HIV needs many more mutations to acquire resistance to darunavir

HIV may need to acquire five to six more mutations (changes) in its protease sequence in order to become resistant to the newest protease inhibitor (PI) drug darunavir (TMC114, Prezista) than it does to become resistant to most other PIs, the Eighth Glasgow International Congress on Drug Therapy in HIV Infection heard last week.

It also heard that of samples sent to a national resistance testing facility, 86% of those with PI resistance had no mutations associated with resistance to darunavir and of those that did, 78% had only one darunavir mutation, which in itself is unlikely to cause treatment failure. This means that only 3% of UK patients currently failing other PI-based regimens have HIV genotypes that predict resistance to darunavir.

Carlo-Federico Perno of darunavir manufacturer Tibotec told a Tibotec-sponsored satellite meeting that the reason for darunavir's exceptional ‘resistance to resistance’ was that it bound much more tightly to the HIV protease enzyme than other PIs, forming two more chemical bonds than most. Its ‘dissociation index’ or tendency to become detached from protease was 100 times lower than other PIs.

Professor Brian Gazzard of the Chelsea and Westminster Hospital said that mutations associated with resistance to darunavir comprised a distinctive list of eleven mutations, most of which were not primary resistance mutations to other PIs. The presence of one or two of these mutations did not confer significant phenotypic (clinical) resistance. For instance the most significant mutation was the one called I50V, which conferred fourfold phenotypic resistance to darunavir. In other words, more than four times as much of the drug was needed to suppress replication in HIV that had this mutation. However darunavir’s ‘cutoff’, the reduction in potency against HIV necessary to start making a real clinical difference in efficacy, was a tenfold phenotypic change.

In the clinical trials of darunavir, it had produced viral load drops of 2.08 logs (a 120-fold drop) in people with less than tenfold resistance, 1.08 logs in people with between 10 and 40-fold resistance and a 0.78 log (less than sixfold) drop in people with more than fortyfold resistance. Less than 45% of people with more than three darunavir resistance mutations had a greater than one long (tenfold) reduction in viral load.

What this meant in practice was that the degree of protease inhibitor resistance seen in patients with three darunavir mutations (since these mutations confer resistance to most other PIs too) was equivalent to nine of the PI mutations listed on the International AIDS Society database of common resistance mutations,

Gazzard said that T-20 (enfuvurtide, Fuzeon) could make a difference in treatment response to people with medium-level resistance to darunavir. In the darunavir clinical trials in treatment-experienced patients no difference in treatment response according to T-20 use was identified in patients with a less than tenfold phenotypic resistance increase against darunavir – in both cases 53% of patients responded to darunavir. Nor was there any difference in treatment response in patients with more than 40-fold resistance to darunavir, where 12-13% responded. But when it came to patients with between 10 and 40-fold resistance to darunavir, then 59% of those taking T-20 for the first time responded compared with 13% of those who did not.

However Gazzard warned that T-20 intake had not been randomised in any of the darunavir trials, so patients who were naïve to it were also likely to be less PI-experienced than patients who had already taken it.

How common, currently, is resistance to darunavir? Clive Loveday, who operates his own resistance-testing facility as a charitable trust in the United Kingdom, analysed PI resistance in 885 UK patients whose samples were sent for analysis.

Of these patients 60% were currently failing a PI-based regimen; 21% were on non-PI-based regimens but had taken PIs before, and 19% were off treatment.

Eighty-six per cent (126) of these patients had none of the 11 signature darunavir resistance mutations, and of these 126, 98 (78%) had only one darunavir mutation, which should not be clinically significant. There were 13 patients (1.47% of the total) with more than three darunavir mutations and two with more than five. Interestingly, not a single darunavir mutation was seen in the database before 2000, indicating that these mutations will tend to slowly accumulate even in patients on PIs other than daruanvir.

Darunavir mutations confer resistance to most other PIs including lopinavir and fosamprenavir, but there is limited cross-resistance with tipranavir; 40-50% of patients with darunavir mutations will be sensitive to tipranavir.