Two studies presented in poster discussion sessions on Thursday at the Twelfth Annual Retrovirus Conference in Boston confirmed the importance of therapeutic drug level monitoring in monitoring the effectiveness of antiretroviral therapy.
The studies found that measuring levels of protease inhibitors (PIs) and non-nucleoside reverse transcriptase inhibitors (NNRTIs) in patients can be used to ensure that drug levels are high enough to suppress HIV and to guide decisions on dose adjustments when high levels of drug lead to toxicity. It is also more useful in patients with higher body weight and taking certain antiretroviral drugs.
Researchers from France performed therapeutic drug monitoring on 115 PI-naive individuals who had recently begun highly active antiretroviral therapy (HAART). Forty-two patients were taking indinavir (Crixivan) boosted with ritonavir (Norvir), 38 were on a ritonavir-boosted lopinavir (Kaletra)-based regimen, and the remaining 35 were taking nelfinavir (Viracept) as 625mg tablets.
All PIs were dosed twice daily, and drug levels were measured at weeks 2, 8 or 16, 24, and 48 after initiating HAART. During the first 24 weeks, PI doses were adjusted if trough concentrations were outside of the manufacturers’ recommended range.
After 48 weeks, failure, which was defined by either two consecutive viral load tests above 200 copies/ml after week 16, or a PI-related adverse event, was observed in three, five, and twelve participants taking indinavir, lopinavir and nelfinavir, respectively.
In an intent-to-treat analysis, 70% of the patients taking indinavir, and 69% of the lopinavir group were successfully treated by week 48. In contrast, only 44% of the patients in the nelfinavir arm were successfully treated.
The researchers note that a majority of the participants taking nelfinavir had suboptimal levels early on in the study, with 62% being outside the therapeutic range at week 8. Consequently, ritonavir was added to ten of the participants’ regimens. This, say the researchers, was well tolerated and efficiently increased the concentrations in six of the ten.
The researchers concluded that early systematic therapeutic drug monitoring is an efficient method for improving virological outcomes and toxicity in individuals on boosted indinavir or Kaletra but recommended that boosted nelfinavir be evaluated further for individuals with low drug levels.
In the second presentation, researchers from California set out to determine whether therapeutic drug monitoring could be targeted at specific patients, finding that 38% of study participants needed drug level adjustments.
In this substudy of CCTG 578, a randomised controlled, factorial study that compared therapeutic drug monitoring and standard of care, an expert committee reviewed blinded drug level, viral load, CD4 cell count, and toxicity data from 177 of the 199 participants and recommended regimen changes if they were deemed necessary.
Drugs levels were measured just before, and at two and four hours after a directly observed dose two weeks after the participants had begun a new PI- or NNRTI-based HAART. The most commonly-used PI was Kaletra, while efavirenz (Sustiva) was the commonest NNRTI.
At baseline, 29% of the participants were antiretroviral naive, 33% were treatment experienced on a current regimen, and 38% were treatment experienced but on a treatment interruption.
The expert committee recommended that 67 (38%) of the patients change the dose of at least one drug. However, only three (2%) of these recommendations were to decrease exposure.
In multivariate analysis, risk factors that independently predicted the need for a change in drug dose were body weight (odds ratio [OR] 1.16, p = 0.003) and use of efavirenz (OR 4.6, p = 0.001) or lopinavir (OR 4.6, p = 0.008). However, age, gender, baseline viral load and CD4 cell counts, prior use of antiretroviral therapy and adherence were not associated with a recommendation to change drug dosage in univariate or multivariate analyses.
In the poster discussion session, the researchers were praised for combining assessment of the patients’ drug levels with information on toxicity and virological and immunological effectiveness in deciding whether a dose adjustment was necessary. However, the researchers concluded that future research should aim to refine the identification of candidates for therapeutic drug monitoring, since their study revealed that it was not necessary in the majority of their study population.
Mentré F et al. Prospective trial to evaluate how therapeutic drug monitoring of protease inhibitors increases virologic success and tolerance of HAART (COPHAR2—ANRS 111 Trial). Twelfth Conference on Retroviruses and Opportunistic Infections, Boston, abstract 639, 2005.
Haubrich R et al. Determinants of the need for therapeutic drug monitoring: rates and predictors from CCTG 578. Twelfth Conference on Retroviruses and Opportunistic Infections, Boston, abstract 640, 2005.