The pandemic has introduced a lot of people to the idea of rapid antigen tests, which can quickly and conveniently reveal the presence of an infection. But in many parts of the world, rapid tests are a central feature of health care. If you don’t have easy access to a testing lab infrastructure—and many in the developing world don’t—rapid tests can provide a quick way of screening for common problems. In a number of countries, rapid test results are what determine whether people are given anti-malaria treatments or not.
But that may be causing a unique problem. A new paper suggests that the malarial parasite may be evolving so that it can’t be recognized by the most commonly used rapid tests.
Make it quick
Most rapid tests detect the presence of one or more proteins on the surface of a pathogen. We can mass-produce antibodies that recognize this protein and couple them to a molecule that is colored. When a pathogen isn’t present, the antibodies remain diffuse, and the color is imperceptible. When the pathogen is around, its protein and the antibodies aggregate, bringing enough of the colored molecule together that we can see it. The result is something like a red bar appearing at a specific location on the test hardware.
In the case of malaria, there are a variety of proteins that are recognized by different rapid tests. But arguably the most effective test kit recognizes two proteins, called histidine-rich protein 2 and 3 (pfhrp2 and pfhrp3). Because this test produces consistent results, a number of countries have made it their standard diagnostic for malaria. Those with symptoms are screened using it and, if the test comes back positive, they’re put on anti-malaria drugs.
One of the problems on basing a test on the presence of specific proteins is that we don’t really know what the proteins do normally. (This is hinted at by their name, which tells you what the proteins look like, rather than what they do.) And a number of malarial parasite strains have been isolated where the genes have gone missing, indicating they’re not essential for infections. So, it’s possible that some strains can be missed entirely by a commonly used test.
This problem is compounded by the fact that the tests are used to determine whether to begin anti-malarial drugs. These drugs typically kill the parasites, or at least stop them from reproducing. As such, they provide a selective pressure that can drive evolution. That selection can potentially produce resistance to the drug. But it can also provide selection against the proteins that caused it to be present in the first place—the ones recognized by the rapid test.
Sweeping away a gene
Ethiopia is one of the countries that uses a rapid test that recognizes the presence of pfhrp2 and pfhrp3. So, a team of Ethiopian researchers decided to find out whether the test was influencing the evolution of malarial parasites in their country. To do so, they collaborated with a number of US researchers, forming a large international team.
They recruited over 12,500 people with malarial symptoms and gave them two different rapid tests: one that recognized pfhrp2 and pfhrp3, and one that recognized a different protein. A total of 2,714 of these participants received a positive result from at least one of the tests. Among those, over 350 (13 percent) were negative based on the test that recognized pfhrp2 and pfhrp3, suggesting these individuals could have damaged versions of the genes that encode these proteins.
Checking the DNA, the researchers found that the genes were missing entirely—they’d been deleted at some point in the parasite’s past. As noted above, though, the genes don’t appear to be essential for infections, and so they could have been lost for reasons unrelated to the diagnostic test.
There are two reasons to think this isn’t the case. The first is that the deletions aren’t specific for the genes that encode pfhrp2 and pfhrp3. They’re big enough to take out nearby genes as well, and some of those do seem to be rather important. One of them binds to red blood cells, one of the cell types targeted by the parasite. Another is involved in allowing the parasite to get inside those red blood cells. So having these genes deleted would seem to be damaging to the parasite, which suggests the deletions wouldn’t be present unless they’re selected for.
The second reason has to do with the DNA surrounding the deletions. If the deletions have been around a long time, you’d expect new mutations and recombination between chromosomes to scramble the sequence nearby, causing that DNA to look similar to the DNA found in the same location in strains without the deletion. That was the case with the pfhrp3 deletions.
But the pfhrp2 deletions were different. Instead, the DNA near the deletions all looked the same, mostly different from that seen in strains that lacked the deletion. This indicates that the pfhrp2 deletions had arisen recently and hadn’t had time for mutation and recombination to scramble the DNA sequences nearby. So, all the parasites that carry the pfhrp2 deletion seemed to have inherited it from the same ancestor, that ancestor was very recent, and there was a selective pressure that was driving the expansion of parasites that inherited the deletion.
All of that is consistent with what’s called a selective sweep, where evolution drives part of a chromosome to high prevalence because it contains a favorable mutation.
There might be another aspect of evolution by diagnosis going on here. Remember that these deletions also take out proteins that may help the parasite infect cells? The researchers suggest that the loss of these might tone down a malarial infection so that it doesn’t cause the severe symptoms associated with infections that reach the brain. If that’s the case, then it may be that people infected with these deletions don’t seek a diagnosis, so they won’t even be given the rapid test in the first place.
While none of this evidence is decisive, it clearly suggests that our ability to diagnose the presence of a pathogen is influencing that pathogen’s evolution. Given that the diagnosis leads to treatment in this case, it’s clear that there’s a selective pressure that is consistent with this idea.
What does that mean for other rapid tests, such as the ones being used for the coronavirus? The answer is, unfortunately, “it’s complicated.” If people changed their behavior in response to the results of a rapid test, then it’s possible that the test could lead to a lower infection rate and thus create some selective pressure there.
There are some clear differences, though. The malarial parasite is a complex cell with lots of proteins on its surface; it can clearly survive losing a few of them. SARS-CoV-2 is a virus and has a grand total of two proteins on its surface, and it can’t lose either of them and still function. Rapid tests against it are based on antibodies recognizing the virus’s spike protein, so there’s a chance that changes to spike could make the test ineffective.
But spike is also the protein that enables the virus to infect new cells, and there are limits to how much change it can tolerate while still performing that function. Plus, at the same time, it’s under selective pressure to evolve ways of avoiding the antibodies made by the immune systems of those who are vaccinated or previously infected. Given all that, it’s not clear whether a diagnostic test will exert any significant influence on the virus’s evolution.
So, even if this risk does get confirmed with malaria, it may not apply to tests for other pathogens. Things will have to be evaluated on a case-by-case basis.