PIVOT study: further analysis from five-year PI/r monotherapy strategy study

BHIVA 2014 included an expanded analysis from the MRC-sponsored PIVOT study whose primary endpoint was the preservation or loss of treatment options after three years. [1]

PIVOT randomised 587 patients on stable ART to either maintain standard triple therapy or switch to ritonavir-boosted PI monotherapy.

In the last issue of HTB we reported top-line results from the PIVOT study, presented as a poster at CROI 2014, which noted a significantly higher rate of virological rebound (three viral load results >50 copies/mL) in the monotherapy group (35% vs 3%: difference 31.8%, 95% CI 24.6 to 39.0%; p <0.001). This difference was partially driven by a low rate of viral rebound in the standard therapy group (certainly compared to most published studies). [2]

Median duration of follow-up was 44 months, with only 2.7% of patients lost to follow-up or discontinuing over five years. Although the choice of PI/r was an investigator decision, data emerging from other studies supported greater safety from using darunavir/r or lopinavir/r. In PIVOT, darunavir/r was used by 79% (n=231) of the monotherapy group, 14% used lopinavir/r, 6% used atazanavir/r and one person used saquinavir/r.

Viral rebound was defined as three consecutive results > 50 copies/mL, including an option to retest the initial sample, and with a confirmatory test 4-8 weeks later. All patients with viral rebound resuppressed, following the addition of dual-NRTIs, switching back to the previous combination, or spontaneously.

New drug resistance to one or more class (the definition used for the primary endpoint of “loss of treatment options”) occurred numerically more frequently during PI/r monotherapy. However, resistance was uncommon in both arms [2.1% vs 0.7%, difference 1.4% (-0.4 to 3.4%), at three years] and the between-arm difference was not significant (p=0.15).

These results supported a conclusion of non-inferiority for the PI/r monotherapy strategy. The one person who developed PI resistance in the monotherapy arm was using atazanavir/r (now known to be insufficiently potent for use as monotherapy) who developed I50L/I. There were no significant differences in other secondary clinical endpoints, or in results at five years.

Grade 3/4 events were more common in the triple therapy arm (46% vs 55%, p=0.043) with this driven by laboratory (p=0.002) rather than clinical (p=0.12) events. There were more deaths in the monotherapy arm (6 vs 1) but these were not related to treatment or strategy (suicide, pulmonary embolism, breast cancer, lung cancer, glioblastoma and anal cancer with PI monotherapy and metastatic adenocarcinoma in the control group).

No differences were seen between the two strategies in changes in neurocognitive function from baseline (NPZ-5 score from verbal learning tests, colour trails and grooved pegboard). Given the concern about CNS penetration for patients using PI/r monotherapy, it is important and reassuring that there appear to be no marked functional CNS consequences.

Comment

This is the largest HIV study to be run in the UK for 20 years (since DELTA). It is also notable that only 3% of patients on the triple therapy arm had viral rebound over five years.

Although the primary endpoint shows non-inferior results well within the predefined upper margin of error, the significantly higher rate of viral rebound may complicate patient management and limit enthusiasm for doctors in the UK to use this strategy.

Cost effectiveness analyses are planned and will need to model this additional care required for people with viral rebound, especially as dual nucleoside components may soon become available as generics.