By Harvey S. Bartnof, MD, Staff Physician at AVERI, AIDS Virus Education and Research Institute in San Francisco, California

PART 6: Lopinavir Reports
      Triple PI Regimens (see Part 1)
      Absorption Increased and Variability of Blood Levels Decreased with Food
      Blood Concentrations are Associated with Body Weight
      "Inhibitory Quotient" Correlates with Viral Undetectability

There were several presentations at the Workshop about lopinavir (co-formulated with PI [protease inhibitor] drug ritonavir as Kaletra). (Note: see Part 1 for reports about triple PI combinations with lopinavir/ritonavir/amprenavir.)

Lopinavir absorption is increased when taken with food in HIV positive patients, as has been reported for healthy volunteers, according to Dr. Robert Bertz of Abbott. The trough (pre-dose) concentration increased by 30%, while the minimum (lowest) concentration increased by 44%, when compared to historical information without regards to food. Also, the variability of those levels was reduced by 50% by co-administration with food. Yet, the "AUC" (area-under-the-curve or total exposure) concentration increased by 12%. (Most studies to date from Abbott have correlated anti-HIV efficacy with lopinavir minimum levels, not AUC.) The meal was a low-fat breakfast, with a muffin, banana, low-fat milk and juice. Nineteen patients were included in the study, with 32% women, 47% African American, and 26% Hispanic. The mean inhibitory quotient (IQ, trough concentration ÷ protein-adjusted EC50 [effective inhibitory concentration to block 50% of wild-type {non-mutated} HIV growth]) for the 19 patients was 102, with all having an IQ greater than 40. (A higher IQ represents greater anti-HIV potency.) Dr. Bertz concluded, "Kaletra should be dosed with food to maximize bioavailability and decrease variability." Also, "If low lopinavir trough levels are observed during therapeutic drug monitoring, confirm that [the] patient is taking Kaletra with food."

Reference
Bertz R and others. Steady-state pharmacokinetics of Kaletra (lopinavir/ritonavir 400/100 mg BID) in HIV-infected subjects when taken with food. Abstract and poster presentation 3.10 at the 2nd International Workshop of Clinical Pharmacology of HIV Therapy; April 2-4, 2001; Noordwijk, the Netherlands.

Body Weight Affects Blood Levels of Lopinavir in Healthy Volunteers
Researchers from Abbott attempted to determine whether race-ethnicity, gender (sex) and age had any effects on the blood levels of lopinavir (LPV). The lead author was Dr. Robert Bertz. A total of 194 healthy (HIV negative) patients from Abbott studies included 26% women, 10% Blacks, 9% Hispanics. The mean weight was 75.7 kilograms (167 pounds) with a mean age of 32 years. Volunteers were given one dose of Kaletra. The results showed that patients with a lower body weight tended to have higher lopinavir blood levels, including maximum concentration and total exposure (AUC, area-under-the-curve concentration). For a 25% decrease in body weight, the AUC would be expected to increase by 20%. Neither age nor gender (sex) had any significant association with the AUC or maximum concentration of LPV. Minimum concentration was not reported. The "half-life" (time until an original amount is reduced by half) was calculated not to be significantly affected by body weight.

Interestingly, Blacks had a lower AUC concentration and lower maximal blood levels of LPV than Caucasians. The AUC concentration was 14% lower among Blacks (p=0.053, borderline statistical significance), while the maximum concentration was significantly lower (14%, p=0.02) than Caucasians. (It is possible for a 14% decrease for both measurements to have marginal significance for one and be significant for the other.) However, Dr. Bertz said that the differences "are probably therapeutically inconsequential." The reported reason was that an equal percentage (91-92%) of 67 Blacks and 160 Caucasians achieved viral undetectability (limit 400 copies/mL) in the phase III Study M98-863 at 24 weeks, using an on-treatment analysis. Obviously, those patients were HIV positive. However, drug levels from patients in that study were not reported. Analyses in the current report for Hispanics were not presented, probably due to a lower percentage. Interestingly, minimum blood levels of saquinavir when co-administered once daily with ritonavir were non-significantly lower among Blacks than Caucasians (HIV positive, see part 4). However, AUC and maximum levels were not reported.

The above analysis should be undertaken in HIV positive patients, including several race-ethnicities and at "steady-state" (not just one dose) to confirm the findings. Current lopinavir dosing recommendations do not incorporate body weight. If the above findings are confirmed in HIV positive patients, that would add "weight" to the potential benefit of therapeutic drug monitoring, particularly when a patient weighs significantly more or less than average.

Reference
Bertz R and others. Effects of gender, race, age and weight on the pharmacokinetics of lopinavir after single-dose Kaletra (lopinavir/ritonavir) in healthy adults. Abstract and poster presentation 3.11 at the 2nd International Workshop of Clinical Pharmacology of HIV Therapy; April 2-4, 2001; Noordwijk, the Netherlands.

Lopinavir "Inhibitory Quotient" Associated with Viral Undetectability
There were several presentations at the Workshop that addressed the "Inhibitory Quotient" (IQ), which represents the relationship between the minimum blood level of drug measured in the patient and the degree of drug resistance of that patient's HIV isolate, usually the "IC50" (concentration to inhibit 50% of HIV growth in the laboratory), "IC95" or "EC50" (effective inhibitory concentration in the patient). In the current study, the IC50 was expressed as a "fold-change" or phenotype resistance of the patient's HIV isolate, when compared to "wild-type" (no resistance) HIV, and was corrected for "protein-binding." The higher the IQ, the more potent the drug is against the patient's dominant HIV "strain."

Dale Kempf, PhD of Abbott expanded his work that he presented at the recent 8th Conference on Retroviruses and Opportunistic Infections last February in Chicago. He showed that in PI (protease inhibitor) drug-experienced patients, the IQ was a significant predictor of viral load response. In Study 957, 57 multiple PI drug-experienced, NNRTI (non-nucleoside drug)-naïve (never took NNRTI drugs) patients were enrolled and took Kaletra (lopinavir/ritonavir) plus efavirenz (Sustiva, NNRTI drug) and NRTI drugs. The median IQ was calculated to be 9.9, ranging from 0.05 to 279. The results showed that IQ was significantly associated with virologic outcome after 24 weeks. For those with an IQ greater than 15, 94% achieved viral undetectability (limit 50 copies/mL). Whereas, for those with an IQ less than 4, only 48% achieved undetectability.

In a similar analysis for Study 985 with ritonavir (RTV, Norvir) "intensification" (added) to an indinavir (IDV, Crixivan, PI drug) mono-PI drug combination regimen, a "virtual phenotype" was substituted for IC50 when computing what Dr. Kempf called a "virtual IQ." (The virtual phenotype from Virco uses a database of more than 15,000 patient samples with both genotype and phenotype resistance results to predict a phenotype from a given genotype.) In the intensification study, IDV and RTV were each taken 400 mg twice daily with NRTI drugs, after standard IDV dosing with 2 NRTI drugs was associated with a detectable viral load or "failure." The results showed a significant association between the virtual IQ (vIQ) and viral load response after 24 and 48 weeks. For those with a vIQ greater than 2, the 48-week response was 62% (limit 50 copies/mL or 0.5 log decrease in HIV RNA). Whereas, for those with a vIQ less than 2, the 48-week response rate was zero. Interestingly, the vIQ was a better predictor of virologic response than the IDV virtual phenotype, trough levels of IDV or the RTV vIQ.

Dr. Kempf concluded, "The use of inhibitory quotients to assess the antiretroviral activity of protease inhibitors in vivo merits additional study." The take-home message is that the inhibitory quotient uses both "pharmacologic" (drug level in patients) information and drug resistance information to help determine the likelihood of achieving an undetectable viral load. While much more research is needed, it appears thus far that the IQ provides more information than either of the two components alone that are used to determine the IQ.

Reference
Kempf K and others. Evaluation of the inhibitory quotient as a pharmacodynamic predictor of the virologic response to protease inhibitor therapy. Abstract and oral presentation 7.3 at the 2nd International Workshop of Clinical Pharmacology of HIV Therapy; April 2-4, 2001; Noordwijk, the Netherlands.