After achieving an undetectable viral load, missing between 11% and 30% of cumulative doses of antiretrovirals during the course of a year leads to the greatest risk of viral rebound due to resistance, according to the results of a US study due to be published in the October 15th issue of Clinical Infectious Diseases (now available online).
These results concur with a previously reported study at the IAS conference in Paris that found that taking between 60% and 90% of prescribed antiretrovirals was the surest way to acquire resistance.
Both studies suggest that 90% or better adherence is needed to prevent the emergence of clinically significant resistance. This is the equivalent of missing no more than five doses of a twice daily regimen in a four week period, or three doses of a once daily drug in a month.
Although this study also attempted to gauge the significance of taking doses on time or more than two hours late, as well as the importance of adherence to food restrictions (e.g. fasting on didanosine, ddI), a limited sample size prevented the authors detecting these effects on viral rebound.
A total of 195 patients attending an HIV outpatient clinic in Baltimore, Maryland were recruited into a year-long observational study in 2000, for a total of 1188 visits. There were a median of seven clinic visits during the study, and the median time between study visits was 49 days. At each visit blood plasma samples were obtained and stored for genotypic resistance testing. Adherence was measured by patient self-report of how many doses they had missed, taken late or without heeding food restrictions over the three days prior to their study visit.
Study participants had to be on a HAART regimen that kept their viral load below 500 copies/mL. Most were on either a PI- or NNRTI-containing regimen, and 58% were treatment experienced prior to their current HAART regime, although none had documented HIV mutations that would have led to clinically significant resistance.
At baseline, 39 participants (20%) reported having missed one or more doses in the preceding three days. This was a statistically significant (p
During the study period, 28 (14%) developed clinically significant resistance. Multivariate adjusted relative hazard (RH) found that certain degrees of cumulative adherence were found to be a statistically significant relative hazard for treatment failure. With 80-89% of doses taken the RH was 2.71 (95% CI 1.02-7.19; p = 0.045), and with 70-79% it was 3.60 (95% CI 1.15-11.23; p = 0.027). Taking 90% or more or less than 69% of cumulative HAART was not associated with clinically significant resistance.
The most statistically significant factor, however, was having missed a clinic visit in the previous 30 days (RH 2.4; 95% CI 1.13-5.09; p = 0.023). The greatest relative hazard was found to be 3.66 (95% CI 1.05-12.75; p = 0.042) which was associated with having had a nadir (lowest) CD4 cell count of less than 200 cells/mm3prior to starting HAART. Both of these factors have previously been shown to lead to an increased risk of HAART failure.
Twelve study participants developed clinically significant resistance despite having 100% cumulative adherence. The authors note that self-reporting can overestimate adherence to treatment. However, no therapeutic drug level monitoring was undertaken in these participants, and so it is not possible to rule out suboptimal absorption as a reason for HAART failure.
The authors conclude that HAART adherence of 70-89%, along with recently missing a scheduled clinic visit are the two fluctuating factors associated with the greatest risk of viral rebound with clinically significant resistance, “indicating a need to clinic staff to assess the underlying reason for missed appointments...[and] the need to set high adherence goals.”
Further information on this website
Adherence, drug levels and baseline viral load associated with resistance - news story July 2003
What level of adherence is necessary? - menu of information
Resistance - booklet in the information for HIV-positive people series (pdf)
Sethi AK et al. Association between adherence to antiretroviral therapy and human immunodeficiency virus drug resistance. Clin Infect Dis 37, electronically published 19 September, 2003.