interestingly, 14 samples initially positive by RealStar became negative upon retesting. Without a third independent assay, it is difficult to determine whether these differences resulted from testing variability or false positives. I think this deserves more discussion because it directly affects how we interpret the reported sensitivity and specificity.
The study was funded by Aldatu Biosciences, the developer of PANDAA LASV, and the corresponding author is affiliated with the company. This does not necessarily undermine the findings, but independent validation would make the results more convincing!

Minor issues

The authors mention that PANDAA could reduce costs, turnaround time, and laboratory workload. However, these outcomes were not actually measured.

Competing interests

The author declares that they have no competing interests.

Use of Artificial Intelligence (AI)

The author declares that they did not use generative AI to come up with new ideas for their review.

Read the original source