Concerning the OptiGene Direct LAMP assay, and it’s use in at-risk groups and hospital staff.

Dear Editor,

The COVID-19 pandemic has triggered an unprecedented demand for diagnostic tests. Severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) is highly contagious in the pre-symptomatic period, when the viral load is high. In the effort to reduce transmission, for the first time in infection diagnostics history, testing is being aimed not only at symptomatic, but also at asymptomatic individuals, both in the health care setting and in the community.

Shortages in PCR reagents and platforms, related to the demand for large-scale testing, fuelled the effort for alternative diagnostic solutions. We have read with interest the Journal of Infection (JoI) article, regarding the laboratory based assay chosen by the British government, for SARS-CoV-2 mass testing (Operation Moonshot), a loop mediated isothermal amplification (LAMP), produced by OptiGene. , The testing of saliva by OptiGene direct ORF1ab LAMP (with no RNA extraction), in newly built laboratories under the management of acute NHS Trusts, is costing the taxpayer over 400 million British pounds and has resulted in the movement of key staff away from essential roles in acute NHS Trust diagnostic laboratories, at short notice.

The OptiGene assay on saliva samples is intended for the testing of asymptomatic NHS staff, to help prevent hospital acquired outbreaks. The assay however, failed to detect more than 50% of saliva positive cases in a pilot in Greater Manchester, when compared to a “gold standard” qPCR (polymerase chain reaction) assay. In contrast, recently published data from the Department of Health and Social Care (DHSC) report 70% sensitivity for swabs and 79% for saliva for this assay.

To understand the performance in more detail, 86 PCR positive nose and throat swabs, with a variable viral load, collected from patients and symptomatic or asymptomatic members of hospital staff, were tested in parallel by direct LAMP and in-house CDC N1/N2 qPCR (Limit of Detection (LOD) 156 digital copies/mL or 2•6 digital copies per reaction, Qnostics) at the Southampton Specialist Virology Centre. These samples were stored at 4 °C, prior to being frozen at −80 °C within 36 h of sample collection. The SARS-CoV-2 Molecular Q Panel (SCV2MQP) from Qnostics was used to quantify the qPCR results in digital copies/mL.

Consistent with results from the Manchester Regional Virus Laboratory, which evaluated freshly collected saliva samples from asymptomatic individuals, we have been unable to replicate the sensitivity (see Table 1 ) reported by the DHSC. For this reason, we would like to raise a number of points on which we seek clarity.

Table 1

Direct OptiGene LAMP (ORF 1ab) compared to N1/N2 CDC PCR (the Southampton Specialist Virology Centre in-house diagnostic assay).

SARS-CoV-2 viral load digital copies/mL (dc/mL)		CDC SARS-CoV-2 N1/N2 PCR samples	OptiGene direct LAMP
			Number of positive samples	Positivity rate compared to PCR (%)
1000 – <10,000	103–104	18	0/18	0
10,000 – <100,000	104–105	20	0/20	0
100,000 – <1,000,000	105–106	17	6/17	35
1,000,000 – <10,000,000	106–107	25	18/25	72
>10,000,000	>107	6	5/6	83
Total		86	29/86	34

A sample of >10,000,000 (>107) digital copies/mL (dc/mL), collected from a symptomatic NHS staff member, swabbed on day one of symptoms, produced a negative result when tested by the OptiGene direct LAMP method. Similarly, a swab from an asymptomatic staff member, a target group for this assay, with 2,500,000 (2.5 × 106) dc/mL was missed. A false negative result, with a high viral load, is a patient safety risk, particularly if the sample belongs to an asymptomatic member of staff.

The sensitivity of a diagnostic test is assessed in comparison to a gold standard, in this case OptiGene direct LAMP compared to qPCR. However, the performance of individual PCR assays varies, depending on a number of factors, including the choice of viral RNA amplification targets, among many other technical variations, which impact on PCR efficiency. For this reason the comparison between assays should be based on quantitation standards and not on Ct values, as published in the evaluation. Which criteria were used for “sufficient sensitivity” for an assay used to isolate infected NHS staff from patients? The experiments relating to the Limit of Detection (LoD) of 103 copies/mL in the DHSC evaluation were flawed. The lack of difference in LoD between extracted and direct LAMP, with the quantitated NIBSC samples, is artificial and not seen in patient samples in the evaluation or in the Journal of Infection paper that states an LoD of 100,000–1,000,000 (105–106) dc/ml. ,

Critically, the inclusion of an internal control, added before nucleic acid extraction, is a vital quality requirement for molecular diagnostic work. Lack of amplification, due to a number of reasons, such as a LAMP platform failure, in individual reaction wells, or presence of inhibitory substances in the clinical sample, is flagged by the use of an internal control, thus preventing false negative results from being reported. The OptiGene direct LAMP method does not contain an internal control, despite being present in alternative SARS-CoV-2 LAMP assays. ,

It has been asserted that nucleocapsid protein (N-) gene targets for qPCR are not valid for assessment of the performance of the OptiGene direct LAMP assay. The reason provided is that N-gene qPCR detects subgenomic mRNA that may persist, and be detectable, beyond the period of viral replication, with an inability to differentiate between RNA from replicating virus and residual RNA after a resolved infection. It is important to stress that duration of detection of subgenomic mRNA, and its implications for infectiousness, are currently debated. Why is this being asserted to be scientific fact, along with the theory that OptiGene direct LAMP assay only detects infectious virus and those not detected are non-infectious, for which no evidence is provided and the evidence available indicates otherwise? , ,

The choice of sample type for use with the OptiGene, for asymptomatic screening for NHS staff is saliva, a heterogeneous sample, which suffers from a reduced sensitivity in comparison to the higher sensitivity of a nose and throat swab. Meta-analyses of saliva testing studies have consistently shown that the sensitivity is inferior to nose and throat swabs, at 83–85% when compared by qPCR. , The lower sensitivity of saliva, in addition to an assay with significantly inferior sensitivity, is of concern to us. Saliva may be acceptable for community mass testing, but not in the health care setting, where a missed positive sample can lead to an outbreak with significant consequences for patients.

We would be grateful for further data to enable the clinical virology and healthcare community, to understand the rational for prioritising the use of this insensitive assay, which lacks an internal control.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper