A weekly explanation of the emerging science behind COVID-19 and its infectious agent, SARS-CoV-2.
The evidence for chloroquine is pretty thin gruel
Chloroquine and the related drug hydroxychloroquine have been in the news a lot lately, promoted as potential allies or even cures in the fight against COVID-19. This is an excellent (and unfortunate) example of a grain of sand quickly snowballing into a house of worship.
We have known that this pair of drugs can harm viruses in the lab since the late 1960s, and there’s even been evidence that it has a similar activity in mice against viruses like H5N1 influenza and the Zika virus. But cells in a culture flask and laboratory animals are not the same as human bodies, and we start running into problems when we simplistically extrapolate virus-fighting powers from one model to another. For example, chloroquine and hydroxychloroquine can block the growth of the SARS virus in cultured cells but not in mice. The drugs do not work to prevent humans from contracting the flu, nor are they useful as a treatment for dengue. And then there’s the worrying paradox of what chloroquine does against the chikungunya virus. In Petri dishes, the drug looked promising, but it turns out that it makes the disease worse in primates and humans, possibly by slowing down part of the immune response, something which can’t be tested for with simple cells in the lab. A recent commentary on the use of chloroquine against COVID-19 summarized the drug’s virus-fighting abilities quite pithily: “To date, no acute virus infection has been successfully treated by chloroquine in humans.”
But surely, we are told, there is evidence for chloroquine and its cousin, hydroxychloroquine, being useful against COVID-19! If you’re looking for intellectual sustenance here, you will find a watered-down oatmeal. First, there was a research letter about the drug’s effect on cells in the lab, not in humans. Then a news briefing from the State Council of China mentioned positive data on over 100 patients but so far, these data have not been released. There have been expert opinion statements and guideline documents weighing in favour of the drug (as summarized here), but these pronouncements are not data points. At least twenty-three Chinese clinical trials on the use of chloroquine against COVID-19 are currently on-going but, to date, no result has been released. This leaves us with a pair of papers from a team in France. Scientists had a field day digging into these papers, some with fine scalpels to study their innards but most using chef knives to expose how spoiled the meat was. Dr. David Gorski of Science-Based Medicine wrote two detailed examinations of these studies, which I recommend to fellow academics who want to understand the extent of their ineptitude. Suffice to say that even in these critical days when speed prevails, you need garden shears to cut corners so clumsily. The lead scientist behind these studies, Didier Raoult, has himself come under scrutiny lately. He has apparently co-signed 3,000 scientific papers, a feat that raises many eyebrows. It has been reported that some of these publications contain questionable figures that may have been altered using image-editing software. Finally, some of his employees denounced his behaviour back in 2017, reporting threats, humiliations and “violent verbal altercations.” Scientific findings do not appear out of the void: as such, Raoult’s unsettling reputation casts doubt on his integrity.
As the FDA nonetheless approves the use of the drug to treat severely ill patients, it’s important to remember that this medication is not harmless. A man died after ingesting chloroquine phosphate sold as an aquarium cleaner. It is easy to overdose on chloroquine because the gap between the smallest therapeutic dose and a toxic dose is actually quite small. The drug is associated with a risk of heart muscle problems. It can also have negative effects on the central nervous system, it interacts with a long list of medications and it can, with long-term use, lead to irreversible damage to the back of the eye. As the drug continues to be championed as a “game changer” in the absence of solid evidence, we must also not forget that many patients with conditions like lupus and rheumatoid arthritis, for whom the drug has been prescribed, are already finding themselves unable to access their medication thanks to a drug shortage caused by hype. Desperate situations call for desperate measures, but let’s be clear: the evidence for (hydroxy)chloroquine for COVID-19 is thin gruel at the moment. Future studies, if done well, should clarify its status in the fight against this pandemic.
The virus appears fairly stable and the talk of two different strains is not accurate
A recent paper suggested there were two different strains of the coronavirus, L and S, leading many to speculate that one might be worse than the other and wondering if this meant the virus could mutate quickly and thus potentially thwart our attempts at containing it. The reality is both more reassuring and nuanced.
It turns out the word “strain” with regards to viruses means different things to different people. To many doctors, two strains of the same virus have to display unique, visible characteristics that are stable under natural conditions. Maybe one is deadlier than the other or its “keys” are better at opening the “locks” at the surface of our cells. By this definition, strains behave differently. But to many scientists interested in genetics, strains simply have different mutations, different letter changes in their blueprints, regardless of how they behave. So when we hear talk of different coronavirus strains, it’s easy to mistake information about the virus’ genetic code for information about the virus’ behaviour when it infects us. The two are not the same.
The good news is that, while over 1,150 genetically distinct strains of the coronavirus have been identified so far, these versions are remarkably similar to each other: an analysis of 95 of them showed them to be 99.99% identical at the genetic level. A fair number of mutations have been catalogued in all strains, but each strain only has up to 20 of these letter differences out of a total of roughly 30,000 letters. And before we panic at the mention of mutations, they are a normal part of living and many of them have no consequences. Our own DNA harbours mutations. Imagine being tasked with recopying a manuscript by hand. You would make a mistake once in a while. The enzyme that makes copies of a virus’ genetic blueprint similarly has an error rate. For RNA viruses like coronaviruses, it tends to be one to ten errors per 10,000 letters, which is considerably more than the enzyme our own bodies use to make copies of our DNA. But coronaviruses have the equivalent of correction fluid: like ours, their enzyme can go back and fix some of these mistakes. This is very useful because, with a large genome, coronaviruses would drive themselves to extinction if they made as many mistakes as other RNA viruses with smaller genomes.
While the virus could theoretically acquire a mutation that would mess with the vaccines being developed, or that would make it more infectious, or that would require a slight reconfiguration of the testing kits, or that would make it more resistant to an antiviral drug we might use against it, the virus appears to be fairly stable at the moment. This is reassuring news.
Take-home message:
- The evidence for the use of the drugs chloroquine and hydroxychloroquine to treat COVID-19 is deficient and the two French studies on their clinical use suffer from serious problems
- The coronavirus appears fairly stable and the talk of two different strains is not accurate
