A rapid coronavirus test at the heart of Boris Johnson’s mass-testing strategy missed more than 50% of positive cases in an Operation Moonshot pilot in Greater Manchester, the Guardian can reveal.
The 20-minute tests, on which the government has spent £323m for use on hospital and care home staff with no symptoms, identified only 46.7% of infections during a crucial trial in Manchester and Salford last month.
This means that many of those carrying Covid-19 were wrongly told they were free of the virus, potentially allowing them to infect others.
The test is one of those due to be used in the UK’s first city-wide mass-testing initiative which starts in Liverpool on Friday. It has also been rolled out to a number of other hospitals in England.
Scientists with Greater Manchester’s mass testing expert group (MTEG) raised significant concerns about the accuracy of the OptiGene Direct RT-Lamp tests this week, and said the technology should not be widely used as intended in hospitals or care homes.
The findings pose significant questions about one of the main tests in the government’s mass-screening strategy, which Johnson heralded this week as the UK’s main route back to normality. The prime minister told MPs on Monday that the government was “rolling out testing of all NHS staff as fast as we possibly can” and that it wanted to introduce rapid, regular testing for hospitals, care homes, schools and universities.
The Salford trial was billed as the first step in the Operation Moonshot mass-testing scheme but was drastically scaled back after just six weeks, in part due to concerns about the accuracy of the OptiGene test, the Guardian understands.
Local leaders had asked the government for clinical validity data behind the technology, but it is understood this data has not been provided.
The Department of Health and Social Care said the test had been validated in three other trials which differed from Manchester’s findings, however it has not made this data public.
The MTEG, which reports to the region’s health chiefs and the Greater Manchester mayor, Andy Burnham, revealed their concerns about the Operation Moonshot findings to Martyn Pritchard, the chair of the city region’s testing strategy group on Monday.
In a letter seen by the Guardian, they wrote: “The current available data from the Manchester Pilot shows low sensitivity (46.7%) of the Direct RT-Lamp platform,” which is favoured for use in hospitals because of its ability to produce results in around 20 minutes.
They wrote that “a high proportion of samples collected from infected individuals in a ‘real world’ setting would not be detected” using the technology and added: “MTEG has significant concerns and do not feel the data supported the investment in the large scale roll-out of Direct RT-Lamp saliva testing in any of the proposed clinical settings considered (hospital staff, care staff, community settings) at this time.”
The Greater Manchester experts conclude that “technical limitations of the system were also a cause for concern”, including the lack of internal controls, and that if not addressed “these may pose a significant risk of errors”. They added: “It should be noted that solutions to these issues are not available at this time and are likely to take considerable resource and time to address.
“It is also of concern that the scale of investment needed [at a ‘new department’ level] to deploy this test at a scale that would be useful and sustainable is unlikely to be cost-effective.
“We remain fully committed to wider asymptomatic testing in the health and care sectors as well as in community settings, however we are not convinced that this is the right platform for widespread deployment across GM.” The experts said they would assess the tests again if there were fixes that “materially improve sensitivity”.
The hospital test is made by OptiGene, a small biomedical firm in Horsham, Sussex. The government has paid £323m for 600 of its newly-developed Genie HT machines and 90m testing kits and chemicals. Each machine can reportedly deliver up to 200 tests an hour.
The DHSC confirmed that the Genie HT machines would be the primary technology deployed to test NHS staff and patients who do not present with coronavirus symptoms.
They do so using two methods: the rapid Direct RT-Lamp, where a saliva sample is added straight to the machine after being made safe by heating, and the RNA Lamp, which takes longer to process because the nucleic acid has to be extracted from the sample.
Prof Mark Wilcox, the co-chair of the DHSC’s technical validation group, said: “The Direct Lamp tests used in Manchester have been validated in other laboratories and in real-world testing for use in different settings.
“It is incorrect to claim the tests have a low sensitivity, with a recent pilot showing overall technical sensitivity of nearly 80%, rising to over 96% in individuals with a higher viral load, making it important for detecting individuals in the infectious stage. The challenge now is to understand the reasons for the difference in claimed sensitivity in one evaluation versus those in multiple others.”
The health secretary, Matt Hancock, announced a six-week trial of OptiGene’s tests at the end of May, which he said “could change the way that we control Covid-19 across the country”. The results of the trial, however, have never been made public. A separate study published last month found that it identified only 67% of infections, meaning it missed one in three cases, and advised anyone who tests negative to take a separate follow-up test to verify the result.
OptiGene referred the Guardian to the DHSC.
Questions have also been raised about a separate rapid saliva test which will be offered to everyone in the community in Liverpool. The public data available on the Innova test is for nose and mouth swabs only, not saliva tests.
The instructions for use say “negative results do not rule out Sars-CoV 2 infection” and should not be used for treatment decisions or “infection control decisions”. The government’s Porton Down laboratory evaluated the test and approved it in August, but only for swab samples and no detailed results have been published.
Johnson earlier told the Commons that the rapid test to be used in Liverpool would be one “that you can use yourself to tell whether or not you are infectious and get the result within 10 to 15 minutes”. The Innova test, however, is for people who already have symptoms and is not approved to be self-read. Any positive case should be confirmed with the gold-standard lab-based PCR swab test.
Experts, including those in Liverpool, are understood to be anxious about the performance of the tests chosen for use in the pilot scheme and to have asked for reassurance from government. There is concern that doubts over the results will undermine public confidence and affect people’s willingness to come forward for testing.
Bill Esterson, the Labour MP for Sefton in Merseyside, asked the government on Thursday to urgently publish the test’s clinical validity data and said: “It’s very worrying if unreliable tests are being used. It’s crucial that the Liverpool mass-testing pilot succeeds.”