How Does Hydroxychloroquine and Azithromycin Combination Therapy Measure Up For COVID-19 Treatment?

Segment Description: Jason Pogue, PharmD, BCPS, BCIDP, clinical professor of infectious diseases, at the University of Michigan College of Pharmacy, speaks with Contagion®’s Senior Editor Michaela Fleming to share his thoughts on the recently published study “Hydroxychloroquine and azithromycin as a treatment of COVID-19: results of an open label non-randomized clinical trial.”

Interview transcript (modified slightly for readability):

Contagion®: Thanks for joining me to talk about this really important topic. We're going to jump into the article that was published in the International Journal of Antimicrobial Agents about hydroxychloroquine and azithromycin as a treatment for COVID-19. Why don't we talk about this? It's getting a lot of attention. We can start with you telling us a little bit about the study and what exactly the investigators were looking at here.

Pogue: Yeah, absolutely. And thanks for having me. Pleasure to be here.

Just to orient, this study included 36 patients, who were SARS-CoV-2 positive. These weren't necessarily infected patients, a few of them were asymptomatic. Some had upper respiratory tract infections, some had lower respiratory tract infections. There were 2 groups in this study; 20 patients got hydroxychloroquine and 16 patients got standard of care. So just supportive management, no therapy that was actually directed toward the viral infections.

The authors were investigating was the impact of hydroxychloroquine. The dose that they gave in this analysis was 200 milligrams every 8 hours. I think it's notable that this analysis actually isn't over yet. This is just some interim data. What they were looking was the impact of this therapy on clinical outcomes. So things like clinical resolution, mortality and safety, and viral eradication. This publication focuses on the viral eradication piece. So that kind of came in 2 different buckets in this publication.

One is just at their defined end point of 6 days, do the patients eradicate the virus? And then the second is of time to resolution or time to eradication -- so was anything associated with quicker eradication? I think it's important for the audience to note that this study is still ongoing. They're still working on these data. They actually did not present any of the clinical data. But, due to what they described their results as being so significant, they reported on interim data on eradication rates in these different patient groups. And so I think it's important to keep that context in place.

Contagion®: Okay, so significant results sound like a good thing. What exactly did the authors demonstrate?

Pogue: yeah, we're definitely looking for significant something right now.

So their timeframe was a 6 day study time period. What they found was that on study day 6—the last endpoint that they looked at, in this analysis—70% of patients who received hydroxychloroquine, had a negative PCR, which means they eradicated the virus, at least for the test that was used in the study. So, 70% of patients, compared to only about 12% of patients who didn't get any therapy. Even though they were small numbers, it was actually a statistically significant finding.

Contagion®: So that sounds good for the hydroxychloroquine. But what about the azithromycin? Where does that come in?

Pogue: Right, the title of the manuscript is hydroxychloroquine plus azithromycin. I think this is where a lot of the debate and where a lot of the interest comes from. So again, remember we're starting to talk about small subsets of patients when we do this, but in their analysis, there were 6 patients that received azithromycin. They described that for either treatment or for prophylaxis of bacterial infection. There was just a subgroup of hydroxychloroquine patients that also got azithromycin in combination, there was actually only 6 patients. However, at study days 6—remember that was the primary end point that they were looking at—all 6 of those patients had cleared the virus from their nasopharyngeal sample again, so they've eradicated the virus, compared to 8 out of 14 or 57% of those who are on hydroxychloroquine alone. Even though you're starting to dice up small numbers, you saw this 100% versus 57%. So you can see why a lot of excitement came from that.

In addition to that, there's a really nice figure in the manuscript. I think this is what really got people interested. Because they were getting daily swabs on these patients, they looked at eradication rates on day 1, 2, 3, and so on and so forth. What you can really see is that around day 3, we really start to see this apparent demarcation between monotherapy with hydroxychloroquine and then the combination from that standpoint. So, if these data hold true, and if these data are accurate, that's a huge implication, right? You're eradicating virus.

So a couple things with that. One, hopefully the patients will improve, although they did not look at clinical outcomes in this study. But in addition to that, I think it's important that there's potentially a bigger impact on the global spread of this. There's a lot of potential things that could be there, so that's why on the surface, that becomes a really important finding in this study.

Contagion®: So final thoughts on this, should patients be receiving this combination therapy then?

Pogue: Yeah, so simply put, my answer would be no. But I think that I have to elaborate on that a little bit. Although these data look good on the surface, I think there are a number of limitations to this data set that we need to talk about. Again, we have to be able to critically assess these data as they come out.

The first thing that I would comment on is this. Remember I said at the beginning that there were 20 patients in the hydroxychloroquine group, and 16 patients who got standard of care. But, if you actually look at the methods in the study, you'll see that there were actually 26 patients who got hydroxychloroquine. For this analysis, remember, the primary thing was eradication at day 6. In order to be assessable, they still have to be analyzable at day 6. There were actually 6 patients in the hydroxychloroquine group that dropped out of the study because they were no longer assessable.

I think it's important for viewers to note that if you actually look in the methods, these 6 patients, there were 3 of them that were taken out of the study because they decompensated and went to the ICU. Every single one of those patients was still PCR positive. There was 1 patient that passed away, that was actually PCR negative, but passed away. There was 1 that went home who was actually PCR negative. Then 1 actually stopped due to nausea—a side effect from the hydroxychloroquine—who was still PCR positive as well. In my opinion—although I understand why the authors did what they did—at least 5 of these patients, I would consider failures. You have patients go into the ICU, you have a patient that passes away, you have someone who had to stop due to an adverse event.

If you think about the impact that could have... remember, we start off and we said 70% versus 12%, that 70% would come down a little bit. It still might be a significant difference, and I ran the numbers, it still is a pretty nice difference. But I think it's important to note that that's 1 piece of the story. And the second piece related to that is we don't know what which of those patients were on azithromycin, which weren't on it and so whether that impacted the combo/mono thing is really unclear. That's 1 really important thing.

The second one, though, in my opinion, is much more important. That's really getting down to the azithromycin finding. This is what's gotten people really interested and excited. First and foremost, if you look there's a really nice table. One thing that I really like about the publication is there's they really put the data out there for us to look at. So right now, we're no longer comparing to the control group, we're talking about the comparison between patients who got hydroxychloroquine monotherapy, and those who got that combination.

If you look at that, the first thing that jumps out is that there are some virological differences between the 2 groups. Let me explain what I mean by that. I won't get into numbers and cycle thresholds or anything like that, because it's not going to add a whole lot to this conversation.

But bottom line, what you'll find is that there's a subset of patients in the hydroxychloroquine monotherapy group that had higher viral loads. Okay, so just logically, those patients would require more out of a drug therapy to be able to eradicate the organism in that situation. So when I look at these data, and we assess these data, if you take out those patients who have higher viral loads, and you try to make the monotherapy and combination therapy more on even terms, what you'll find is that the differences between the 2 completely go away.

Hydroxychloroquine monotherapy does really well in that situation. You'll see very similar eradication rates at day 6. But, remember that we talked about before that we were looking at that time to clearance on day 2, on day 3, and day 4 and so on and so forth. You'll see that they pretty much over overlap.

I think there's a huge limitation that the benefit of combo over mono might just be driven by differences between the patient population. When you get rid of that, when you kind of limit it to like patients, at least from a biological standpoint, the difference goes away. So I think that viewers need to think about that and read and critically think through that process. When you see that type of a finding, then you have to kind of keep walking down that pathway, right? The next one is what I expect azithromycin to help out in this situation?

The bottom line is there's no evidence outside of this publication that would suggest that there's going to be some bang out of giving azithromycin to our patients. There's no in vitro data against any coronaviruses, let alone SARS-CoV-2. There's no clinical data. There's no animal data.

When you start to walk through that process, in my opinion, it gets really hard to get to a logical place that azithromycin is playing a role here. Especially when you limit the study to like patients, there's absolutely no difference between the 2 groups. If you kind of keep walking that story through, [there's] not a real likelihood of benefit, and then you have to think is there a risk if I give my patient both of these therapies in combination? And I would argue that there absolutely is.

Both drugs can be associated with QT prolongation. If you give them both, that's an additive adverse event that you could see in a patient. So without a real good reason to think that there is a benefit. Again, there's no other data that supports it.

We talked about the limitations and what happens when you compare like therapies. It's my opinion that there's no rationale to give this combination right now, and the risk absolutely outweighs the benefits. I think that's an important thing to think of when you're assessing how to apply these findings to our patients.

Now, there are a couple other limitations to this publication as well related to the detection limits, related to the fact that the controls were at 1 site and they seem to be tested a little bit differently, whereas the actively treated patients were at a different site; very small numbers, and so really not a lot to go on there.

In my opinion, there's nothing in this study that really supports that combination therapy would be a good thing. That being said, despite all of that stuff we just talked about, if you take out the combination therapy piece of the story, I think the data are encouraging for hydroxychloroquine. Again, when you use those patients who didn't have that higher viral load, success rates in both arms are pretty good. Now, again, there are some limitations to these data too. They should not be taken as definitive.

Remember, we're just talking about viral eradication at different time points. We don't know if there's a clinical meaning to that. We don't know if these patients tolerated the drug. There's a lot of things we don't know. But I think it's fair to say that it's at least encouraging that there might be something there with this therapeutic option for our patients. I think it's an important way to think through the data.

Contagion®: Thank you so much, Dr. Pogue for that very thorough analysis. We always appreciate you taking the time to talk with Contagion.

Pogue: It's my pleasure. I love what you guys do.