Resource Logo
Being Alive

When A HAART Attack Leaves You Flatlined: Studying "Non-responders" to Anti-HIV Therapy


Being Alive 1997 Jun 5: 6

The count was two strikes. Having briefly tried and stopped the drugs saquinavir and ritonavir due to severe side effects, Kevin McCarthy started the only other protease inhibitor on the market, Merck's indinavir (Crixivan). "I can remember the exact day," he recalls. "April 23rd. This time, I didn't have any side effects." Unfortunately, McCarthy didn't seem to get any activity from the drug, either. His viral load-a measurement of the amount of hiv genetic material in the blood- remained around 220,000 for at least four months. Finally, his count dropped to around 70,000. "I discussed it with my doctor I had pretty much come to the conclusion that my viral load was too high and I needed something else." Still, Kevin wanted to be sure it was time to switch. "Not wanting to jump ship, I had another [viral load] done, and it came back at 21,000." Apparently, the Crixivan had worked. But most experts would expect a drug like Crixivan to work in four weeks, not four months. Was this a fluke? aids clinicians and patients have shown great excitement in the past year over the arrival of the protease inhibitors. These new drugs form the backbone of potent combination therapy, also called highly-active anti-retroviral therapy (HAART). HAART often results in 100- or even 1000-fold reductions in the viral load. A lower viral load count has been shown to predict slower progression to aids in untreated patients; many doctors and patients believe lowering the viral load with drugs will also slow down the disease process.

It has become clear, however, that not everyone responds to the new combinations as they do in the laboratory. Some small portion of hiv+ people-apparent "non-responders"- do not initially attain even a 2-fold reduction in viral load, despite trying many different combinations of drugs. These patients often seem to be those who have had low T-cell counts.

"Non-responders" (perhaps more correctly called "delayed responders") should not be confused with patients who respond to, but quickly fail, protease inhibitor-based HAART combinations. The reasons for failures may be somewhat less mysterious than for delayed responses. Excluding people who cannot take the drugs due to side effects, poor responses might result from not taking the drugs on schedule, which is particularly important for Crixivan. Some people may not absorb the drugs from the gut, leading to inadequate levels in the blood. Finally, some people may have pre-existing drug resistance, either due to previous drug exposures or to simple bad luck. (Certain mutations associated with protease inhibitor resistance are known to occur naturally without previous drug exposure.) The existence of delayed responders-and the possibility that they may still benefit from anti-hiv drug treatment-has important implications for ongoing aids research. These atypical responses to HAART cannot be easily explained by the current model of hiv infection developed by Dr. David Ho and colleagues at New York's Aaron Diamond Center. According to the Ho team, the rapid viral load drops and T-cell boosts seen by most patients on HAART indicate an intense rate of hiv reproduction and T-cell destruction and replacement. They believe viral load measures cell-free hiv particles in the blood, and that 99% of these particles are made by newly-infected T-cells. By blocking the infection of new cells, HAART results in a 99% drop-off in viral load in as few as seven days.

If the Ho model holds true, then why do some people show such a lag in their responses to HAART? Most doctors and researchers would reply, "Drug resistance." But if these patients carry drug resistant hiv, their viral loads would be expected to increase or hold steady during drug treatment, not eventually go down.

Dr. Steven Deeks, a San Francisco physician with a large aids practice, believes the phenomenon of delayed responders is a real, albeit rare phenomenon, accounting for no more than 5% of patients. In comparison, patients who have an initial response to HAART therapy and fail relatively quickly-within a year of starting therapy-may make up another 15�20% of protease users, he believes.

Deeks believes delayed responders tend to be people with long-term infection, whose virus has had ample time to establish footholds in various body tissues, which become "reservoirs" of virus. "The kinetics in a patient like that would be very different," says Deeks-meaning that their viral load response to drug treatment will not look like the classic examples presented by the Ho team. Such patients "can respond to therapy, but they need a lot of 'coaching'," says Deeks, who believes such patients deserve intensive study.

Dr. David Senechek, on the other hand, is skeptical about how common delayed or non-responders truly are. He says in his own practice of over 600 patients, he has seen no one who was totally new to protease inhibitors totally fail therapy, and he would estimate that such patients comprise no more than 3% of protease inhibitor users. Senechek believes pre-existing drug resistance due to prior treatment with single drugs, however, is a serious problem.

Still, Senechek admits that not all patients respond to treatment the same way, and comments, "Maybe all viral [load] levels are not created equal." Senechek agrees that the establishment and burden of hiv in body tissues might be one factor explaining different responses to treatment. Another factor may be the rate at which the virus reproduces-Senechek believes variations in viral genes such as tat, nef, and rev could result in patients having distinct hiv strains that reproduce more or less quickly.

Thus, Senechek hypothesizes that two people with the same viral load might have two very different levels of infection. One person might have a heavy burden of virus throughout lymph tissues, but the virus reproduces slowly. Another person might have less infection of the tissue, but what little virus exists might be reproducing furiously.

Another take on the "viral reservoir" mystery comes from Dr. Carlos Paya, an aids researcher at the Mayo Clinics in Rochester, Minnesota. "I think the reservoir may not be a particular tissue," says Paya, "but rather a cell, one that is universally found throughout the body." Specifically, Dr. Paya believes hiv may "hide out" in a type of white blood cell called a "monocyte." Monocytes in the blood transform into another type of cell, called macrophages, which creep into tissues. Macrophages are known to be infected by hiv, and unlike T-cells, they do not quickly succumb to the virus. Instead, hiv-infected macrophages become little factories, producing infectious hiv particles for weeks. Monocyte-derived cells are also found in organs believed to harbor hiv, such as the brain, spleen, and lymph nodes.

Importantly, most classes of anti-hiv agents do not work against chronically-infected macrophages; even protease inhibitors require concentrations 100 times as great as those effective for other cells. Because of the way protease inhibitors work, however, the particles produced might not be infectious. If the infected macrophages do eventually die out, as all cells do, and if the particles they produce do not infect new cells, one would expect the viral load eventually to decrease.

New advances in testing technology may allow exploration of the mystery of HAART delayed responders. Understanding their atypical responses to anti-hiv treatment may shed light on hiv reproduction, and suggest better ways to treat the disease. Perhaps "non-responders" should simply hold on with their current regimen, or maybe they would be better off trying an experimental treatment, such as radiation therapy. Only further research can find the answers.

Rick Loftus is a Research Associate at the Gladstone Institute of Virology & Immunology in San Francisco and a member of ACT UP Golden Gate. He can be reached via e-mail at . He also heads the Protease Inhibitor Response Project (PIRP), whose web site is compiling personal reports of protease inhibitor "failures." PIRP's Internet address is


Information in this article was accurate in June 5, 1997. The state of the art may have changed since the publication date. This material is designed to support, not replace, the relationship that exists between you and your doctor. Always discuss treatment options with a doctor who specializes in treating HIV.