Anti Retro Viral Therapy

Treatment Advances in HIV Disease and New Antiretroviral Strategies Michael Johnson, M.D.

Significant changes have taken place over the past few years in our understanding of HIV infection, our ability to monitor infected patients, and our ability to successfully treat HIV-positive patients. I. Viral Dynamics A. The virus and infected CD4 cells turn over very quickly. HIV replicates rapidly at all stages of illness. 1.

About 1 x 1010 virions are produced daily in an infected patient.

2.

The t1/2 of plasma virions is about 6 hours.

3.

CD4 cells are produced and destroyed in HIV-infected persons at 10100 times the normal rate.

4.

The gradual fall in CD4 count during clinical latency appears to be due to an inability of CD4 production to match HIV-mediated CD4 destruction.

5.

There is no virologic latency. HIV replication proceeds relentlessly, and the body eventually fails to keep up.

6.

The amount of HIV in the bloodstream at any given time is the result of a dynamic equilibrium between production and destruction of new HIV particles.

B. Drug resistance mutations develop easily. 1.

Replication errors are common in RNA viruses like HIV, due to the absence of DNA proofreading mechanisms. It is estimated that one mutation occurs in every 3 newly produced HIV genomes. Given the daily production of 1 x 1010 virions, it is likely that every possible mutation, and many combinations of mutations, exist in a given patient's HIV population, even in the absence of selective pressure. HIV, therefore, has an enormous ability to become drug-resistant.

II. Viral Load A. It is now possible to measure the number of HIV particles in the bloodstream at any given time. This number, which is the end result of the battle between HIV and the body's immune system, is called the HIV viral load or viral burden. The value varies from < 20 copies/ml to > 1 x 106 copies/ml. Two techniques are commonly used to measure viral load: 1. Quantitative RT-PCR: amplifies the nucleic acid. Currently sensitive to < 400 copies/mi. 2. Branched chain DNA: amplifies the signal from the nucleic acid. Currently sensitive to < 500 copies/ml. The RT-PCR assay typically gives values about twice those of the bDNA assay. It is, therefore, important to know

When somebody is initially infected with HIV there is rapid viral replications, so that within a few weeks of getting infected there is a very large amount of virus in the blood. At the same time the CD4 cell count drops rapidly and during this period of time the person may well have a mononucleosis like illness characterized by fever, lymphadenopathy, headaches, rash, and diarrhea. Eventually the immune system takes over to some degree and the CD4 cell count comes back up often not to the same levels it was at before, but it certainly rises from where it was during the acute infection. The amount of virus in the bloodstream comes down. The person at this point then enters a period of clinical latency after the acute infection. This period can last on an average somewhere between 8 and 10 years, so there is wide variability. During this period of time the person feels completely well. There is nothing you can find on exam to show HIV infection, the virus is anything but latent during this period. HIV is replicating very rapidly at every stage of HIV infection, including during this period of clinical latency. About 10 billion new variants are produced every day in an HIV infected person. This is somebody who is feeling completely fine and every day; they turn out 10 billion new HIV particles. The half-life of these particles is a very small, 6 hours, so that there is just a constant production and destruction of HIV in the bloodstream. Along with the CD4 cells, which are one of the main targets of HIV infection, being produced and destroyed at 10 to 100 times the normal rate. There is really an ongoing battle going on between HIV production and destruction, even while the patient is feeling really completely well. The net result of this battle is the balance between the production and the destruction of HIV particles is what determines a patient’s viral load. The viral load is simply the amount of HIV in a person’s bloodstream. The higher the viral load the more rapidly the patient’s CD4 cells are destroyed and the more rapidly the person progresses to AIDS. The other thing that is very important to know about HIV replication is that it is very error prone. HIV, being an RNA virus, does not replicate with the same fidelity as DNA does. There are no proofreading mechanisms involved. It has been calculated that 1 and 3 new HIV particles have at least 1 point mutation in it. Because you are getting 10 billion new particles a day, basically every single base pairing mutation, and many combinations of mutations in one virus, are created on a daily basis. What this means is that the virus has an enormous ability to become resistant to antiretroviral agents. If you are treating a patient with a single drug, for example, and there is 1 point mutation that is necessary to create resistance to that drug. You will, within a very short period, often weeks, have a viral population, that is completely resistant to that one antiretroviral drug because of this ability of HIV to rapidly mutate. Viral load testing. Viral load is simply the number of HIV particles in every ml of blood. It varies enormously from patient to patient. Newer assays can now measure down to about less than 20 in some experimental assays. From very low levels to over a million or sometimes more than a million, 2 million and 3 million particles per ml. Viral load is typically measured in clinical labs using one of two techniques. The first is the quantitative reverse transcriptase PCR assay, which is simply a PCR which amplifies the nucleic acid and then compares it to known standards and can give you a quantitative measurement of the amount in your patient’s sample. The second is a branch chain DNA assay, and this is simply an assay that amplifies the signal from the nucleic acid in the patient’s bloodstream as opposed to amplifying the nucleic acid itself. The lab you use may use one or both of these techniques. Both tests are reasonable to use in your day to day practice, but they don’t give exactly the same answer. It is important to know that the RT-PCR assay gives results that are about twice as high as those from the BDNA assay. It is important not to switch back and forth between assays. If you are forced to switch between assays at least be aware of which assay you are using and the fact the PCR values will typically tend to be higher and make a person look slightly sicker then the BDNA assay. It is also important to know that the assay is extremely imprecise, so that any change under half a log or a 3-fold change is really not thought to be clinically significant. What we are looking for when we look for changes in somebody’s viral load is really changes of at least 3 fold. If somebody’s viral load goes form 10,000 to 20,000 that

which assay you are using and to avoid directly comparing results between the 2 types of assays. B. Changes in viral load of < 0.5 log (< threefold change) are not thought to be clinically significant. C. Viral load is an excellent predictor of a patient's long-term prognosis. 1. Viral load alone is an excellent predictor of both survival and AIDS-free survival:

Viral Load

Median AIDS-free period (yrs)

5 year survival (%)

< 4500

> 10

95

4500-12,000

7.7

90

13,000-36,000

5.3

75

> 36,000

3.5

51

Viral load remained highly predictive of outcome, even when adjusted for CD4 count. 2. Combining information from both the patient's viral load and his CD4 count provides even better prognostic information. The course of HIV infection can be likened to a train driving towards a cliff, which represents either AIDS or death. The viral load measures the speed of the train, while the CD4 count measures the distance of the train from the cliff. D. Treatment-induced drops in viral load correlate with improved prognosis. In patients treated with one or more nucleoside reverse transcriptase inhibitors, a decrease of 1.0 log in the viral load was associated with a 65% reduction in the risk of AIDS or death over one year. The goal of antiretroviral therapy should be to keep the viral load as low as possible for as long as possible. E. Recommendations for viral load testing 1. Every HIV-infected patient should have his viral load measured as soon as possible after diagnosis. Many experts recommend obtaining 2 values, 2 weeks apart, and taking the average. This initial testing is vital in terms of estimating prognosis, and planning antiretroviral therapy. 2. In stable patients, on or off therapy, repeat the viral load every 3-4 months. 3. Check viral load about 4 weeks after initiating or changing therapy to assess response. 4. Do not check viral load during an acute illness or within a few weeks after vaccination. Any immune stimulus can transiently elevate the viral

is really not considered a significant change. However, if it goes from 10,000 to 100,000, then this really is a significant change. The first way we use viral load is in helping us determine a patient’s prognosis. The CD4 count is not great prognostic indicator. If somebody came in with a high CD4 count it was really hard to tell them how long it might be before they would progress to actually developing AIDS and possibly dying. The viral load however, turns out to be a much better prognosticator. This involved was looking at samples that had been obtained back in 1984 and 1985 and it was something called the, which was a longitudinal prospective observational study of men who had HIV infection. When they enrolled in the study during this time period blood samples were drawn and simply frozen away. When viral load testing technology came along years later the samples were retrieved, the viral load was measured, and they correlated it with how the patient’s had done since they had given the ample. Since the testing was done about 10 years after the samples were originally obtained they already had the clinical follow-up and how the patient’s and done after the sample had been drawn. What they found was really very remarkable and that was the lower the patient’s viral load the better the person had done. You can see here the patient’s with the lowest of viral loads, those who in this study were under 4500 had a median AIDS free period of over 10 years and 95% of them were alive after 5 years. Compared to those who had the 25% highest viral loads, over 36,000 here had a median AIDS free period of 3 ½ years and half of them had died within 5 years. There was really a good ability of this test to discriminate between people of different prognostic groups. This is a graph showing that same data. This is proportions surviving compared to time and years and 10 years here. These are the patients with the lowest viral loads and these are the patients with the highest viral loads. You can see that the higher you’re viral load got the worse you were likely to do. If you compare that to survival in terms of CD4 count, CD4 count wasn’t nearly as good as a predictor. The patients with the lowest 25% of CD4 counts didn’t do as well as the rest, but the patient’s who had CD4 counts over 322 you couldn’t discriminate their prognosis at all based on CD4 count. The viral load was just dramatically better. This is an attempt to show that viral load and CD4 count are really independent predictors of how a patient will do. If you look at the top graph here this represents the patient’s in the study who had CD4 counts over 500. They divided them into those who had viral loads over about 10,000 and those who had viral loads under about 10,000. You can see that here those who had the lower viral loads had much better survival then those who had the higher viral loads, but the 2 groups had almost exactly the same CD4 count, which was 781 in one group and 787 in the other group. If you just had the CD4 count you couldn’t have predicted any difference in these groups, but the viral load has a very big difference. Viral load and CD4 count together as it turns out actually provide the very best prognostic information that you can provide to your patient today. The analogy that unfortunately I didn’t make up and I wish I had, but which is a very good one to think about is to think about having HIV infection as like being on a train. The train is driving towards a cliff and the cliff is either developing AIDS or it is dying of AIDS depending on how you choose to look at it. You can think of the CD4 count as simply the distance of the train from the cliff. It is very high and you have a long ways to go. However, the viral load really measures the speed of the train, so knowing those 2 factors really gives you a wealth of information about how a patient is likely to do. Looking at the combined effect of CD4 count and viral load in a larger group of patients from the same study that I was mentioning earlier. The point of this is imply to again show that the CD4 count and the viral load together give you the very best prognostic information about your patients. Down here is the viral load and you can see as the viral load goes down the percentage of patients that will get AIDS within 3 years goes down. As the CD4 count goes down the percentage of patients that re likely to develop AIDS goes up. Having both numbers of CD4 under 200 and the viral load over 60,000 is very bad and having a high CD4 count and undetectable virus you have essentially no chance of going on to get AIDS in the next few years. In addition to prognostic information we also use the viral load as a means to assess patient’s response to treatment. It has now been well shown that treatment induced reductions in viral load correlate really very well with an improved clinical outcome of the patient on antiretrovirals. There are a number of studies that have shown this and I’ll show you data from just a couple of them. The first was by published in the New England Journal and again the reference is in your syllabus. They found that in patients who were given one of a number of different regimens of nucleoside antiHIV drugs one log decline in viral loads a 10 fold decline was associated with a 65%

reduction in the risk of AIDS or death over one year. This was a study using not very powerful antiretroviral agents, but it was very clear that if you reduce the viral load the patients did better. This is simply a table from that paper. Looking at the decrease in viral load after 8 weeks of therapy the patient’s who had more than .63 log decline in viral load are here and those with less than .63 log decline are here. You can see that 7% of the patients with a decline in viral load progressed to either AIDS or death during the study versus 17% who did not have a decline in viral load progressed to AIDS or death. It seemed to be a very good surrogate marker for clinical outcome.

load. III. Antiretrovirals A. Three categories of antiretrovirals are currently available: 1. Nucleoside reverse transcriptase inhibitors a. zidovudine (AZT, ZDV, Retrovir®)

The Veterans Cooperative Study published by Obrien looked at the proportion of patients again in a trial of different nucleoside analogs. The proportions were remained AIDS free over the months of the study going out to 60 months. They looked at viral load changes and CD4 count changes. What they found is those patients who had rising CD4 counts and dropping viral loads had the best outcome. Those who had neither change had the worst outcome and having either a rising CD4 or a dropping viral load, but not both had an intermediate outcome. They are both things that you want to look at in terms of determining how patient is responding, but viral load is probably the most key. That is not shown by this study, but we’ve learned that in other areas. This leads us to a rather simple statement that the goal of the antiretro viral therapy is to reduce levels of circulating viruses as much as possible and for as long as possible. This makes a lot of sense, but the problem is that is not always an easy task.

b. didanosine (ddI, Videx®) c. zalcitabine (ddC, HIVID®) d. stavudine (d4T, Zerit®) e. lamivudine (3TC, Epivir®) f. 1592U89 2. Non-nucleoside reverse transcriptase inhibitors a. nevirapine (Viramune®) b. delavirdine (Rescriptor®) c. efavirenz (dmp-266, Sustiva®) 3. Protease inhibitors a. saquinavir (Invirase®) b. ritonavir (Norvir®) c. indinavir (Crixivan®) d. nelfinavir (Viracept®) B. General principles of antiretroviral use 1. Combination therapy is superior to single agent therapy. 2. Protease inhibitor-based regimens are more potent than 2 nucleosides a. Virologic Efficacy: After 24 weeks of therapy: Regimen

Mean CD4 increase

Mean HIV RNA decline

Patients with RNA < 500 copies/ml

ZDV/3TC

46

0.83 log

0%

IDV

101

1.24 log

43 %

IDV/ZDV/3TC

86

1.77 log

90 %

In terms of just nuts and bolts of viral load testing what should you be doing with your HIV infected patients. I would say that everybody with HIV needs to have 1 and probably 2 viral load measurements done when they enter care. If somebody comes into your practice with HIV this is absolutely one of the most important tests that you can draw on that person. Typically what I will do is draw a viral load on the initial visit and then get another one in a few weeks and compare the results. Because as I said earlier they are not extremely precise and it does help to have 2 numbers as a base line before you decide whether or not to treat or to use as a base line to assess the person’s response to treatment. In stable patient’s whether or not you’ve put them on antiretro virals we recommend repeating the viral load every 3 to 4 months. That is really the longest time period you should let people go by without having their viral load checked. The idea of monitoring somebody’s viral load is to be able to catch it if it starts to rise and try to do something about it before the rising viral load results in immunologic damage. If somebody has been put a new antiretro viral therapy or is starting for the first time one generally checks the viral load about 4 weeks after starting a treatment. That is definitely long enough to see a response if the patient is going to respond to the therapy. It may not be long enough to see the maximal response, but if you see no change after 4 weeks this means either the patient isn’t taking the drugs or the drugs are not working for their particular virus and maybe they need to be on a different regimen. It is important not to check the viral load within a few weeks of an acute illness or a vaccination. This is because anything that stimulates the immune system will also stimulate HIV replication and most of the time when people are in the throws of Pneumocystis pneumonia let’s say, the viral load will rise much higher than it was at the base line before they got PCP. It will fall again after the PCP is treated. This actually also holds true for the CD4 which may drop somewhat during an acute illness. Let’s say a patient comes in the hospital with new PCP and they were not known to be HIV infected, it is not a good idea to check the viral load while they are in the hospital with PCP. The thing to do is to treat the PCP and get them back into the office. Once they are better check the viral load and that will give you a much better indicator of where things stand. Antiretroviral therapy. There are now 5 reverse transcriptase inhibitors and 3 protease inhibitors, we now have 100 licensed agents and 2 more that I have listed here with stars which are available by expanded access and a few more that are waiting in the wings and will probably be released in the next few years. We now have 5 FDA approved nucleoside RT inhibitors, which are zidovudine (AZT), didanosine (DDI), zalcitabine (DDC), stavudine (d4T), and lamivudine (3TC). There is a new category of drugs, the non-nucleoside reverse transcriptase inhibitors. There are 2 of those available now, and they are nevirapine and delavirdine, and there is a much more potent one, which is likely to become available in the next year, called efavirenz (DMP-266). Then there are now 4 protease inhibitors that are currently available. The RNA is reversed transcribed into DNA in order to insert the DNA into the host viral genome. Reverse transcriptase inhibitors act right here at this point inhibiting obviously the reverse transcriptase, so that the RNA just can’t be reversed transcribed and you don’t get any integration of the genome. Both the nucleosides and the non-

Some patients on this and similar studies have now been treated for over 2 years with "triple therapy" and the response to treatment seems to persist. Not uncommonly, CD4 count continues to rise slowly over the first year or 2 of treatment. Similar results are seen with ritonavir and nelfinavir. Saquinavir, in its current formulation, is poorly bioavailable, and does not give as impressive results when used as the sole protease inhibitor. A new saquinavir soft gel capsule, with improved bioavailability, will soon be available. b. Clinical Outcome ACTG 320 (Hammer et al, NEJM 1997;337:725-33) randomized 1146 HIV+ pts with CD4 < 200 (mean 87) and prior ZDV therapy to either ZDV/3TC or ZDV/3TC/indinavir. After a median 38 weeks of followup, the rate of progression to AIDS or death was 11% in the 2 nucleoside arm, but only 6% in the 3 drug arm. The rate of death was reduced from 3% to 1%. Both differences were statistically significant. c. Epidemiologic Data In the United States, between 1995 and 1996, the death rate from AIDS declined 23%, while the incidence of AIDS-related opportunistic infections declined 6%. These changes are thought to be due primarily to the use of protease inhibitor-based antiretroviral therapy. 3. The goal of antiretroviral therapy is to keep the viral load as low as possible for as long as possible.

Any significant decline in viral load, even if virus is still detectable, is associated with an improved clinical outcome in patients with AIDS. However, if the viral load can be brought down to undetectable levels (< 400-500 copies/ml in current assays), the chances of resistant virus developing are much lower, and the drugs may remain effective for a much longer period of time. Therefore the best long-term outcomes will probably be in those patients with undetectable viral loads on treatment. 4. Adherence is critical. Patients who adhere closely to a prescribed antiretroviral regimen are more likely to achieve a persistently undetectable viral load than are pts who frequently miss doses. Constant suppression of viral replication, and thus the development of resistance mutations, is critical. In one study 90% of pts who were closely adherent achieved a viral load < 200, while only 66 % of those who regularly missed doses had a viral load < 200. 5. Patients respond best to the first antiretroviral regimen they receive. Once a pt's virus has been exposed to antiretrovirals and has had the opportunity to develop resistance mutations, subsequent antiretroviral regimens are less likely to be effective, due to cross resistance and other poorly-understood factors. 6. Starting or changing antiretrovirals should be done very carefully. a. Highly effective antiretroviral therapy can prolong patients' lives and prevent opportunistic infections. However, taking a sub-optimal drug

nucleosides act against this enzyme. The difference between the 2 categories is just simply their mode of action against the enzyme. The protease inhibitors on the other hand act really at the other end of the HIV life cycle. After the DNA has been made it is transcribed and then translated into a long polyprotein. That polyprotein has to be broken up into smaller functional proteins, which are then inserted into the new HIV particle. The protease is what is responsible for breaking up this long polyprotein into the smaller proteins. Inhibiting this doesn’t allow the release of functional viral particles, so this is where the protease inhibitors act. Recent developments in antiretrovirals combination therapy is really better than single agent therapy in treating HIV. Any sort of monotherapy is very clearly contraindicated in HIV infected patients. The first trial that showed this was ACTG175, which was a very large trial of almost 2500 patient’s with CD4’s between 200 and 500. It compared AZT to DDI + DDI to AZT + DDC. What this trial showed was that AZT monotherapy was clearly the inferior arm of the 4, so that if you looked at an endpoint of patients who had more than a 50% decline in CD4 count, or AIDS, or death during the course of this trial, 32% of the patients who received AZT had a clinical endpoint. This was compared to only 18% of those on AZT and DDI, 20% on AZT and DDC, and 22% on DDI. The second development which came along relatively quickly after this was the finding that regimens with protease inhibitors in general are dramatically more effective than any combination of 2 nucleoside analogs, AZT-DDI, AZT-3TC, and any other combination you can think of. That is in terms of viral load response, CD4 response, and in terms of clinical endpoints. One study that has been widely reported especially in the media is a study looking at different combinations of AZT/3TC/Indinavir. This is study sponsored by Merck, which they call the #035 study. This is a study where they look at 97 AZT experienced patient’s who had a wide range of CD4 cells and a viral load of over 20,000. These patient’s were given 1 of 3 regimens. They were either given AZT/3TC, which was probably arguably one of the best combinations of nucleosides that is available, or Indinavir alone, which is a protease inhibitor, or Indinavir/AZT/3TC. After 24 weeks what they found was that the triple drug regimen was really superior to either of the other regimens. In terms of CD4 cell count you can see that they rose by 86 points in this regimen and 100 points in the Indinavir regimen, but only 46 in the nucleoside arm. In terms of RNA decline you can see that the log declines here and most impressively looking at the patients who had achieved undetectable RNAs on the AZT/3TC arm nobody had an undetectable RNA. Undetectable simply means under the smallest amount that you can measure with your assay. It does not mean that the HIV is gone or that the person has been cured of HIV. Indinavir alone 43% had an undetectable RNA, but on the 3-drug regimen an amazing 90% had undetectable RNA after 24 weeks, which was really dramatic and had never been achieved with any other class of drugs against HIV. This is just looking at the CD4 count, you can see the superiority of either of the regimen’s with Indinavir compared to the 2 nucleosides. This is looking at the proportion of the patient’s who had viral loads under 500 and now out to 48 weeks and you can see that the patient’s on the 3-drug arm rapidly 80% or so of them developed viral RNA’s that were under 500. What was very interesting is how that was sustained over these 48 weeks of therapy. Once they developed these low viral loads the viral loads stayed down. The arms didn’t do nearly as well especially the AZT/3TC arm. Patients on Indinavir had AZT/3TC added and those on AZT/3TC added Indinavir. That happened somewhere during this time period. They just continued to follow the people who were already on 3 drugs. At 100 weeks it is a smaller number of patients of 28 in that arm, but still 80% of them have undetectable viral loads. There really hasn’t been any detectable change since those viral loads first went undetectable back here. These regimens really have the potential to have a very long term sustained affect on patient’s infections and to really suppress the virus for many years, though obviously we don’t know that at this point because the drugs haven’t been around long enough to know much more than 100 weeks. The one thing I want you to notice here which we are going to come back to in a minute is what happened to the patient’s on the other arms of Indinavir monotherapy and the AZT/3TC combination therapy when they were switched to the triple therapy. You’ll notice that they didn’t go up to this 80% undetectable viral RNA. The patients on AZT/3TC had a boost so that 40% of them undetectable, but nothing even really happened to the patients on the Indinavir monotherapy arm when you added the AZT/3TC. Sequentially adding these drugs didn’t have the same affect as starting all 3 drugs initially at the same time.

regimen or not adhering closely to a regimen can lead to the development of resistant virus and long-term difficulties in ever successfully suppressing the patient's viral load. C. International AIDS Society recommendations for antiretroviral use 1. Initiation of therapy a. Therapy is recommended for: (1)

All patients with RNA level > 5000-10,000.

(2)

All patients with CD4 < 500, and especially those with CD4 < 350.

b. Therapy should be considered for all HIV-infected patients with detectable viral loads. c. For patients at low risk of progression, based on viral load and CD4, it is reasonable to withhold therapy, while monitoring the patient every 3 months. 2. Initial Antiretroviral Regimens: a. Three-drug regimens are now recommended for all patients. Most patients should receive 2 nucleoside analogs and a protease inhibitor. An alternative is to use 2 nucleoside analogs and an NNRTI. This latter option is less well-studied, and is probably somewhat less effective. b. Patients who are not candidates for triple-drug regimens may benefit from a dual nucleoside regimen, however viral load suppression is more likely to be sub-optimal. 3. Changing Therapy a. Indications for changing therapy: (1)

Treatment failure, as measured by confirmed rising viral load, falling CD4, or clinical disease progression.

(2)

Unacceptable medication side effects or toxicities.

b. When therapy is changed due to treatment failure, ideally all the drugs in the regimen should be replaced with new drugs. Failing this, at least 2 new drugs should be used. Never add one new agent to a failing regimen, because this will likely lead to the rapid emergence of resistance to that agent. c. If viral load suppression is complete, but therapy must be changed due to medication toxicity, it is acceptable to simply replace the offending agent with a different drug.

Protease inhibitor containing regimens have also now been associated with better clinical outcome. A recently published trial which looked at this was the ACTG 320 trial, which looked at 1146 AZT experienced patient’s who had low CD4s of under 200 with a mean of 86. These patients were given either AZT/3TC or AZT/3TC/Indinavir. There was a median follow-up of 38 weeks and the study was actually terminated early because of such a big difference in the 2 arms. What was found was that the rate of progression to AIDS or death was 11% on the 2 nucleoside arm and 6% in the 3-drug arm, which was highly statistically significant. There is a very dramatic affect on mortality and on the development of AIDS by using these drugs. From 1995 to 1996 there was a 23% decline in the death rate from AIDS in the United States. This is probably mostly due to the advent of these very effective antiretroviral regimens and protease inhibitors in particular. The goal is to keep the viral load as low as possible for as long as possible. We know that any significant decline in viral RNA is associated with a better clinical outcome, but if the viral load can be maintained at undetectable levels for a long period of time, drug resistance is much less likely to develop. The regimen will be more durable. Our goal really for treating people is to get viral loads down to undetectable levels. The reason for that is simply that if there is no replicating virus or very little replicating virus the virus can’t develop resistance mutations and the drugs will continue to work. If there is replicating virus, 10,000 HIV particles, there is a significantly increased chance that those viral particles will be able to develop resistance to the drugs. The patient will break through the regimen and those drugs will fail. The ideal goal with treating patients is to get the viral load under 500 or under the lowest limit of the assay that you are using. That is what is likely to result in long term suppression of HIV. We know for this to happen that adherence to the drug regimen by the patient is absolutely critical. Patients who take their medications faithfully are much more likely to get a persistently undetectable viral load than patients that don’t adhere well to their regimens. One small study looking at patients who are on a dual protease inhibitor regimen showed that patients who were adherent to their doses, had a 90% chance of getting a viral load under 200. There was a 97% chance of getting a viral load under 1000 compared to patients who didn’t take that their doses was only a 66% chance of getting under 200 and a 73% chance of getting under a 1000. It is really critical that the patients take their medications regularly and don’t take them on and off. Because any time that there is a drug free period the virus starts replicating again. So the virus picks up again right away if there is not drug there. It will start replicating and then you re-challenge it with a drug. The ideal scenario for creating HIV resistance to the drugs that you are using is continually re-exposing replicating virus to the drugs at hand. That is the thing you don’t want to have happen. We also now know that patients really respond best to the first antiretroviral regimen that they receive. When a patient has a sort of a virgin virus, with no resistance mutations, is when you are going to have the best chance of getting them down to an undetectable viral load and really keeping it there. You want to do this right the first time around. Once resistance mutations have begun to develop, there is a lot of cross-resistance to future therapies. It is clear that it happens and it is relatively common. The ideal situation if you are seeing naive patients or drug naive patients is to get them on a very effective regimen the first go around. Starting or changing antiretrovirals really needs to be done extremely carefully. Because if you expose a patient’s virus either to a suboptimal regimen, let’s say by giving them AZT/DDI in a situation where maybe they would be better off with a 3-drug regimen, or if the patient doesn’t take the drug regularly, that can lead to resistant virus and long-term problems. The AZT/3TC/Indinavir study where some of the patients got the 3-drug regimen up front shown in green and some of them were switched from 1 or 2 drugs to the 3-drug regimen later on. The patients who had the additional drugs added here after 24 weeks never got the same response that the patients who started on the 3-drug regimen got. The message from this is that probably during this period of time these patients were developing some mutated virus and some resistance. When you added Indinavir to AZT/3TC, the AZT/3TC couldn’t work as effectively in that combination because there was already some resistance as they could down here when all the drugs were started simultaneously. When you start somebody they really should get a very effective regimen right up front. Make sure you talk to the patient about the importance of staying on the drugs that unlike a lot of other medications if you take half the dose you don’t get half the effect. In fact you get a negative effect by generating resistance. Recommendations on who should be treated. The International AIDS society recommends that patients with viral loads greater then 5000 to 10,000, or patients

D. Protease Inhibitors 1. These drugs are all metabolized by the cytochrome P-450 system. Ritonavir is especially problematic in terms of drug interactions. None of the protease inhibitors should be given with rifampin, carbamazepine, phenytoin, phenobarbital, dexamethasone, or astemizole. Rifabutin can be problematic, and recommendations for its use vary between the protease inhibitors. Midazolam (Versed®) and some other benzodiazepines are much more slowly metabolized in patients on protease inhibitors. Many other hepatically-metabolized drugs are potentially problematic. The full list is too long to include here Check for potential drug interactions before starting a protease inhibitor. 2. Prescribing protease inhibitors: Saquinavir:

Dose:

600 mg po tid, taken within 2 hours after a meal Side effects: generally well-tolerated. Can cause diarrhea, abdominal discomfort, and nausea.

Ritonavir:

Dose:

600 mg po bid with meals. Should gradually increase dose from 300 mg bid to full dose over 4-7 days to reduce side effects.

Side effects: common. May include nausea, anorexia, abdominal pain, circumoral paresthesias.

Indinavir:

Dose:

800 mg po rid on an empty stomach or with a very light meal. Patients must drink at least 48 oz fluids/day. Must be separated from ddI by 1 hour.

Side effects: nephrolithiasis, mild hyperbilirubinemia, ?skin rash Nelfinavir:

Dose:

750 mg po tid with a snack or meal.

Side effects: mild to moderate diarrhea, usually controllable with anti-diarrheal meds E.

Prophylaxis against opportunistic infections in patients on highly active therapy antiretroviral Despite the rise in CD4 that is seen in many patients on potent antiretroviral regimens, discontinuing prophylaxis that a pt was previously taking is not recommended. The "new" CD4 cells may not have the same ability to fight off a

with CD4 counts under 500, and especially under 350, should all get antiretroviral therapy, or at least you should recommend antiretroviral therapy to those patients. Some may choose not to take it. You should consider antiretrovirals for all HIV patients who have a detectable viral load no matter what there CD4 count is, but it may be reasonable to simply watch some patients who have high CD4 counts and low viral loads. We know from the earlier data that those patients have very good long-term prognoses and it may be better to leave them with a naive virus that has not been exposed to any of these drugs. They should be observed closely and begin treating when they start to progress, rather than to take somebody with 700 CD4 cells and a viral load of 5000 and put them on treatment right away when you know that person would have a good prognosis even without the treatment right now. If you don’t treat somebody however, you still want to bring them into the office every 3 months to get a viral load and a CD4 count so you’ll know when it is time to start treatment. The initial regimens generally it recommended is a 3-drug regimen, which is usually 2 nucleoside reverse transcriptase inhibitors and a protease inhibitors. There is some discussing about using 2 nucleoside RT inhibitors along with a non-nucleoside RT as initial therapy, especially in patients who aren’t too advanced. There are nucleoside combinations that are better than others. Typically they involve either AZT or D4T. These have all been shown to be clinically effective. Those are then combined with one of the 4 protease inhibitors. Typically the best protease inhibitors to use are probably Indinavir, also know as Crixivan, which is taken 3 times a day on an empty stomach. It has a risk of kidney stones, but is otherwise quite well tolerated. Nelfinavir just taken 3 times a day with food and that is often the way the one decides between these two. Nelfinavir can cause diarrhea in a significant percentage of patients. In fact, most people have at least some loosening of stools, but most people will tolerate that reasonably well. Ritonavir is as effective as Indinavir and Nelfinavir, however, has a high incidence of side effects when this is given in its full doses. I’ve probably had to pull probably at least half the patients that I’ve out on Ritonavir back off the drug. Saquinavir is a good drug in vitro, but is very poorly absorbed so that generally we don’t use this as the soul protease inhibitor in a regimen. There is a new formulation that came out a couple of weeks called Saquinavir soft gel caps. It is better absorbed, but still probably doesn’t work as well as Indinavir, Nelfinavir, or Ritonavir. Another thing you will see people using is a combination of Ritonavir and Saquinavir. The reason for that is that you can use lower doses it turns out in combination of Ritonavir and it is well tolerated in lower doses. The Ritonavir boosts the patients Saquinavir blood levels about 30 fold, so that using the 2 drugs together you actually will get a very effective antiviral effect from both the Saquinavir and Ritonavir. In terms of when to change therapy, which should be done if the treatment is failing. If the viral load is rising, the CD4 count is falling, or the disease is progressing and if the viral load or the CD4 count looks like it is worsening on treatment, always check the numbers again before you change the regimen. It is not too uncommon to see one isolated number that doesn’t fit with the rest. Somebody who is always undetectable in their viral load and suddenly goes to 2000, that is not the time to switch their therapy, it is time to get them back in and repeat the viral load. If it is still 2000 maybe they are starting to break through treatment, but if it is back under 500 you probably don’t need to anything. The other reason to change is if the medication side effects are unacceptable. When you are changing due to treatment failure, you want to change as many drugs as possible simultaneously. It is ill advised to add one new agent to a failing regimen. If somebody is on AZT/3TC, you don’t add Indinavir. You change the AZT/3TC to something else like D4T/DDI, and then you add in the protease inhibitor. The idea is not to continue to use drugs to which the virus has proven itself resistant. If your viral load suppression is complete the regimen is working, but if the patient doesn’t tolerate one of the drugs, then it is fine to go ahead and simply replace that drug with a different drug. If the patient doesn’t tolerate protease inhibitor #1, but it is working, you can just switch that to protease inhibitor #2 and leave the rest of the regimen alone. With protease inhibitors, think about drug interactions. These drugs are metabolized by the chrome P450 system by which many other drugs are also metabolized, and they have varying effects in this system, and they will inhibit the metabolism of some drugs and they will speed up the metabolism of other drugs. There are a lot of drug interactions. In patients on protease inhibitors, look through their list of medication and check to make sure there is nothing that is hepatically metabolized that might be a problem.

variety of infections as a similar number of CD4 cells in someone whose count has not yet dropped. Therefore, until further research data are available, patients should receive prophylaxis appropriate for the lowest CD4 count that they reached before beginning antiretrovirals. References Viral Dynamics: 1.

Havlir DV, Richman DD. Viral dynamics of HIV: Implications for drug development and therapeutic strategies. Ann Intern Med 1996;124:984-994.

2.

Ho DD, Neumann AU, Perrelson AS et al. Rapid turnover of plasma virions and CD4 lymphocytes in HIV-1 infection. Nature 1995;373:123-126.

3.

Perelson AS, Neumann AU, Markowitz M, et al. HIV-1 dynamics in vivo: virion clearance rate, infected cell lifetime, and viral generation time. Science 1996;271:1582-6.

Viral Load: 1.

Saag MS, Holodniy M, Kuritzkes DR, et al. HIV viral load markers in clinical practice. Nature Medicine 1996;2:625-629.

2.

Mellors JW, Rinaldo CR Jr, Gupta P, et al. Prognosis in HIV-1 infection predicted by the quantity of virus in plasma. Science 1996;272:1167-1170.

3.

Mellors JW, Munoz A, Giorgi JV, et al. Plasma viral load and CD4+ lymphocytes as prognostic markers of HIV- 1 infection. Ann Intern Med i997; 126:946-954.

4.

Katzenstein DA, Hammer SM, Hughes MD, et al. The relation of virologic and immunologic markers to clinical outcomes after nucleoside therapy in HIV-infected adults with 200 to 500 CD4 cells per cubic millimeter. N Engl J Med 1996;335:1091-8.

5.

O'Brien WA, Hartigan PM, Martin D, et al. Changes in plasma HIV-1 RNA and CD4+ lymphocyte counts and the risk of progression to AIDS. N Engl J Med 1996;334:426-31.

Antiretroviral Therapy: 1.

Carpenter CC, Fischl MA, Hammer SM, et al. Antiretroviral therapy for HIV infection in 1997. Updated Recommendations of the International AIDS Society-USA panel. JAMA 1997;277:1962-1969.

2.

Department of Health and Human Services. Draft guidelines for the use of antiretroviral agents in HIV-infected adults and adolescents. 1997. (Final version should be published in the near future.)

3.

Hammer SM, Katzenstein DA, Hughes, MD, et al. A trial comparing nucleoside monotherapy with combination therapy in HIV-infected adults with CD4 cell counts from 200 to 500 per cubic millirooter. N Engl J Med 1996;335:1081-90.

4.

Gulick RM, Mellors JW, Havlir D, et al. Treatment with indinavir, zidovudine, and lamivudine in adults with human immunodeficiency virus infection and prior antiretroviral therapy. N Engl J Med 1997;337:734-9.

5.

Hammer SM, Squires KE, Hughes MD, et al. A controlled trial of two nucleoside analogues plus indinavir in persons with human immunodeficiency virus infection and CD4 cell counts of 200 per cubic millimeter or less. N Engl J Med 1997;337:725-33.

6.

Deeks SG, Smith M, Holodniy, Kahn JO. HIV-1 protease inhibitors. A review for clinicians. JAMA 1997;277:145-153.

7.

Threlkeld SC, Hirsch MS. Antiretroviral Therapy. Med Clin N Am 1996;80:12631282.