Guidelines for the Use of Antiretroviral Agents in HIV-Infected Adults and Adolescents, January 28, 2000.
FAQ answers by John G. Bartlett, M.D.

Q: Why aren't there more PI-PI combinations advocated for initial therapy since this is what many physicians are doing?

A: The DHHS guidelines have certain requirements to achieve the status of a recommended regimen in the "preferred category." Although not stated, the Panel would like to see a good clinical trial with a sufficient sample size, analysis by as treated and by intent-to-treat, comparison with another regimen that is in the preferred category, virologic response using the threshold of 20-50 c/ml, and an observation period of at least 24 weeks and preferably longer. The only combination of protease inhibitors to satisfy these guidelines is saquinavir/ritonavir. We have mentioned the potential attractiveness of ritonavir in combination with indinavir or amprenavir based on information regarding pharmacology, but there are very limited data from clinical trials that oblige the requirements noted above. Please note that these are "evidence based guidelines" or should be whenever possible since they establish standards. The liabilities associated with premature placement in the rank order of preference include the possibility that it is simply wrong and also the possibility that the proper research will never be done because it is considered unnecessary.

Q: Many authorities think that therapy based on CD4 cell count can be delayed until the level is <350/mm³. The DHHS guidelines seem to take a strong stand at the 500/mm³ threshold and this may obviously lead to early therapy with loss of options at the time when the patient actually is most in need.

A: The DHHS guidelines provide substantial flexibility with respect to starting therapy based on any CD4 cell count or viral load. The intent of the "threshold" is that consideration or discussion of antiretroviral therapy should begin with the CD4 cell count is 500/mm³ which may be an appropriate starting point for some patients but not others. It is also noted in the guidelines that "some authorities prefer to start therapy when the CD4 cell count is below 350." The most important issue concerning the initiation of therapy is patient readiness. The decision is obviously a joint one involving both physician and patient input. In the presence of a low viral load, there does seem to be agreement that a CD4 cell count above 500/mm³ is too high for treatment, and a CD4 cell count below 350/mm³ is sufficiently low to justify treatment in virtually everyone. That middle zone is a "gray zone"; gray zones are very common in medicine, and we shouldn't feel that this one is any different.

Q: When will resistance testing be included in the guidelines?

A: There is already reference to resistance testing in the guidelines, but they will be formalized with a separate section dealing with the issue in a guideline version that we anticipate in late January or early February. This section has been written and is in the process of panel review at the present time.

Q: No detectable virus is defined as a viral load of <500/ml, but most clinical trials and most physicians are now using the ultrasensitive viral load test and prefer a threshold of 20-50 c/ml. This makes the DHHS guidelines seem antiquated.

A: The original DHHS Guidelines were published in April 1998 which predates the availability of the ultrasensitive test. There has been a subsequent modification stating that no detectable virus using an assay with a threshold of 20-50 copies is preferred. The reason we have been reluctant to make this a standard guide for decisions regarding changing therapy is that most studies in the "real world" suggest that the lower threshold is hard to achieve. Many clinical trials show that 50-80% of participants have <20 c/ml at 24 or 48 weeks, but results in HIV clinics seem to indicate that, in this setting, only about 30-50% achieve this goal. In other words, most of the patients will be virologic failures with a threshold of 20-50 copies. The concern is that patients may lose all of their antiviral options relatively early in the course with attempts to achieve the lower threshold. There are also studies demonstrating that clinical outcomes are often as good with partial suppression as with complete suppression despite theoretical reasons to favor the latter. Thus, the Panel seems to endorse the lower threshold in theory but feels that it is unrealistic for many patients and rigid adherence to the guidelines might actually be deleterious to some patients because they may exhaust drug options. The bottom line is that the goal of therapy is to drive the viral load as low as possible for as long as possible, and this has been the stated goal of treatment since the guidelines were initially written.

Q: Do most AIDS physicians actually need guidelines?

A: Probably not. The guidelines are written for a much broader audience than the experts who provide care. They serve some providers who, for a variety of reasons, have only a small number of AIDS patients, they tell third party payers the standards of care that require monetary support, and they establish standards by which care is measured in quality assurance programs. These latter issues are particularly important in the era of managed care and reduced resources for healthcare.

Q: The guidelines say that HIV care should be supervised by an expert but does not define an expert. Who's an expert?

A: There is no official mechanism to define expertise in the area of HIV care. The closest analogy with the current system of board certification is with the Infectious Disease sub-specialty board, which includes 15-25% of its content dealing with HIV. No other primary care or sub-specialty discipline has more than 5%. Providers of HIV care are rather diverse in terms of medical background including many who are in primary care or specialists in fields other than infectious diseases. For this group, there is no current mechanism to verify expertise. There are ongoing discussions about a certification method for this group. In the meantime, according to several sources, the best method defined so far is based on "panel size," the number is HIV-infected patients under care by the provider. Using this criterion, HRSA has used a panel size of 25 active HIV-infected patients to define a "HIV expert," but it is also noted that a panel size of 50 is preferred. Some organizations have used the panel size of 50 but combine it with a CME requirement and/or chart audits as well.

Q: I think treatment should be individualized, but the guidelines seem to state that everyone should have the same treatment.

A: This is a misunderstanding of the guidelines. The time to initiate therapy and the specific regimen used is quite variable depending on the patient's willingness to accept therapy, probability of compliance, and the severity of disease at the time the decision is made in terms of viral load and CD4 count. It is clear that "alternative regimens" are "preferred regimens" in selected patients because of these variables. The "preferred" term is based on probability of achieving no detectable virus regardless of individual circumstances. The need to individualize therapeutic plans is a critical component of all diseases in medicine including this one.