People living with HIV (PLWH) don't have the luxury of making these choices. Antiretroviral medications extend lives, but only if they are taken consistently as prescribed. Since it's best to take medications before HIV erodes the immune system too much, PLWH may have to suffer through treatment side effects without enjoying the benefit of overcoming any noticeable bodily complaints in the early stages of their infection. Which side effects are merely bothersome, and which are serious? What is science doing to try to limit side effects? Can your choice of medications make any difference? This article will lay out some of the key challenges and changes relating to HIV treatment side effects. In some instances, the evidence is conflicting because there's often no way to know whether the development of progressive ailments is actually triggered by medications or is merely a natural occurrence of living longer.
Patients living with other chronic diseases such as cancer have said that they would be willing to suffer through more side effects if the medications were known to extend life.1 To see if the same preferences would hold with HIV disease, Loren Millier's team at the UCLA Medical Center asked a small group of HIV-positive patients to compare their preferences for treatment regimens according to four different criteria: side effects, pill burden, regimen inconvenience, and regimen potency.2
Obviously, patients would not be willing to suffer more side effects, pill burden or inconvenience for a treatment plan that was actually less potent. So this last category was included to determine how much potency could offset the negative impact of the first three criteria. It should be noted that in the first category, Miller's team measured only "side effects that were bothersome but not severe enough to necessitate drug discontinuation." Generally, this meant things like headache, fatigue, nausea, and diarrhea, but not potentially life-threatening side effects like organ shock or heart disease.
The comparative choices were presented to patients in pictures, with larger or smaller drawings depicting better or worse side effects, pill burden, regimen inconveniences or treatment potency. The results? "Most though not all participants reported that they would want a regimen that was most effective at fighting HIV and prolonging life, regardless of side-effect severity, complexity, inconvenience or pill burden." In other words, these PLWH are just like patients facing other chronic diseases: their top priority is getting strong treatments.
The study also found that the patients were less bothered by side effects than many physicians tend to believe. In interviews with patients, many reported that side effects were most severe when they had begun a new regimen. After a period of time, their bodies "had grown accustomed" to the meds, or else the patients developed strategies to minimize the side effects, such as timing the dose with or without food, drinking more water, or not taking doses as soon as they wake up.
However, side effects were still more troublesome to patients than were inconvenient dosing or higher pill burden. Patients "preferred regimens with fewer side effects to those that were more convenient" in dosing schedule. Preference for fewer pills was the lowest among the four domains. The importance of pill burden may have declined because today's regimens tend to be so effective that few PLWH have to suffer through a daily series of additional prophylactic pills to ward off specific opportunistic infections.
Suspicious eyes have been cast towards protease inhibitors because it was soon after their development that doctors started reporting significant new cases of "protease paunch" and "buffalo hump," bizarre accumulations of fat in the abdomen and neck regions. Around the same time, inexplicable fat loss began showing up in PLWH, a condition just as stigmatizing. Known as "lipoatrophy," the condition leads to shrunken arms and legs, and sunken cheeks, even when overall bodyweight is unchanged.
As time went on, studies called into question the hypothesis that protease inhibitors are the culprits behind what true fat loss or fat accumulation does occur. Patients who had never taken PIs were found to develop the same bodily changes over time. That led researchers to speculate that all HIV treatments (not just PIs) may eventually cause bodily changes, due to the toxicities, or to the ways that the medicines interact with bodily hormones, disrupt cytokines, or trigger insulin resistance. This year's d4T vs. tenofovir study (Zerit vs. Viread) implicated d4T in spurring fat accumulation, for example. Patients taking d4T lost over six pounds overall, and over 18 pounds of limb fat specifically, while patients in the tenofovir arm had negligible overall weight change and lost less than 12 pounds of limb fat.3 Another study compared patients taking only nucleoside analog drugs, one arm receiving AZT (Retrovir) and 3TC (Epivir) while the other took d4T and ddI (Videx). Again, the d4T patients began losing limb fat sooner (and lost more limb fat overall).4
However, a new study in the Journal of AIDS questions whether medications have been proven as the sole cause of lipodystrophy. The authors note that all previous lipodystrophy studies "have evaluated the time period after the introduction of highly active antiretroviral therapy (HAART) regimens."5 As a result, the development of lipodystrophy might at least partially be a coincidental correlation with the advent of HAART, and not always be caused by it.
To try to answer this question more firmly, the authors drew on patients in the seven clinics nationwide that make up the HIV Outpatient Study (HOPS) cohort. Patients were examined in the fall of 1998, and then again in the summer of 2000. Of patients present for both surveys, 546 had no lipoatrophy during the first visit, but 13% had developed moderate or severe lipatrophy by the second survey. This study also included data on patients as far back as 1994, allowing researchers to investigate other possible causes of lipoatrophy that they observed. Many factors seemed to correlate with risk of developing lipoatrophy. Not surprisingly, patients with low CD4 counts (less than 100) were at higher risk of developing lipoatrophy. The researchers came to the same conclusion for patients who had ever had what they defined as high viral load (more than 1,000). (The value of this correlation is uncertain since nearly all PLWH will have had viral loads more than 1,000 at some point in their infection, and those who appeared not to may simply not have had a measurement at the time when they actually had higher levels.)
However, when researchers accounted for differing CD4 counts in their patient comparisons, only three factors were shown to be independently significant. People at greatest risk of developing lipoatrophy were those who were white, had a low body mass index (BMI), or had bounced back from CD4 counts that were previously less than 50. The researchers could not explain why race would play a role in risk for lipoatrophy, and they noted that white race probably captures many other social and treatment dynamics that cannot be easily itemized.
The reference to low BMI, meanwhile, may seem surprising. Body mass index is the measure of muscle relative to overall body weight. People with less fat typically have higher body mass indices. So it may seem counterintuitive that the researchers noted low BMI (i.e., higher fat content) as a risk factor for abnormal and localized fat loss. Though the mechanism that causes this phenomenon is unclear, it may be the case that a moderate exercise routine could preserve not only muscle but also appropriate body fat. As for the correlation to rebounding CD4 counts, the authors postulated that either the severity of the illness itself or the mechanism of immune reconstitution could be responsible. The renewal of T-cells might inadvertently "produce unopposed proinflammatory cytokines (e.g., tumor necrosis factor alpha) that may" cause the loss of the body fat.
The study found no relationship between use of antiretroviral medications and the abnormal loss of body fat. Neither total time on medications, time of drug initiation, drug continuation, or drug discontinuation presented any clear relationship to lipoatrophy. The authors did not completely rule out the possibility that medications may "exacerbate an underlying predisposition to lipoatrophy" but they concluded that, "HIV infection or factors associated with immune reconstitution may play a greater role ... than the use of any specific medication." In fact, their findings suggest that beginning antiretroviral medications in a time when the disease is "less advanced" may stave off lipoatrophy by preventing either the excessive loss of CD4s or the possible side effects of immune reconstitution.
Meanwhile, a small Italian study this year suggests that the frequency of lipodystrophy problems may have been overstated. PLWH may judge their bodies harshly, and perceive lipodystrophy when none can be found through more objective measures.6 The authors compared patients' and doctors' often-conflicting assessments of body fat redistribution, either fat accumulation, fat loss, or both. The authors noted disagreement between the patients' and physicians' interpretations, and also pointed out that common tests used to define body fat redistribution (waist-hip ratio) are open to subjective interpretations. This does not mean that lipodystrophy is not a real and vexing problem, only that fears may be magnifying the reality of the problem.
To control this debiltating side effect, physicians may prescribe numbing agents, such as 5% lidocaine gel, or anticonvulsant agents such as lamotrogine or gabapentin. Many PLWH also feel that acupuncture helps diminish neuropathy, though the scientific proof of this effect is dubious.10 Recombinant human growth hormone showed more evidence of reducing pain associated with neuropathy; yet this treatment did not impact the underlying nerve damage, and so is not widely used.11 With so many causes for neuropathy and so many manifestations, doctors struggle to find the tools to allow PLWH full mobility free of pain.
However, the risks have been challenged by other studies, including a much larger study of 36,766 veterans living with HIV, which was presented at last year's CROI. In fact, that study found that admissions to hospitals for cardiovascular events or deaths from heart disease dropped from 1993 to 2001.13 The Kaiser Permanente Medical Group conducted an observational study to see whether antiretroviral therapy in general, or protease inhibitors in particular, increased the rates of coronary heart disease. This study focused on over 4,000 HIV-positive men old enough to have some chance of developing heart disease (35 to 64), but who had not yet personally had any coronary heart disease incidents. The patients on antiretroviral medications in general or protease inhibitors in particular were found not to be more likely to experience coronary heart disease during the four year follow up, though the authors acknowledged that differences might arise if a longer follow up was conducted. By comparing the 4,000 HIV-positive patients to 40,000 HIV negative men of the same age and under care at the same clinic, the study confirmed that PLWH do face higher risks of cardiovascular disease, just not that it's related to their medications.14 Again, what health challenges facing PLWH may simply be the byproduct of the disease itself, and may have become more noticeable now that patients are living much longer -- long enough for progressive heart disease to manifest itself. In fact, new data from the DAD study (Data Collection on Adverse Events of Anti-HIV Drugs) found that antiretroviral medications help prevent myocardial infarction. The study followed 17,600 patients over three years, and concluded that those who continued their medications faced a six-to-11 percent chance of death, while those who discontinued their meds faced a 22-29 percent chance.
On the other hand, the significance of elevated cholesterol is not always apparent. Retrospective studies can point to heart problems patients developed, yet these studies may not prove that HIV medications caused high cholesterol, or that the elevated cholesterol was the cause of the heart disease. A recent study in the journal AIDS tried to settle these questions by tracking cholesterol levels in HIV-positive patients, and then by comparing these levels both to those in HIV negative persons and amongst subgroups of persons taking different treatment regimens.17
The patients taking antiretroviral medications were divided into groups: those taking only nucleoside analog drugs (nukes), those taking nukes plus a non nucleoside reverse transcriptase inhibitor (non-nukes), and those taking nukes and protease inhibitors. For the purposes of experimental control, the study did not examine patients who were taking both non-nukes and PIs, though such patients obviously exist and might potentially experience a multiplied impact of the side effects studied here.
The study concluded that patients taking either protease inhibitors or non-nucleoside reverse transcriptase inhibitors were more likely to show elevated total cholesterol, LDL-cholesterol, and HDL cholesterol as well as elevated triglycerides (for protease patients) or apoliprotein A1 (for non-nuke patients).
So what does all of that mean? If you remember the butter vs. margarine debate of the 1980s, you know that all cholesterols are not created equal. Less than a third of the study patients reported to have elevated levels of total cholesterol also had elevated levels of the so called "bad" cholesterol, LDL. On the other hand, patients on any treatment showed significant increases in HDL cholesterol (the so-called "good cholesterol" that is believed to protect against heart disease). In fact, their HDL levels were generally higher than those in untreated HIV-positive patients. So the study could not say that any class of medications, or any specific medication, was likely to promote heart disease. From a cholesterol standpoint, the increase in HDL attributed to treatment may generally offset the increases in LDL and total cholesterol that are common side effects of medications.
Fortunately, these side effects also seem to be the most controllable. Most PLWH develop personal strategies that help mediate these effects. To control nausea and vomiting, some PLWH choose lighter meals with their dosing, while others find a weighty breakfast settles their stomach. Some do better if they drink a lot of water with their dose. One universally accepted rule about diarrhea: it's easier to prevent it than to stop it. Many PLWH regularly take a calcium supplement or a fiber bar to ward off diarrhea.
Now that eradication theory has been soundly disproved, providers and PLWH need to communicate as partners devising a manageable plan. For years, many PLWH were voting with their dosing -- secretly skipping doses of medications they found were not palatable. Yet everyone knows the important of medication adherence; a study at last year's International Conference on AIDS showed that PLWH who took more than 90% of their medications on time were nearly four times more likely to live through the decade than were patients who took less than 90% of their prescribed doses.19
First generation protease inhibitors faced a major challenge: the body cleared them out (through the liver's P450 enzymes) before they could complete their task. To counter this natural bodily response, pharmaceutical companies had to produce treatments with heavy and frequent doses, which only increased patients' side effects. Ritonavir (Norvir) was particularly singled out for unpalatable side effects in its full dose formulation. But then researchers discovered that ritonavir not only inhibits HIV's protease, but also inhibits the P450 enzyme, allowing for a more steady stream delivery of another protease inhibitor into the bloodstream. Many second generation PIs are now "ritonavir-boosted," combining a lower dose of ritonavir with another PI. The first formulation to deliver two PIs in the same pill was Kaletra, which adds lopinavir to low dose ritonavir. Initially this drug was used almost entirely as a salvage option for patients who had suffered failure on other PIs, though now physicians recognize that the drug has durability to support its use as an initial treatment. Today, many medications are prescribed with a ritonavir boost, as the strategy allows for reduced dosing (once or twice daily instead of three times daily) and reduced overall pill burden.
As new treatment options continue to improve life expectancy for PLWH, the issue of side effects becomes more important than ever before, because patients need regimens that they can tolerate during ever-longer lives.
Stephen Fallon, Ph.D. is the President of Skills4, a Florida-based health consulting firm specializing in improving health care and disease prevention efforts nationwide through technical assistance, grant writing, program development, and public education. His clients include the National Minority AIDS Council, the Centers for Disease Control and Prevention, Engender Health International, the Gay & Lesbian Community Center of South Florida, the North Broward Hospital District, Abbott Laboratories, Bayer Diagnostics (previously Visible Genetics), the U.S. Office of Minority Health, GlaxoSmithKline, Bristol Myers Squibb Immunology, the Council of Community Clinics, and the Departments of Health of Florida, Arizona, Kentucky, Tennessee, Arkansas, Montana and Ohio.
Article sponsored in part by an unrestricted grant from Abbott Virology.