LIVING WITH COVID-19
Using our diagnostic testing options wisely now and in the future
- New Zealand and Australia’s COVID-19 strategy has shifted from elimination to suppression, resulting in different testing requirements.
- Many testing options have been developed, with most using three core types of testing methodologies: PCR testing, antigen testing, and serology testing. The sampling method also needs to be considered, as not all test types are compatible with all sampling methods. There are pros and cons for each testing and sampling methodology.
- A strong strategy will consider how each testing option can be utilised to best effect to strengthen our defences against the spread of COVID-19 in the community. This article lays out our recommendation for the ongoing analysis of the evolving data on COVID-19 diagnostics in New Zealand and Australia.
As New Zealand and Australia both move away from elimination and towards suppression of COVID-19, the strategies and tools for limiting the impact of this deadly disease on our population continue to evolve rapidly. There is no doubt that diagnostic testing will remain central to any strategy, but the core question on the minds of many across government, businesses, schools, and event organisers alike is what are the right diagnostic testing solutions that meet their specific needs. There has been a lot media and political attention on sampling and testing methodologies, and in some cases individual products, but little or no critical analysis of what information each can offer, and how they are best utilised. In this brief report, we take a step back and set out a thorough landscape of the breadth of diagnostic technologies available for COVID-19, explore their advantages and limitations, evaluate what the data says about the most appropriate usage situations, and discuss how they are being used internationally.
In the first phase of both Australia and New Zealand’s COVID-19 response, the diagnostic testing strategy was very clear. With a very low rate of COVID-19 community transmission (and a correspondingly low rate of positive tests), both countries chose to rely almost exclusively on the gold-standard approach of nasopharyngeal swabs for collecting a sample from individuals, and RT-PCR (Reverse Transcriptase Polymerase Chain Reaction) to test the sample in a certified laboratory for the presence of viral genetic material. However, as the Delta strain has taken hold, first Australia and now New Zealand have shifted away from the initial ‘elimination’ strategy to one of ‘suppression’.
To meet the clear needs of the global population, a multitude of testing options have been developed. Many are different manufacturers’ versions of one of the three core types of testing methodologies (discussed below). Added to this there are almost daily announcements of new innovative approaches being trialled. For this incredible pace of innovation, the many public and private entities involved (and in many cases partnerships) must be applauded. The pace of innovation and evidence generation is not slowing down either – if anything it is accelerating. The strongest public health response must incorporate the most suitable technologies, so staying on top of this rapidly evolving landscape is crucial.
Let’s start by breaking down the diagnostic landscape into some of its constituent parts,
and discussing some of the methodologies in a little more detail.
Nucleic-acid testing, the basis of RT-PCR
The typical gold-standard laboratory method for identification of COVID-19 is to test for the presence of the viral RNA – its genetic material – using Polymerase Chain Reaction (PCR) methodology. More accurately in this setting, the technology is Reverse Transcriptase Polymerase Chain Reaction, or RT-PCR. This methodology has been central to both Australia and New Zealand’s testing strategy thus far.
The PCR testing process starts by breaking open cells or viral particles present in a sample to expose the nucleic acid chains. Primers are used to select a particular type of nucleic acid chain – in this case the COVID-19 RNA. The selected chain is then repeatedly replicated, producing thousands of copies of the original chain of interest.
Because of its process of amplifying the amount of target sequences of RNA, it can identify low levels of virus with a high degree of sensitivity (identification of true positives) and specificity (identification of true negatives). Because of this profile, it is widely considered the ‘gold-standard’ testing option to which all other testing methodologies are compared. As such, the World Health Organisation (WHO) and most countries use this methodology for testing symptomatic individuals, or those who are known to have been exposed to COVID-19, where a high degree of accuracy in the results is critical.
PCR can be used with nasal, nasopharyngeal and saliva samples (more information follows regarding sample types), but the choice of sample can affect the methodology a laboratory must use for preparation and processing, and this methodology (and the laboratory’s use of it) must be appropriately validated.
How a sample is collected can also impact the accuracy of the results. Samples collected by trained healthcare professionals have been shown in studies to be the most accurate. However, in some countries (such as the UK and the USA), kits have been approved for self-collection of samples that can then be couriered or mailed to an approved laboratory for PCR testing. This extends the potential reach of PCR testing by reducing the barriers to access and the resource intensity of their use. While PCR technology is a cornerstone in molecular diagnostic laboratories, the standard process is time consuming and requires skilled laboratory technicians who are trained in sample preparation and operation of the PCR machines. The drawbacks therefore, are the timeliness of the results, which can take up to 24 hours, and the requirement for all samples to be processed through centrally located laboratories. In most jurisdictions internationally, including Australia and New Zealand, the laboratory must have a specific accreditation for processing clinical samples (as opposed to samples for research purposes).
Antigen Tests, also known as Rapid Tests or Lateral Flow Tests
Overview of Antigen Testing
Antigen tests identify the presence of specific proteins on the outer surface of viral particles. The name ‘antigen’ denotes a foreign particle which may be recognised by our own immune system, triggering the normal immune response, and resulting in the production of antibodies. This type of test uses laboratory-produced antibodies to bind viral antigens and create a visual indicator of its presence. After being mixed with reagents to create a solution, a few drops of the sample are placed into the end of the test kit, and the liquid is pulled through via capillary action. If the viral antigen is present, it binds to the antibody, and will light up as a band of colour on the test (see Fig 2, above). Most people will be familiar with this type of test because of its similarity to a pregnancy or ovulation test. It is a numbers and probability game. The antibodies must “catch’ the antigens on their way past and there must be sufficient numbers of them to show as a band of colour. COVID-19 antigen tests typically require nasal or nasopharyngeal swabs. There are products internationally that do use saliva samples, but these are not well-evidenced or widely utilised.
Antigen tests have two major advantages in comparison to PCR. Firstly, they are quick, with results available within minutes. Secondly, they require no transport of samples to a central laboratory, and indeed no lab equipment at all, to process the test result. Because of these features, they have been identified by WHO as being particularly useful where speed of result is crucial and there is a low chance of the test subject being positive (i.e. in surveillance testing). Markets such as the USA, the UK, and Singapore, have introduced antigen testing for screening those entering school, university, or work campuses, and for entry into crowded public spaces and large-scale events.
There are, however, some drawbacks. The first is their sensitivity, i.e. their ability to accurately identify whether someone is infected or not. Broadly, the data appears to show that antigen tests are good at identifying those with high viral load, i.e. when people are in their most infectious state and may be actively unwell. However, they are not as good at identifying those with a lower viral load – perhaps earlier in their infection cycle where they may not even be aware that they are infected.
As a result of this, while an antigen test in an asymptomatic person can inform someone of the fact that they are COVID-19 positive, a negative test should not necessarily be taken to mean that an individual is not infected. This makes antigen testing a poor choice for test-and-trace programmes.
When considering which of the options might be selected for use, it is important to consider not only if the test has been validated, but how that validation was carried out.
Secondly, there is the issue of sample collection. In the validation process for many rapid antigen tests, samples are taken from test subjects by trained healthcare professionals. When the sample is taken by a non-healthcare professional or by the test subject themselves, the accuracy of the result drops markedly. In the USA, some antigen tests are only approved for use in a setting where the sample collection is monitored by a healthcare professional to ensure it is being taken correctly.
Thirdly, there is also the issue of central data collection. If a test is being taken in the home setting there is no way to ensure a positive test result is reported, or that the infected person has medical support if required.
Considering test accuracy and validation data
A wide variety of antigen tests from different manufacturers are available internationally, each having been developed and validated separately. Not all will be of equal quality. While other countries (including Australia) have established regulatory systems with formalised review processes for in vitro diagnostic tests, New Zealand has not. This puts New Zealand at a distinct disadvantage when it comes to ensuring that the rapid antigen test products made available by manufacturers are of sufficient quality.
When considering which of the options might be selected for use, it is important to consider not only if the test has been validated, but how that validation was carried out. Aspects to review are:
- what was the usage setting, and the estimated likelihood of those being tested having an infection (e.g. were they symptomatic, an asymptomatic contact, or a randomly selected person)?
- who took the sample, and what level of training did they have?
- what was the sensitivity of the test compared with PCR? (how many of the true active infections did the test identify)?
- has the test been re-validated as being sufficiently accurate against all of the current prevalent strains?
This will be particularly important if the new Omicron strain were to take over from Delta as the prevalent strain in the community.
As an example of the need to carefully review data on an ongoing basis, we can consider a specific rapid antigen test that has been approved for importation into New Zealand, and approved for general sale in Australia to the public. Validation data produced by the manufacturer indicate high sensitivity, both for use by healthcare professionals (>98%) and self-testing (>95%). However, recently published results of a large independent study, analysing both symptomatic and asymptomatic people referred to a COVID-19 testing centre in Spain, showed that while the test has reasonable sensitivity in symptomatic people (~80%), it was poor in asymptomatic people (~56%). This would mean that in a setting where someone was asymptomatic, nearly one in two positive cases would be told they were ‘negative’ despite the fact they were actually positive. This example clearly shows the need for very careful analysis of the entire dataset related to a particular product, and for authorities to constantly be reviewing their suitability.
This example also calls into question whether this test could be regarded as suitable for testing asymptomatic individuals because of its performance in this setting, and as there is a distinct chance of someone delaying further testing despite emerging symptoms because of a negative test result and putting others at risk. This is not an isolated example – US regulatory authorities have previously banned use of a different brand of test on the basis of a similar disparity between manufacturer and ‘real-world’ trial performance.
With that said, the chance of identifying even a fraction of the undiagnosed cases entering into settings such as schools and hospitals could well justify the use of rapid antigen tests more widely.
Serology Tests, also known as Immunity Tests and Antibody tests
The third major category of tests is the serology (blood) test, which tests for the presence of antibodies to the virus. More simply put, they identify whether a person has developed an immunity from either a prior infection (natural immunity) or from a vaccination-triggered immune response. People with immunity build a reservoir of virus-specific antibodies (and antibody-producing cells) in the bloodstream. This ‘process allows for more rapid detection and response to subsequent COVID-19 infections (see Fig 1, above). Because serology tests measure the body’s reaction rather than detect the virus, they do not provide evidence as to whether someone is currently positive (has an active infection) or negative for COVID-19.
Serology tests will typically require a small drop of blood, which can be self-collected if required, and the testing process is rather like that for rapid antigen tests – with results returned within minutes.
Serology tests have potential in several settings. In Australia and New Zealand, they have been used to provide additional evidence as to whether infections identified as having a low viral load are historic, or whether they are in fact in the early stages of a new infection. There is also potential that they could be used to confirm whether an individual does have immunity, discern whether this immunity is from a prior infection or vaccination, and even provide a quantitative measure of the strength of this immunity. With the growing focus on vaccine passports for travel, this could provide an additional layer of certainty that travellers crossing the border do indeed have immunity.
Important to note, though, is that many key jurisdictions do not have support from regulatory bodies for the widespread use of serology tests. In general, it is recommended that they are only used under the guidance of a healthcare professional who can appropriately interpret the results.
Novel testing methodologies
As well as these established testing methodologies, there are a variety of new technologies in varying stages of development that have the potential to offer new advantages and overcome the limitations of existing technologies.
Isothermal Amplification tests identify viral nucleic acid sequences using specially designed primers, much as PCR does. The difference is that these tests do not require laboratory processing. Instead, they employ a single-step operation that can take place on a small desk-top machine suitable for use on a ward or in a clinic. They can deliver results nearly as rapidly as antigen tests – in minutes – and with observed sensitivity and specificity that is in some instances close to that observed with laboratory-based RT-PCR. A number of commercial Isothermal Molecular Test machines and kits are available. There is potential that this methodology could be utilised to reduce the burden on laboratory services during major outbreaks. It could also be used as a surveillance tool in high-risk settings where speed of results can be crucial to support appropriate patient care, followed by more comprehensive RT-PCR results to confirm the initial test result. Current kits largely utilise nasal or nasopharyngeal swab samples.
CRISPR (clustered regularly interspaced short palindromic repeats) is a genetic splicing technology, and it has received considerable attention for its potential to offer a simple, rapid and high-sensitivity and specificity way to test for the presence of viral nucleic acids. Simply put, CRISPR technology acts like molecular scissors to cut out specific sequences of viral genetic material, to which isothermal amplification can then be applied. This could lead to a test that is as accurate (or more so) as RT-PCR, but as fast as antigen tests. CRISPR technology is not yet commercially available for COVID-19 diagnostic use, but is under development by several groups.
when reviewing the emerging … technologies it is important to carefully consider how the data that validates its use has been collected
And finally, several countries have reported early success in the use of highly specialised sniffer dogs, such as those already used in airports and by police teams to smell out those infected with COVID-19. This could enable officials to identify potentially infected individuals before entry into public spaces without the need for any sampling, thus minimising the number of people who would be required to take further confirmatory tests. As with antigen testing, when reviewing the emerging data for these technologies it is important to carefully consider how the data that validates its use has been collected, and what any independent ‘real-life’ testing has shown. The same four questions, outlined in the discussion on antigen testing, are a good place to start.
Nasopharyngeal (NP) Swab: NP swabs are where the sample is taken from the back of the nasal cavity where it meets the throat (pharynx), and it is the gold standard technique. Most PCR testing, and many rapid antigen tests rely on this swabbing technique. This location has the highest viral load in the respiratory tract. However, it may not be the most acceptable technique for some, including children, frail elderly people, and those with disabilities and special needs such as autism. It is also possible that the thought of this technique is putting people off from being tested. Comparisons of NP swabs taken by trained healthcare professionals vs self-testing samples largely show that trained professionals are more likely to collect a quality and suitably sized sample. Some rapid antigen tests try to correct for this by having the self-tester swab another location (such as the nasal passage) as well.
Nasal Swab (NS): Another option is a swab taken from the nasal passage. Here there is a lower viral load, so there is a lower chance of detecting an infection even if a person is infected. This can potentially be corrected for by increasing the length of time with which the swab has contact with the nasal passage. This sampling approach may be more acceptable and more realistic for people to collect themselves – either in a home setting, or under the supervision of a healthcare professional.
Saliva Swab: The other option that has generated much debate is the use of saliva. Like nasal swabbing, an advantage for saliva swabbing is the acceptability of the sampling method to test subjects. However, similar to nasal swabbing, there can be a lower chance of detecting an infection because of a lower viral load in the saliva. Effective collection technique (overseen by a healthcare professional) can increase the likelihood of a good sample.
implementing saliva testing at scale may not be as easy as it seems at face value
Accuracy can also be affected by whether the person has had a drink, eaten, or brushed their teeth within the period of around 30-60 minutes before the sample is taken. What is not as well known is that different laboratory techniques and protocols are required to extract the viral genetic material from saliva, and therefore implementing saliva testing at scale may not be as easy as it seems at face value. Validation of the protocol and the testing laboratory is crucial for saliva sampling to be an effective sampling tool.
International usage of testing methodologies
Like the many aspects of the pandemic response, usage of testing has been highly variable from country to country.
The national health service has extensively utilised both PCR testing, largely using nasal or nasopharyngeal samples, and rapid antigen testing.
Where an individual is symptomatic, PCR testing is used. As well as samples taken by healthcare professionals, at-home testing kits have been approved whereby a person is sent a sampling kit, and self-collects the sample, before posting it back to a laboratory for processing, with results delivered back electronically.
For those who are an asymptomatic contact, or for workplace, school or university surveillance testing, specific brands of rapid antigen tests procured by the NHS are freely available for home delivery or collection from a pharmacy. Serology tests have not been widely used to date.
Australia and New Zealand
In both Australia and New Zealand the focus has been on PCR-based testing. It is freely available for any individual who is symptomatic, or who may have been a contact of a positive case. In several Australian states, sampling for PCR is exclusively with nasal or nasopharyngeal swabs, with saliva testing specifically not allowed.
Recently both countries have approved the use of the Therapeutic Goods Administration (TGA)’s antigen tests for surveillance monitoring of employees in essential services, for pupils in schools, and more recently for self-testing. There are still restrictions for use. In New Zealand they will be used under the guidance of a healthcare practitioner (most likely a pharmacist) and some Australian states maintain a total ban on their use.
A number of serology tests have also been approved by the TGA. As with rapid antigen tests, states have made their own decisions regarding their use. Two Australian states maintain a complete ban, whilst others allow their use for providing information on possible historical cases, but only when administered by healthcare professionals. While earlier in the pandemic the TGA approved antigen and serology diagnostics under special emergency legislation, they have retrospectively engaged a laboratory to undertake an independent evaluation. A number of tests have subsequently been removed from the market when they were judged to be not of sufficient quality.
Singapore has pursued a COVID-zero approach for much of the pandemic but, like Australia, high vaccination rates have enabled a slow opening up and the move to a suppression approach.
Similar to other countries, PCR testing has been central to their response, and testing of all symptomatic individuals using this approach has been widespread. They require regular PCR testing for a range of workers in what they deem as higher-risk settings (transportation, construction, those living in dormitories, anywhere serving unmasked patrons) on a frequency determined by vaccination status. Food and beverage workers have been required to take an antigen test every two weeks, supervised by a healthcare professional, with the results uploaded into a central swab registration system. The cost of this testing regime is now being transitioned to employers.
Antigen tests are now becoming part of COVID surveillance monitoring especially as kindergartens, schools and workplaces begin to open up. A selection of self-testing kits (all requiring nasal or nasopharyngeal swabs) have been approved by health authorities. A number were distributed for free to all households, and they are also available for purchase directly by the public.
ONGOING ANALYSIS OF THE EVOLVING DATA ON COVID-19 DIAGNOSTICS
Diagnostics are instrumental for our ongoing response to COVID-19 as it becomes endemic in our communities and the pandemic continues to rage globally. Now is the time to consider how we can use the available tools most effectively. We need to stay on top of the evolving diagnostic landscape to ensure our COVID-19 management strategy continues to be world-leading. This applies to both the types of technologies and the evidence to validate their use. This work would enable strong and informed choices, as we execute the COVID-19 diagnostic strategy.
There are no binary choices regarding diagnostic technology. Each of the types of tests have their advantages and disadvantages, and situations where they are suitable for use (and situations where they are not).
A strong strategy will consider how each testing option can be utilised to best effect to strengthen our defences against the spread of COVID-19 in the community. Within each existing category of testing and sampling methodologies, there are choices to be made – different protocols, commercial kits, and products – each validated in a slightly different way. Even with existing products, the data must continue to be updated as the virus evolves, as is the case now with the new Omicron strain.
For Australia, New Zealand, and indeed all countries, staying on top of the evolving data landscape, and ensuring that approved products and providers continue to be of the highest quality, is not a done-once job; it needs to be ongoing into the future.