Project Inform - September, 1997

For regular readers of PI Perspective and other Project Inform publications, the new Guidelines offer no surprises. The Guidelines mimic, often word for word, the strategies that have appeared in PI publications for the last year and half. They underscore and support all the key points made in earlier PI strategy recommendations, including:

• The need to achieve "undetectable" viral load as the measurable objective of antiviral therapy;
• The use of triple combination therapy as the minimum baseline for all antiviral treatment, regardless of the stage of disease;
• Discouragement of the use of the standard formulation and dosing of saquinavir, recognizing that the available version of the drug lacks the potency of other protease inhibitors;
• The need to change two or more elements of a combination therapy when treatment fails;
• The recognition that science does not at any one moment have all the answers to the most difficult individual questions such as "when is best time to start therapy?" People must learn to make reasonable choices, acknowledging the uncertainties, risks and benefits and not expect to find rigid cookbook solutions.

History of the Guidelines

Long before data about the first protease inhibitors were submitted to the Food and Drug Administration (FDA), community groups called upon the various federal agencies to convene a task force to address the issues this new class of drugs was likely to raise. Studies had made it clear as early as 1994 that the new drugs offered an unprecedented level of hope but also triggered a new set of problems. Although a Task Force was convened that united community representatives, researchers, government regulators, physicians and pharmaceutical companies, the Department of Health and Human Services never acted upon the group’s recommendations. Early in 1996 when the promise of aggressive combination therapy was clearly proven, Project Inform and others called upon Secretary of Health and Human Services Donna Shalala to prepare physicians to use the new drugs wisely. Her office initially assigned the task to the new Office of AIDS Research (OAR). An initial OAR effort using an outside contractor failed, leading to creation of a committee in the fall of 1996.

By then Secretary Shalala had already initiated a second, independent process to accomplish the same task, headed by Dr. Eric Goosby of the National AIDS Office and co-chaired by Dr. Anthony Fauci of the National Institute of Allergy and Infectious Diseases. In the end, the OAR committee wrote a document describing the general principles of therapy, while the Goosby/Fauci panel wrote the specific recommendations. Both reflect the input of a wide range of experts in AIDS research, community activists, clinicians and other health care providers. To the credit of the government people working on the project, they did a fine job incorporating the many different views into a single, coherent document.

Nuts and Bolts of the Guidelines

The complete Guidelines will be distributed to physicians nationwide and will be available from Project Inform, government offices, websites and many AIDS service organizations. Much of the content of the Guidelines can be summarized in a few points:

• The pace of HIV disease progression is tightly linked to a patient’s individual level of HIV replication.
• For now, the evidence makes it clear that HIV disease is best treated by suppressing HIV replication to the lowest level possible, ideally below the limit of detection on the available assays (RT-PCR and bDNA tests).
• HIV RNA levels (viral load) and CD4+ cell counts should be used to determine the timing, course and effectiveness of therapy.
• At any stage of disease, treatment should be based on a powerful 3-drug combination with a high likelihood of achieving undetectable viral load. This now discourages the use of previously recommended 2-drug regimens, such as AZT plus 3TC even early in the course of disease.

Who Should be Treated?

The Guidelines make a fairly strong assertion that most people who are HIV infected should make use of the new antiviral drugs. However, they acknowledge that several factors must be considered when deciding to initiate therapy, including personal preferences and willingness to commit to what might be a difficult treatment regimen. Other factors to consider, in addition to lab data, include side effects, the long-term course of therapy and the need to extend the use of available drugs over many years. Specifically, the Guidelines suggest that physicians offer treatment as the recommended option for:

• All people with symptoms of HIV infection;
• All people with CD4+ counts consistently below 500;
• People with CD4+ counts above 500 who have significantly detectable viral load (above 10,000 copies on the bDNA test or 20,000 copies on RT-PCR).

The only time the Guidelines recommend "observing" rather than "treating" a patient is when people have CD4+ cell counts above 500 and viral load below 10,000 copies on the bDNA test or below 20,000 on the RT-PCR test. By "observe" the Guidelines mean to repeat CD4+ and viral load testing at 3 to 4 month intervals. The Guidelines also acknowledge that some physicians may choose to treat even this group, based on observations about the constant rate of HIV replication regardless of disease stage.

A broader description of the treatment recommendations is found in Table 1.

Table 1: Recommendations Based on Stage of Disease

The Guidelines recommend that physicians also consider the relative risk of disease progression over time as a factor in determining when to initiate therapy, particularly regarding asymptomatic people with CD4+ counts above 500. Table 2 represents data from the Multicenter AIDS Cohort Study (MACS) and indicates the relative levels of risk for three groups based on CD4+ cell counts and HIV viral loads. Consideration of these data may give physicians and their patients a sense of the relative risks associated with delaying treatment decisions. These figures represent a change from previously published data, primarily a correction for the age of the stored samples used. Earlier versions of this data showed the same outcomes over time, but associated them with lower levels of viral load, based on errors in the testing process caused by use of older, stored blood samples. Researchers recently conducted new studies to determine how long-term storage had affected the blood samples. The new study found that storage of the blood samples caused the viral load levels to shrink somewhat as the blood sample aged. Thus, the real viral load level associated with a particular outcome, using fresh blood samples, would be somewhat higher that the numbers reported earlier. In addition to this important observation, the new study also showed that the aging of blood samples affected the two most commonly available PCR tests (bDNA and RT-PCR) quite differently. Thus, two sets of numbers are given – one which is relevant for people using the bDNA test and the other for those using the RT-PCR.

Table 2: Multicenter Aids Cohort Study

Progression Rates by CD4 and Viral Load Category

Group 1: CD4+ = <350

Group 2: CD4+ = 351-500

Group 3: CD4 = >500

* Values given are adjusted from the original MACS data to correct for the presence of heparin and the effects of prolonged storage.

** NA= not available. N= number of people in the group in the MACS study

What to Use in Treatment

In all instances, whenever treatment is initiated, the Guidelines suggest that it should employ a triple drug combination using one of the highly active protease inhibitors (indinavir, nelfinavir, or ritonavir) and two nucleoside analogues as described in Table 3. For the most potent results, all three drugs should be new to the patient (not previously used). Saquinavir in its current, hard-gel formulation is not recommended as a first-line protease inhibitor because of its poor retention in the blood stream. Recent studies suggest that people would need to use 8 to 9 times the current standard dosage of saquinavir to achieve antiviral activity equivalent to the other protease inhibitors (see the article XXXX on page ##). Some researchers believe that triple combinations employing nevirapine, rather than a protease inhibitor may be a reasonable choice in some situations as an alternative to the protease inhibitor. They argue, correctly, that there is insufficient data from clinical trials to prove that a protease inhibitor always makes for a more potent combination. No specific mention is made of the role of other drugs in the same class as nevirapine.

There is insufficient data at this time to recommend any one of the three main protease inhibitors or any of the various combinations of reverse transcriptase inhibitors over any others. . Each drug, however, has highly individual characteristics that will impact people in different ways. Individuals and their physicians should study the options and choose according to specific needs and preferences.

Table 3: Putting Together an Effective Combination Therapy

Choose one drug from Column A and one combination from Column B

Certain combinations are not recommended, based either on evidence from clinical trials, toxicity, or the resistance characteristics of the drugs. The combinations specifically not recommended include AZT + d4T; ddC + ddI; ddC + d4T; ddI + 3TC and ddC + 3TC.

When to Change Therapy

The Guidelines suggest six criteria that may signal the need to change therapy. These include:

1. Failure to achieve at least a 1 log (10 fold) reduction in HIV RNA within 4 weeks of initiating a new treatment regimen. (Though not mentioned in the Guidelines, Project Inform recommends caution in applying this point to people with advanced disease, who sometimes show a slower response to therapy due to concurrent conditions or other factors).
2. Failure to suppress HIV RNA to undetectable levels within 4-6 months after initiating therapy. This is a general but not an absolute requirement, since some people may start with such high viral loads that it is almost impossible to achieve undetectable levels. Physician judgment is required in implementing this point.
3. Return to detectable levels of HIV RNA after they had become undetectable, indicating the development of drug resistance. (This should be determined by repeated tests, not a single test result, since any single test might be misleading). Return to very low viral load levels, such as 5000, may not always warrant changing therapy, but should signal a need for close observation and more frequent viral load testing to determine whether resistance is beginning to develop.
4. Any reproducible (by repeated tests) increase of HIV RNA of 3-fold or greater above the lowest level achieved. Care must be taken, however, to be sure that such an increase is not due to other causes, such as concurrent on-going infections, use of vaccines or changing test methods, such as switching from bDNA to RT-PCR versions of the viral load test.
5. Persistently declining CD4+ count, as measured on at least 2 occasions.
6. Clinical deterioration, unless the person was already severely immuno-compromised prior to initiating the antiviral therapy.

While any of these conditions may indicate at least the beginning of antiviral treatment failure, they are often not the only factor to consider when deciding to change therapy. The Guidelines remind physicians that patients do not have an unlimited list of antiviral drugs from which to choose. Thus, the present and future available options must also be considered in making the decision to change treatment. In some cases, no optimal choice may be available for an individual at the moment but one might be on the near horizon as new drugs move toward FDA approval. In such cases, it may be wise to delay changing treatment until the better regimen becomes available.

What to Change to

Very little data are currently available to guide physicians and patients in making decisions about what to do when a triple combination therapy fails. No clinical trials have yet been completed for this situation, although some are just getting underway. Therefore, the Guidelines offer only "expert opinion" on this issue. "Expert opinion" in this context may almost be translated as "try whatever is left," as there is very little data to support any of these recommendations.

Commentary

Many recommendations from these new Guidelines are sound and well supported by clinical trial data. This is particularly true for the more general points, such as the need to strive to suppress HIV below the limit of detection and the need to initiate therapy with three drugs instead of two. The data about specific drugs, however, are less certain. While a few points are clearly established, such as the inferiority of the current hard-gel version of saquinavir, there is little hard substance to many of the other recommendations made about specific drugs. For example, most of the impressive data about triple combination therapy comes from studies which matched a highly active protease inhibitor with AZT and 3TC, yet the Guidelines seem to imply that any two nucleoside analogue drugs will do just as well. This is very likely not the case. The importance of 3TC is probably underplayed here. At least when a person first begins to use it, it is the most potent of the approved nucleoside drugs and it likely is a critical factor in achieving high rates of success. But, since it is prone to early development of resistance, it must be used for the first time in a triple combination in order to preserve its potency.

The Guidelines are equally fuzzy about the use of nevirapine. People on the committee responsible for the Guidelines were divided on the use of this drug. Some felt it had not been shown to be an adequately effective partner as a replacement for a protease inhibitor in a three-drug combination. Others felt it had clearly distinguished itself. The problem behind such debates is that few of the studies done for any of these drugs are directly comparable to the studies done for any others. It is difficult to determine the relative contributions of each drug except when one is dramatically weaker, such as saquinavir. This lack of comparability in clinical trials also affects the ability to evaluate the relative potency of nelfinavir, and will soon have a similar effect regarding the new, improved version of saquinavir.

In a broad sense, the general thrust of the Guidelines that treatment must always include a protease inhibitor plus two nucleoside analogues is an overstatement. This is largely an artifact of the way recent studies have been done, not a broad comparison to all the other possible strategies. There is no data to suggest that there is any magic to such combinations or that they are inherently more effective than other possible combinations, such as two protease inhibitors or a protease inhibitor plus one each of the two different kinds of reverse transcriptase inhibitors. The Guidelines favor "one protease plus two nucleosides" only because such combinations have been proven more effective than two-drug nucleoside combinations such as AZT plus 3TC. But there is already growing evidence that dual protease therapy, such as the ritonavir/saquinavir combination, can be at least as effective and durable as any triple combination therapy. It is likely that, as more data accumulates, we will find that what matters most is the total potency of the individual drugs used and their patterns of resistance. The more potent the drugs and the less sensitive they are to resistance, the fewer will be needed.

For now, it is best to view these guidelines as strong generalities, but not as the final word on individual therapies. The text sections, not reprinted here, address these and many other issues. Interested parties should read the entire document.

Project Inform will continue to conduct its own assessment of present and future clinical trial results in the interests of building upon and extending the knowledge presented here. Earlier work allowed PI to accurately present the strategies for antiviral treatment more than a year before government was able to achieve consensus. We had previously urged the use of combination therapy at least five years before it became the standard-of-care. We hope to maintain a similar lead in future developments.

©1997. This document is copyrighted by Project Inform, 205 13th Street, #2001, San Francisco, CA 94103. Treatment Hotline: 800-822-7422 (toll-free) or 415-558-9051 (in the San Francisco Bay Area and internationally) All Project Inform materials may be reprinted and/or distributed without prior permission. However, reprints may not be edited and must include the following text: "From Project Inform, for more information contact the Project Inform National HIV/AIDS Treatment Hotline, 800-822-7422." For permission to edit any Project Inform material for further publication, contact David Evans at the Project Inform office.

Project Inform, established in 1985 as a national, non profit, community-based HIV/AIDS treatment information and advocacy organization, serves HIV-infected individuals, their care-givers, and their healthcare and service providers through its national, toll-free treatment hotline, the PI Perspective and other information publications, educational Town Meetings, on-line services and research and drug access advocacy programs. All information is available free of charge; donations are strongly encouraged. For more information, contact the Project Inform National HIV/AIDS Treatment Hotline. Email: web@projinf.org; Website: http://www.projinf.org.

The original of this article can be found at http://www.projinf.org/pub/22/FedGuidelines.html