Posted on by Dr. Francis Collins
During the pandemic, it’s been critical to track in real time where the coronavirus is spreading at home and abroad. But it’s often hard for public health officials to know whether changes in the reported number of COVID-19 cases over time truly reflect the spread of the virus or whether they are confounded by changes in testing levels or lags in the reporting of results.
Now, NIH-funded researchers have discovered a clever workaround to detect more accurately where COVID-19 hotspots are emerging. As published in the journal Science, the new approach focuses on the actual amount of virus present in a positive COVID diagnostic test , not just whether the test is positive or negative. What’s even better is these data on a person’s “viral load” are readily available from polymerase chain reaction, or PCR, tests that are the “gold standard” for detecting SARS-CoV-2, the virus responsible for COVID-19. In fact, if you’ve been tested for COVID-19, there’s a good chance you’ve had a PCR-based test.
Here’s how a PCR test for COVID-19 works. After a person provides a nasal swab or saliva sample, any genetic material in the sample is extracted and prepared for the PCR machine. It uses special nucleic acid primers that, if any genetic material from SARS-CoV-2 is present, will make millions more copies of them and result in a positive test result. PCR is an enzymatic reaction that works by running many cycles of heating and cooling; each cycle results in doubling of the genetic material present in the original sample.
But it turns out that PCR can go beyond a simple “yes” or “no” test result. It’s also possible to get some sense of how much coronavirus is present in a positive sample based on the number of cycles required to make enough copies of its genetic material to get the “yes” result. This measure is known as the “cycle threshold,” or Ct, value.
When a sample is run with lots of virus in it, the PCR machine doesn’t need to make so many cycles to reach detectable levels—and the Ct value is considered low. But, when the virus is barely present in a sample, the machine needs to run more cycles before it will reach the threshold for detection. In this case, the Ct value is high. This makes the Ct metric a bit counterintuitive: low Ct means a high level of infection, and high Ct means a low level of infection.
In the new study, researchers in Michael Mina’s lab, Harvard T. H. Chan School of Public Health, Boston, including James Hay and Lee Kennedy-Shaffer, wanted to use Ct values to understand better the overall trajectory of the spread of SARS-CoV-2. Their idea was a little out of the box, since Ct values weren’t being factored into a diagnostic testing process that was set up to give people a yes-or-no answer about COVID-19 status. In fact, Ct values were often discarded.
The team members had a hunch that the amount of virus in patient samples would vary based on whether an outbreak is increasing or declining. Their reasoning was that during an outbreak, when SARS-CoV-2 is spreading rapidly through a community, a larger proportion of infected individuals will have recently contracted the virus than when it is spreading more slowly. The researchers also knew that the virus reaches its peak level in humans soon after infection (generally a couple of days before symptoms begin), and then falls to very low but still detectable levels over the course of weeks or sometimes even months. So, when viral load within samples is highest—and Ct values are lowest—it suggests an outbreak of SARS-CoV-2 is underway. As an outbreak slows and cases fall, viral loads should fall and Ct values rise.
The researchers found that just 30 positive PCR test results on a single day were enough to give an accurate real-time estimate of the growth rate of SARS-CoV-2 infections based on Ct values. With Ct values from multiple time points, it was possible to reconstruct the epidemic curve and estimate the true number of people infected. They found that even Ct values collected from a single location at a single point in time could provide extremely valuable information about the growth or decline of an outbreak.
The findings suggest that these data can now be captured and put to good use as a key metric for decision-making and gauging the success of the pandemic response going forward. It’s also important to note that the value of these data are not unique to COVID-19 and the ongoing pandemic. It appears this can be extremely useful new way to monitor the course of other viral outbreaks, now and in the future, in a way that’s less susceptible to the vagaries of testing. The hope is that this will mean even greater success in capturing viral outbreaks and mobilizing resources in real time to the places where they are most needed.
 Estimating epidemiologic dynamics from cross-sectional viral load distributions. Hay JA, Kennedy-Shaffer L, Kanjilal S, Lennon NJ, Gabriel SB, Lipsitch M, Mina MJ. Science. 2021 Jun 3.
COVID-19 Research (NIH)
Michael Mina (Harvard T. H. Chan School of Public Health, Boston)
NIH Support: Common Fund, National Institute of General Medical Sciences; National Cancer Institute