Dr. Herazo-Maya talks about blood tests that can predict the severity of COVID in patients and explains how this application is unique to treating COVID.
What are the biggest like problems with treating COVID?
HERAZO-MAYA: With COVID the main issue that we’re having, at least from my standpoint, is trying to recognize who is the patient, who’s going to get really, really sick and try to distinguish that from the patient who is going to do fairly good in their hospital course and be able to be discharged home in good health. So, I think that’s the main issue we deal with is how to distinguish those group of patients. And because obviously the patients who will develop a rocky course, so to speak, they tend to stay in the hospital for weeks. They get lots of complications. What patient is going to get very sick from COVID from the patient who’s going to do really well during their hospitalization is the main challenge, to my point of view. Because unfortunately, the patients who will get very sick will need ICU, will need mechanical ventilation, and may need to stay in the hospital for a number of weeks. Those patients would require more care. So that’s the trouble I’m personally facing is how do you distinguish those two classes or subgroup of patients?
And right now, are you doing that by determining other conditions?
HERAZO-MAYA: Correct. Yeah. So right now, we’ve seen the Hispanic patients, for example, or African Americans, they tend to do worse if they have comorbidities. But there’s not a real good predictor of, you know, this phenotype of patients.
But you’re trying to change that.
HERAZO-MAYA: Correct. Exactly.
And how are you doing that?
HERAZO-MAYA: We analyze the expression of genes of the human genome in each one of the patients. The human genome has about thirty thousand genes. So, what we came up with is an analysis to try to distinguish the genes that can predict what are the patients who are going to progress and die from the disease.
And this has a resemblance that COVID does to lung disease.
HERAZO-MAYA: Correct. So, when the pandemic started, I noticed that COVID patients, the very sick ones, ended up with very scarred lungs. And in the past, I’ve studied idiopathic pulmonary fibrosis, which is a fibrotic lung disease or scarring in the lungs. So, the hypothesis was, you know, we didn’t know anything about COVID, right? So, if we could use the knowledge of IPF and apply it to COVID-19 to try to expedite the development of a good predictor test or maybe even treatments, then that would have saved lives, right? And time and money, too. So, we use that knowledge in IPF that was extensively validated and applied in COVID-19.
With the IPF, how many genes do you find? Because I think I read it’s like 50 genes for COVID, right?
HERAZO-MAYA: So, in idiopathic pulmonary fibrosis or IPF, we found about fifty-two genes that predicted mortality. And although this was 2017, so for COVID-19, we use about 50 of those fifty-two. So, we use basically just two genes less. And what was very impressive is that the performance was exactly the same in both diseases that are completely unrelated, right? So idiopathic pulmonary fibrosis is not caused by a virus, but COVID-19 is caused by a virus. So, the performance is the same. Essentially, you can identify a risk group of patients who are going to die from COVID-19 or are going to need intensive care unit admission or mechanical ventilation from those patients who will, you know, have a normal hospital course and be able to discharge home.
And this is by a simple blood test. How does it work?
HERAZO-MAYA: Yeah. So, this is a very simple blood test. Basically, it’s a phlebotomy. And we bring the blood to the lab. We extract RNA, which is a byproduct of DNA or the genome. And then we analyze the expression of those 50 genes.
And is it all 50 genes have to match? Or is it 20 percent of those genes or … ?
HERAZO-MAYA: Right. So, one of the things that we’re trying to do in the study we published this summer, we used all the 50 genes. But preliminary data on the follow-up study suggests that maybe we could use less. All right? Because the idea is that we can use this blood test and apply it in any country in the world. You know, analyze the expression of 50 genes will require some sort of a more complicated or sophisticated technology. But the idea is to pack it up in maybe a smaller number of genes so that we can do it in Africa, in South America, anywhere.
With your study, how accurate was it?
HERAZO-MAYA: So, we measure it in something called the area under the curve. This is about point seven three percent, it’s about 73 percent accurate to predict mortality, which is a pretty good number. Seventy percent is acceptable. Point seven three percent is within the range that we expect. So basically, out of every four patients, you can predict mortality correctly in three.
How would your treatment change for someone and when would you give this test and how would your treatment change?
HERAZO-MAYA: Essentially, anybody with COVID could have the test whether it’s in the hospital or at home. Right? So now the study was focused on hospitalized patients. So, let’s talk about the hospital. If a patient comes to the hospital, we do the test. And we see the patient is in a lower risk group, then we know that the patient can be maybe watched for maybe a day or two or discharged home from the E.R. because you have a good idea that the patient is going to do well based on the gene test. If the patient shows a high-risk profile as a patient that you right away want to send to the intensive care unit for close monitoring. And basically, you can allocate resources to those group of patients. For example, you need to have ventilators available because you know that for the blood test predicts also a high use of mechanical ventilation. And one of the things that we have not done well, I think, in COVID-19 is that the clinical trials use the same drug with the same dose for everybody, right? So, this is a principal of something called precision medicine or personalized medicine where not everybody should be treated the same way. If you have two patients with the same disease but one will do poorly, one will do well, well, the clinical trials should be done considering that some patients will do worse than the others because otherwise the effectiveness of the drug will be different, right? I think that the blood test could also help identify or manage using the clinical trial in a different way.
So how have you seen this work? Have you seen a patient come in and be not so bad, be predictive of being in that three out of four people?
HERAZO-MAYA: Yeah. So, we’ve seen it. And it’s very disheartening. Unfortunately, a lot of times we see, you know, young unvaccinated people that we do the blood test. We actually do some preliminary analysis. And we can see the risk profile. But unfortunately, we can’t act on it. But we see how it correlates, unfortunately, how the patient tends during the hospitalization. So, you definitely see it.
And so how fast do you think this could get out into the real world and be used?
HERAZO-MAYA: We’re in a phase of validation. So, in science research, you’ve got to research again and again and again. The initial study was a retrospective analysis. So now we’re doing it in a prospective fashion. We collect the samples. We follow the patient. And we need to reproduce it to make sure this is valid. I mean, the study was published. And it shows that it’s valid, but we want to be comfortable that the findings are there. And like I told you, we’re trying to pack it in a smaller amount of genes so that we can do it in basically in any country.
Now just to put people’s minds at ease because I know this will be a question that pops up. They’ll say, oh. Well, if it shows I’m definitely off the charts going to die, am I just not going to be cared for?
HERAZO-MAYA: Well, not really. I mean, it’s the other way around, right? If you have a high-risk profile, that means that we have to be more aggressive with your care and not the other way around, right? So, I wouldn’t necessarily see it that way.
Does it come down to like maybe finding immune cells as well that has changed because of covid?
HERAZO-MAYA: Correct. Exactly. So, the blood test, what we do is we measure the expression of these genes in immune cells. And unfortunately, what happens with COVID patients is some of the patients have this very exaggerated immune reaction to the disease, right? And that’s why certain patients don’t have that exaggerated immune reactions and do well. And some of the patients have this exaggerated immune reaction. So, the test can identify the immune cells that basically trigger this exaggerated response.
Anything I’m missing?
HERAZO-MAYA: I don’t think so.
I think it’s pretty exciting because everything is about personalized medicine. And then we have this vaccine that pops up out of basically nowhere. And already we’re getting it to the point where we want to treat the individual and not, you know, it seems like it’s taken decades for that for cancer or other things but we’re able to do that so much quicker now.
HERAZO-MAYA: Right. Correct. I mean, I think that the key is to create also awareness that not everybody with COVID has the same response to the disease and not everybody should be treated the same way. And if we can identify those who are not going to do well before they decline or decompensate and be very aggressive, maybe using higher doses of the treatments, maybe get them earlier into clinical trials will change the pandemic, right? It will save countless of lives. So, let’s say you have you been diagnosed with COVID. You’re at home. And let’s say we can create this test that you can do at home, for example. And it’s shows that you have a high-risk chance of dying. Are you going to stay at home? You’re not. You’re going to come to the hospital. Right? On the other hand, if it shows that you are at low risk, then, you know, at least you have peace of mind that you’re not going to do so poorly. Now, one thing that we’re also trying to investigate with the test is how it changes over time. Right? You know, those are the risk profile gets better with treatments or not. So that’s another part of personalized medicine, right? Because let’s say you have a high-risk profile, and we get you treated. And the profile changes to a lower risk. That means you’re going in the right direction. But let’s say you have a high-risk profile, and we treat you and the profile doesn’t change. Well, that means that the drugs we’re giving you are not the right drugs. Again, the same principle of precision medicine. So how the changes of the profile in response to treatment can also be informative of how the patient will do. So, and I think that’s another exciting part that we’re also trying to investigate in this study.
Interview conducted by Ivanhoe Broadcast News.
END OF INTERVIEW
This information is intended for additional research purposes only. It is not to be used as a prescription or advice from Ivanhoe Broadcast News, Inc. or any medical professional interviewed. Ivanhoe Broadcast News, Inc. assumes no responsibility for the depth or accuracy of physician statements. Procedures or medicines apply to different people and medical factors; always consult your physician on medical matters.
If you would like more information, please contact:
Sarah Worth
University of South Florida
Sign up for a free weekly e-mail on Medical Breakthroughs called First to Know by clicking here
