Many abstracts focused on evaluating new variations of combination therapy, usually designed to improve convenience, or to compare different drugs as part of the "recipe." Of real importance are studies of ways to make regimens easier to take, such as twice daily protease therapy, or once daily NNRTIs. One of the themes in the hallways at this meeting was the need for researchers and companies to be very clear in how they analyze studies, and to use conservative "intent to treat" methods as often as possible. You will notice that this gives lower estimates of the proportion with undetectable viral load than you are used to hearing about -- but numbers which are much closer to what we usually see in "real life."

Poster Session 12224: Longterm comparison of BID and TID dosing VIRACEPT (Nelfinavir) in combination with stavudine (d4T) and Lamivudine (3TC) in HIV patients

This is followup of a study funded by Agouron in Europe to compare d4T (ZERIT) and 3TC (EPIVIR) with Nelfinavir (VIRACEPT) given either as 1250 mg (5 pills) twice a day, or as the original dose of 750 mg three times daily. Actually, the study started with 3 different twice daily doses, but all of the twice daily patients were changed to 1250 mg early in the study. All patients were naïve to d4T or 3TC and protease inhibitors. 211 patients received twice daily therapy, and 76 received the three times a day ("tid") dosing. The median viral load was quite high: 5.0 log or 100,000 copies and the mean CD4 count was about 260. Bottom line: Nelfinavir 1250 twice daily was at least as effective as three times daily. With about half the patients followed to 48 weeks, the standard analysis showed 80% in both groups remained less than 400 copies and 65% were less than 50 copies in both groups. By the conservative (and more correct) intention to treat method, about 75% and 55% were below 400 and 50 copies. Drug levels at trough were about equal, and peak levels higher in the twice a day group. Side efffects were similar, with diarrhea the major problem. This is a well designed equivalence study, now with reasonable followup data, which makes a pretty strong case for twice a day nelfinavir.

Poster Session 12305: Women First: A study of the effects of HIV positive women with combination Nelfinavir (NFV), Saquinavir (SQV), Stavudine (d4T) and Lamivudine (3TC)

I liked this study for several reasons. It used an interesting 4 drug combination, it helped to confirm that Nelfinavir can be used twice a day, and the investigators (all women) decided to do a women only study. It might seem exclusionary, but remember that about 25% of people with HIV are women in the US but usually only 10% of study subjects. Fifty five women, naïve to protease inhibitors and to either d4T or 3TC, were randomized to d4T, 3TC, nelfinavir and saquinavir (hard gel, or the old invirase) give either as 1250 mg twice a day of nelfinavir and 1000 mg twice daily of saquinavir HGC, or conventional three times a day dosing. Pharmacokinetic studies were included, which helped confirm that handling of these drugs was similar in women. Baseline viral load was about 4.8 log (70,000). The quadruple therapy was well tolerated and potent. At 24 weeks, about 75% of patients were below 400 copies and 62-63% were also below 50 copies. There was no difference between the all twice daily regimen and the more complex regimen. Combined with the Spice study of nelfinavir and saquinavir soft gel (FORTIVASE), it appears that these two protease inhibitors combined with two nukes is a very potent 4 drug combo. Interestingly, diarrhea was significantly less common in the Women First study than in the pivotal 511 trial of nelfinavir, which included a majority of gay men. This may be one gender-specific difference. We may need to design more studies which reach directly out to groups that are often not trial volunteers.

Poster Session 12226: Stavudine (d4T), didanosine (ddI), and nelfinavir combination therapy in pretreated HIV-patients: Antiviral effect and safety in an ongoing study

This was a small single arm study in Germany looking at the combination of nelfinavir, ddI and d4T in antiviral treated patients. 53 patients were enrolled, all of whom had been on nucleosides, and the majority had been on at least one protease inhibitor. Nelfinavir was given the conventional way, and ddI at 200 mg once daily. At 24 weeks, the early results looked reasonable, with a viral load drop of 2.2 log, but only 30% were below 400 copies. The authors stated that many of the poor responses were in patients who had trouble sticking with the regimen. In this relatively sick group (median CD4 162, median viral load almost 100,000 copies) there were 3 cases of pancreatitis and 3 cases of elevated lipase without symptoms. Two lessons from this study seem to be that, as we keep seeing, results in heavily pretreated patients with conventional 3 drug regimens are a bit disappointing, and, despite the excellent results of the d4T, ddI combination with a variety of other drugs, we must keep watching for pancreatitis, particularly in sicker patients. Longer term followup of two European studies of once a day ddI were presented here, which continue to show that ddI given once daily (usually 400 mg at night) is equivalent to twice daily.

Poster Session 12275: Stavudine (d4T), Nelfinavir (NFV) and Nevirapine (NVP): Suppression of HIV-1 RNA to fewer than 50 copies/ml during 5 months of therapy

This small open label study from Brown University is interesting because it is one of several "triple class" studies, that combine a protease inhibitor with one nucleoside (d4T) and one non nucleoside (NNRTI, in this case nevirapine). The study was primarily designed to look at the pharmacological interaction. D4T and nelfinavir (three time a day dosing) were given for 7 days, drug levels were measured over an 8 hour interval, then nevirapine was added and drug levels were again measured. Twenty five patients started the study; 20 have been on the regimen for 30 to 50 weeks. Half were drug experienced with past AZT, d4T, ddI or 3TC. Median viral load was 4.5 log (30,000 copies) and median CD4 count was 280 (pretty average for study volunteers). The antiviral efficacy so far has been impressive, with at least 85% (intention to treat) below 400% and 50% below 50 copies. This suggests that triple class regimens are quite potent. We should keep in mind the small size of the study, though, and the concern about what would be left to use after this regimen. Whether triple class regimens are better than protease plus NNRTI combinations (such as efavirenz indinavir or nevirapine nelfinavir) must be determined by head to head comparisons with long term followup.