ABT-378/r (Lopinavir): defining a resistance profile and the effects of resistance on virological response in experienced patients

ABT-378 is a protease inhibitor being developed by Abbott Laboratories. When combined at 400 mg with ritonavir 100 mg (ABT-378/r) and dosed BID it displays good oral bioavailability and a long half-life.

Being a peptidomimetic compound it is likely to share similar genotypic resistance patterns with currently available PIs. The high trough concentrations relative to the EC50 (at least for wild-type HIV) have led many to believe that ABT-378 might retain activity against some PI-resistant HIV mutants. Data from studies in patients with past history of treatment failure on other PIs who have subsequently received ABT-378/r is difficult to interpret. So far, prior PI-experience has been limited and the initiation of an NNRTI at the same time as ABT-378/r in NNRTI-naive subjects makes it unclear how much of the antiviral effect observed may be due to the ABT-378/r or the NNRTI.

A series of presentations on ABT-378/r attempted to define further both the resistance profile of this compound, and its activity in subjects experiencing virological failure on multiple prior protease inhibitors.

Dale Kempf of Abbott Laboratories presented the results of a study which identified genotypic correlates of reduced in vitro susceptibility to ABT-378 [8]. A total of 112 clinical isolates from patients failing either a single prior PI (n=56) or multiple PIs (n=56) were both genotyped and tested for phenotypic sensitivity to ABT-378. Susceptibility was expressed as fold EC50 compared to the wild-type standard virus. Various methods of statistical analysis were then applied to determine which mutations were associated with reduced susceptibility. The relationship of phenotypic susceptibility to the number of mutations was also assessed using linear regression. The results revealed that:

  • Eleven mutations in HIV protease were statistically associated with reduced in vitro susceptibility to ABT-378 in HIV selected by other PIs (at amino acid positions in the protease gene 10, 20, 24, 46, 53, 54, 63, 71, 82, 84 and 90).
  • Across the whole panel of isolates there was a mean 2.8-fold (range 0.7 – 26.0) higher EC50 in single PI-experienced subject isolates and a mean 16-fold (range 0.5 – 96.0) higher EC50 in multiple PI-experienced subject isolates.
  • An ABT-378 ‘mutation score’ defined as the total number of mutations associated with either 4 or 10-fold decreased susceptibility (n=6 at codon 10, 56, 63, 71 82 and 84) or 20 or 40-fold decreased susceptibility (n=2 at codon 20 and 53) was constructed and found to be a better descriptor of reduced susceptibility than the total of genetic changes from the wild-type sequence.
  • The median fold EC50 of ABT-378 increased incrementally with ABT-378 mutation scores of 4 and above.
  • The 16 viruses that displayed >20-fold change in phenotypic susceptibility to ABT-378 all contained mutations at positions 10, 54, 63 plus 82 and/or 84. In addition, they contained a median of 3 mutations at positions 20, 24, 46, 53, 71 and 90.

It was concluded from this study that the ABT-378 mutation score constitutes a virtual phenotype for ABT-378 and can be used to predict phenotypic sensitivity. It was further surmised that as the through plasma levels of ABT-378 are >75-fold above the serum adjusted EC50 against wild-type HIV that the accumulation of at least 6-8 mutations might be required to achieve clinically relevant phenotypic resistance to ABT-378/r.

In a related presentation, Eugene Sun of Abbott Laboratories presented an analysis of response rates by genotype and phenotype for subjects who had experienced virological rebound on prior regimens including at least 2 PIs (Study M98-957) [9]. Subjects were na•ve to NNRTIs and initiated therapy with a regimen containing efavirenz, ABT-378/r and investigator selected NRTIs. After 24 weeks the virologic response to therapy was analysed with respect to baseline phenotype and genotype, and the corresponding ‘clinical breakpoints’ for ABT-378/r in this patient population were identified.

Phenotype and genotype was obtained for 56/57 baseline viruses and the EC50 of ABT-378 ranged from 0.5 to 96-fold above the EC50 against wild type HIV. The prevalence of >10-fold and >40-fold change in ABT378 susceptibility was 24/56 (43%) and 8/56 (14%) respectively. When using the ABT-378 mutation score the majority of baseline isolates (79%) had mutation scores of 4 or higher.

Fifty two of the 57 patients enrolled in this study qualified for the analysis of week 24 response with 42 (81%) of these having had HIV RNA ²400 copies/mL at week 24. Genotype and phenotype showed the following effects on response

Table 1.

Baseline phenotype for ABT-378 Virological response ²400 copies/mL at week 24
<10-fold reduced susceptibility 93%
10 to 20-fold reduced susceptibility 78%
20 to 40-fold reduced susceptibility 67%
>40-fold reduced susceptibility 50%
Baseline ABT-378 mutation score
0 – 5 96%
6 – 7 76%
8 – 10 33%

Using a virological response rate (<400 copies/mL) of 50% at 24 weeks the investigators concluded that ABT-378/r may be expected to achieve this response rate in patients with phenotypic resistance to ABT-378 at baseline of up to 40-fold reduced susceptibility and/or containing up to 7 of the 11 mutations making up the ABT-378 mutation score. They hypothesise that these response rates are likely due to the high plasma ABT-378 levels relative to the EC50 for wild-type HIV (in excess of 75-fold when dosed at 400/100 mg BID).


The role of these agents still needs to be further defined. Their utility and activity in extensively treatment experienced patients is crucial and studies combining a number of these agents and others (such as entry inhibitors) should proceed rapidly. Availability to patients in these true salvage situations is hindered by lack of compassionate access.

It remains unclear to what extent ABT-378/r contributes to viral control in experienced patients.

Again data from use in those patients with extensive experience of both PIs and NNRTIs as well as multiple NAs is needed to guide clinical choice now that ABT-378/r has limited availability for compassionate use.


  1. Kempf D, Isaacson J, King M et al. Genotypic correlates of reduced in vitro susceptibility to ABT-378 in HIV isolates from patients failing protease inhibitor therapy. 4th International Workshop on HIV Drug Resistance & Treatment Strategies. 12-16 June 2000, Sitges, Spain. Abstract 38.
  2. Kempf D, Brun S, Rode R et al. Identification of clinically relevant phenotypic and genotypic breakpoints for ABT-378/r in multiple PI-experienced, NNRTI-naive patients. 4th International Workshop on HIV Drug Resistance & Treatment Strategies. 12-16 June 2000, Sitges, Spain. Abstract 89.

Links to other websites are current at date of posting but not maintained.