covid rapid antigen as good for omicron as delta

 A recent NIH study  (Testing Us At Home study) found that rapid antigen testing was as accurate for the omicron as the delta variants, in a pre-print, pre-peer reviewed article (see covid rapid antigen test accurate medrvix2022 in dropbox, or https://www.medrxiv.org/content/10.1101/2022.02.27.22271090v2.full.pdf or doi.org/10.1101/2022.02.27.22271090 ). Of note, this was NOT a pre-specified study analysis, but developed to answer another question (not being a pre-specified analysis might diminish the statistical rigor associated with a new study)

Details:

-- 5,506 participants without symptoms were enrolled from 18October2021 to 24January2022, and received rapid antigen (Ag-RDT) and nasal PCR tests every 48 hours for 15 days, all nasal tests self-administered (with a 15-minute break between samplings)

    --Ag-RDT tests used (all approved EUA tests): 

        -- BD Veritor At-Home, Quidel QuickVue At-Home, Abbott BinaxNOW

    -- PCR testing: all positives from December and January were to have whole genome testing to identify the SARS-CoV-2 variant

-- study exclusions: those with Covid symptoms in prior 14 days or any reported positive test in past 3 months

-- outcome: sensitivity of Ag-RDT, comparing those subsequently infected with the delta vs omicron variant, with sequencing done on the first PCR positive test

    -- subsample analysis was done on the participants' results on the second PCR done within 48 hours of the first one

-- 39,615 days of testing data was available for assessment

--four groups were assessed:

    -- all patients (153) with negative PCR during the study prior to a subsequent positive one, and the following subsets:

        -- those with a second PCR positive within 48 hours: 86 people (56%)

        -- those with a second PCR that was negative within 48 hours ("isolated PCR"): 38 people (25%), about twice as common in those with the delta variant

        -- those who did not perform a second PCR within 48 hours for whatever reason: 29 people (19%)

--similar proportions of the total group were not vaccinated (21% for delta and 29% for omicron)

Results:

-- 153 people were PCR-positive: 61 delta and 92 omicron

-- 36 (23.5%, CI 16.8%-30.7%) were Ag-RDT positive on the same day as the first positive PCR; 84 (54.9%) were AG-RDT positive within 48 hours

-- sensitivity for Ag-RDT (not statistically significantly different between delta and omicron):

    --same-day positive Ag-RDT and PCR: 

        -- delta: 16.4% (8.2%-28.1%)

        -- omicron: 28.2% (19.4%-38.6%)

    -- Ag-RDT positive within 48 hours of PCR test: 

        -- delta: 45.9% (33.1%-59.2%)
        -- omicron: 60.9% (50.1%-70.9%)

    -- 86 participants had consecutive 48-hour positive PCRs (ie, not just a single test being positive, but isolated positive):

        -- delta: 79.3% (60.3%-92.1%)
        -- omicron: 89.5% (78.5%-96.0%)

-- 38 participants who had isolated PCR positive results (ie, the second PCR within 48 hours was negative): all remained negative by Ag-RDT, independent of the variant

-- assessment of Ag-RDT sensitivity by PCR cycle threshold (Ct, effectively inverse of viral load):

    -- same-day Ag-RDT and PCR positive, and Ct <30: 83.3% for delta and 87.2% for omicron

    -- 48-hour positive antigen vs PCR, and Ct <30: 90.9% for delta and 95.8% for omicron

--graphs of Ag-RDT sensitivity by PCR Ct:

Commentary:

-- this study found:

    -- Ag-RDT testing was not inferior in those infected with omicron vs delta variants (though not significant statistically, the omicron edged out the delta both on same-day positivity between Ag-RDT and PCR, as well as within 48 hours after positive PCR)

     -- Ag-RDT sensitivity improved with serial testing (as found in prior studies), with repeated Ag-RDT testing 48 hours after a positive PCR increasing sensitivity from 23.5% to 54.9%

        -- AND, if one looks only at those who had 2 positive PCRs within 48 hours, these numbers for Ag-RDT increased from 36.1% to 86.0%!! 

        -- AND, further, those with initial PCR Ct <30 and Ag-RDT repeated within 48 hours, 33 of 34 people had positive Ag-RDT (and in the single case of these 34 without a positive Ag-RDT, the first Ag-RDT was not done within the first 24 hours of the positive PCR and there was no additional Ag-RDT done. so probably should not be counted so much...).

        -- those with isolated positive PCR tests (ie, the 48-hour test was negative) uniformly had negative Ag-RDT, independent of the variant. And, the antigen tests are really quite specific to a past Covid infection (about 100%)

-- why did those with only a single positive PCR but a negative retest 48 hours later uniformly have negative Ag-RDT tests? this finding does raise the possibility that an isolated PCR test may not be so specific to Covid infection

    -- other studies have confirmed that there are limits to the PCR sensitivity and specificity: https://www.cap.org/member-resources/articles/how-good-are-covid-19-sars-cov-2-diagnostic-pcr-tests#:~:text=Analytic%20performance%20of%20many%20SARS,and%2098%2D99%25%20specificity. , though the specificity does depend on the amount of virus present (eg, high Ct values by PCR might actually reflect contaminants amplified by the repeated PCR cycles needed to get a positive value; or perhaps some of those with high Ct values had very small amounts of dead/nonviable SARS-CoV-2 but their immune systems were able to control the infection early on and subsequently clear these remnants, leading to a repeated PCR test being negative???)

        -- some studies have suggested that the best correlation between PCR Ct and viral transmissibility is when Ct <27 instead of their 30 cutpoint (see http://gmodestmedblogs.blogspot.com/2022/01/covid-infectivity-of-variants.html ). And there is a clear relationship between the PCR Ct values and sensitivity of the Ag-RDT test as in the graphs above (there are overlapping confidence intervals, though especially so for omicron, and with the 48-hour lag; Ct of 27 does look like a better indicator of antigen positivity, and, i'd wager, the trend of the relationship between Ct values and a positive AgRDT test is highly significant)

    -- is this single positive test and subsequent negative one related to the timing of when the PCR happened to be done, since there was a 48hr lag between tests? see below in the limitations

    -- this question of the validity of a solo PCR positive is a particularly important clinical issue since:

        -- we in the community do not get a quantitative PCR result. the quantitative cutpoint for a positive vs negative is not clear and it may depend on the lab/may vary by different areas of the country (i'm not sure how standardized this is). for example, last year the Kansas Health and Environmental Laboratories used a Ct cutoff of 35 for a positive test, which is really pretty high (eg, see the graph above) and may substantially decrease the specificity of the PCR test as a measure of communicable SARS-CoV-2 infection (see https://www.coronavirus.kdheks.gov/DocumentCenter/View/1505/SARS-CoV-2-COVID-19-PCR-Ct-Cutoff-Values-PDF---1-7-21 )

        -- and, we in clinical practice do not routinely repeat PCR testing within 48 hours to see if the initial result even fell into the "isolated PCR group"

limitations:

-- no breakdown by the different types of Ag-RDT tests. were any of them more sensitive than the others?

-- there was no information on the number of people who developed symptomatic vs asymptomatic infections. so, no info on differences in outcomes. is the quantitative PCR different between symptomatics and asymptomatics? did the AgRDT sesitivity track as well in both groups?

-- did vaccination status matter? they noted similar numbers were vaccinated. but would be important to have more granular data/breakdown for both variants in how vaccination might have produced differing results

-- there was a study suggesting that nasal swabs were less accurate for omicron (see https://www.medrxiv.org/content/10.1101/2021.12.22.21268246v1 ). AgRDT for omicron did quite well in the above study. perhaps would have been even better if a salivary antigen test were done???

-- these were self-administered nasal swab tests. were there errors in technique limiting the results? what was the order done? were PCR swabs done first and that decreased pickup of the Ag-RDT tests that were only done soon thereafter? perhaps people were not even waiting the full 15 minutes between tests, as requested by the researchers. also, the PCR tests were assessed in a standard lab, but the AgRDTs were read at home. were there more errors by the latter self-reported assessments?

--would the results have been different if the repeat testing were done in 24 hours? some of the suggestions for Ag-RDT testing is to repeat in 24 hours to improve accuracy. And the omicron variant may be more accurately detected with sooner retesting: more rapid rise in viral load but lower peak and quicker clearance

so, overall a reassuring study to me that the antigen tests have pretty high sensitivity, especially with serial testing. and this seems to be true for the omicron variant. we will certainly need to continue to monitor the antigen test sensitivity (and specificity) for upcoming variants, as soon as these variants develop, to assure us that our current antigen tests are still working effectively 

geoff

 

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@uphams.org

 

to get access to all of the blogs (2 options):

1. go to http://gmodestmedblogs.blogspot.com/ to see them in reverse chronological order

2. click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

3. or you can just click on the magnifying glass on top right, then  type in a name in the search box and get all the blogs with that name in them

 

or: go to https://www.bucommunitymedicine.org/ , a website from the Community Medicine section at Boston Medical Center.  This site does have a very searchable and accessible list of my blogs (though there have been a few that did not upload over the last year or two). but overall it is much easier to view blogs and displays more at a time.

 

 

please feel free to circulate this to others. also, if you send me their emails, i can add them to the list