The Body: The Complete HIV/AIDS Resource
Follow Us Follow Us on Facebook Follow Us on Twitter Download Our App 
Professionals >> Visit The Body PROThe Body en Espanol
  • Email Email
  • Printable Single-Page Print-Friendly
  • Glossary Glossary

Stop, Start, Switch

Community Forum Summary

June 2000

A note from The field of medicine is constantly evolving. As a result, parts of this article may be outdated. Please keep this in mind, and be sure to visit other parts of our site for more recent information!

Speakers: Mark Harrington, Treatment Action Group, Dr. Bruce Polsky, Memorial Sloan-Kettering Cancer Center

From "Hit early, hit hard," to "Hit hard, but only when necessary." Most would agree that the fundamental message of HIV treatment has changed significantly since highly active antiretroviral therapy (HAART) was introduced five years ago. Practitioners and people with HIV are increasingly delaying the initiation of HIV therapy, waiting to start until immune function is decreased, as indicated by a CD4 cell count around 200-350. Why this change of "HAART"? First of all, there has been no data to support the theory that HIV can be eradicated from the body. Eradication of HIV was one of the main goals of "hit early, hit hard." Also, we know more about the toxicity associated with anti-HIV treatments. From body fat redistribution to peripheral neuropathy to pancreatitis -- long-term therapy is no walk in the park. Adherence issues are also a consideration, as delaying therapy means delaying complicated regimens. If anti-HIV medications are not taken as prescribed, resistance may rear its ugly head, rendering therapy ineffective.


Mark Harrington, the first speaker of the evening, reviewed the overall goals of treating HIV infection. Stated simply, they are to:

  • maximize health
  • prevent AIDS
  • preserve life
  • preserve quality of life

With these goals in mind, Mr. Harrington discussed different strategies for starting therapy. He noted that there have been no large-scale, randomized clinical trials addressing the issue of when to start. Two studies from Merck can be compared in order to shed some light on the subject, however.

In study Merck 060, participants with more than 500 T-cells and a viral load of 1000 or higher were started on an AZT/3TC/Crixivan combination regimen. The participants, who had never taken anti-HIV medications, had an average increase of 184 T-cells, with 83% maintaining an undetectable viral load, after a year on medications. In another study, Merck 035, participants with between 50 and 400 T-cells and a viral load greater than 20,000 copies initiated therapy with the same regimen. After a year, the average T-cell increase was 187, with 80% of participants maintaining an undetectable viral load. The results of these studies imply that there is no benefit to starting anti-HIV medications early in HIV infection, as the individuals in Merck 035 who started therapy in intermediate infection had comparably favorable results to those who started in early infection.

Mr. Harrington mentioned that the immune system has the capability to mount a response against HIV if immune system cells are exposed to the virus. When anti-HIV medications reduce the level of HIV in the blood to low levels, however, the cells do not have the exposure to the virus that they need to generate a response. Individuals who have had an undetectable viral load for several years on therapy may not have the benefit of their own immune system working against HIV.


With that point in mind, Mr. Harrington discussed the possible benefits of stopping HIV medications temporarily, and allowing the virus to reproduce in the absence of medications. These interruptions take several forms: non-adherence (failure to take medications as they are prescribed), drug holidays (stopping medications randomly for a few days), and structured treatment interruptions (also known as STIs -- stopping medications for a longer period of time under close medical observation). Non-adherence and drug holidays offer no benefit: they foster resistance and lead to treatment failure. But there may be benefits to structured treatment interruptions. It has been theorized that STIs can:

  • Preserve/restore the immune system's response to HIV by exposing immune cells to an individual's own virus
  • Improve/reverse resistance
  • Improve/reverse side effects

Mr. Harrington warned: Don't try this at home. STIs may be safe if they are closely monitored and if an individual has a high T-cell count and a low viral load when an STI is initiated. If these conditions do not exist, the drop in T-cells that occurs when medications are stopped could be more harmful than helpful. The long-term impact of STIs has not been established, and more clinical research is necessary.


People with HIV and their care providers usually consider switching a medication regimen or a component of the regimen if there are dangerous or intolerable side effects or if virologic failure occurs. Definitions of virologic failure vary among individuals. For someone who has never had an undetectable viral load, the threshold for failure will likely be higher than for someone who has been undetectable for a prolonged period. Mr. Harrington cautioned against changing a regimen based on a "blip" in viral load, meaning a one-time measurement that is higher than usual. A trend of increasing viral load measurements may indicate that it is time to consider changing medications, however.

When is the right time to make the switch? Mr. Harrington compared switching early, when an increase in viral load is observed, with switching later, when a T-cell count decrease is observed. Switching early decreases available treatment options, while switching later may result in higher levels of resistance or even cross-resistance. Cross-resistance is resistance to all medications in a drug class, for example, the non-nucleoside reverse transcriptase inhibitors (NNRTIs). Individuals who develop resistance to one member of this class, such as Sustiva, will likely be resistant to Viramune and Rescriptor, the other drugs in this class. The decision of when to switch is highly individualized, and should be based on both immune system status and available treatment options.

A Doctor's Perspective on Stop, Start, Switch

Dr. Bruce Polsky, the second speaker of the evening, addressed issues related to optimal HIV treatment at all stages of infection. He stressed the importance of close adherence to a medication regimen in the success of that regimen. Clinical research has shown that individuals who are older, male, have had AIDS, and who have a doctor who is experienced in HIV are most likely to adhere to their regimens. Strategies to improve adherence in other populations, such as women and current/former intravenous drug users are necessary to improve outcomes.

Dr. Polsky discussed an interesting study performed by Dr. Margaret Fischl in Miami to emphasize the importance of adherence in good clinical outcome. Dr. Fischl compared the viral loads of a group of prisoners, whose medication intake was closely regulated, with the viral loads of a group participants in AIDS Clinical Trials Group (ACTG) studies, who received support from clinic staff but whose medication intake was not so closely regulated. She found that 100% of the prisoners maintained a viral load less than 400 copies (with 85% at less than 50 copies) after a year of therapy. In comparison, 68% of the study participants maintained a viral load less than 400 copies (with 50% at less than 50 copies). In even starker contrast, Dr. Polsky shared viral load data gathered from several typical large urban HIV clinics. After 3-7 months of therapy, 48% of patients had undetectable viral loads, with that percentage dropping to 38% after 7-14 months of therapy. Clearly, there are benefits to taking anti-HIV medications on a tight schedule.

Salvage Success

"Salvage" therapy is a term used to refer to a treatment regimen attempted after an initial treatment regimen is no longer effective. In order to have several viable treatment regimens, Dr. Polsky recommends initiating therapy with a regimen containing nelfinavir (Viracept). He also noted that triple-nucleoside therapy (AZT/3TC/Ziagen) has been shown to be as effective as protease inhibitor-containing regimens, while preserving protease inhibitors for future use.

Dr. Polsky described several strategies for salvage therapy:

  • Switch medications early. When an initial regimen is no longer suppressing viral load, switch early to avoid high levels of resistance to failing medications. This may contribute to future treatment success with the failing medications.
  • Use a class of medications that the patient has never taken. For example, a patient who has only taken protease inhibitors (PIs) may have success with a regimen containing an NNRTI.
  • Use a combination of medicines from all three classes of anti-HIV drugs.
  • Use an STI to decrease resistance and then rechallenge with medications that a patient has taken before. This tactic may be effective in patients who have taken all available medications.
  • Use Mega-HAART. Mega-HAART is a combination of 4 or more anti-HIV medications.
  • Optimize pharmacokinetics (levels of drug in the bloodstream). This requires special tests (available in clinical trials) to determine if dose adjustments are necessary for the medications to work as well as they can.

Both speakers emphasized the individualized nature of all treatment decisions. The information presented above should be used to initiate discussion with HIV care providers, with whom all decisions to start, stop, or switch should be made.

A note from The field of medicine is constantly evolving. As a result, parts of this article may be outdated. Please keep this in mind, and be sure to visit other parts of our site for more recent information!

  • Email Email
  • Printable Single-Page Print-Friendly
  • Glossary Glossary

This article was provided by AIDS Community Research Initiative of America. Visit ACRIA's website to find out more about their activities, publications and services.
See Also
More on Switching Treatment