Sex Differences in Antiretroviral Therapy Toxicity: Lactic Acidosis, Stavudine, and Women

Lactic acidosis, a serious complication of treatment with nucleoside analogue reverse-transcriptase inhibitors, was first reported in patients with HIV infection <15 years ago [1, 2]. Initial reports included patients treated with didanosine and zidovudine, but it later became clear that treatment with stavudine was most closely associated with this complication. A common finding in the literature describing case series of nucleoside analogue reverse-transcriptase inhibitor—related lactic acidosis has been the over-representation of women (who accounted for 83% of cases in 1 series), especially obese women [3,4,5–7]. Sex-specific event rates for lactic acidosis were not calculated in earlier series because of limited prospective data; however, given the relatively low proportion of women in HIV care in the mid-1990s in the United States and Europe, one can imagine that the female-specific incidence rates were very high. Because of the fact that <50% of people with HIV infection in sub-Saharan Africa are women, the high rates of lactic acidosis being reported in recent data from antiretroviral treatment programs in South Africa should not come as a surprise. Geddes et al. [5] reported 14 cases of lactic acidosis among 891 patients receiving treatment with stavudine, lamivudine, and a nonnucleoside reverse-transcriptase inhibitor. In their series, all 14 cases of lactic acidosis occurred among obese women with a median weight of 81 kg. The overall incidence rate was 19 cases per 1000 patient-years; however, this rate included both men and women in the denominator. The mortality rate was 29%. In this issue of Clinical Infectious Diseases, Bolhaar and Karstaedt [6] extend these earlier findings with a retrospective review of 1735 patients initiating antiretroviral therapy in Soweto, South Africa. In their series, the incidence of lactic acidosis was 16.1 cases per 1000 patient-years in women, compared with 1.2 cases per 1000 patient-years in men. Again, the mortality rate was 30%. This new study also reports rates for symptomatic hyperlactatemia, defined as a lactate level ⩾4.5 mmol/L with a normal serum bicarbonate level or pH. The incidence of symptomatic hyperlactatemia among female patients was 27 cases per 1000 patient-years, compared with 8.7 cases per 1000 patient-years among male patients. In both reports, the authors emphasize that these rates may be an underestimate of the true incidence rates, because they were derived from retrospective data. Both groups also report outcomes for the small number of patients who have recovered from lactic acidosis and who subsequently underwent successful treatment with zidovudine in place of stavudine; outcomes were consistent with earlier experience in the United States reported by Lonergan et al. [8]. The mechanism for the marked sex difference in rates of lactic acidosis rates has not been elucidated. In most cases, stavudine is dosed on the basis of weight, with 40 mg administered twice daily for patients weighing <60 kg. Women who weigh 60–70 kg might be more likely to receive a higher dosage per kilogram of body weight than men who weigh 80–90 kg; however, it is not clear that this small difference in weight-based dosing explains the higher rates of lactic acidosis, mostly because the median weight of women in the Geddes series was 80 kg. In addition, in a preliminary report by Osler et al. [9], the risk of lactic acidosis appeared to increase with weight <75 kg, and the OR for women in this case-control study was 33-fold the OR for men. There is growing support for the use of the lower dosage of stavudine (30 mg twice per day), including a recent addendum to the World Health Organization guidelines for the use of antiretroviral therapy; however, it is not known whether this will reduce the incidence of this serious complication. Another possible explanation for the sex difference in rates of lactic acidosis is that obese women are predisposed to hepatic steatosis, which may increase their risk for developing lactic acidosis. Alternatively, sex-linked mitochondrial genetic differences could increase susceptibility to the development of lactic acidosis. Before the availability of alternative first-line nucleoside analogues (e.g., tenofovir and abacavir), studies examined the role of routine monitoring of lactate levels as a means of reducing the rates of this dreaded complication of treatment. Unfortunately, routine monitoring of lactate levels in the United States did not appear to be beneficial, likely because of the infrequency of lactic acidosis and the problematic nature of measuring lactate levels. Stavudine is no longer favored as a first-line treatment agent in developed countries as a consequence of both short-term toxicity (lactic acidosis) and long-term toxicity (lipoatrophy and neuropathy). However, because of its relatively lower production costs, compared with those for alternative drugs, stavudine remains a mainstay of generic fixed-dose combination regimens currently employed in most resource-limited settings. The data reported by Bolhaar and Karstaedt [6] underscore the urgency of ongoing efforts to secure lower-cost alternative first-line treatment regimens to replace stavudine. Clearly, the factors that influence treatment procurement are complex and evolving. These recent data from 3 sites in South Africa highlight the importance of considering the costs of toxicity monitoring and the treatment of patients who experience complications at the program level. Although a greater awareness about the risk of lactic acidosis has led to efforts to assess patients presenting with symptoms suggestive of this syndrome, these activities are associated with increased costs and are not sufficiently accurate to completely prevent morbidity and mortality. It is expected that the cost of first-line nucleosides (tenofovir and zidovudine) will decrease during the next 18 months. In the mean time, when limited supplies of zidovudine or tenofovir as alternatives to stavudine become available, women—and, in particular, obese women—should be the first in line to receive them. Finally, it is notable that nevirapine, the other mainstay of first-line treatment in the developing world, is also associated with a sex difference in toxicity. A higher rate of symptomatic hepatitis among women with CD4+ cell counts <250 cells/mm3 was noted in post—marketing surveillance data several years after the drug was approved [10]. The delay in identifying this finding was, in part, attributable to the fact that the drug was not initially studied in large populations of women, especially not among women with higher CD4+ cell counts. Collectively, the data on the higher rates of 2 important toxicities among women provide strong evidence for the need to closely examine drug safety in adequately powered clinical trials with adequate enrollment of women as early as possible in drug development. Potential conflicts of interest. J.S.C. has received recent research grants from Merck, Tibotec, Theratechnologies, and GlaxoSmithKline and is a consultant and/or a member of advisory boards for Bristol Meyers Squibb, Gilead, GlaxoSmithKline, Abbott, Vertex, and Tibotec.