Report # 6 from Glasgow, Scotland...last report before returning to the Big Apple.

Glasgow Highlights, Report # 6: last report before returning to the Big Apple

From Jules Levin, NATAP

Thursday, November 12, 5:30pm

CHEESE Study: Indinavir vs Saquinavir SGC

For the first time publicly, preliminary 40 & 48 week data was presented on the CHEESE STUDY. This study is a 48-week randomized open-label comparison between Saquinavir SGC 1200 mg TID +AZT 200 mg TID + 3TC 150 mg BID vs Indinavir 800 mg TID + AZT 200 mg TID + 3TC 150 mg BID. 70 participants were antiretroviral naïve but were permitted <12 months AZT experience. Mean baseline CD4 was 300 in each arm, and HIV RNA was 100,000 copies/ml.

In the final analysis at week 24 using an intent-to-treat analysis JWT Cohen Stuart reported that 82% in the saquinavir group had <50 copies/ml, and 75% in the indinavir group had <50 copies/ml. Using an on-treatment analysis of the observed data at week 40 83% in the saquinavir group and 80% in the indinavir group had <50 copies/ml. Stuart said they did not have enough data yet to present 48 weeks. Using an intent-to-treat analysis 70% in both groups had <50 copies/ml.

There were 5 early discontinuations in the SQV SGC arm: 1-nausea; 1-pancreatitis; 1-death; 2-noncompliance. And 7 in the IDV arm: 1-AZT intolerance; 1-impaired kidney function; 2-deaths; 1-recd wrong medication; 1-grade 4 anemia; 1-noncompliance.

Following are the frequency of adverse events which did not lead to discontinuation, and the rate is about similar in both arms. In the saquinavir arm (n=35): 11-gastrointestinal complaints; 1-headache; 1-fatigue. In the IDV arm (n=35): 6-gastrointestinal complaints; 2-headache; 1-arthralgia/myagia; 1-fatigue.

There were 5 virological failures in the 2 arms together: 4 in the saquinavir arm and 1 in the IDV arm. Failure was defined as HIV RNA levels >400 copies/ml at two or more consecutive study visits. Three of the four SQV virological failures were found to have low SQV trough levels although they were compliant with taking their medications. One person never achieved undetectable viral load and it is believed it's because their baseline viral load was over 6 million copies/ml. The one person in the IDV arm who had virological failure was not compliant with taking their medication.

Although HIV-brain experts generally say PIs cannot penetrate the brain, the incidence of HIV related brain dysfunction has declined in the PI era. Still, it's been reported that indinavir penetrates the CSF. In a study presented at Glasgow it was reported that drug concentrations of saquinavir in the CSF were low or not detected, I forget which. HIV causes immune deficiency as well as immune dysfunction. One explanation for the decreased incidence of HIV related brain dysfunction is that immune dysfunction caused by HIV may be becoming normalized. The brain is affected by cytokines which are part of the immune system. HIV appears to cause disregulation of cytokines and other parts of the immune system.

New Key Data Reported on Efaviranz+AZT/3TC

Although 36 week data was reported at ICAAC for the DuPont 006 study of efaviranz (EFV) plus AZT/3TC, they did not present the percentage of patients below 50 copies/ml who had a baseline viral load >100,000 copies. At ICAAC they reported overall in the study 88% taking EFV+AZT/3TC had <50 copies/ml using an observed data analysis. Using the most stringent intent-to-treat analysis (non completer=failure) 64% had less than 50 copies/ml at week 36. They also reported that whether one had viral load at baseline above or below 100,000 copies/ml the percentage below <400 copies/ml was about the same.

For the first time Schlomo Staszewski of Frankfurt Germany, reported at Glasgow that the percent below 50 copies/ml was the same whether a person had above or below 100,000 copies/ml at baseline, using either the observed data analysis (88%) or the most stringent intent-to-treat (NC=F) analysis (64%).

What is Treatment Failure? and When To Switch a Person off a Regimen When Viral Load is Detectable

There was a good deal of discussion about this subject at Glasgow. Several researchers discussed how a good percentage of their patients have had viral load rebound from undetectable to detectable but the person's CD4s remained elevated. As well, they reported that upon rebound some patient's viral load appears to stabilize and remain at a relatively low level such as 1,000 or 5,000 copies/ml. If these patients have had extensive pretreatment the doctors reported preference for leaving them on their current regimen. They reported that in some instances when they discontinued treatment viral load would increase suggesting that although their viral load was detectable the drugs were holding their viral load at a lower level. One researcher said that if he cannot compose a regimen that will protect the drugs on board he would prefer to leave the person on their holding regimen. For example, if a person has not used an NNRTI yet but has developed resistance to a few PIs and several NRTIs, there may not be enough drugs to combine with the NNRTI to create a regimen that would offer durable viral load suppression. Their reasoning is if you were to give such a person an NNRTI with two drugs to which they would be resistant, resistance would develop quickly to the NNRTI and the person would lose that class of drugs.

However, one cannot generalize because treatment decision making must be based upon an individual's situation. For example, a person's viral load may not stabilize and remain at a low level after becoming detectable. It may skyrocket. In which case remaining on the same regimen may not be a reasonable option for that person. In some cases doctors and patients are choosing to take drug holidays.

It is important to realize that by remaining on a PI regimen while replication is ongoing a person may be accumulating PI mutations. If so, they may not be able to respond to any of the PIs in development which may have a somewhat different resistance profile such as: ABT-378, amprenavir, BMS-232632, or tipranavir. To preserve the ability to respond to these new PIs it may be important not to allow mutations to accumulate by remaining on a PI while viral load is detectable.

If a person is taking their first PI regimen (meaning they should have treatment options) and was undetectable, but is becoming detectable it may be crucial to switch that person to another PI immediately. Several studies have shown data that switching the PI while viral load is low offers the best opportunity to respond well to the 2nd PI. In order to detect rising viral load one must monitor viral load frequently, as often as every 4-6 weeks. If viral load becomes detectable upon a 2nd confirmation test changing the PI regimen may be the only opportunity to prevent resistance from building up due to staying on a failing PI. I reported in a previous highlight report that at the Glasgow meeting Rob Murphy presented 60 week data from ACTG 347 in which individuals who failed amprenavir were immediately switched to indinavir+nevirapine+2 NRTIs. After 60 weeks 80% remained undetectable. This positive result may have occurred for two reasons: the immediate switch while viral load was low; and, possibly amprenavir is not completely cross-resistant to indinair.

Read NATAP web site for follow-up reports on Glasgow