High levels of drug changes
The HIV Implementers’ Meeting heard of high levels of drug changes in people taking first-line therapy, due to peripheral neuropathy caused by d4T.
In Uganda, Willy Were of CDC-Uganda reported that 10% of patients receiving treatment through a home-based care programme developed severe peripheral neuropathy, and this side-effect was vastly more frequent than any other severe side-effect. Seventeen per cent of patients switched from d4T during 18 months of follow-up, compared with 4% who switched from nevirapine due to adverse reactions (chiefly rash).
Participants in the study had a 73% probability of remaining on their original three drug, first-line regimen after 18 months, and a 16% probability of experiencing a severe adverse event by this point.
Severe peripheral neuropathy was significantly more likely to occur in people aged 35 years and over (hazard ratio 2.88 (confidence interval 1.22 – 2.71), and in those taking TB treatment at the same time as d4T.
A similar pattern was seen in Cote d’Ivoire, in a report from Dr Eugene Massou. He told the conference that among 2012 patients receiving one of three first-line regimens (d4t/3TC plus nevirapine (32%) or efavirenz (25%), or AZT/3TC/efavirenz (38%)), the incidence of peripheral neuropathy was also vastly greater than any other serious side effect.
17.9 cases per 100 patient years of d4T treatment were observed, compared with 3.9 cases of anaemia per 100 patient years of AZT treatment and 6.3 cases of rash per 100 patient years of nevirapine treatment. Median follow-up was 16.9 months.
Patients receiving d4T and nevirapine were more than three times more likely to modify their treatment than patients taking AZT/3TC/efavirenz (34 changes per 100 patient years compared to 10.5 changes per 100 patient years).
In Rwanda 83% of changes reported at two health facilities were from d4T, almost entirely due to peripheral neuropathy.
In Kenya however peripheral neuropathy was not reported as a major cause of treatment changes by Lillian Kocholla of Mgabathi Hospital in Nairobi. She reported on 486 changes made to first-line therapy among more than 2000 patients receiving antiretroviral therapy. Sixty-five per cent of changes were due to lipodystrophy and only 11% due to virologic failure of first-line treatment, with peripheral neuropathy leading to treatment changes in less than 5%.
Need for common definition of toxicities
The findings led some audience members to ask about the definition of peripheral neuropathy used for switching treatment in Kenya, and Dr Eric van Praag of Family Health International remarked that his field experience in Tanzania had shown that conduct of a neurological examination for peripheral neuropathy was a skill that tended to get forgotten very quickly by doctors after basic training.Prof. Charles Gilks said that reaching common definitions of all the toxicities being seen by treatment programmes would be very important for pharmacovigilance programmes. “We need common definitions because because I’m not sure what Mgabathi hospital is describing is the same as what’s coming out of Rwanda. I also think more training is needed so that health care workers know what to look for.”
Without common definitions and diagnostic procedures it would be difficult to define the true incidence, he said. WHO will convene an expert consultation to reach agreement on definitions within the next few months, he went on.
Differences in the incidence of side effects may reflect real differences between populations rather than observation bias, however. Lactic acidosis, another life-threatening side-effect of d4T, has been observed at an unusually high frequency in women with higher body weight (> 75kg), possibly due to fat accumulation in the liver as a result of greater body weight. Much higher body weights have been seen in South Africa in women starting treatment.
Prof. Gilks told aidsmap that evidence and programme management were beginning to favour the use of AZT/3TC as the nucleoside analogue backbone in resource-limited settings not only for reasons of safety, but also because the use of this backbone would allow a smaller formulary to be maintained, thus avoiding stock-outs and simplifying procurement. He pointed out that AZT/3TC could also be used in prevention of mother to child transmission and in post-exposure prophylaxis for health care workers and following sexual assault, both major problems that needed to be addressed within treatment programmes.
References
Were W et al. Clinical toxicity to highly active antiretroviral therapy in a home-based AIDS care program in rural Uganda. HIV Implementers’ Meeting, Kigali, Rwanda, abstract 1134, 2007.
Messou E et al. Main reasons of modification of the first-line antiretroviral regimen in adult patients who intiated HAART in the International Family Health Initiative in Abidjan, Cote D’Ivoire. HIV Implementers’ Meeting, Kigali, Rwanda, abstract 1215, 2007.
Turate I et al. Observations on 406 adverse effects necessitating drug changes in 1st line ARV regimens in patients from two health care facilities in Rwanda and recommendations for patient care. HIV Implementers’ Meeting, Kigali, Rwanda, abstract 493, 2007.
Kocholla L et al. Reasons for switching highly active antiretroviral therapy regimens among HIV/AIDS patients in low-resource settings: Mbagathi Hospital, Kenya. HIV Implementers’ Meeting, Kigali, Rwanda, abstract 685, 2007.