Evaluating Susceptibility to ABT-378 (Kaletra)

Scott Brun, MD, with Abbott Labs, presented an oral slide discussion called  "Analysis of Viral Isolates following Viral Load Rebound on Therapy with  ABT-378/r". This talk discussed data that was originally reported at the  Resistance Workshop this Summer in Sitges, Spain. The talk was a review of observations of changes in genotype and phenotype (or lack thereof) during viral rebound on ABT-378/r therapy. The observations were interpreted in the context of a concept from Abbott called the ABT-378 Mutation Score. It is a proposed way to evaluate a person's ability to respond or their susceptibility to ABT-378 based on their genotype prior to beginning therapy with ABT-378. At the end of this report is a model Abbott used to test how useful the ABT-378 Mutation Score is in evaluating susceptibility to ABT-378. The reports on the presentations at Sitges presented by Dale Kempf and Eugene Sun from Abbott can be read on the NATAP web site in the Conference Summaries from the Resistance Workshop (www.natap.org). Later in this article is information excerpted from the NATAP Sitges Summaries, where Kempf and Sun discuss their observations in using resistance testing in evaluating susceptibility to ABT-378.

Kaletra is a co-formulation in one capsule of 400 mg of ABT-378 also called lopinavir with 100 mg of ritonavir. The addition of ritonavir enhances the blood levels of lopinavir, increases its oral bioavailability, and achieves high plasma concentrations of lopinavir. Phase 2 and 3 studies show ABT-378 has potent anti-viral activity in naŒve and PI experienced individuals. A question is how does prior PI use effect the benefit or anti-viral activity of ABT-378? What is the impact of varying degrees of prior PI resistance on a person's response to Kaletra? Brun's talk as well as the previous presentations at the Resistance Workshop & Durban address this question.

Brun reported on an analysis of three phase 2 studies: M97-720 (100 treatment-naive; 96 weeks; d4T+3TC); M97-765 (70 single-PI experienced, NNRTI-naive; 72 weeks; nevirapine+NRTIs); M98-957 (57 multiple PI experienced, NNRTI-naive; 24 weeks; efavirenz+NRTIs). In the naive study at week 96 using the less stringent on-treatment analysis, 97% had <400 copies/ml and 92% had <50 copies/ml; using a more stringent ITT analysis 83% had <400 copies/ml and 78% had <50 copies/ml. Individuals with >100,000 copies/ml appeared to perform better than individuals with <100,000 copies/ml (72%  40/55 vs 88% 40/45 had <400 copies/ml). And, in the multiple PI study 63% had <50 copies/ml at week 24 using the ITT analysis.

The objectives objective of this analysis were was to assess the impact of baseline

genotype on the response to therapy, and the development of resistance at rebound. The study is limited as isolates from viral rebounding patients were limited to 23 patients out of 227 enrolled in the phase 2 studies. Patients included were those who failed to respond to ABT-378/r and those with a sustained rebound in viral load to >400 copies/ml while on ABT-378 therapy. Baseline phenotype resistance was evaluated by Virologic's Phenosense test (M97-720 & M98-957) and Virco's Antivirogram test (M97-765). Baseline genotyping was evaluated by population sequencing.

Kempf reported previously on 11 genotypic mutations observed in patients who had prior PI experience before they started ABT-378 therapy and was associated with baseline phenotypic resistance to ABT-378 in vitro: 82, 84, 24, 10, 20, 63, 71, 90, 53, 54, 46. Brun reported that this list of mutations was arrived at by statistical analysis of over 100 viral isolates from patients who failed other protease inhibitors, but had not yet received ABT-378. See the NATAP report of Kempf's talk at Sitges for a more detailed discussion about the mutations observed. At Sitges, Kempf said that these findings do not eliminate the possibility that there may be other  unidentified mutations associated with ABT-378 resistance, or that there is some unique resistance profile or pathway leading to ABT-378 resistance. Abbott has reported that they have not been able yet find ABT-378 resistance from treatment-naive studies in individuals with virologic failure. So, to this point there is no resistance profile associated with ABT-378 resistance. Upon failure in these studies resistance to 3TC has been noted.

At ICAAC, Brun reported Abbott found 3 distinct patterns during viral rebound

or failure to respond to ABT-378. Brun said these patterns could be differentiated by the baseline ABT-378 Mutation Score. Group A consisted of 10 patients (4 from naive studies and 6 from the single PI experienced study) who had 0-2 genotypic mutations (from the list of 11). No additional protease mutations were observed after viral rebound and no phenotypic resistance to ABT-378 was seen. Patients entered study with essentially no ABT-378 resistance and still had no additional resistance at viral rebound.

Of these patients, 7/10 developed genotypic or phenotypic resistance to either 3TC (naÔve patients) or nevirapine (PI-experienced patients). Of the 3 patients who did not develop resistance to either 3TC or nevirapine, two patients experienced a temporary increase in viral load with subsequent resuppression to < 400 copies/mL and the third patient had documented noncompliance.

At the other end of the spectrum was Group C with 8 patients from the multiple PI experienced study. They had 6-9 of the 11 mutations associated with ABT-378 reduced susceptibility at baseline, and all had failed at least two prior PI regimens and were more heavily treatment experienced. In addition, these patients entered study already with reduced phenotypic susceptibility to ABT-378, as they all had 10-fold or greater phenotypic resistance at baseline. At viral rebound little or no evolution in protease resistance was seen although efavirenz resistance evolved. This is presumably because substantial resistance to ABT-378 was already present before therapy, and the virus needed primarily to mutate to become resistant to efavirenz (these patients were all NNRTI-naive).

Brun said Group B was the most interesting. The 5 patients in this group were from either the single or multiple PI groups (4 single and 1 multiple PI experienced) and had 4-5 mutations at baseline. This was the only group who at viral rebound they saw the further

accumulation of protease mutations, and subsequently the development of ABT-378 phenotypic resistance. After viral rebound, the range of phenotypic resistance to ABT-378 was from 9 to 99-fold relative to wild-type virus. Nevirapine and efavirenz phenotypic resistance also evolved after viral rebound. In this group, after viral rebound an additional 1-4 protease mutations accumulated as well as "polymorphisms" of unknown significance.

Brun discussed a patient from the 957 multiple PI failure study who had prior PI experience with ritonavir, indinavir and saquinavir. At baseline the person had 2.8 fold ABT-378 resistance and was sensitive to efavirenz. At day 112 the person had 99-fold ABT-378 resistance and efavirenz resistance (>333-fold). And, 2 new PI mutations emerged (V82A, A71V). This highlights to me the importance of using enough drugs in a new regimen to which the patient is adequately sensitive, so full viral suppression occurs and viral rebound does not occur. You don't want to lose new drugs. In this case because

adequate viral suppression did not occur, the patient lost two key  drugs--ABT-378 and efavirenz.

Brun described 4 patients admitting it is a small data set but he wanted to illustrate some points. Two of the 4 had prior experience with indinavir or nelfinavir but not with saquinavir or any other PI, and were sensitive to saquinavir at baseline. They remained sensitive to saquinavir although developing ABT-378 resistance (>30-fold). After day 140 in one patient and day 500 in the other patient the 2 patients were still sensitive to saquinavir (<4-fold compared to wild-type). A third patient had prior saquinavir experience and resistance at baseline, and was equally resistant to saquinavir at day 340.

The fourth patient had prior saquinavir, indinavir and ritonavir experience, but at baseline there was no saquinavir resistance. However, at day 153 the person had almost 60-fold saquinavir resistance. All 4 patients remained <8-9 fold resistant to amprenavir and were fully sensitive to tipranavir. Brun said that Abbott will be studying the use of these 3 protease inhibitors for individuals who fail ABT-378 in studies.

Brun offered an explanation for these observations regarding the development of resistance in groups A, B and C. He said they can be explained in terms of the "genetic barrier" to ABT-378 resistance which involves the interplay of genotype, phenotype, and drug levels in the patient. As mutations accumulate there is a reduction in phenotypic susceptibility. This phenotypic reduction in susceptibility does not become clinically relevant until drug levels in the patient are no longer able to keep their virus suppressed. In theory, since ABT-378 achieves high blood levels, you need the accumulation of a significant number of mutations to develop phenotypic loss of susceptibility to ABT-378 and clinically relevant resistance to develop. Patients entering studies with low numbers of mutations have to accrue a large number of mutations before clinically significant resistance occurs. In Group C, individuals already had high numbers of mutations and phenotypically reduced susceptibility to ABT-378. Since the genetic barrier was already compromised, there is not much more pressure for more mutations to accumulate

In other words, as significant resistance is essentially present prior to initiation of ABT-378/r therapy, selective pressure for the accumulation of additional protease mutations is low compared to that for efavirenz, and initial evolution of further resistance to ABT-378 is not observed

In the intermediate Group B, they already had some mutations and only needed to accumulate a few more, and so there was high selective pressure for more protease evolution to occur.  With high selective pressure and a low genetic barrier, resistance to ABT-378 evolves during rebound

The introduction of the ABT-378 Mutation Score appears to be a natural evolution based on developments in the understanding of resistance and resistance testing. But using this information gives one a preliminary context in which to evaluate a person's ability to respond to ABT-378. The data and theories are based on early results and need further exploration. Abbott feels the reason its difficult to develop resistance to ABT-378 is because of the high blood levels they are able to achieve of lopinavir by adding 100 mg

of ritonavir. So, in theory if you can keep viral replication low or keep viral load <50 copies/ml resistance to ABT-378 should not develop easily and should keep a person undetectable. Of course, this means that in general you need to include in a new regimen with ABT-378 enough other drugs to which the patient is sensitive.

ICAAC UPDATES of DATA REPORTED AT RESISTANCE WORKSHOP IN SITGES, SPAIN (June 2000)

NATAP Report: Identification of Clinically Relevant Phenotypic and Genotypic Breakpoints for ABT-378/r in Multiple PI-experienced , NNRTI-naÔve Patients (this article has many interesting details including Kempf's findings on the effect of specific mutations)

Link to article:

http://www.natap.org/june_2000/spain_resistance/Spain_rpt_1_abt378_61500.htm

At ICAAC Abbott added new data on <50 copies, where previously at Sitges only <400 copy data was reported (see below).

Virologic Response at Week 24 with Respect to Baseline Genotype & Phenotype (Multipke Pi-Experienced Study 957- patients were NNRTI naÔve and also added efavirenz)

Virologic Response by Number of ABT-378 Mutations at Baseline-

Virologic Response Week 24 by Phenotypic Resistance at Baseline

It was generally agreed among the researchers at Sitges that around 7 mutations or somewhere just below 40 fold may be the cut-off where reduction in viral load is caused by efavirenz and ABT-378 was likely not having much if any benefit.

At Sitges and again at ICAAC Abbott presented-

Model Suggesting Utility of ABT-378 Mutation Score; Parameters Associated With Virologic Response

Three stepwise regression models were performed to assess the effect of baseline phenotype and genotype in the context of other parameters that might impact virologic response. Baseline phenotype (p=0.016), genotype (p=0.024) and inhibitory quotient (trough/EC50, p=0.068) were associated indenpently to be predictive of  virologic response.

In the first model, baseline phenotypic susceptibility and genotype at baseline were considered and phenotypic susceptibility remained the most closely associated with the virologic response at week 24. In model 2, in which the baseline phenotypic susceptibility was omitted, the ABT-378 mutation score remained a significant predictor of virologic response at week 24. I the third model, ABT-378 pharmacokineticparameters were considered along with the other parameters considered in models 1 and 2. Baseline phenotype was incorporated as part of 4 exploratory inhibitory quotients (IQ) representing the ratio of Ctrough, Cmin, AUC and Cmax to the serum-adjusted EC50 for each baseline isolate (IQCtrough, IQCmin, IQAUC and IQCmax, respectively). Baseline genotype entered model 3 first but was ultimately displaced by the IQC trough which remained associated with week 24 virologic response. Other parameters considered which appeared to trend to association with being predictive of response were: NRTI susceptibility , number of new NRTIs, years since HIV diagnosis, baseline patient weight.

Sun concluded that these results provide a framework for the interpretation of HIV phenotypic and genotypic testing.