Boosted PIs appear less vulnerable to imperfect adherence than other regimens

[Guest guest]

Boosted PIs appear less vulnerable to imperfect adherence than other regimens

By Mark Mascolini

February 08, 2006

Regimens based on ritonavir-boosted PIs may allow less-than-perfect adherence without virologic rebound, according to results of a British Columbia cohort study presented at the 13th Conference on Retroviruses and Opportunistic Infections. However, less than 95% adherence to a combination based on an unboosted PI or NNRTI heightened the risk of viral breakthrough in this cohort.

The study also confirmed that the highest rates of adherence provide the best chance of sustained viral control with either PIs or NNRTIs.

With colleagues at the University of British Columbia, Vancouver, Canada, and 2 US centers, Gross calculated adherence through pharmacy refill records for 1634 adults who began antiretroviral therapy between July 1996 and June 2003 and reached an HIV RNA load < 500 copies/mL. All HIV patients in British Columbia receive free antiretrovirals distributed through designated pharmacies.

The study group began treatment with a median CD4+ cell count of 200 cells/mm3 (interquartile range [iQR]: 80-350 cells/mm3) and a median viral load of 120,000 copies/mL (IQR: 40,000-309,500 copies/mL). Unboosted PI regimens, prescribed to 46%, were the most common regimen used, followed by NNRTI-based combinations in 39% and combinations that included a boosted PI in 15%.

After a median follow-up of 29 months, Dr. Gross calculated the risk of < 95% adherence and viral breakthrough, defined as 2 consecutive HIV RNA viral loads > 1000 copies/mL. The risk proved highest among people taking an unboosted PI (unadjusted hazard ratio

---

: 1.92; 95% confidence interval [CI]: 1.52-2.42), somewhat less for those taking an NNRTI (HR: 1.58; 95% CI: 1.09-2.29), and was lowest for those taking a boosted PI (HR: 1.28; 95% CI: 0.57-2.86).

A multivariate analysis of factors linked to viral rebounds determined that < 95% adherence raised the rebound risk by 66% with unboosted PIs (adjusted HR [AHR]: 1.66; 95% CI: 1.35-2.10) and by 47% with NNRTIs (AHR: 1.47; 95% CI: 1.01-2.14). But < 95% adherence did not raise the breakthrough risk significantly with boosted PIs (AHR: 1.05; 95% CI: 0.46-2.42).

People with a history of injecting drug use experienced a higher breakthrough risk with boosted PIs (AHR: 1.69) than with NNRTIs (AHR: 1.47) or unboosted PIs (AHR: 1.37). But wide confidence intervals (0.86-3.34) around the AHR for boosted PIs undermine the reliability of that finding.

The clinically relevant adherence threshold for boosted PIs may lie somewhere below 95%, Gross and colleagues concluded. But they cautioned that a formal test for the interaction between adherence to boosted PIs and rebound was not significant and that pretreatment resistance may confound their results.

Reference

Gross R, Yip B, Wood E, et al. Boosted PI are more forgiving of suboptimal adherence than non-boosted PI or NNRTI. Program and abstracts of the 13th Conference on Retroviruses and Opportunistic Infections; February 5-8, 2006; Denver, Colorado. Abstract 533.

Regards, Vergelpowerusa dot org"The great tragedy of life is not that people set their sights too high and fail to achieve their goals but they set their sights too low and do."Michelangelo