Important note: Information in this article was accurate in October 2000. The state of the art may have changed since the publication date.A reminder of where we've come from: in 1987 at the first International AIDS Conference that I attended, there were two things I remember very clearly. First, in the exhibit area there was a huge map of the United States which said on the top: "The Problem: AIDS. The Solution: Banish All Homosexuals." It had little pushpins on all the major cities of the United States indicating how many homosexuals lived in each. Now the fact that (a) this made it into the exhibit area at an international AIDS meeting, and that (b) no one, save one brave gentlemen, tried for the whole week to push the table over (he was quickly whisked out), says to me something amazing about where we were then.
My other memory from that Conference was then-Vice President George Bush addressing the assembly. The people who attended AIDS conferences in those days were mostly white men in suits, and as soon as he got up and started to talk, they started to boo him. (Remember -- this was 7 years into the Reagan Administration and it would be another year before Reagan uttered the word AIDS.) When people started to boo, Bush stepped back from the podium a bit, turned around and said to the Secret Service agent standing behind him (thinking he was turned away from the microphone), "It must be a gay group." The words reverberated throughout the auditorium of The Washington Hilton.
Now that's some International AIDS Conference history!
There were several studies at the conference about mother-to-child transmission (MTCT), which remains an especially big issue in Africa. (One study, published last year, looked at mothers during pregnancy and instances of transmission to offspring, according to various levels of viral load, with and without AZT in each category. It showed two things: first, in most of the subgroups without AZT, there was more transmission than with AZT. But what's really significant about this study is that no mother on this study with a viral load lower than 1,000 transmitted to her baby. I think it's really important when we look at cofactors, whatever the issue is (including evaluating people for post-exposure prophylaxis) that we consider the degree of viral load an extremely important co-factor.
A couple of things that came out at Durban regarding MTCT. AZT is the standard that we've used for many years. We know that, as a single drug, it can cut transmission by two-thirds. Over the last several years we've been in the habit, rightfully so, of giving most pregnant women triple combination therapies. But we've never had any data on d4T or ddI, although many women have taken it, until Bristol-Myers Squibb, which manufactures both of these drugs, did a study comparing AZT versus d4T versus ddI versus d4T plus ddI. Each drug was given, not during the last two trimesters of pregnancy, but rather started just prior to labor, given throughout labor, and also given to the infant for 6 weeks after birth. The cost was under $100 per pregnancy, and overall, for all four regimens, there was no statistically significant difference in transmission. (Overall transmission rate was small -- only 3.6% across the four groups). This is more evidence that antiviral therapy, whatever it is, can help reduce transmission and the drugs we use are reasonably safe -- with the exception of efavirenz? (Sustiva), which is not recommended during pregnancy (because of published reports of fetal abnormalities in monkeys).
The Saint study, which received a lot of press over the last year, compared nevirapine to AZT plus 3TC. The non-nucleoside nevirapine was given as a single dose -- one time only -- to a woman about to deliver, and whose baby then got one dose when about two days old. We're talking about something that costs about $4, and basically cuts the transmission rate in half during each treatment arm, (to 14% and 10.8% respectively for nevirapine and AZT/3TC). The point this study makes is that even if a woman gets no treatment for HIV all the way through pregnancy, there is something that can be done at the very end of pregnancy that would still reduce transmission. This is phenomenal evidence, which takes us far beyond Yokohama six years ago, when we were looking at AZT through the last six months of the pregnancy. It also underscores the importance of such a treatment in a place like Africa where we have to talk about costs in pennies and dollars, and where women don't have the facilities for prenatal care. (There was some concern about a single dose of nevirapine being associated with resistance, but that's not been validated yet.)
Other studies presented in Durban indicated that breastfeeding is associated with a 1.8 times higher risk of transmission -- not surprising -- and higher viral loads are associated with transmission. In a major session on breastfeeding and HIV transmission, it was found in the Durban cohorts that infants who were formula fed -- no breastfeeding -- had a much lower risk of transmission than those who were breastfed. But infants who got both have the highest transmission rate of all (19.4% formula versus 27.4% breast versus 35% both). This is not understood, although there are several hypotheses being considered.
Circumcision may be one of the major interventions we can make in a place like Africa when you think of things that aren't medication and that are safe. We have to be sensitive to people's cultural beliefs, but this study looked at two places where there are very high circumcision rates (Yaounde, Cameroon: 99% and Coutonou, Benin: 99%) and two places where there are very low circumcision rates (Kismu, Kenya: 26.8% and Ndola, Zambia: 7%). The HIV prevalence was much lower in the two places with high circumcision rates as compared to the prevalence where circumcision was minimally practiced (3.8% and 4.4% versus 21.9% and 25.9%, respectively). We've seen similar data before but I think it's more crucial now than ever. The impact of circumcision on HIV transmission is pretty well understood. The glans of the uncircumcised penis is covered by the foreskin. This means that the glans stays a little spongier, a little more porous, and if it's bathed in fluid that has the virus, it might be more likely to admit the virus than a circumcised glans, which, because of exposure gets a little bit tougher.
Should we be giving anti-viral therapy to people who are not known to be HIV-infected but who have had a potential exposure? This is a question we've been asking for about four years now. Several studies in Durban showed zero seroconversions, which is good news. What we don't have from these studies is a single standardized regimen -- they all used different combinations. One used triple ddI, d4T and nelfinavir; one used Combivir (AZT/3TC)? for one month; and one used mostly three drugs -- Combivir? plus a protease inhibitor. So although we still don't really know the most appropriate regimen, timing and duration, if indeed there be one, one of the most important things we can take from this is that if somebody may have been exposed, it's worth speaking to a health provider immediately -- not four days later. It's recommended that post-exposure prophylaxis be started as soon as possible -- certainly within a day or two. We know from healthcare studies that the risk of actually getting HIV from a needle stick from someone who already has HIV is about one in 250 -- which is not very high. The three studies we looked at here totaled 330 people who all got some kind of post-exposure prophylaxis, so zero out of 330 is not yet statistically significant. (The 331st person could conceivably acquire HIV!) But certainly, evidence is mounting that post-exposure prophylaxis is a good option to consider. As a practicing physician, I would not turn down a patient who came to me and said, "I had unprotected sex with someone whom I know to be HIV+, or I really think is HIV+, and I want to go on prophylaxis."
There were two studies which basically looked at the same issue. Both asked whether, when viral load has gone from undetectable (meaning under 50) and now has increased slightly -- to 58, or 250 -- does that mean a regimen is failing? In a small group of 32 studied, 75% went back down to under 50, most of them on the very next blood draw. Only 8 out of 32 (25%) failed to reach resuppression and went on to other regimens. In a larger study of 240, 10% had a blip and the blip was usually not associated with viral rebound. I think that evidence is growing that viral blips are certainly not cause for panic. As we'll see when we look at structured treatment interruptions, there may actually be benefits to having blips, which is the whole premise of the STI theory.
The Atlantic study finally presented one year data at Durban. We're continually asking "Is there a best initial regimen?" In this study, people who were antiretroviral naive with T-cells over 200 were all given d4T and ddI, and then were randomized so that one-third got Crixivan? (indinavir, a protease inhibitor or PI) as the third drug, one-third got nevirapine (Viramune, a non-nucleoside reverse transcriptase inhibitor), and a third got 3TC (Epivir, a nucleoside analog, or nuke). In 298 people, of those who achieved viral load under 50 (meaning undetectable) in the stricter "intent-to-treat" analysis at one year, the results for all three drugs basically look the same. If we look at the people as treated (those who stayed on the drug), those who remained undetectable were 91% of the Crixivan group, 82% of the Viramune group, and 57% of the 3TC. (The difference between the 3TC group compared to the Crixivan group is the only one that's statistically significant.) What it appears we got from the Atlantic study is simply that in antiretroviral naive patients with a median viral load which was relatively low (around 15,000 copies), in the first year of therapy, the PIs and the non-nukes are comparable, but the potency of the triple nuke regimen may not be as great.
One study presented in Durban took Combivir and Crixivan, and compared it Combivir and Ziagen (abacavir). Glaxo Wellcome, who makes Combivir, also makes abacavir, also a nucleoside RTI, (i.e., in the same class as AZT and 3TC). If you want to spare the PI and the non-nuke classes at the same time, and save them both for later, one of the big questions is whether you can do so by using three nukes. This study basically shows that comparing the two groups, there was no statistically significant difference in the people who are achieving undetectable viral load, nor for the subgroup who are over 100,000 viral load to start. Numbers are also comparable as to the change in T-cells. This data is after only 24 weeks, and one of the things that gets people a little bit nervous is, just as the Atlantic study showed 3TC falling off after a year, what would we see at a year or two years of abacavir? For some, it feels counterintuitive to use three drugs that work at the same place in the HIV replication process. Then again, if they suppress viral load just as well, and allow sparing of the other two classes, might this not be a reasonable course to take? Stay tuned...
FYI: Glaxo Wellcome is taking these two drugs, AZT and 3TC, which are already in one pill (Combivir), and adding Ziagen to it to make a new drug called Trizivir, which may be FDA approved sometime over the next six months or so. If so, it will be one of the few times in history that the FDA has approved three medications in one pill. That would mean that a full HIV regimen in this case could be a single pill in the morning, and a single pill in the evening! Now from an adherence perspective, that's very exciting. The two fears about it are (a) this regimen, despite what we see in the beginning, might not be as potent in advanced disease or over the long haul, and (b) abacavir is a drug which has a potentially very serious -- theoretically fatal -- hypersensitivity reaction associated with it, so much so that every physician in the United States has received letters over the last year from Glaxo informing us of it. Yet we know what a major factor the number of pills is in determining adherence to antiretroviral regimens. It's an interesting situation, and we should know more by the time of the Retroviral Conference in Chicago next February.
Several studies provided some clarity in Durban, and the clarity is not good for hydroxyurea. One study looked at 98 people with over 100 T-cells: some were antiretroviral-naive and had never been treated, and some were experienced. All got ddI once a day, d4T twice a day and either hydroxyurea or placebo. At close to a year (48 weeks), the proportion of people who went undetectable was comparable whether they got hydroxyurea or placebo -- meaning the hydroxyurea did not give any additive boost to the ddI in terms of potency. In addition, the study was stopped due to increased toxicity for peripheral neuropathy from the hydroxyurea. In summary, there appeared to be no antiviral benefit in adding hydroxyurea for treatment-naïve patients, and although there may be more of a possible benefit for the treatment-experienced, one must bear in mind the possibility of increased peripheral neuropathy.
| See also: May 2000: ddI, d4T and Hydroxyurea Side Effect Warning! To read this article en Español, click here March 1998: Hydroxyurea-More, But Is It Enough? |
With three drugs, you can take 60% - 80% of people on the average triple regimen at one year and show that they have an undetectable viral load. That's good, but you may ask, what about the other 10% - 30% who aren't undetectable? It may be an adherence issue, but might it be an insufficient potency issue? One study that was reported in Durban enrolled 152 antiretroviral-naive people and took one of the most widely used regimens in the United States -- Combivir -- and the protease inhibitor nelfinavir, and compared it with Combivir, abacavir and amprenavir -- a quadruple regimen (which spares the non-nuke class). At one year, the percentage who achieved undetectable viral load was greater in the 3-drug group than the 4-drug group. The 4-drug group had many more toxic failures, much greater incidence of nausea and emesis and a much greater incidence of rash. Also, although not statistically significantly, a greater number on the 4-drug withdrew from the study. So we may be on the verge of accepting that we can do all right with three drugs, but we have to learn how to really use the three well, and how to improve adherence.
Two large studies provide data that say that lack of adherence is associated with failure of the regimen. A study of 736 people defined "good adherence" as taking more than 90% of medications. Over a period of almost two years (93 weeks), 68% were good adherers. In this cohort, there were only two factors which predicted death: an AIDS diagnosis had a 64% higher risk and poor adherence had a 67% higher risk. 1.8% of the "good adherers" died, versus 8.4% of the "poor adherers." In another study (from Canada) of 950 patients studied over a 13 month period, the risk of death was also greater for people with any of three factors: an AIDS diagnosis, lower T-cell count, or poor adherence.
Lipodystrophy Update There were several studies, and for many, not a lot of new information. I think we clearly understand that you can get lipodystrophy with non-nukes, with nukes alone, even with no treatment at all, but it is more likely with protease inhibitors. In an Italian study of 321 subjects, the development of lipodystrophy was more rapid for people who are on a first regimen. In the Swiss cohort, (a very well respected study of almost 1400 people), 42% of people said they had some degree of fat redistribution. This study found d4T and ddI were more likely than AZT and 3TC to cause the problem. The HOPS study, an American cohort of more than a thousand people, found that around 50% have some degree of fat redistribution. Factors associated with a greater likelihood thereof were: older age, duration of PI therapy greater than two years, AIDS diagnosis greater than seven years, and greater time since HIV diagnosis. It's pretty clear after Durban that we can say you're more likely to develop lipodystrophy if you're older. A Spanish study of 118 found no difference between AZT or d4T in incidence of fat redistribution, although many have observed anecdotally that d4T is more likely to cause facial wasting than AZT or the other nucleosides. (If this does turn out to be true, i.e., more facial wasting with d4T, we must remember that it is also the only one of the 14 approved anti-HIV medications for which we virtually never see high-grade resistance.)
Before people start getting afraid of HAART therapy because of anticipating body fat change, let's look back for a moment to January 1996 in Washington at the Third National Retroviral Meeting. This was right after the first protease inhibitor, saquinavir, had been approved -- we didn't have ritonavir, indinavir, nelfinavir or amprenavir yet. A study was presented in which more than a thousand persons with AIDS, all with less than 50 T-cells (and thus, a life expectancy of months to maybe a couple of years) were randomized so that half of them had ritonavir (in a clinical trial) simply added on top of their existing nucleoside regimen (and we know today many were likely resistant to one or more of the drugs they were already taking). The startling finding was that within months, the ritonavir-treated group started showing a statistically significant decline in new complications, and ultimately, their survival started to increase as well. This of course was the heralding of the HAART era -- but the findings, which made the audience nearly four years ago (myself included) gasp with excitement, are still relevant today: Simply adding a PI to a probably failing regimen improved outcome even within the first months. HAART therapy works! And while the fears about body fat changes are well-founded and need to be dealt with appropriately (including finding better-tolerated regimens), those fears should be balanced with a healthy respect -- humility, if you will -- for what these drugs can do!
The Switch Studies Does lipodystrophy occur more frequently with the use of protease inhibitors? Does it help to switch to a non-nuke nevirapine, efavirenz, etc. A meta-analysis -- an analysis in which other previously done studies are "blended" and collated together -- was presented at Durban. The meta-analysis took 11 studies (so you get a good number -- in this case, 446 patients) in which people were switched from a PI to nevirapine (a non-nucleoside). In total, over 90% remained undetectable at six months, indicating the switch from a PI to a non-nuke means a good chance of retaining undetectable viral load, (although 7+% didn't). The mean cholesterol and triglyceride levels both decreased slightly on the nevirapine, but this drop was not statistically significant. And, perhaps most importantly of all, there was no consistent change in body shape. Now, this is a meta-analysis and we must be mindful that every individual has her or his own experience. Some people have reported dramatic changes in restoring former body shape. So, on the basis of this analysis, we can't definitively say that if you switch from a PI to a non-nuke, it's going to make lipodystrophy better, or it's going to make your lipids better, but at least it's a reasonably safe thing to do in terms of virological suppression.
Lactic Acidosis As protease inhibitors and non-nukes take a hit for lipodystrophy and fat changes, the nukes such as AZT and d4T have taken a big hit over the last year for lactic acidosis. Lactic acidosis is a condition in which the normal mitochondria -- the little organelles inside each cell that engage the metabolism -- get poisoned, basically. Mitochondrial toxicity may be responsible for neuropathy, kidney disease, pancreatitis, muscle atrophy, anemia and a host of other known complications. One of the things that happens when the mitochondria can't work anymore is that the body shifts its metabolism to another pathway, and a chemical called lactic acid, which we all have inside of us in low amounts, tends to accumulate. The symptoms of too high a lactase level include loss of appetite, abnormal liver function tests, and most of all, fatigue. Most HIV providers today know this, but I think many generalists do not. If your provider is not well versed in HIV, he or she may not know about the potential association. The good news is that it's a really easy thing to diagnose -- a fairly simple blood test can be done when you have your other tests. I think that for people who have been heavily treated, (with nucleosides especially) and constantly feel awful, the lactic acid (lactate) level should be measured. If elevated, the most we know to do now is to stop antiretroviral therapies and see if symptoms disappear and levels normalize. (Other studies in progress are looking at specific therapies for lactic acidosis.)
Is there a better drug to use first for your third drug? Is a PI or a non-nuke better? Although it's not the first time the issue has been raised, one of the interesting studies at Durban found that among 164 people who are already on nucleosides -- in this case, over a three-year period -- a third of them were hypersensitive to a non-nuke. (This concept of "hypersensitive" is a good thing, because it means that a little drug goes a long way, or the same amount of drug goes even further in suppressing viral activity.) In addition, 11% were hypersensitive to all three non-nukes that are currently available. They also found that the longer the nucleoside RTI exposure has been, the more likely that one is hypersensitive to non-nucleosides. It's too early to say beyond a shadow of a doubt, but we can make a reasonable hypothesis here -- that if you use a PI first, meaning you get more time to have been on nucleosides (since these are virtually always used as part of a pi-containing regimen), you may get a better "bang" from your non-nuke when you eventually use it. It's an interesting avenue for exploration but we need more data.
These tests are not yet FDA approved but we've been using them quite a bit in the last year, particularly the genotyping test. With genotyping, your virus is studied by a computer program and we see which mutations it has acquired, which can help us infer which drugs you may not respond to. In a phenotyping test (which costs two to three times as much and takes two to three times as long to do, and is not as far developed as genotyping), you actually put your virus in company with a particular drug and see how well the drug suppresses it. The phenotypic test is really much more direct, the genotyping more indirect. One of the things that we've been asking is whether these tests help at all. There have been ample studies published showing that genotyping does confer an advantage in terms of making your provider more likely to use drugs that are going to help you. Similar data for phenotyping were presented over the last year. But here's a study that for the first time in my memory actually compared all three in the same study: phenotypic testing, genotypic testing and SOC, which means standard of care (that is, your provider tries to make the best decision without benefit of either test). In this test, out of 541 who were failing multiple HAART regimens, the percentage who got down to under 200 (we aren't shown what the starting viral load is) at three months was basically the same, whether the providers got phenotypic advice, genotypic advice, or no advice at all. Of interest, 50% of the people in the SOC group were prescribed more than three new drugs, versus only 20% of the people in the other groups who were prescribed more than three new drugs. What I think that may indicate is if we don't have the typing information, we may make more changes in finding the right regimen for you, but the end result may be the same. I think we'll see more studies about the relative merits of these tests, especially given the large sums of development money behind them.
Regarding the popular theory that protease inhibitors cause diarrhea, we should also remember that diarrhea is a symptom of HIV infection in some people, and that one of the advantages of starting HAART therapy is that the diarrhea, the neuropathy, the thrush, the KS, etc. stop for many people. One study of new protease inhibitor prescriptions filled by 1,144 people in San Francisco (over half got nelfinavir, others indinavir or ritonavir) indicated that about a third of people with a new PI prescription also filled prescriptions for antiarrheal medicine. But of the 228 who had diarrhea before the protease and were already on antidiarrheal meds, the number who subsequently needed treatment for diarrhea diminished substantially. It's important to remember, when we want to blast drugs for diarrhea, that there are some people whose diarrhea has improved. We need to look at the same issues with neuropathy, which can be a symptom of HIV infection. For some people on HAART therapy, neuropathy improves. Of course, for others, especially those on ddI or d4T, neuropathy may get worse.
Several smaller trials included a test of Source QITM, a Chinese herbal medicine, which showed a modest decrease in stool among 16 people (not quite statistically significant). Sb300, which is available over the counter and made by Shaman Botanicals, provided further validation of its decreased stool rates. Another interesting small trial indicated that nelfinavir (Viracept), the most prescribed PI, which is definitely associated with diarrhea, had increased stool output correlated with lactose intake (lactose is the sugar found in milk and dairy products). The trial only studied eight people, but it may be if you're on Viracept? and you get diarrhea, that you consider diminishing your lactose intake or maybe taking lactase pills to help you.
One study showed that 22% of patients on PIs had erectile dysfunction. A second study showed a statistically significant greater chance of decreased libido/interest in sex in people on PIs versus people not on PIs. The study didn't look at non-nucleosides and didn't look at other regimens. (One of the reasons may be that PIs get used the most and so also get slammed the most because they're the most studied!)
We're still trying to figure out the best thing to do with Serostim, a very expensive pharmaceutical, priced at upwards of $70,000 a year (although there's a cap of $36,000 now for any one person). The irony of history is just after Serostim was FDA-approved for AIDS-related wasting, protease inhibitors came out and, essentially, people stopped wasting. The data we've had up to now on its efficacy has been based largely on three-month trials, which showed a (desired) loss of fat as well as a gain in muscle. In Durban, however, we finally got some data that goes beyond three months. One study showed an average five kilogram gain of lean mass and equal loss of fat (again, not new information), but this study went out to half a year -- and found there was no further change between week 12 and week 24 in body fat and muscle. In addition, once people were taken off the Serostim, the beneficial changes were rapidly reversed. So, we still ask, if somebody does actually get AIDS-associated wasting, for which growth hormone is an FDA-approved treatment, what will be the actual benefit (clinically, i.e., in terms of decreased progression and increased survival) of a medication as expensive as this?
This is certainly the name of the game today. There were several studies comparing ritonavir/indinavir combinations. One study compared ritonavir plus indinavir dosed once a day with the more standard indinavir three times a day. There was good viral control in both regimens but perhaps increased kidney stones in the indinavir thrice daily group. The Nice study looked at ritonavir and indinavir at a commonly used 400/400 mg twice daily versus standard indinavir (two pills = 800 mg thrice daily) and found that there were increased side effects in the combination group but better adherence (i.e., no food restrictions, fewer dosings). I think there's ample evidence that fewer pills -- or, maybe more importantly, fewer dosings -- is associated with better adherence, and some people will even tolerate more side effects if they can take fewer pills and/or fewer doses. The BEST study (name of trial, not anyone's comment on it!) looked at a different dose of ritonavir, 100 mg twice daily, plus indinavir, 800 mg twice daily -- a dosage used in many practices today -- and once again compared it to standard indinavir treatment, 800 mg thrice daily. It found a similar tradeoff.
There was also a study of ddI once a day, nevirapine also once daily, and d4T twice a day, which showed a comparable viral load reduction at one year. In this case, two out of the three meds were taken as a single dose.
One study looked at nelfinavir with ritonavir once daily. (Ritonavir is a PI which has perhaps found its greatest use for its pharmacokinetic property of suppressing the metabolism of the other protease inhibitors and therefore causing them to remain in the bloodstream longer. The non-nuke delavirdine -- which has been minimally used by the treating community -- seems to share with ritonavir the ability to raise levels of other PIs.) This pharmacokinetic study had four arms: the typical nelfinavir, 1,250 mg twice a day -- by the end of 12 hours, the drug level in the blood had dropped down, which is why it's taken twice a day. One of the other dosings was one ritonavir a day with 2,500 mg nelfinavir once daily (At 11 pills, it's not an easy thing to take, but it's only once a day.) At 24 hours out, there was still drug action in the bloodstream. Viracept is being reformulated and by about six months to a year from now, there will be a 625 mg pill, so we could be talking about five pills (four nelfinavirs at 625 mg, plus one 100 mg ritonavir) once a day for the complete PI portion of an antiviral regimen.
I want to summarize a study, not from the International Conference, but just released, about which Bristol-Myers Squibb wrote a letter to doctors throughout the US. BMS a1454-148 compared AZT/3TC and nelfinavir versus d4T/ddI and nelfinavir. The ddI was given once a day (versus the other drugs being taken twice daily). There was an indication of some possible inferiority in the ability of the ddI/d4T regimen to keep the viral load suppressed, as compared to the AZT/3TC regimen. So, in a very responsible way, Bristol Myers said the preferred dosing is twice daily for ddI. This is an appropriate response. What I do in my practice is to follow patients' viral loads. If somebody's been on once daily ddI and the viral load remains under 50 (undetectable), and they love taking it just once a day, at this point many of my peers and I won't switch people back to twice a day but will keep an eye on the viral load.
Perhaps this was the leading treatment story to come out of Durban (ironically so, in Africa, a place where so many treatments are unavailable -- and maybe the most exciting thing on the horizon for developed nations is treatment interruptions). Our whole concept of structured treatment interruptions may hinge on the concept of CTLS: cytotoxic t-lymphocytes. We know that HIV, or almost any germ we acquire, causes the body to produce proteins, called antibodies, in response. We know that with HIV, unlike with measles, hepatitis, or many other infectious conditions, the antibody itself is not enough to quell the progression of the infection. It seems that the body, in addition to antibodies, needs certain lymphocytes -- a specific type of white blood cells -- that have been specially programmed to kill HIV infected T-cells. These are the CTLS. Now the only way we get CTLS, since nobody's born with them, is to have virus present to stimulate the lymphocytes and turn them into CTLS. So, the theory goes that if the viral load is suppressed way down and kept there, you may be doing okay but you're not going to get CTL activity. And many people think that CTLS are a necessary part of long term control of HIV. Thus, if you start with a suppressed viral load, but then stop the meds for a while and let the virus bounce up a little, this will enable an increase in CTL activity. But then you would go back on the medicine because you don't want too much virus to be floating around for too long. So you start this rhythmic pattern of interruptions, and over time your CTL function rises. Now, we all ask, for how long do you interrupt? And how high do you let the viral load go? And does this really work in people besides the famous Berlin patient? We don't know.
A study reported earlier this year at the National Retroviral Conference in San Francisco looked at 238 subjects who stopped antiretroviral therapy for various reasons, (usually side effects and/or patient preference). This was not a controlled, randomized, prospective study -- they simply went back and looked at these people. They found that only 5.5% remained at undetectable after three months, so the generally accepted idea for now is that if you stop your medications, the virus will return. (But as mentioned above, given the STI theory, this could possibly be put to good "immune" use.)
Dr. Tony Fauci presented in Durban very preliminary results of two very small trials. The first protocol (of fewer than a dozen subjects) studied two months of HAART therapy followed by one month off. During the month off, viral load did rebound, as expected. He reported that the successive rebounds were looking lower, but there was not enough data to approach statistical significance. The second protocol (of seven subjects) was one week on and one week off (which has many people nervous because of the potential for resistance developing due to repeated instances of partial drug levels in the blood), and found no rebound of viral load in six and a slight increase in the seventh of the participants. It's unlikely that one week off was enough time for most of the people to have any rebound. Fauci was extremely cautious about anybody doing this other than in an experimental protocol
The well-respected Swiss cohort had a larger group of 122 people who were on a slightly different protocol -- eight weeks on therapy, then off for two weeks, repeated five cycles -- so almost an entire year. By the Conference, 56 had completed four cycles. There are four possible outcomes: (a) no rebound at all, (b) each rebound comparable to the last one, (c), the rebounds become higher each time, or (d) the rebounds become lower each time (which is what we think we might see if CTL activity is being appropriately stimulated). What this study showed is that all four scenarios occurred, spread over the 56 patients, with no single outcome showing statistical superiority over the others. This is clearly still a wide-open subject of study and discussion.
Dr. David Ho spoke at a symposium and stated that entry inhibitors could be the next exciting place to study. There are currently three different kinds of drugs being studied that can block the virus from getting into the t-cells: attachment inhibitors, co-receptor inhibitors and fusion inhibitors. The reason Ho is so excited about these is that people have wondered for years about treating virus that's floating around the bloodstream, as opposed to virus already in the t-cells, which is what we've been treating. Treating virus before it even gets into the target cell would be well before our already established therapies (nucleoside analogues, NNRTIs, and PIs) work -- since these target virus already within the cell.
ABT-378, now known as lopinavir (brand name Kaletra) is about to be FDA-approved, probably in the next few weeks. [See also FDA Approves New AIDS Drug Kaletra (AP, 09-15)]. It'll be our fifteenth antiretroviral drug and is a powerful protease inhibitor. Abbott Pharmaceuticals hopes it will be perched, because of its demonstrated potency, to be used as a first-line therapy. (Personally, I'm not sure this will happen, at least right away -- many HIV providers tend to be, with good justification, cautious about how to incorporate new therapies.) Lopinavir is going to be combined with a little bit of ritonavir, Abbott's own drug, in each pill to boost its levels (to well above what we've seen with any of the already available anti-HIV drugs). The dose of Kaletra will be three pills twice daily, preferably but not mandatorily to be taken with food.
I'd like to mention an earlier study presented at the Resistance Meeting held just prior to this Conference. Usually with the use of protease inhibitors, by the time a person acquires four or five mutations, they're considered pretty resistant. What this study indicated is that people who have six or seven mutations still have some degree of viral activity from ABT-378, so there is some emerging data to confirm Abbott's belief that this drug might be able to be better withstood without the development of resistance. In the development studies, there is an incidence of diarrhea of more than 20%, which, as usual, might rise once the drug gets on the market. I will tell you anecdotally that I have had some very good results with this drug for a patient who had never been able to achieve undetectable load before. So, we may be entering a new, exciting era that will solve some of the problems we've been facing. But I do have a personal, intuitive feeling that many treatment specialists will not be starting clients on, for example, Combivir and Kaletra? up front, but that we will save that as a later (salvage?) regimen. I realize you can argue and ask why -- if it's so strong, why not use it right up front? I don't have a scientific answer for you on that. I think we need to give the drug four to six months and see what happens once it is formally released.
With interleukin-2, the substance that helps make more t-cells, we're still at the same place that we were years ago. In a study of 78 reported in the Journal of the American Medical Association early in July, half had IL-2 added to their HAART regimens. After a year, there was a 112% rise in t-cells in the IL-2 group, versus 18% in those treated on HAART alone. Good news, but we must consider: IL-administration makes most people feel sick, like having a bad flu. It will not be re-labeled by the FDA for use in HIV until we get survival and progression outcomes. In other words, does creating more t-cells mean that it will take longer for people to get sick, or will they not get sick at all, and live longer? For a drug which costs in the neighborhood of $10,000 a year (added to the average HAART regimen, which is generally more than $10,000 itself), it will virtually double the cost of HAART therapy. Everybody with HIV wants more t-cells, but scientifically speaking we'll have to prove that it's really going to make a difference before this therapy will be approved.
There's been some talk the last few months in certain clinical settings as to whether the reversal of OIs we witnessed with the onset of HAART therapies in 1996 is now itself being reversed, i.e., that we are on the threshold of a recurrence of persons with HIV becoming seriously ill. One small presentation in Durban might validate this unproven observation. This Yale study compared a number of cases of Pneumocystis, mac and CMV -- three opportunistic infections we used to see all the time which we rarely see anymore -- and showed an increase in their occurrences from 1997 to 1998. I would add that the people who got mac were mostly not on prophylaxis This is a very treatable as well as preventable infection. People with 50 or 75 t-cells or less need to be on prophylaxis. What this tells me is that we shouldn't let our HAART-induced euphoria stand in the way of our using scientifically validated treatments when they are called for!
There was a striking study presented in Durban regarding our progress on KS (Kaposi's sarcoma). Think back to 15 years ago -- KS along with Pneumocystis occupied center stage in people with AIDS. People died from it; we treated it like a cancer, although we have since found out that it is caused by a virus, the so-called HHV-8 (human herpes virus 8). This study took 45 people who had KS on their skin only, not internally, and who had had no prior antiretroviral or KS therapies. All of them began a regular triple HAART regimen which included a protease inhibitor. At one year, a third of them had a complete remission, i.e., the KS was gone. Another third had partial remission. That's two-thirds improvement or remission from KS in people on HAART therapy alone. To me that's absolute evidence that these antiviral drugs do something for people taking them, especially when you think back to where we were with KS not too many years ago, blasting it with chemo. It's true that some people still need chemo, but it's clear we've come a long way.
Let's look at the median survival after a diagnosis of AIDS is made (not after HIV infection occurs, but an actual AIDS diagnosis). The median survival, by the way, is the point in time at which 50 out of 100 people would have died and 50 would still be alive. In 1982, the median survival with AIDS was less than six months (i.e., of 100 people with an AIDS diagnosis, half of them died within six months). By 1987, with better medical care but no real drug regimens, it had grown to about six months. In 1992, by which time we were using around two nucleosides, the median survival had grown to almost two years, and this may have largely been due to prophylaxis. To tell you now the median survival for persons diagnosed in, let's say, 1995, we would have to find that point in time at which 50% of those diagnosed with AIDS back then had died. But in Los Angeles County, that hasn't yet happened! More than 50% of people diagnosed with AIDS in 1995 are still alive, meaning the median survival rate is now at least five years -- we don't know how long it will stretch out to. Now that's progress!
If you look back over the last few years...
Successes:
Failures:
Mark Katz, MD is the Regional HIV/AIDS Physician Coordinator for Kaiser Permanente of Southern California.
001001
BA001001
Copyright © 2000 - Being Alive. Permission granted for noncommercial reproduction, provided that our address and phone number are included if more than short quotations are used. Subscription lists are kept confidential. Being Alive, 621 N. San Vicente Blvd., West Hollywood, CA 90069 TEL: 310.289.2551; FAX: 310.289.9866; Email: BeiAlive@aol.com http://www.beingalivela.org/
ÆGiS is made possible through unrestricted grants from Roxane Laboratories, Inc., iMetrikus, Inc., the National Library of Medicine, and donations from users like you. Always watch for outdated information. This article first appeared in 2000. This material is designed to support, not replace, the relationship that exists between you and your doctor.
ÆGiS presents published material, reprinted with permission and neither endorses nor opposes any material. All information contained on this website, including information relating to health conditions, products, and treatments, is for informational purposes only. It is often presented in summary or aggregate form. It is not meant to be a substitute for the advice provided by your own physician or other medical professionals. Always discuss treatment options with a doctor who specializes in treating HIV.
Copyright ©1990, 2000. ÆGiS & the Sisters of Saint Elizabeth of Hungary. All materials appearing on ÆGiS are protected by copyright as a collective work or compilation under U.S. copyright and other laws and are the property of ÆGIS and the Sisters of Saint. Elizabeth of Hungary, or the party credited as the provider of the content.