Sapan Gai, CCO at Sovereign Metals, discusses their superior graphite test results. Watch the video here.
London South East prides itself on its community spirit, and in order to keep the chat section problem free, we ask all members to follow these simple rules. In these rules, we refer to ourselves as "we", "us", "our". The user of the website is referred to as "you" and "your".
By posting on our share chat boards you are agreeing to the following:
The IP address of all posts is recorded to aid in enforcing these conditions. As a user you agree to any information you have entered being stored in a database. You agree that we have the right to remove, edit, move or close any topic or board at any time should we see fit. You agree that we have the right to remove any post without notice. You agree that we have the right to suspend your account without notice.
Please note some users may not behave properly and may post content that is misleading, untrue or offensive.
It is not possible for us to fully monitor all content all of the time but where we have actually received notice of any content that is potentially misleading, untrue, offensive, unlawful, infringes third party rights or is potentially in breach of these terms and conditions, then we will review such content, decide whether to remove it from this website and act accordingly.
Premium Members are members that have a premium subscription with London South East. You can subscribe here.
London South East does not endorse such members, and posts should not be construed as advice and represent the opinions of the authors, not those of London South East Ltd, or its affiliates.
Thanks celtics.
Thanks Celtics
Another really good infomrative post celtics. Thanks its appreciated.
These are the posts we want to be seeing not the $hite that is posted too often (including by me)
I wanted to talk about the association of CT values and Infectiousness and its relevance on the Clinical results disclosed by Avacta. Alastair explained that there is no binary cutover between the Point of “being infectious” and “being NON-infectious”. The reason for this is that each COVID case is unique – length of exposure, environmental conditions etc.
Studies have been completed that correlate successful isolation of the virus in a cell culture and the CT value.
Lower viral loads are indicated by a higher number of amplifications (CT value) in a PCR test.
In other words, the evaluators are trying to find out what point is the viral load low enough (CT value high enough), where the ability to culture the virus is no longer possible. The ability to positively culture has served as a proxy for Infectiousness.
Study 1
https://www.nejm.org/doi/suppl/10.1056/NEJMc2027040/suppl_file/nejmc2027040_appendix.pdf
Supplementary Appendix - https://www.nejm.org/doi/suppl/10.1056/NEJMc2027040/suppl_file/nejmc2027040_appendix.pdf
Within the Supplementary Appendix on Page 20.
At CT values 28-30, viral culture was positive in 2 out of 14 tested viral cultures.
At CT > 30, ZERO positive viral cultures were cultivated from 38 samples.
The PCR Test used in the evaluation was Allplex™ 2019-nCoV Assay on page 23.
Study 2
https://wwwnc.cdc.gov/eid/article/27/5/20-4688_article
This was part of a Clinical evaluation on Roche SD Biosensor Antigen Test.
At CT > 30, only 1/27 (4%) could be positively cultured.
Figure 2 provides an excellent visual representation - https://wwwnc.cdc.gov/eid/article/27/5/20-4688-f2
The PCR test /brand used is not mentioned.
Why is this relevant?
Antigen LFTs need to be as Sensitive as possible. AffiDX has robustly demonstrated Sensitivity up to and including CT Value 30. With 19 positive samples between CT values 29-30, ALL correctly identified.
In Avacta’s presentation (slide 18), explicitly noted is the PCR test engaged in the Clinical evaluation - Applied Biosystems TaqPath COVID-19 CE-IVD RT-PCR Kit.
Not all PCR tests are equal. There are various assay efficiencies. CT values are not consistent across labs using different PCR tests. Michael Mina commented on this - https://twitter.com/michaelmina_lab/status/1345917801052180480 (point 20/30). This is also an excellent thread about the ability to culture.
TaqPath is a product from Thermofisher Scientific (TMO). TMO is a world leader in medical instruments. I would expect TaqPath to have excellent assay efficiency. Therefore, a CT of 30 on TaqPath assay may in fact be equivalent to 30+ on another lab using a different PCR test provider.
In summary, with Positive samples up to and including CT value of 30, being demonstrated as “Infectious” by studies, LFTs should be Sensitive enough to detect to CT value of 30 in order to be employed as a “Test to release”. Avacta has just done that, not many competitors can say the same.
Celtics.
The original statistical calculation by Celtic is correct if we are talking about a totally random glitch that occurs 1 in a 100 times, regardless of what is done with the swab. However, I suspect a control of 100 blank tests (that is, not actually swabbing the person) would not just randomly show a positive in every 100th test. It is more likely something in an actual swab that is causing it, which perhaps hasn’t been considered by the manufacturer, and which could therefore increase the likelihood of a second test coming up with an FP too.
Ok, we’ll all agree with you. Now what?
Think about solutions, not problems for a change - who cares why. The innova test currently has better clinical specificity and targets the N protein. Keep a box spare at home. If you get a positive on your twice weekly 1 in 100 S protein AffiDx test, confirm it with a 1 in 1000 false positive innova N protein test. All done quickly at home. If you’re falsely isolating by then, you’re really unlucky as that’s 1 in 100,000. At that point you’d get a confirmatory PCR test anyway. The additional costs would be tiny - providing an Innova test for every 100 AffiDx tests on average. This cost will reduce if the 99% improves to 99.x%.
With the above, the use case for the exceptionally sensitive AffiDx test remains the same - full confidence to mix socially as you’re not infectious and the Innova tests are there to provide confidence you’re not isolating for no reason - which wouldn’t happen with PCR testing being used to confirm anyway. Problem solved.
Responding to the cause of FPs. FPs can be caused by non-specific binding or cross reactivity. Non-specific binding and cross reactivity are NOT the same.
Cytiva outline some solutions to False Positives that arise from non-specific binding.
https://www.cytivalifesciences.com/en/us/solutions/lab-filtration/knowledge-center/lateral-flow-assay-troubleshooting-and-how-to-switch-membranes
On a Cytiva Labroots Presentation in 2020, David Wilson (Avacta Commercial Leader) briefly spoke about the trade-off between Specificity and Sensitivity @ 36.20 mins https://www.youtube.com/watch?v=9sWbhA8nZR8
David stated explicitly the bias towards Sensitivity and NPV, as well making an implicit reference to Likelihood ratios "ideally high at both levels". It is important to produce a test that is "useful" across a broad range of disease Prevalence.
thanks,
Celtics.
Nope, that isn’t how it works unless it’s an undiagnosed true positive. False positives are random ‘glitches’ with a test - contamination, error in use or formulation etc.
PAH I think raising the sensitivity is probably the wrong phrasing but the likelihood of two false positives in a row at 99% Is 1% x 1% or 0.01% which is 1 in 10,000. So confirmatory testing reduces the potential of a false positive having such an impact.
If that 1 was an outlier, which AS alluded to in the prezzo, then yes larger sample sizes dilute the impact of that single outlier and increase the S&S
Brilliant Celtics - really appreciate your contributions here. Please share with DHSC/PHE ;-)
BUMP - thanks for the excellent post Celtics. I complete agree that a 100% sensitive test for the infectious is perfect for Ready, Test, Go. As suggested, the use of testing stations ahead of events will ensure compliance and allow events to go head with full confidence and without any social distancing. The same can be said ahead of flights.
No, the false positive is a randomly segregating event. Test someone who isn't infected 100 times and get 1 false positive or test 100 negative individuals and get 1 false positive. If you test an individual twice then the chance of a false positive result is 1% of 1%, ie specificity is 99.99%.
The NHSBT use this methodology all the time when testing for blood borne viruses; every donation is tested twice to improve the outcome specificity of the tests used.
Can someone please share this post with Deeks
"Economically, it’s better to take a second LFT immediately after, and the probability of receiving a First FP immediately followed by a Second FP is 0.0096% (101/102 x 101/102)"
Does this necessarily hold true? What are the reasons for the false positive, is it caused by the reagent binding to some antigen or compound other than coronavirus, in which case the second test might yield the same result. I genuinely don't know the answer to this, so putting it out there.
Thanks for that Celtics, I feel I learnt something from your post!
The market is under-appreciating the LFT Clinical results due to inability to interpret. Will help the group to understand the Clinical significance using Predictive Values and importantly Likelihood Ratios used by Medical Statisticians. We can confidently say 98.96% Cumulative Sensitivity [95 TPs / (95 TPs + 1 FN)] up to and including CT Value 30 is robust. 19 Actual positives within CT 29-30, were ALL correctly identified as positive.
NPV = 99.02% [101 TNs/(101 TNs + 1 FN)]. PPV = 98.96% [95 TPs/(95 TPs +1 FP)], up to an including CT of 30.
Predicative Values like NPV and PPV are affected by Disease Prevalence. As Disease Prevalence lowers, a higher % of Positives are FPs. As Disease Prevalence increases, a higher % of Negatives are FNs. Manufacturing capacity potentially available to AVCT in UK has an initial Upper limit of 30M LFTs/month. Using that figure, with a Clinical Specificity of 99.02%, there would be 294,118 FPs/month. Currently, LFT positives are confirmed by PCR. If we price PCR at 50 GBP/test, that confirmation is 14.7M GBP/month cost. Economically, it’s better to take a second LFT immediately after, and the probability of receiving a First FP immediately followed by a Second FP is 0.0096% (101/102 x 101/102). This is GCSE Math (independent probabilities). From 30M tests/month, 2,884 persons would have a FP immediately followed by a Second FP.
Specificity and PPV is important when thinking about “a test to isolate”. Sensitivity and NPV is critically important when thinking about “a test to release /enter”. Test to release is important to safely allow a return to freedom, a Human need. Likelihood ratios (LRs) are NOT affected by Disease Prevalence. LRs inform us of the usefulness of a Test - https://www.uws.edu/wp-content/uploads/2013/10/Likelihood_Ratios.pdf. Positive LR = Sensitivity / (1-Specificity). Mathematically Positive LR is between 1 to 1000, with being as close as possible to 1000 indicating a more useful test and 1 useless. Negative LR = (1-Sensitivity) / Specificity. Mathematically Negative LR is between 0 to 1, with being as close as possible to Zero indicating a more useful test and 1 useless.
A Test to Release approach requires a very Low Negative LR i.e. as close as possible to Zero. Up to an including CT value of 30 AffiDX has Negative LR of 0.011. Whilst, on a larger sample set, Innova is claimed by Jon Deeks as having a Negative LR of 0.25 - https://twitter.com/deeksj/status/1326135461178388480
AffiDX shows evidence to support an effective and safe “Test to Release” approach in my opinion. I think the UK Gov’t will push Test to Release /Access to Enter using Professional Use only tests only with an auditable QR code/24 hr. Negative Test Certificate (using national pop up Testing stations). We’re seeing this in ongoing trials – Snooker, Nightclub, Wembley. In my opinion, this is the best way to safely keep prevalence low and not with home tests since sampling error increases.
Celtics.
Tue - Avacta announced CV RNS
Wed - ODX RNS they will start producing the Avacta test
Thu - Avacta discuss all the potential deals lined including how BBI/Abingdon fit in.
Results disclosed today exceeded Avacta's self defined "high performing" test threshold.
https://avacta.com/how-diagnostic-test-performance-is-measured/
Sample Size - 200
1) Sensitivity (PPA) - how many of the positive cases are detected = 97.96%
= 96 True Positives / 98 Actual no. of Positives
2) Specificity (NPA) - how many of the negative cases are detected = 99.02%
= 101 True Negatives / 102 Actual no. of Negatives
3) Positive predictive value (PPV) - "how confident are we that a positive result is actually a true positive" = 98.97%
= 96 True Positives / (96 True Positives + 1 False Positives)
4) Negative predictive value (NPV) - "how confident are we that a negative result is actually a true negative" = 98.06%
= 101 True Negatives / (101 True Negatives + 2 False Negatives)
Note - Avacta have produced these excellent results where 65 out of 98 Actual Positives in the sample population had a CT value 26 to 30 (which is low viral load).
These results demonstrate why Mologic (GAD) offered capacity to Avacta instead of reserving that capacity for their own Antigen test. I think there is an exciting path ahead for Avacta and its Shareholders. Today was a significant milestone for Avacta, its employees and equally its manufacturing partners (GAD, BBI and Abingdon). Thank you!
Celtics.