Weight Gain in Patients with AIDS-Related Cachexia: Is Bigger Better?

When I was serving recently as the teaching attending in a general medical ward team, the most heated disagreements were about the nutritional support of a patient with the acquired immunodeficiency syndrome (AIDS). He had decreased appetite and oral intake leading to a modest loss of body weight in association with other typical complications of AIDS. He was awaiting—no, demanding—placement of a central venous catheter for hyperalimentation. Aware that weight loss was associated with shortened survival in AIDS, he was adamant about delaying his death by increasing his caloric intake. As an oncologist who works with patients with cancer who have cachexia caused by malignancy and have wasting from multiple disease-related causes, I was aware of the controversies surrounding the level of supportive care of the terminally ill. I was unprepared, however, for the passionate debate that this patient with AIDS engendered among the ward team. What right did this patient have to demand an expensive, complex therapy? How exactly could treatment improve his life? Could the complications of total parenteral nutrition shorten his survival? Where were the data? Although we all had opinions about these medical and ethical issues, we had few data on which to base our opinions about the expenses and the legitimate outcomes of preventing weight loss in this patient.

In this issue, two randomized, placebo-controlled clinical trials [1, 2] are reported in which high doses of megestrol acetate were given to patients with AIDS. The most common side effect of progestational agents in patients with metastatic breast cancer is weight gain [3]. A previous pilot trial [4] of megestrol acetate in patients with AIDS suggested that this undesirable side effect in women with breast cancer could be exploited to alleviate the effects of AIDS-related cachexia. Both of the trials reported in this issue chose weight gain as the primary objective end point of treatment, and both included a questionnaire to assess patients' perception of well-being. Whereas the study by Von Roenn and colleagues [1] addressed dose-response by randomly assigning patients to receive placebo or three different dose levels of megestrol acetate, the study by Oster and colleagues [2] simply compared the highest dose level in the Von Roenn trial, 800 mg per day, with placebo. Eligibility criteria were similar in the two trials, and each group of investigators had the same difficulties entering patients into the study and having patients complete the full 12-week course of treatment. Forty-nine of the 371 participants in the two trials failed to meet the stated entry criteria, and a relatively high rate of attrition occurred during therapy, typically caused by comorbid illness common in this study population.

The results of the two trials are remarkably similar. For the primary end point of weight gain, the trial by Von Roenn and colleagues showed that 800 mg per day of megestrol acetate was superior to placebo and to lower doses of drug, and the study by Oster and colleagues documented a nearly identical weight gain of approximately 3.63 kg (8 lb) at this dose level. Both studies showed an increase in daily caloric intake of somewhat greater than 600 calories at the highest dose of megestrol acetate. Body composition measurements showed that this weight gain was not related to water retention but primarily to increased levels of stored fat and, to a lesser degree, to increased lean body mass. In addition to objective evidence that megestrol acetate led to increased caloric intake and body weight, the nine-question scale used to assess patients' perception of well-being also consistently favored megestrol acetate to placebo, particularly at the highest doses. These perceptions were more favorable for the specific measures of appetite and weight gain and less so to overall well-being and quality of life. This perceptual difference during the course of the 12-week trial may simply reflect the development of additional substantial complications of HIV infection that might have occurred in the study population.

Because megestrol acetate was generally well tolerated in these two trials and has been shown to be relatively safe when used for longer periods of time in patients with AIDS-related metastatic breast cancer, are there any reasons not to consider megestrol acetate the treatment of choice for AIDS-related cachexia? There are no obvious or easy alternatives: Intensive oral nutritional support is usually unsuccessful, and total parenteral nutrition is expensive and complicated. The chief drawback to megestrol acetate therapy as given in these trials is the cost of the medication. Although not nearly as costly or complex as total parenteral nutrition, the cost to the patient for megestrol acetate suspension in our hospital pharmacy is approximately $250 per month ($3000 per year). The generic oral preparation (which would require twenty 40-mg tablets daily) is even more expensive at nearly $4400 per year. For a patient with AIDS, who may need many other oral medications not reimbursed by health plans, megestrol acetate simply may not represent a legitimate treatment option for AIDS-related cachexia.

These trials raise several issues that must be studied before all patients with AIDS-related cachexia who have “persistent weight loss of 5% or more of ideal body weight” are treated with megestrol acetate, as endorsed in the concluding sentence of the Von Roenn [1] article. Like the cancer anorexia-cachexia syndrome, AIDS-related weight loss is multifactorial in origin and may not only be a correlate of poor prognosis related to progression of the basic disease process but, possibly, may also be an independent factor in survival [5]. The study samples in these trials were highly selected in an attempt to exclude patients with identifiable causes of weight loss, such as intractable diarrhea and untreated infections. It would be interesting to know, therefore, how many patients were excluded from the studies because of these criteria. Without this information and without a close reading of these articles, it may be difficult for the casual reader and for the lay press to assess how broadly these results may be applied to the larger population of patients with AIDS.

The measurement of success in these trials was unique, in that weight gain and perception of well-being were considered to be desirable end points of treatment, independent of more traditional measures such as survival. This innovation is laudable, standing in contrast to many oncology clinical trials in which patient perceptions of treatment benefit have been rarely systematically evaluated with the same vigor as more traditional measures of success, such as toxicity and survival. More recently, cancer clinical trials have begun to include quality-of-life questionnaires that collect information on patients' overall perceptions of treatment and the specific effects of therapy on functional aspects of their lives. The psychosocial scales used in the two trials reported in this issue address psychological well-being but not other recognized quality-of-life measures, including functional adaptation. A broader approach to the measures of physical functioning, including self-care, mobility, and activity and role limitations, has been applied to the use of megestrol acetate in patients with advanced breast cancer [6]. A similar battery of psychosocial tests, therefore, would allow for a more comprehensive evaluation of treatment with supportive measures, such as megestrol acetate.

Future trials of megestrol acetate in patients with AIDS-related cachexia should continue to ask questions about weight gain and perceptions of well-being but should also investigate which patients are most, or least, likely to benefit from reversal of cachexia, by evaluating additional measures of quality of life and functionality. In addition, efforts should be made to determine the optimal selection of patients and the timing and duration of therapy, so that the added cost of this treatment is distributed equitably to those who are likely to benefit the most.

• Copyright ©2004 by the American College of Physicians