Transcript (slightly modified for clarity)

 Hello, and thank you for joining us for this 

 entitled, “Optimizing Outcomes in the Treatment of HIV.” Substantial advances have been made in using antiretroviral therapy for prevention and treatment of HIV infection. Newly available regimens are safer, more effective, and more convenient than ever before. In this 

 panel discussion, I am joined by a panel of infectious disease specialists who are experts in treating HIV infection. Together, we’re going to talk about the most recent recommendations for long-term management of your patients who are at risk for, or are already infected with, HIV. In addition, we’re going to talk about emerging agents that are coming to the clinic soon.

My name is Dr Peter Salgo. I’m a professor of medicine and anesthesiology at Columbia University College of Physicians and Surgeons and an associate director of Surgical Intensive Care at New York-Presbyterian Hospital.

Joining me for this discussion are: Dr Eric Daar, chief of the Division of HIV Medicine at Harbor-UCLA Medical Center and professor of medicine at the David Geffen School of Medicine at UCLA in Los Angeles, California; Dr Joseph Eron, professor of medicine at the University of North Carolina [at] Chapel Hill in North Carolina; Dr Ian Frank, professor of medicine at the Hospital of the University of Pennsylvania in Philadelphia; and Dr Paul Sax, clinical director for the Division of Infectious Disease at Brigham and Women’s Hospital and professor of medicine at Harvard Medical School in Boston, Massachusetts.

Guys, thank you for joining. I’m going to start right now with our first question, which is the obvious one: what are the most recent recommendations for HIV screening? Who wants to start us off? Ian?

 The CDC has created some guidance for HIV testing in the United States. They recommend that everybody between the ages of 13 and 64 be tested for HIV at least one time if you live in a neighborhood where the HIV prevalence is not less than a tenth of 1%—which I don’t think exists in the United States.

 Yes. I was going to say that for me, as somebody who was in practice at the beginning of the HIV epidemic (which was not an HIV epidemic then), it was whatever you wanted to call it. That’s an astounding statement. Basically, they want to test everybody. Why?

 There are a couple of reasons why the CDC decided to do that. One of them is obvious: you want to be able to find people who are infected before they get sick. That way, they can get life-saving therapy and not transmit the virus to other people. But the other reason they did it was actually kind of more psychological. They said that if we test everyone, the stigma around getting tested will be reduced. And that’s actually been very effective, because when you look at, it’s more important to test high-risk people than it is to test everybody. This was kind of a de-stigmatizing effort by the CDC.

 I remember that in the early days, you would say, “Well, I know who to test. I can tell.” Right?

 We don’t know who to test, and the range is a little funny. It’s not really clear why you should stop, because older people can get HIV. In fact, we’re seeing that. But I think Paul made a critical point, which is that people who are at highest risk need to be tested more often. In fact, we were trying to get away from what’s called “targeted testing,” but that’s exactly what we need. And that’s what’s missing. We need to consider the younger people who don’t touch healthcare. They’re not in healthcare. They don’t touch the system.

 So, I’m going to screen everybody in the United States, which is effectively what I heard you say. Is it cost-effective?

 It’s cost-effective when you’re testing people above a certain background prevalence.

 But that’s not what he said. He said pretty much everybody.

 The bottom line is that the test is very inexpensive, and the years of life that can be saved from finding an undiagnosed case of HIV are so huge that it ends up being cost-effective.

 And the other point, too, is that the tests are exquisitely sensitive and specific. It’s not like a test where you get a whole bunch of false-positives and you waste a whole bunch of money doing prostate ultrasounds.

 There’s another benefit, though, that’s really important. It’s not just for the individual, too. There’s always been this concern that 15% to 20% of people in the United States are infected and don’t know it and that those people may be disproportionately driving the current epidemic.

 That number, though, is much improved compared to the number 5 years ago.

 Right now it’s probably about 15% of the people infected in the United States don’t know it. And the only reason they don’t know it is because they’ve never been tested.

 So, again, you test everybody to avoid stigmatizing the people that you really should be testing more often?

 It’s not a bad idea, frankly, if it’s cheap and it’s easy.

 Plus, you do find people. But it’s just not a very efficient find. In North Carolina, the number of tests increased by 2- to 2.5-fold. And the number of new diagnoses went up around 10%. But you do find people, so it’s valuable to find them.

 We did a study in our emergency room of testing people in whom testing would not otherwise be indicated. We were actually screening everyone, and the yield was understandably very low. We did find a few. Interestingly, we found a couple who were actually already known to be infected but had “forgotten,” or were in denial. So, it was actually quite useful.

 Before we move on, I think we should talk about who should be tested, rather than just suggest that we test everybody.

 That was my next question, anyway, because you did allude to this. There are some people that you really want to drill in on, if you will—high-risk individuals. How would you define those?

 I think we would define them as men who have sex with men, individuals with multiple concurrent sexual partners, individuals who do intravenous drugs, or folks who have certain indicator diseases that can be associated with HIV that are often overlooked—psoriasis, herpes zoster, tuberculosis.

 Zoster is a big one. We just had a woman with zoster a year ago—a young woman who came in with HIV, a low CD4 with PCP (pneumocystis jirovecii pneumonia).

 Anyone who’s had a sexually transmitted infection?

 Thrombocytopenia. There are a whole bunch of conditions that normally are not thought to be HIV-related and yet are early clues.

 De novo, so-called idiopathic thrombocytopenia, patients should get an HIV test done.

 That should be, unequivocally, part of the workup.

 So, those are your high-risk individuals. Of course, if you know that they’re thrombocytopenic, you’ve tested them for something else anyway.

 The current guidelines are targeting people who are encountering the healthcare system for other reasons. Certainly, everyone is encouraged to seek out testing sites, but the guidelines are really focusing on people who encounter the healthcare system for any reason.

 We have 2 kinds of compliance here, right? There’s the compliance of people who present and would like to be tested. You offer them the test. Then, you’ve got compliance for primary care physicians or other physicians who should be proactively offering it. Why don’t we divide that into 2 groups? How do we get people to want to be tested? Anybody have a brilliant idea?

 You have to get the word out of how effective therapy is. That is number one. Because you add 30, 40 years of life to someone who is 20 years old. If they weren’t treated or weren’t diagnosed, they might live 20 years. Instead, they can live 50 or 60 years and can, essentially, have a normal life expectancy. So, that’s one obvious way. I think de-stigmatizing the test is really important.

 In infectious disease practice, the way we do it is we say to people who come in who haven’t been tested, “We test everyone,” and they go, “Okay,” because they’re everyone. They’re part of that.

 But does every doctor do that? Is it part of every doctor’s practice?

 The recommendations are over 10 years old, and clearly everybody isn’t doing it.

 It’s sort of an unfunded mandate, if you will. There’s no money that came with it. It takes a lot of effort, although they tried to make it easier. They got rid of the consent process, which was huge. All of the pre- and posttest counseling that used to take up a lot of time, they got rid of all of that. But not everybody has the ability to spend the time with the patients and simply say “We’re going to do this as part of routine care.” And then the other big thing is dealing with the fallout when you find a positive. A lot of centers, whether it’s a private doctor’s office or even an emergency department, have got to move on to the next patient. They’re not looking for new problems.

 But I think some of it is just assumptions that we make. You’ve been taking care of this gentleman for many years, and you think you know their sexual habits, and you say, “Oh, they’re not at risk for HIV.” And I think that that’s what happens. Like Paul said, if your approach is that we test everybody for HIV infection, when you’re ordering any kind of test (eg, routine blood work), you can say, “Hey, it’s our standard practice to test everybody.” The counseling doesn’t need to be more complicated than that. It doesn’t need to take a lot of time.

Who Should Be Tested for HIV?

 Now, in 2017, where are we in terms of disease prevention? How do we prevent this disease in the first place? Pre-exposure prophylaxis, how do we deal with that?

 We spend a lot of time with voluntary counseling and testing. We have invested billions of dollars in the richest country in the world, and the number of new infections is estimated to be, essentially, unchanged in the United States over the last decade or more.

 You could argue that it could have been a lot worse had we not made the investment—but it wasn’t enough if we’re still having 40,000 to 50,000 new infections a year. So it wasn’t enough. People started looking at biologic ways to manage this that doesn’t require, entirely, people-changing behavior. We always look to vaccines, and so far, vaccines have been a bust.

Then, the next big step, and we actually learned this over 20 years ago in preventing pregnant women from transmitting HIV to their baby, is using antiretrovirals. And the big advancement that has had a big impact, globally, is the recognition that when we treat somebody with HIV and suppress their virus, the risk of them transmitting to their partner is extraordinarily small.

 That’s because the actual particle count is nil.

 Exactly. That’s been, by far, the biggest advance in prevention. It is virtually 100% protective, based on the data that’s available (if somebody is on therapy and their virus is suppressed).

 In fact, in several cities where a high proportion of people are on therapy, the number of new infections has dropped.

 Let’s hit on a couple of quick bullet points. For women who are trying to conceive, can they be prophylaxed?

 I think there are 2 ways to look at this. So, you’re assuming now that the male partner is HIV-infected and the woman is uninfected and that they’re trying to conceive and have a biologic child. There’s lots of pieces to this, right? Because there are alternative means in which one would accomplish this feat. So, the first step, based on the data that’s available, is that if the male partner’s viral load is undetectable and has been for at least 6 months, the likelihood of them transmitting to the partner should be virtually zero.

 Okay. So, the bullet point on that is you treat the male partner?

 What about unborn children? Prophylaxis?

 Well, prevention of HIV from mother to child—we know how to do that. You treat the mother. Treat the mother, treat the mother, treat the mother. I don’t know what it’s like in Massachusetts or Pennsylvania, but in North Carolina, maybe we see 1 or 2 transmissions in the entire state from mother to child.

 It’s gone away. Pediatric HIV is done.

 I would like to point out that when I said it was a bust, that was before you said that. That’s huge.

 People who specialize in pediatric HIV need to find a new specialty.

 The whole network needs to find a new thing to do.

 I will say one thing. There is still an open research question as to what the safest treatment is to give to pregnant women. We can prevent transmission of HIV, but the babies born to HIV-infected moms who don’t have HIV themselves appear to have health issues. We’re not sure whether it’s that they’re getting some toxicity from the drugs or something else.

 I know it’s brief—pre-exposure prophylaxis (PrEP). We have to mention PrEP because there may be a lot of people out there listening to this that are primary care doctors. This has to do with what Eric was trying to say: about using antiretrovirals in those very high-risk groups that Ian mentioned&mdash;the men who have sex with men and injection drug users, which is going way up. It’s like back in the bad old days. Remember PrEP, right?

 And that’s for the masses, right? So, first I talked about what the million or two in the United States can do to prevent from transmitting to their partners. PrEP is really for the millions and millions of uninfected people that allow them to protect themselves.

 And it’s a reason why people may want to volunteer and get tested. If individuals know that they’re at risk for acquiring HIV and if they’re aware that there’s a way that they can prevent becoming infected by taking a pill a day, then that’s incentive to be tested to find out if you’re negative and then to request PrEP.

 Okay. What about post-exposure prophylaxis? And this is of particular interest, of course, to healthcare providers. But it’s also directed toward anybody who suddenly finds out. “Oh my gosh, I was exposed.” What do you do?

 We give 28 days (4 weeks) of therapy. In occupational exposures, it makes sense. It probably works. We give it, also, to people who have, for example, sexual assaults. In that setting, I think a lot of what we’re doing is just trying to do something rather than nothing, because often we don’t know the source or if the patient has infection at all. So there’s a tremendous amount of overtreatment for postexposure prophylaxis. Nonetheless, we do it. It is endorsed in the guidelines, and it’s probably one of the reasons why, at least in occupational exposures, there has not been occupationally-transmitted HIV in this country for decades.

 Decades. The advance is that, now, it’s safe and well tolerated. Twenty years ago you were taking zidovudine (AZT) every 4 hours. You’re vomiting. And now it’s really very different.

 You can say there has not been an occupational exposure that has led to HIV acquisition in the United States in decades.

Pre- and Post-Exposure Prophylaxis in HIV

 Let’s talk about initiating treatment. Somebody has HIV and you’re going to initiate treatment. What are you looking to do? What are your goals?

So the goals are, I think, really straightforward. It’s to either maintain, or restore, health. That’s the goal. To do that, we focus on getting people’s viral loads suppressed. And you want to do it with the safest, most tolerated therapy possible. But we’re really just trying to either maintain health (if their CD4 is already high) or restore health (if their CD4 is low). I think that’s the simple answer.

 Okay, that’s it? We want to make them better?

 You’re basically choosing a regimen that’s going to suppress viral replication. We’re lucky. We have a phenomenal surrogate marker for HIV health, and that is that suppression of viral replication with antiretroviral therapy correlates so strongly with good outcomes. If you can do that, you’re golden.

 And that’s why everybody is focused on viral load. The patients are focused on viral load. The doctors are all focused on viral load. But my point is just that the goal is to keep people healthy.

 There’s always these 2 markers that everybody is talking about. One is the CD4 count and one is the viremia.

 Viral load. What’s the difference? Virus in your blood. Viral load. What’s the difference?

 I think that the providers would say viral load. But you’re right. Viral load is the amount of viremia.

 Is it better to look for viremia (or viral load) or CD4 count? Or do you use both?

 The thing is, you can’t really do anything about CD4 cell count. You can suppress the viral load, and in most people, their CD4 count goes up—and that’s a good thing&mdash;or virtually everybody is stable. There are some people who you suppress their viral load and their CD4 doesn’t go up or it goes up as much as you expect.

 We do know what to do with CD4s, and we know that when they’re below certain thresholds, we give antibiotics to prevent them from developing opportunistic infections.

 But Joe is right. I was invited to give a lecture at another academic medical center in Boston. They said, “We want you to talk about people whose viral load is suppressed, but whose CD4 cell count hasn’t recovered.” And I said, “That’s going to be a very short talk. We don’t know what to do about those people. Changing therapy doesn’t really help.”

 So, you initiate therapy when there is virus in the blood or when the CD4 count is low?

 Wait. We are no longer considering CD4 count when making the decision to initiate therapy. Right now, the standard procedure is that everybody with HIV—if their viral load is measureable, which is almost everybody with HIV&mdash;should be started on therapy.

 We want to get the CD4 count because, initially, we’d like to understand if they’re at risk for opportunistic infections. We don’t have to choose between getting a viral load or a CD4 count before starting therapy—we can do both, and we do. The CD4 count tells us if you are at risk for an opportunistic infection now. And the viral load is telling us, “Okay, here’s the starting point.”

 They’re 2 different classes of information?

 The CD4 cell count also gives you a sense of urgency. So, when we have a newly diagnosed patient and their CD4 cell count is, for example, very depleted (ie, under 100, under 50), even if they’re feeling relatively well, we essentially tell those patients, “This is your job to start antiretroviral therapy. This is your top priority.” Because in the old days, those are the people who had a prognosis kind of like acute leukemia. They were the ones who died within 6 to 18 months. They have to go on treatment immediately. So, for the patients with higher CD4 cell counts, we like to start them soon, as well, but it’s not quite so easy.

 Why would you not start them on the day of diagnosis?

 Well, that’s a great topic and an area of research. There are several studies now that suggest that if you actually start people on their day of diagnosis, maybe they actually return to care better, suppress faster, and retain care better. You’re telling someone, “You have a serious illness” and “Oh, by the way, we’ll start therapy the next time we see you.” But the flip side to that is, of course, that it’s complicated to start therapy. You need to know some stuff about people, and you want to make sure they take it correctly. So, it’s a big step, and you have people that push back.

 What about starting therapy after infection, prior to antibody detection?

 That’s a different story. I’d like to just mention one other thing about same-day starts. I think the reason we didn’t start therapy on the same day as diagnosis in the old days was because the regimens were very complex. An entire industry was built creating education about taking antiretroviral therapy and what it means to take medicines with and without food, as well as what time of the day to take them. There were these counselors. You have to come back and go to adherence class, and you have to come back and understand what viral load and CD4 means. All of that stuff, today, is antiquated—we don’t need that anymore. Basically, people can take 1 or 2 pills a day and they’re fine. So right now, same-day starts make a lot of sense.

 That’s astounding. There’s no excuse not to get on the ball here.

 There are practical issues, though, that we all have to deal with. Not everybody comes to us with insurance. Not everybody comes to us who can literally have a prescription written and get the drugs that day. And not everybody who comes in, often having just found out about this diagnosis, is ready to leave with pills that day. So, there may be advantages for select patients to do this, if you can. Otherwise, it’s as soon as it’s feasible to get people on treatment.

Initiation of Antiretroviral Therapy in HIV

 What factors do you consider when you start considering a therapeutic regimen and who to put on what?

 There’s one important virologic factor, which is resistance. Some people get infected with resistant virus. Based upon the drugs we use, that doesn’t necessarily influence a choice of therapy as much today as it had in the past, but it occasionally may. A lot of the factors have to do with the patient. Are they on other medications which could be associated with drug—drug interactions that could influence our choice? Do they have another disease that could influence our choice? Do they have heart disease? Do they have kidney disease? Do they have hepatitis B infection? Do they have hepatitis C infection? Occasionally, the viral load may influence our choice because there are a couple of drugs we may not want to use if the viral load is greater than 100,000—although it’s not my practice to use those drugs. I tend to prefer to use drugs that are as potent and effective irrespective of the viral load or a CD4 count. Those have been some factors that influence decisions.

 Personal approach and initial ART (antiretroviral therapy): what’s your standard? Like, all-comers, what would you choose?

 It’s interesting. Since the beginning of potent, antiretroviral therapy about 20 years ago (in the mid-90s), it was 2 nucleoside analogs and a third drug. And here we are, 20 years later, and it’s still (for the most part) 2 nucleoside analogs and a third drug.

 The third drug had traditionally been a protease inhibitor or a non-nucleoside reverse transcriptase inhibitor. And then, more recently with the rollout of more and more integrase inhibitors, integrase inhibitors have really taken over as the predominant third drug for most people’s first-line regimens.

 How do you select the patients if you’re going to do all of these various complicated things? Run through some of these drugs and how you would parse them out.

 If you look at the current guidelines, at least in the US guidelines, one of the guidelines only recommends integrase-based regimens. The integrase inhibitors vary a little bit. Some are once-a-day regimens, and some are twice-a-day regimens. Then, the biggest difference is often what they’re co-formulated with. Because, in addition to the discussion about how easy therapy is, how forgiving it is, and how there’s little reason to jump through all the hoops we used to, there’s also more and more single-tablet regimens where we’re not just treating people with easy regimens. We’re literally giving them 1 pill a day. But they’re not all co-formulated with all of the options.

 I think what scares a lot of the primary care doctors is a list like this.

 The list is shrinking. It used to be 2 pages long.

 Dolutegravir, abacavir, and lamivudine. I’m reading these off and I can barely pronounce them. Dolutegravir, tenofovir, alafenamide, and emtricitabine. Look at this menu.

 Sure. To be a little more critical, some of these are better than others.

 I think the best integrase inhibitor is dolutegravir, and I don’t think that there’s a lot of doubt about that. It’s a once-a-day regimen, it’s very potent, and it has a remarkable resistance barrier, meaning people who fail treatment don’t get resistance to it. And you could combine it with any of the nucleosides, including with tenofovir, alafenamide, and emtricitabine, which is my personal favorite combination.

 think Paul’s point is that there really are maybe 2 or 3 initial regimens that you would pick from. Dolutegravir and Descovy, the problem with that regimen is it’s 2 pills and some people don’t want to take 2 pills. And then there’s dolutegravir/abacavir/lamivudine. That is 1 pill, but not everybody is as happy with abacavir as they are with TAF (tenofovir/alafenamide). Finally, there’s the combination that includes elvitegravir, which is the third integrase inhibitor. And the issue with that is it’s boosted. You have to give an inhibitor of P450, and that’s not the best thing to give people, necessarily.

Now, Paul’s going to fix this because Paul is working on a medication, which we’ll get to at the end, that actually has a lot of the advantages of dolutegravir, but is co-formulated with this tenofovir/alafenamide/emtricitabine. So, we’ll get to that at the end, but things are even getting better.

 But you do understand that for somebody looking at a list like this, it turns into word salad.

 Yes, but the list used to be a page-and-a-half long. Now, if you read the guidelines, it’s down to 4 choices and they’re very, very similar. You could throw a dart at them and pretty much get it right.

 I understand what you’re saying. If someone were to show me, for example, the drugs for rheumatoid arthritis, how could we choose that without using them every day? That’s really what you’re talking about.

 And there’s certainly subtleties. So, when Paul said, “This is definitely what I would use, this is my preferred option,” there may be certain situations in certain patients where he might not.

Nuances of First-line Antiretroviral Therapy in HIV

 So what about patients in whom nucleoside reverse transcriptase inhibitors are not an option? What do you do then?

 Well, I can tell you there are very few people where that’s not an option. It really is about people who you don’t really want to give them nucleosides for one reason or another. And it keeps getting easier. With this newer drug, tenofovir/alafenamide (TAF), we used to get into trouble when people had modest renal dysfunction because we weren’t as sure what to do and the nucleosides got us a little bit. But now, with this tenofovir/alafenamide, you can go down to a creatinine clearance of 30. I think the hardest group are people that have real renal dysfunction; that’s an issue. Or transmitted drug resistance that’s multi-drug resistant, which is really uncommon. I can’t really think of too many other scenarios.

 What comes up sometimes is that despite the fact that most of the time the integrase inhibitors are really well tolerated, there are some people who have neuropsychiatric, psychostimulatory reactions.

 You started them on the integrase inhibitor, but then you have to back off. But you don’t know that in advance.

 You cannot pick those people, but they’re still drugs. They’re not placebos.

 Right, sure. I completely agree with you.

 Let me run down a couple of other problematic cases. How do you approach patients with very high viral loads (100,000) or low CD4 counts (less than 200)? Are they special? Are there specific drugs you like?

 There are not anymore. There used to be specific drugs we tried to avoid in those people. For the overwhelming majority, it was a nonissue. And all of those drugs that we try to avoid are no longer generally considered first-line therapy. So, unless you have some particular attraction to a certain drug, it rarely matters.

 But there are some misconceptions out there. You go to meetings, and even some experienced HIV treaters believe that if you have a low CD4, you should get a protease inhibitor. And the evidence for that is pretty much zero.

 Yes. I think it has to do with the fact that the first life-saving drugs in the mid-90s were protease inhibitors.

 So they remember that. The clinical outcomes were first demonstrated in that famous study, where protease inhibitor-based therapy goes up and the deaths from AIDS go down. So, you know that figure went into everyone’s minds: “You’ve got to use a protease inhibitor for patients with AIDS.” But it’s not true.

 How about this case—antiretroviral therapy in HIV-infected women during pregnancy?

 That’s a tough one, and Paul mentioned that right in the beginning because we just have less information. For some of these new drugs, like dolutegravir and tenofovir/alafenamide, there really are no data for them in pregnancy or the data are very limited. The treatment array shrinks, but we have pretty good advice. Raltegravir, an integrase, is okay to use during pregnancy. So, you can certainly pick a reasonably well tolerated, very effective therapy, but it might not be the top-of-the-shelf choice that Paul was picking off of earlier, and that’s a shame, actually.

 It’s interesting. For example, consider tenofovir/alafenamide, which is the safer version of tenofovir/disoproxil fumarate. There’s really no experience with that drug in pregnancy at all. At our institution, where we’re treating pregnant women, we use the old-fashioned version of tenofovir plus raltegravir, whereas we would not choose that regimen for a newly diagnosed patient in any other context.

 Of course, there were people using AZT (azidothymidine) in those patients.

 This is much better than where we were previously.

 It is also worth saying that most of the pregnant women who are newly diagnosed are extremely healthy. The exception are the ones who acquired HIV as babies. They’re extremely challenging, and usually they have bad resistance.

 It sounds like a balancing act to me. I mean, if you’ve got a woman who is infected, you would like to treat her for her benefit and for the benefit of the fetus. On the other hand, you don’t know about the drugs.

 But one of the things that does get done consistently is HIV testing in pregnancy. So, as Paul mentioned, healthier women are more likely to get pregnant. Unhealthy HIV-infected women are probably less likely to get pregnant.

 Let me ask a very quick question. HIV-positive with a positive viral count in a pregnant woman giving birth: if you treat around the time of delivery, and you treat the infant, does the infant become viremic or not?

 Mostly not. Really, the ones that are are the cases where it’s missed entirely. Unfortunately, women can become HIV-infected during pregnancy, and it previously was recommended to have 1 HIV test during pregnancy. Now, it’s actually recommended that they should be tested twice if they come in early because you’re actually at greater risk of infection during pregnancy because of changes in the cervix and that sort of thing. So, that has occasionally happened, where a woman was tested, was negative, gets infected during pregnancy, and has very high viremia, obviously. This is high risk to the infant, and that’s the scenario where we see transmission.

 And there are still rare cases that we all see of women who present in labor that had no prenatal care and are found out to have a rapid HIV test when they come in in labor. And they’ll get antiretrovirals and the baby will get antiretrovirals. We’ll be able to reduce the risk, but it’s not going to be zero.

 If she hadn’t been previously tested during her pregnancy, absolutely.

HIV in Pregnancy

 You mentioned hepatitis B and hepatitis C. How do you approach that? Is it a distinct subgroup of people who are HIV-positive?

 Well, a lot of people with HIV also have hepatitis B or C co-infection.

 The drug that we would prefer to use in individuals with hepatitis B is tenofovir. Tenofovir not only treats HIV, but it also treats hepatitis B. The TAF (tenofovir alafenamide) formulation has recently been FDA-approved to treat hepatitis B as well. So for those individuals with hepatitis B, we definitely want to give them a tenofovir-inclusive regimen. Usually we’re, as Eric said, thinking about a nucleoside combination, which is either going to be TAF/FTC (tenofovir alafenamide with emtricitabine) or abacavir/lamivudine. That’s going to be the nucleoside component. So, if somebody’s got hepatitis B co-infection, we’re going to give them a tenofovir-inclusive regimen.

 In hepatitis C, again, the great thing about the integrase inhibitors is they don’t have a lot of drug—drug interactions.

 Yes. So we can use dolutegravir or raltegravir. We could use abacavir/lamivudine. There’s few drug—drug interactions with the TAF formulation. So, for the most part, these same regimens are pretty safe and effective. Occasionally, there’s a drug–drug interaction that an astute HIV hepatitis C provider will be aware of. But for the most part, hepatitis C doesn’t influence our treatment decision.

 What about comorbidities (ie, patients with bone disease, kidney disease)?

 That’s where it gets a little sticky. You get people that are newly diagnosed, but they’re older. They have diabetes, perhaps, or they have renal dysfunction. Then you really have to be careful and that definitely influences your choices. You have women that have multiple pregnancies and may be really at risk for osteoporosis. We know that tenofovir, especially the TDF form (tenofovir disoproxil fumarate), lowers bone density, but the TAF form appears not to do that. So, you want to think very carefully about the comorbidities. There is the drug—drug interaction issue, and in people that are on some of the more complicated diabetes regimens, maybe there are things you need to consider. But, in general, again, as Ian said about hepatitis C, what we’re choosing now is really so tolerable and the interactions are few. Unless you have really frank renal failure—then you’re kind of a little bit more stuck.

 We face the issue with comorbidities, more commonly, in the people who have been on therapy for years, rather than in newly diagnosed patients. All of us have this panel of patients, now, who are aging with us. They’ve been on HIV treatment, sometimes very toxic HIV treatments for 5, 10, 15 years. And so they’re the ones we sort of struggle with in terms of the comorbidities—more than the newly diagnosed patients. One of the companies tried to do a study of people with newly diagnosed HIV in renal disease, and it’s actually very hard to do because there aren’t that many people who have renal disease.

 We talk about it a lot, but it’s not that common.

When Comorbidities Influence HIV Management

 This is all very encouraging. When you’re talking about all these things, I get a sense that you’re kind of down in the noise. You think about them, but there aren’t all that many people. What about the HIV-positive patients with the opportunistic infections—the ones that I would call tuberculosis (TB), cryptococcal meningitis, PCP (pneumocystis pneumonia), and all these things? We used to say that, “It isn’t the HIV that’s killing them, it’s all these other things.” How do you balance the ART (antiretroviral) therapy with the therapy for the opportunistic infections?

 It turns out that the patients who benefit the most from ART are just the people you described—the ones with opportunistic infections. So, there’s a sense of urgency we have of getting them on to ART as soon as feasible&mdash;much more so for those patients than for any other group.

We recently had a woman in our hospital who was newly diagnosed with cerebral toxoplasmosis. Her CD4 cell count was 14. Yes, we treat the cerebral toxoplasmosis, but the clock is ticking to get them on antiretroviral therapy. Now, you have to put up with some possible drug—drug interactions, some polypharmacy, and maybe some immune reconstitution inflammatory syndrome (IRIS), but those are trade-offs we’re willing to make in exchange for the improved outcome that’s been proven in randomized clinical trials for people who start ART early rather than waiting.

 And that’s really what’s amazing with HIV. We’re so fortunate in that there have been so many large randomized controlled trials that inform all of our decision making. In particular, we have large randomized controlled trials that tell us that people who present with advanced immune suppression and opportunistic infections actually benefit, not just from early therapy, but if you start them within 2 weeks (somebody with PCP), they’ll do better than if you wait 6 weeks. And the same thing is true for people with TB with advanced immune suppression (CD4 less than 50). We have a lot of great data. One of the conundrums has always been cryptococcal meningitis, which is probably one of the most common opportunistic infections people present with in our hospital nowadays. And that’s hard because of this IRIS issue.

 You need to define that. What’s an IRIS?

 This immune reconstitution inflammatory syndrome, where you enhance their immune response and then they have an inflammatory reaction to the antigen. And, unfortunately, the antigen, in this case, is residing in a fixed compartment in the central nervous system.

 They develop cerebral edema, and not only is it a bad place to develop inflammation, but it’s also really common. About 25%, 30% of the people develop cryptococcal meningitis.

 Cryptococcal meningitis is that common. And really, their best hope is a competent immune system. Yet, it’s their immune system that may kill them. What do you do as doctors?

 It’s a delicate balancing act. What I always love to talk to the residents and fellows about, when we have a case like this, is, I say, “Why don’t you read the paragraph in the guidelines?” Because it basically says, “Well, you could start early, but then you might not want to start early, and you could wait until it’s kind of late, but that might not be good either.”

The biggest waffle in the guidelines is, what do we do then? And we still fight about it because the best randomized data comes from a resource-limited setting; it comes from Africa. And those data suggest that if you start it early, you actually increase mortality.

 They don’t suggest it. They definitively demonstrated it in the clinical trial.

 So, what are you saying? Do you go after the cryptococcal meningitis first with an anti-cryptococcal drug and knock the burden down?

The question is, when in that process do you start? And I think that in a first-world setting (resource rich setting), that you should start early. You don’t start day 1 or day 2, but you start around 2 weeks in. That’s what I do. You get ready for this IRIS, you prepare for it, and you prepare the patient for it.

 Four to 6 weeks on anticryptococcal therapy?

 There is an analysis of timing of ART in resource-rich settings that suggests that early is probably okay. It’s not randomized. It was only a small number of randomized patients in one study. Can I mention about tuberculosis?

 Tuberculosis is decreasingly common in the United States, but it still happens. It is extremely challenging to treat because TB treatment includes rifampin, and rifampin is the biggest pain in the world when it comes to drug interactions. So, a lot of the drugs that we commonly use for ART don’t get along well with rifampin. And so it’s sort of this separate world of people with HIV and TB—they get a slightly different antiretroviral approach than people without TB.

You can see some serious IRIS with someone that actually has a temperature for days on end. You have to give them steroids.

 Oh great. You’re going to give somebody who’s immunocompromised steroids?

 But I guess the saving grace is that it’s not a closed compartment?

TB meningitis—that’s as bad as cryptococcal meningitis, perhaps even worse. We just never see it.

 But there, too, you have at least a static therapy. I’m not sure if anything is really cidal against TB at the end of the day. But you can knock it down.

 My early infectious disease professor said, “Antibiotics, shmantibiotics. It’s the white cells that kill everything.” And all you’re doing is holding everything in check for the body to go after it. You’re putting that in steroids when you give them the ART therapy.

 You have to treat bad IRIS with something because, as Joe was alluding to, these are patients who are very sick. They have really hectic fevers. If they have mesenteric adenitis, they have terrible abdominal pain. If they have lymph nodes in the chest, they’re coughing. It can be very difficult.

And we’ve certainly all seen abdominal rupture from mesenteric nodes in the IRIS. So, it’s not just about treating fever.

Managing Opportunistic Infections in HIV

 Let’s go back to history again. For all of my patients with HIV early on, we treated them with prophylactic antibiotics—whether it was with Bactrim or something&mdash;because we knew somewhere down the line they were going to catch something bad. But now you’re telling me that you can control the HIV really well. So where do we stand on prophylaxis? What do you use as an indicator? Is it the viral count? Is it the CD4 count? Where do you go with that?

 It’s basically the CD4 count, but I think we use far less prophylaxis than we used to. I think the only disease I prophylax is pneumocystis in a naïve patient.

 But it’s tough to teach the residents that. It’s so hard.

 Right. Don’t mess up their lives. Get them on antiretroviral therapy.

 It’s just a holdover. People look at these ancient recommendations that haven’t really been effectively updated, I think.

They can’t be. They get the studies that were done. For example, consider mycobacterium avium complex.

That study would have to be gigantic. Who’s going to fund it?

 But the point is that the guidelines are changing. The guidelines do say, for example, for MAI (mycobacterium avium-intracellulare) prophylaxis, you should start them on antiretroviral therapy. But you don’t necessarily need to start.

That’s what one set of guidelines says.

 Okay. Well, that takes care of mycobacteria.

 But not everybody that is taking care of patients and you tell them to start their antiretrovirals does so. In fact, often the people who get admitted to the hospital are not just newly diagnosed. They’re the people who leave and don’t start antiretrovirals.

 But if those people are not taking their antiretrovirals, they are not going to be taking their MAC (mycobacterium avium complex) prophylaxis.

 It’s really HIV therapy. The important thing for providers to understand is that when viral replication is under control, it doesn’t matter what the CD4 count is. It can be 50, it can be 10.

 Are you telling me that somebody whose CD4 count is low and whose viral count is low is not at risk for mycobacterium avium?

 Certainly, the risk is dramatically reduced if people are suppressed, even if their CD4 remains low. However, if you look across a large population, they do have higher risk of opportunistic infections.

 Compared to all-comers with HIV. But I do agree with Ian. It really is probably pneumocystis prophylaxis that I wouldn’t shed.

 For skin test positive, that is people who have been exposed?

 If somebody is on effective antiretroviral therapy, their virus count is low, and their CD4 count is okay, would you or would you not prophylax for pneumocystis? We’re 50/50?

 If they are between 100 and 200, I don’t prophylax.

 Yes. In someone who has poor immune recovery, whose CD4 is between 100 and 200, and whose viral load is suppressed, I do not give them pneumocystis prophylaxis.

 And that’s consistent with the guidelines?

 It’s the recommendations. If their CD4 count is greater than 100 and they’re virologically suppressed, they don’t need to be prophylaxed. It’s also important for folks to understand that really, the situation that I was mentioning before almost doesn’t exist. When you’ve got viral replication control, the CD4 count is not going to be 10. And on these new drugs, CD4 counts often rise 100 cells during the first month of therapy. So they rapidly get their CD4 count above a threshold that we’ve traditionally associated with being at risk for these opportunistic infections.

 Okay. And just to be clear, again, somebody is on effective (however you want to define that) ART therapy?

 We know how to define it—viral load below the limit of detection. That’s actually not an argument. You won’t get an argument there.

 Okay. Viral load below the limit of detection. Are you prophylaxing for mycobacterium avium complex?

 CDC guidelines still include it. And although the data are old that support using the prophylaxis, they’re never going to do the study to show that it’s not necessary. So we’ve just left it in there. But that’s kind of lazy, because the studies were all done in the early 1990s, and that’s not the current era that we’re in for ART.

But to clarify, on the other hand, if someone comes in and their CD4 is 50…

and you start them on ART, I do give them pneumocystis prophylaxis until they get above 100, consistently.

Preventing Opportunistic Infections in HIV

 Let’s talk about monitoring patients with HIV. Laboratory monitoring: how often do you do CD4? How often do you do HIV-RNA monitoring? Do you do it periodically? Is there a schedule? What are you doing?

 It’s something that I think we’ve all laughed at ourselves, in clinics, for years and years, because we’ve routinely ordered these tests every time people came in and we often saw them just as often. But then we would never do anything with the results.

 Especially the CD4. So people sort of revisited this. I think there’s a consensus that you want to know what the CD4 is at baseline to know whether they need to be on any prophylaxis, and usually, you get a baseline viral load. Even that, arguably (unless it’s undetectable), is usually useless, because your goal of therapy is undetectable. And I don’t care whether you started at 1000 or 500,000—the goal is the same. And then we monitor viral loads pretty close at the beginning. And some of this is probably because of the idea that if they aren’t responding well or they’re experiencing failure, we want to know it earlier than later. My guess is it’s more a marker for adherence for us, so we’re using the viral load. We’re not going to do anything based on the CD4 for months and months, so the utility of doing a lot of CD4s early on [is] useless. We do the viral load, and once a person is undetectable, then the only reason we’re doing the viral load is to see if they experience viral rebound. There’s a general consensus and there’s data that suggest that once somebody is undetectable for a long time, they’re not likely to rebound unless something happens.

The Evolution of Antiretroviral Therapy for HIV

Who Should Be Tested for HIV?

Pre- and Post-Exposure Prophylaxis in HIV

Initiation of Antiretroviral Therapy in HIV

Nuances of First-line Antiretroviral Therapy in HIV

HIV in Pregnancy

When Comorbidities Influence HIV Management

Managing Opportunistic Infections in HIV

Preventing Opportunistic Infections in HIV

HIV: Understanding the Laboratory Monitoring Needs

Improving HIV Treatment Adherence

Improving Care: Rationale for Switching HIV Therapies

Switch/Maintenance Therapy in Patients With HIV

Novel Mechanisms of Action in HIV Treatment

Practical Experience With Emerging Agents in HIV

Long-Acting Formulations Emerging in HIV Treatment

The Past, Present, and Future of HIV Management