Click here for free access to our latest coronavirus/COVID-19 research, commentary, and news.

Support nonprofit science journalism

Science’s extensive COVID-19 coverage is free to all readers. To support our nonprofit science journalism, please make a tax-deductible gift today.

George Thompson

University of California-Davis Health

Did an experimental drug help a U.S. coronavirus patient?

On 26 February, what seems like ages ago in the ongoing pandemic, the University of California Davis Medical Center in Sacramento finally got RNA test results confirming that a critically ill patient it had been treating for a week had coronavirus disease 2019 (COVID-19). (It took 4 days before the U.S. Centers for Disease Control and Prevention agreed to evaluate the patient samples because the person did not meet the strict criteria the agency then had in place, and 3 more days for the result to come back.) The patient, who for privacy reasons UC Davis physicans refer to only as they but who has been described as a woman by California's governor and state health officials, was the first likely case of U.S. community spread detected, meaning that the source of her infection was not known: She had had not traveled outside the United States to an infected area or been in known contact with a confirmed case.

The difficulty the medical center faced acquiring a test for its patient received widespread media scrutiny, but her fate largely escaped notice: After her condition declined, the UC Davis doctors secured what’s known as compassionate use permission from the Food and Drug Administration to test an experimental drug on their patient outside of a clinical trial. The drug, remdesivir made by Gilead Sciences, is given by an intravenous drip. Several randomized, placebo controlled trials of remdesivir for COV-19 are now underway in China and the United States and everyone is looking for quick hints on whether the drug works—a new preprint out today on the drug’s use in three COVID-19 patients is raising questions about its ultimate value.

Remdesivir cripples an enzyme called RNA polymerase that is used by many viruses to copy themselves; it does not specifically target SARS-CoV-2, the virus that causes COVID-19.  But it worked well in test tube and animal studies of human coronaviruses, cousins of SARS-CoV-2 called severe acute respiratory syndrome and Middle East respiratory syndrome, that cause similar respiratory conditions. (Ebola also is an RNA virus, but a test of remdesivir in the Democratic Republic of the Congo last year showed that it didn’t work for that disease.) 

George Thompson, an infectious disease specialist at the medical center, was on the team that cared for the California patient and spoke to ScienceInsider about the case. This interview has been edited for clarity and length; additional information added by ScienceInsider is in brackets.

Q:  When did the patient start on remdesivir?

A: From diagnosis to therapy, about 36 hours, which is very short for emergency approval of an investigational drug.

Q: How sick was the patient?

A: We thought they were going to pass away.

Q: Did you do extracorporeal machine oxygenation (ECMO)? [This is a type of artificial lung that removes blood, pumps oxygen into it, and then returns it back to the body. It has been used extensively in China to save some critically ill COVID-19 patients.]

A: We had started those conversations to see how we would do it logistically, and they were very close to needing it, but their health started to turn around. We have some severe influenza cases every year that end up on ECMO, but nobody likes to do it. Those patients are so sick and they require a lot of time, a lot of resources. But we do it if we need to. And we were willing to do it for that patient.

Q: One truism of antivirals for acute diseases is that if you start them late, they don't work. Do you think that the point at which you gave her the treatment was early enough for it to have worked?

A: I think so. One requirement was that the patient had to have a positive test [showing the presence of SARS-CoV-2] right before starting the drug to make sure that they hadn’t spontaneously already cleared the virus. The day after the infusion of the drug, they consistently got better. I can't prove it's related. I wish we had been able to do serial [polymerase chain reaction] PCR testing of their blood, but we couldn't because of lack of resources. With most investigational drugs tested in, say, macaque monkeys, there’s a nice correlation between the administration of the drug and a drop in the amount of virus in the blood. That's what we hoped we could have seen in this patient.

[PCR can amplify a tiny amount of viral RNA so that it can be detected and is the basis of most SARS-CoV-2 tests.]

Q: What do you mean by a lack of resources?

A: The county only had 20 coronavirus tests for the whole county. That patient already had a confirmed infection and was not a priority for using them.

Q: Couldn't you take serial blood samples, store them, and look at them later?

A:  We have all those samples saved in our local hospital and we’re going to be able to do the testing next week from what I've been told. We finally have our own home-brewed [SARS-CoV-2] test up and running and I think that'll tell more of the story. If the drops in their viral loads are temporally related to remdesivir, it's a much more compelling argument than, They lived, so it must have worked. But then again, I've heard stories that many COVID-19 patients just have the virus in their respiratory samples and not their blood.

Q: When would you expect to see that drop in viral load if it was due to remdesivir?

A: Within the first 24 hours you really want to see a [big drop]. But I think we've got a lot to learn about this disease.

Q: Has the patient been discharged?

A: To protect privacy, let’s just say they are doing well.

Q: Have you treated other patients at UC Davis Medical Center who have confirmed COVID-19?

A: Yes, but they had moderate disease and it was mild enough for them to go home. None would have qualified for compassionate use of remdesivir, which is only given to patients with severe disease.

Q: That’s a conundrum though because remdesivir has the best chance of working in patients who are treated early, before the disease becomes severe.

A: For most any infectious disease, I think the earlier we start drugs the better. But it’s a risk versus benefit question. What if this drug causes liver toxicity in 50% of the people, and we've given it to somebody who was probably going to do well without it?  

[Clinical trials underway will look at all stages of disease.]

Q: Are you planning to test other experimental drugs for COVID-19?

A: We've already been contacted by a whole bunch of companies with different compounds. So, I think we'll be able to offer most anybody with confirmed COVID-19 some experimental protocol in a clinical trial. There’s a lot of interest in chloroquine, and interferon with Kaletra.

[Chloroquine is an antimalarial, interferon is an immune system messenger, and Kaletra is a combination of two protease inhibitors used to treat HIV infection.]

Q: Was what you saw frightening?

A: Well, it can be severe. My research lab works on a pathogen, Coccidioidomycosis fungus, that must be studied in a BSL-3 [biosafety level 3] lab. So I'm pretty used to wearing all the gear. [BSL-3 is the second strictest level of containment.]  There's a lot of fear in the hospital. Nurses need to talk a lot about it. And respiratory therapists want to talk a lot about what are you going to do for a patient who’s on a nebulizer [a machine that turns liquid medicines into a mist].

Q: Many people talk about this disease with great authority based on reports they’ve read. My sense of this virus is it's really nasty for people who are older than 80, and for people who have heart conditions, diabetes, hypertension, but for most healthy people, it's going to be like a flu. It typically doesn’t do much of anything to kids. Is that your perception of this having seen the disease?

A: That’s my general perception based on the literature and what we know of the virology, but it's pretty early, right? We've had such poor testing capacity, I don't think we really have clinical experience with the breadth of the disease. Our first couple of patients have been, middle-age, healthy people. But again, you know, if you have some very minor version of it, you're probably not going to seek any care. So that side of the iceberg—I just don't think we'd have clinical experience with yet.

*Correction, 13 March: Due to a transcription error, an earlier version of this story said Thompson studies coxsackievirus.Thompson's references to the patient have been changed to they/their, his original language.