COVID: value of rapid antigen testing

 The CDC just published an evaluation of the Abbott BinaxNOW rapid antigen test for SARS-CoV-2 infection at 2 community-based testing sites in Arizona (see https://www.cdc.gov/mmwr/volumes/70/wr/mm7003e3.htm#T2_down )

 

Details: 

-- the BinaxNOW rapid antigen test received FDA Emergency Use Authorization 

    --it is advertised by Abbott as a "fast, $5, 15-minute, easy-to-use Covid-19 antigen test", and that with FDA EUA, is "ramping production to 50 million tests a month", and is the size of a credit card

-- 3419 specimens were collected from people >10 yo, during November 

    -- age range 10-95yo, with median age 41, 7% age 10-17/55% aged 18-49/22% aged 50-64, and 16% >65yo

    -- 38% male, 75% white/2% Black 

    -- asymptomatic 76%, symptomatic 24% with 19% having symptoms for <8 days: 11% 0-3 days, 8% 4-7 days 

-- symptomatic people: 14% had a positive antigen test result, 20% had a positive PCR 

-- asymptomatic: 2% had a positive antigen test, 5% positive PCR 

-- viral culture was performed in 274 of 303 PCR-positive specimens 

 

Results: 

-- comparing the BinaxNOW rapid antigen test with PCR: 

    -- asymptomatic patients: sensitivity 36%, specificity 99.8%, positive predictive value (PPV) 92%, negative predictive value (NPV) 97% 

    -- symptomatic patients: sensitivity 64%, specificity 100%, PPV 100%, NPV 91% 

-- virus was culturable from 96 of the 274 specimens (35%) including:

    -- 85 (58%) of 147 with concordant rapid antigen and PCR positive test results 

    --11 (9%) of 124 with false-negative antigen test results

        --of these 11 people with false negative antigens (all had positive PCR and cultures), 5 were symptomatic and 6 asymptomatic

    -- none had false-positive antigen results

-- for the specimens with a positive viral culture:

    -- asymptomatic patients: sensitivity 79% 

    -- symptomatic patients: sensitivity 93% 

 

-- for symptomatic people who were within 7 days of symptom onset,comparing rapid antigen test to PCR: sensitivity was 71%, specificity 100%, PPV 100%, NPV 93% 

-- for asymptomatic people: there were 4 false positive antigen tests as compared to PCR, and 142 were false negative antigen tests (63 from symptomatic and 79 from asymptomatic people)

-- for those with false negative antigen tests: median PCR Ct values were much higher (suggesting lower RNA viral loads), than those with concordant values, with median Ct values:

    -- symptomatic people: 34 vs 22 (p<0.001) 

    -- asymptomatic people 34 vs 23 (p<0.001) 

Commentary:

-- rapid antigen tests such as the above are inexpensive and have results within 15 to 30 minutes. These have received FDA EUA approval for symptomatic patients 

-- the BinaxNOW rapid antigen test detects viral protein, requires >40,000 SARS-CoV-2 viruses to detect the presence of virus 95% of the time (ie, is 100 times less sensitive than PCR) ( see https://newsroom.uw.edu/news/rapid-antigen-test-far-less-sensitive-lab-standard )

-- although the rapid antigen test has a lower sensitivity to detect infection, it does have high specificity and a high positive predictive value in settings of high pretest probability. So, this test is especially useful in communities with high prevalence of SARS-CoV-2, or in specific patients with high pretest probability of infection (symptomatic, or close contact with covid-positive person)

--and, the rapid results are really helpful in quickly isolating infected people, decreasing their viral transmission to others. 

--because these tests are easily done (eg can be self-administered) and cheap, serial testing is a real possibility (eg 1-2x/week), which should improve the test sensitivity for picking up infections

--one concern with PCR is the fact that it can be positive for dead viral remnants, or even contaminants and be positive. There is a statistical cutpoint for the PCR quantitation suggesting real virus, with the cycle time (Ct) typically below the 35-40 range.  Much lower Ct’s reflect higher viral loads and are more likely to be true positives. Unfortunately, we in clinic-land (vs research-land) are only getting “positive” or “negative”, without a quantitation or even what the cutpoint is. Makes it much harder to interpret the results....

    --for example, PCR can be positive for months after symptomatic infection, but no study so far has found culturable live virus >9 days after symptoms, despite persistently high viral loads (see https://www.thelancet.com/journals/lanmic/article/PIIS2666-5247(20)30172-5/fulltext). the article reporting the lowest PCR Ct did find that the it was always >24 (which is still a very high viral load) in one study (see https://academic.oup.com/cid/article/71/10/2663/5842165), though it was > 34 in other studies [as a point of reference, in the above rapid antigen study, those with false negative antigen tests had Ct values of 34, those with concordant results it was 22-23]

--a recent article made the strong argument that testing frequency and turnaround time are more important than test sensitivity, using mathematical modeling: see https://advances.sciencemag.org/content/7/1/eabd5393 , or covid antigen test sensitiv less impt than testing freq sciadvances2021 in dropbox.  serial testing also has the advantage of being able to pick up virus at different times, including when there are higher viral loads. the test's lower sensitivity may relate to missing lower levels of virus that are less likely to be transmissible

    --there was a poignant post by Michael Mina (one of the authors in this article suggesting that test sensitivity is less important), comparing PCR with with rapid antigen, noting that they serve very different functions:  PCR is a very sensitive test (to the point of picking up non-transmissible virus) that is useful for a clinician wanting to see is that patient has Covid-19; rapid antigen test is more useful for getting public health information to help stop the spread of the virus. see https://emilyoster.substack.com/p/antigen-testing-guest-post-with-michael?token=eyJ1c2VyX2lkIjozNzczMTI2LCJwb3N0X2lkIjozMTEwMTc5NCwiXyI6ImRwa3ZiIiwiaWF0IjoxNjExMTk1MTc5LCJleHAiOjE2MTExOTg3NzksImlzcyI6InB1Yi0yOTc3MCIsInN1YiI6InBvc3QtcmVhY3Rpb24ifQ.GapZ92623EquGdaFcFIKeO1d_DwoXVzbL9Kgaa1TYf0 

 --it was notable in the rapid antigen study above: the sensitivity of the rapid antigen test was significantly higher when comparing it to viable, culturable virus than with PCR. likely because the antigen test requires lots of live (contagious) virus to produce a positive result (per Michael Mina)

Limitations: 

--Anterior nasal samples were done for the rapid antigen test but nasopharyngeal for PCR (the latter perhaps a bit more sensitive)

--symptomatic patients may have actually been asymptomatic for SARS-CoV-2 but had another infection causing the symptoms  (eg a different virus): ie not necessary a coherent cohort

--the above results apply only to this specific rapid antigen test 

--viral culture can be finicky (depends on the specific viral sequence, or the conditions such as storage and freeze-thaw cycles).  So, a negative culture does not necessarily mean that there was no virus there.  But a positive viral culture is likely very significant 

 

--The CDC suggests that a negative antigen test result in a person at elevated risk (symptomatic or known exposure to the virus) should be confirmed by a highly sensitive nucleic acid amplification test (NAAT)

    --NAAT tests detect genetic material (eg RNA) from the virus, not antigens or antibodies, and can do so before antigens and/or antibodies appear in the blood stream

--and, an asymptomatic person with a positive antigen test but has a high risk for adverse consequences resulting from a false positive (eg, in a long-term care facility), should get confirmatory NAAT testing

so, a pretty strong argument that the relatively insensitive rapid antigen test, when done frequently, is the way to go in terms of rapidly identifying those infected, allowing for rapid isolation and thereby diminished viral spread. and the whole approach of patient self-testing is overall great: empowering patients to understand what is going on in their bodies and able to take direct action to help their families/friends/coworkers/society overall.  these self-tests may also be really useful for those vaccinated to be more certain that they do not have asymptomatic infections (though, would still advise distancing, masks, handwashing. just in case...)

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