Transcript: Daniel S. Fierer, MD, on HCV Reinfections Among MSM With HIV

Amanda Balbi: Hello, everyone and welcome to another installment of Podcasts360—your go‑to resource for medical news and clinical updates. I'm your moderator Amanda Balbi with Consultant360 Specialty Network.

The Conference on Retroviruses and Opportunistic Infections 2020 was a virtual meeting of the minds in infectious disease and public health. A research team who presented their latest study at CROI 2020 identified 3 high‑risk behaviors and evaluated them via patient questionnaire among men who have sex with men (or MSM) in New York City.

The team was led by Dr Daniel Fierer, who is an associate professor of infectious diseases at the Icahn School of Medicine at Mount Sinai in New York, New York. Today I will be talking to Dr Fierer about this research and finding out more insight on this patient population.

Thank you for joining me today, Dr. Fierer. Let's dive into your study.

The title of your paper is, “Sex Not Drug Use is Driving HCV Reinfection Among HIV Infected MSM in New York City.” We've talked about your work in this area before on ID Consultant. This seems like an extension of that research.

Daniel Fierer: Yes, this is, in a way, a direct follow‑up of that same group presented last year at CROI, where we presented the reinfection rate in our cohort in New York City, so that it was a very substantial 4.4 reinfections per 100 years of observation.

100 person years could be summarized as a 4.4% reinfection rate, among HIV‑infected MSM in New York City. We were fortunate to have prospectively collected risk-factor information of a specific type that I'll describe, to try to hone in on behavioral risk factors, therefore an idea on how we could possibly decrease the high reinfection rate.

Amanda Balbi: Let's get into your new study or new analysis. Can you tell us a little bit about how it was conducted, and a little bit about the findings?

Daniel Fierer: The background on our behavioral research in this area, is that some years ago, published in the MMWR 2011, we showed that there were 2 risk factors for primary hepatitis C infection among HIV‑infected MSM in a cohort, New York City.

These were determined on a multivariable analysis, having asked a lot of questions. The 2 behavioral risk factors were condomless receptive anal intercourse, reporting having received semen to the rectum. The second behavioral risk factor for primary, that is the first hepatitis C infection among these men, was being high on crystal methamphetamine at the time of sex. It was specifically methamphetamine, rather than any of the other drugs that are commonly used.

This particular study, we were interested in determining the risk factors for sexual transmission. We excluded those who reported injecting any drugs, for the obvious reason that would complicate analysis of sexual transmission. It felt that use of needles, it's traditional in the field to accept it, drug use is a risk factor of hepatitis C transmission or acquisition. This was just the sexual transmission.

What we were interested in was to assess whether these behavioral risks were important, and perhaps to fine‑tune our previous study when considering these men who had been reinfected. They had been infected once, hepatitis C was cleared through treatment or spontaneous clearance, and then reinfected. What are the behavioral risk factors for that, especially with the impressively high reinfection rate in our cohort of 4.4 per 100 person years?

Behavioral risk factors generally have not been well assessed in European studies. Also the French have recently done a nice job in looking at a number of risk factors for reinfection.

What we did was perform a cohort study of the HIV‑infected MSM in New York, who had clearance of their primary hepatitis C between the beginning of January 2000 and the end of 2018—a fairly substantial period that also crosses the 2 epochs of interferon era, in some ways even preinterferon era 2000 (the very beginning of it) and through the year of treatment with direct acting antivirals, that we call DAA.

We considered that our start of the observation period. It was the end of treatment, not the cure date for this analysis. We did that because the European group had done it that way, and so we could directly compare the results. In the end, the observation was the clinical onset of reinfection or the last undetectable viral load or study visit, whichever came first to those who had not been reinfected.

In each visit, the men were queried about having received semen in the rectum, whether they had sex on crystal meth or whether they had injected crystal meth, and they added this even though it wasn't found in our first study because we had excluded those men. We consider these as the really known risk factors in New York, although injection use is not completely known.

We had follow‑up information and behavioral information on 244 HIV‑infected MSM with cleared primary HCV. This was a reasonable proportion of the larger group. I don't discuss this in our poster at length because there's not much room. In the publication, we'll, of course, address it that there is no difference between ... Then, this was 29 men who had been reinfected compared to 215 men who had not been reinfected. Demographics and other basic characteristics were the same between the groups.

The median follow‑up was 4 visits for those who had been reinfected and those not reinfected, over a total of 600 person years. It's a fairly substantial follow‑up.

What we found was in our major analysis. The way we approach this is a time‑dependent analysis. To do this, we used the Cox proportional‑hazards model. What's important and interesting about this model is that it accounts for changes over time.

We then, in assessing the 3 characteristics, had seen that rectum sex and crystal meth injection of crystal meth had been very common in the hepatitis C month of 29 men who got reinfected. 28 of them reported receiving semen in the rectum, essentially everybody. Well, two‑thirds of those who didn't get reinfected, so there was difference by either sex while on crystal meth, but fully 62% of the men who were reinfected reported that while the 44%, it didn't.

Injection of crystal meth is quite a substantial 45% of the 3 infections reported having injected crystal meth as part of sexualized drug use while 28%—a fairly substantial proportion as well—who didn't get reinfected, had that behavior.

Doing the multivariable analysis, though, which is the crucial thing there to sort out colinear behaviors, that it was only receiving semen in the rectum that was significantly associated with hepatitis C reinfection. This is an analysis as adjusted for age, race, ethnicity, and the year, or we could consider the epoch of hep C clearance that is interferon era or DAA era.

The hazard ratio is 4, the p‑value of .008. It's quite a substantial p‑value, the hazard ratio of 4 with confidence intervals that went from 1.4 to 11. Somewhat wide, but a reasonable translation of that, as I think about it, is that men who continued to receive semen in the rectum over time are 4 times as likely to be reinfected than those who did not continue to have semen in the rectum.

That magnitude, pretty substantial risk while the sexual high and specifically injection use were not significant in multivariable analysis. Then this is how the title of, which just as a result in a sense ... The title of our abstract, it is "Sex Not Drug Use that is Associated with the Behavior Associated with Hepatitis C Reinfection."

Those of our listeners here who will have access to the poster, and that will be linked, will see why we wanted to show this idea in a visual way. You can't graph a Cox proportional‑hazards model result, but to give a visual idea of the importance of semen in rectum over time to the Kaplan‑Meier survival analysis where you divided your group into exposure to semen.

We're going to call it time‑weighted average rectal exposure. That's over time. It's basically exposure to semen in the rectum over time and divided by those who had above the median time‑weighted exposure compared to those with below the median time‑weighted exposure.

There's a very clear separation that happens very early before one year of follow‑up, where those with above median exposure have a significantly higher reinfection probability. In the Kaplan‑Meier survival analysis, the 2 curves separate quickly and substantially, and using a log rank P‑value of .001—again, quite a substantial result. In this case, we did over a period of 5 years, although reinfections did continue to occur out to well over 10 years, although the numbers who were being followed were very small at that time.

This lends further weight looking at another way, and you'll see the digital audit to demonstrate that it is sex, not drug use per se. That is the behavior associated with Hep C reinfection, and since it's sex but receipt of semen into the rectum was condomless in all intercourse.

This is a question that isn't asked in most of the behavioral questionnaires and some I think it's somewhat assumed. We found that that wasn't necessarily the case in our initial analysis but there may be a difference between just simply asking about condomless receptive anal intercourse and receiving semen in the rectum.

We think this is an important result because the previous research that our group and others have done demonstrating hep C in semen makes this association biologically plausible. That's why I talk about receiving semen in the rectum because this is presumably how Hepatitis C infection occurs.

This isn't so much a secondary effect. For instance. how drug use might cause dis‑inhibition and then leading to Hepatitis C exposure. It's the semen in the rectum. That is the Hepatitis C exposure plausibly.

The bottom‑line behavior response, you say, "What do we do with this?" The first thing that's rather obvious is that recommending condom use has not worked to prevent Hepatitis C. It hasn't worked well to prevent Hepatitis C primary infection. That's because it's just not well known at all, I would say.

Even now that the word has been out in the scientific community to some extent and to community of HIV‑infected MSM, I would say that still few if any men actually know this to be true. Somewhat, I think it's because there is still in the Hepatitis C investigation community or what I'm going to really say is an overemphasis on blood and the assumption that if there isn't bleeding, then there isn't going to be hepatitis C transmission and that is simply not true. Certainly not visible but is absolutely not necessary.

What somebody means by microscopic blood, I think, is starting to really split hairs. It is the plasma that gets the blood. It's where the hepatitis C is. These fluids are in equilibrium with human and with rectal fluid.

What this means that the condom use hasn't prevented the primary hepatitis C, arguably because the men at risk didn't know their risk and wouldn't think of condom use to prevent hepatitis C. It is a reasonable assumption, certainly in my curb.

Actually, I can say that these are people who were all under my care. I treated all of them or took care of them as they spontaneously cleared, once we recognize that semen in the rectum was an important risk factor.

At least 10 years ago, I've been telling all the men when they come in to see me, throughout the visits, and then after they're cured, that this is a risk for primary infection, and then no reason to believe that wouldn't be for reinfection.

I'd have to say that all my patients which is all this cohort, who would have been told more than once—I would say with a median visit, number of visits of 4—they've been told the median of four times that hepatitis C is in semen. That in and of itself, you're providing the information about how reinfection may occur. We still ended up with a substantial reinfection.

It is certainly possible that the reinfection rate could have been higher if I hadn't been telling everybody this. We know that that's the case. Instead the condom use has not been ... Well, couldn't get the reinfection rate below 4.4%, which is still pretty high. We need to consider what I described in the poster as novel, or new interventions to prevent deposition of seminal HCV into the rectum.

There are other ways we could phrase it that would keep the hepatitis C in the semen out of the circulation and out of the liver. There are different ways to think about that. So far, I have to say, I propose this is looking for other ideas. I don't have a novel intervention up my sleeve for how to do this. that think that this is a very important barrier to our ability to control HIV and even really significantly decrease the prevalence.

There's reasonably good evidence from that medical modeling from a few different groups—Natasha Martin in San Diego, the really great work of this Swiss cohort that suggests that behavioral interventions are necessary component to expect elimination of hepatitis C among HIV‑infected MSM, and the behavioral interventions to be keeping semen out of the rectum.

This should be an emphasis in ... Obviously drug use is very common here. Our study is evidence that that is not the major problem. Although of course, taking care of drug use interferes with parts of people's lives should be considered.

We may be emphasizing the wrong thing and then to focus back on sex, at least in New York. Other communities have different local cultures, but I would say in a community here where the culture of sex while high being prevalent in almost half of the men who had primary Hepatitis C in the past, that's pretty darn common. I think that it would be surprising to me if other communities had a higher level of drug use and association with sex, higher level of sexualized‑drug use.

Again, I think we need to get back to thinking about what I think has been a neglected area, even specifically in this field, which is specifically the transmission through body fluids that are not blood and how to address that.

I invite colleagues, non‑colleagues, anybody if we have ideas how to look at this and make some proposals for how we can intervene. The ultimate goal is preventing hepatitis C primary and reinfection. I should note that this study was done in a cohort of HIV‑infected MSM.

As noted before, that MSM who are using pre‑exposure prophylaxis against HIV, have been clearly shown to be getting hepatitis C. French study, they have essentially the same rate as HIV‑infected men.

I think it is reasonable to assume that the same risk factors apply to the extrapolation, but I think for public health, we ought to include the MSM on PrEP in this group and start trying to get that information out more clearly.

I think that has really not been communicated well that this is an infection that's sexually transmitted among men on PrEP, and to consider the semen in the rectum for condomless receptive anal intercourse as a significant risk for acquiring hepatitis C.

The most important part that I would like to take this opportunity to re‑emphasize, even though it was not something we looked at specifically for risk factors. The most sobering part of the result, which was that there was no statistically significant difference in Hep C reinfection incidence and the epoch of interferon or DAA, is that since DAAs have been available, we have treated a lot more people.

The prevalence of Hepatitis C and HIV‑infected MSM has felt to be significantly decreased by TRIO DAA. One thing we can say about people who have HIV infection are very closely monitored, including essentially everybody gets a Hepatitis C test essentially the day they enter care. These days, many or most are monitored over time for incident Hepatitis C infection, so we really know who is infected.

This is one part where we've done a really super job. Everywhere in the world where HIV‑infected men have sex with men have been cared for in treating and curing Hepatitis C infection, to the extent that most what we called co‑infection clinics are most HIV clinics, is that they've pretty much treated everybody.

80% is perhaps what you'd call everybody in here. That's certainly true at my institution where we feel we don't need and it's true, many colleagues whom I've spoken to, they don't really even need a specific co‑infection clinic anymore so that they can be taken care of in a more as‑needed basis.

With the prevalence, apparently so much lower from this DA treatment, one would reasonably expect that the incidence rate of new infections would be much lower as well. This is the concept of treatment as prevention. We would want to see a decrease in incidence with a decrease in prevalence in the general community.

To me, the thing that's most distressing about this, no change in incidence at least so far. This was an analysis through about 2018, with the data getting a little sparser into 2019.

That's a pretty substantial number of years and well into the time where we really thought we had treated most everybody which just took a few years 2015, '16, maybe '17 in most big cities. Little bit later in Europe as the DAAs were rolled out later.

We treated this amazing work and the Swiss treated essentially everyone in the country in a very short period of time and then went through and found everybody again wasn't cured and treated them again.

Even then they had re‑introductions from other countries. The same thing happened in Amsterdam. In Holland generally, there was large re‑introductions, because of a certain amount of variability. Europe and the US—you could almost think of those as the proximity is similar to the countries and sizes to states, perhaps.

In the US travel was easy, so re‑introductions from other places that haven't had the same simultaneous treatments, there was no problem.

In New York City I doubt that reintroduction from the outside is a major source of this. Large and complicated geography, demographic and maybe different issues than many large US cities. My suspicion, although I don't have the data to back that up, is that we've missed a significant proportion of people who are, obviously, the main transmitters.

Work that needs to be done, work that we could, in a sense have in the freezer is to look at the phylogenetic relationships between the viruses that are circulating now, compared to those circulating 10 years ago. See if we can make more sense of that. That is a study for me, in search of funding currently, but is readily available.

It's a good way of doing that analysis. We can learn a lot more about why the incidence is so high.

Amanda Balbi: Great. Thank you so much for speaking with me today about your research and sharing this insight.

Daniel Fierer: I hope to have the opportunity to speak with you more about this subject.