An increasingly fierce debate over the merits of ABT-378/r, the new protease inhibitor from Abbott Laboratories, has dominated the Fifth International Congress on Drug Therapy in HIV Infection this week in Glasgow.
ABT-378/r, or lopinavir/ritonavir, has already been licensed in the US under the trade name Kaletra, but awaits the green light from the European drug licensing body, the EMEA.
Dr Margaret Johnson of the Royal Free Hospital and University College Medical School, London, presented 48 week follow-up of a head to head study of ABT-378/r or nelfinavir combined with d4T and 3TC. Data from this study was presented at the 40th ICAAC in Toronto last month, and shows that viral load suppression in the ABT-378/r arm was significantly better than in the nelfinavir arm (67% vs 52% below 50 copies at week 48, by intent to treat analysis). Those who experienced viral load rebound on ABT-378/r did not show any evidence of protease inhibitor resistance mutations, suggesting that the reason for treatment failure in many cases may have been non-adherence or the development of resistance to 3TC.
The absence of resistance to ABT-378/r is argued to be related to the unusually high blood levels achieved with this drug – a trough level roughly 70 times higher than the amount necessary to reduce viral replication by 50%.
In contrast, 20 individuals who failed nelfinavir therapy had nelfinavir-associated resistance mutations.
However Dr Julio Montaner of the University of British Columbia in Vancouver warned that these results should be interpreted with caution.
“It’s too early to tell what the resistance issues are with this drug. We need to accumulate data from hundreds of naïve patients failing Kaletra in order to understand resistance.”
At a satellite meeting organised by Abbott Laboratories, Dale Kempf of Abbott argued that the extremely high levels of ABT-378/r present in the blood posed a much greater barrier to the development of resistance than other drugs.
However, Roche researchers have contested this claim after experiments using lopinavir synthesised in their own labs.
They found that Abbott’s own estimates of the amount of drug which becomes bound to alpha-1 acid glycoprotein were too low, and that Abbott may not have used the best method to calculate the extent to which free drug is mopped up by proteins in the blood, thus preventing it from exerting an antiretroviral effect at the concentrations hoped for.
Roche has calculated that if ABT-378 levels are calculated with a provision for the particularly high levels of alpha-1 acid glycoprotein that may be present in the blood of HIV-positive people, the level of ABT-378 needed to inhibit HIV replication by 50% may be 6.5 times higher than previously published.
Roche is now calling on all protease inhibitor manufacturers to get together to agree a set of standards for presenting information on the pharmacokinetic profiles of their products, in order to allow doctors and patients to make accurate comparisons of the potency of the various PI combinations.
However, Dale Kempf of Abbott told aidsmap that he believes Roche’s research is not borne out by evidence from clinical trials.
“We see no difference between protein binding in HIV-positive people and healthy volunteers. Roche’s experiment in itself is artificial because they only used alpha-1 acid glycoprotein”.
He is also dismissive of Roche’s criticism that the IQ, or inhibitory quotient of ABT-378/r may be much lower than Abbott has calculated.
The inhibitory quotient is defined by Abbott as the ratio of the trough concentration of a drug to the human serum-adjusted EC50. The higher the IQ, the more likely a drug will be to overcome any resistance which develops. An IQ of 1 or less would mean that a drug's trough level could easily fall below the level needed to control viral replication.
“The attack on the IQ is based on false assumptions. The model has to correlate with clinical results and ours does. One way to give yourself confidence in your model is to relate it to what you see clinically.”
On the final day of the Glasgow congress, Ann Hsu of Abbott reported on the relationship between the IQ/trough concentration ratio of ABT-378/r and the likelihood of virological suppression in 56 patients with known reductions in susceptibility to ABT-378/r as measured by phenotypic resistance testing.
All individuals with an IQ greater than 15 (n=21) achieved viral load below 400 copies at week 24, while 90% with an IQ of 4-15 (n=15) and 80% of those with an IQ below 4 (n=16) achieved VL below 400 copies (p=0.026).
“If our model is wrong, why are we seeing these results?”, asks Dale Kempf.
Andrew Hill argues that far from being a vindication of Abbott’s arguments, the study shows a median IQ of 9.9, vastly lower than the IQ stated for wild-type HIV (>75, according to Abbott data), and suggestive of the probability that well characterised reductions in susceptibility to other PIs may also cause much greater reductions in susceptibility to ABT-378/r than hitherto expected.
He also points to findings included in the US data sheet of ABT-378/r, which suggest a huge variation between individuals in the plasma levels of ABT-378/r. Mean trough concentrations of ABT-378/r were found to lie within the range of 1.5 ug/mL to 9.5 ug/mL(the 95% confidence intervals), with a mean of 5.5 ug/mL
Dale Kempf meanwhile remains confident that Abbott’s model is sound, and promises more data by the time of next February’s Retroviruses conference to help understand the differences between ABT-378/r and the other boosted PIs.
Andrew Hill told aidsmap: “I’m happy to be proved wrong, but let’s do comparative trials – it’s the only way to settle the argument.”
An independent comparative study called MaxCmin II will compare saquanavir/ritonavir with lopinavir/ritonavir, and will begin recruiting shortly in the USA. In Europe recruitment to the study will begin when lopinavir/ritonavir is licensed in Europe, in the first half of 2001. It is likely that some 48 week data will be available by the time of the 2001 ICAAC conference next September.
First line use of ABT-378/r still debated
Although ABT-378/r is already available on a named patient basis in the UK for salvage therapy in PI-experienced patients, it is not yet licensed for first line treatment. Much of the activity at the Glasgow conference by all PI manufacturers was unashamedly focussed on dealing with the threat of a potent new product arriving on the market within the next few months.
Roche, the company most directly threatened, has been keen to highlight the possibility of sequencing other PIs after nelfinavir, due to the drug’s distinct resistance profile, and evidence from clinical studies in which higher success rates have been seen with second-line PI treatment after failing nelfinavir compared with other single PIs.
Merck Sharp and Dohme appears less perturbed by the arrival of ABT-378/r, possibly because the indinavir/ritonavir combination has not been found wanting in a head to head comparison (unlike nelfinavir). Merck’s more relaxed attitude is also connected to the fact that its own labs have tested indinavir/ritonavir plasma levels and found a Cmin for indinavir up to 79 times higher than the IC95 of the assay used by the company (which equates with the IC50 levels calculated by Abbott, according to Dale Kempf).
One UK doctor told aidsmap “I don’t think that Merck and Abbott fundamentally disagree on methodology in this debate”.
There is also a keener appreciation of the potential weakness of ABT-378/r as a second or third line protease inhibitor, given the results cited above in which 20% of those with an IQ below 4 failed to achieve a viral suppression below 400 copies despite concurrent use of efavirenz, to which they were naïve. Taking these results into account, would ABT-378/r perform any better than other boosted PIs in patients with experience of all three classes of drug?
“It’s better to use ABT-378/r as first line therapy” says Dale Kempf of Abbott – our research shows that the IQ is much greater and so the risk of resistance is much less. So far, of 20 failures, we have seen no primary mutations, whereas in PI-experienced failures on ABT-378/r we see the stepwise accumulation of further mutations.”
Charting a potential future role for ABT-378/r in salvage therapy yesterday, Dr Julio Montaner told a symposium sponsored by Boehringer Ingelheim that he would be interested to see the testing of ABT-378/r and tipranavir as a combination for multi- PI-experienced patients, on the grounds that the ritonavir component of ABT-378/r would also provide a 1200% boost to tipranavir levels (although it’s not known how such massive levels of tipranavir might induce lopinavir metabolism, thus reducing levels of that drug).
Craig C et al. Antiviral efficacy and effects of protein binding on HIV proteinase inhibitors (PR) inhibitors (PI). Fifth International Congress on Drug Therapy in HIV Infection, Glasgow, abstract P290, 2000.
Hill A et al. Prediction of drug potency from Cmin/IC50 ratio: false precision? Fifth International Congress on Drug Therapy in HIV Infection, Glasgow, abstract 233, 2000.
Hsu A et al. The Ctrough inhibitory quotient predicts virologic response to ABT-378/r therapy in treatment-experienced patients. Fifth International Congress on Drug Therapy in HIV Infection, Glasgow, abstract PL9.4, 2000.
Johnson M et al. ABT-378/ritonavir versus nelfinavir in antiretroviral-naïve subjects: week 48 comparison in a phase III blinded randomised clinical trial. Fifth International Congress on Drug Therapy in HIV Infection, Glasgow, abstract PL6.5, 2000.