Nelfinavir - Major Trial Results at Retroviruses Conference

Nelfinavir (VIRACEPT(R)) is a protease inhibitor which is expected to be approved soon by the FDA, probably within a few weeks; it is already fairly widely available through expanded-access programs and clinical trials. It appears to have fewer side effects than the protease inhibitors now in use; also it has a somewhat different resistance profile, so that persons whose virus has already become resistant to other protease inhibitors may be able to benefit. Also, it offers new options for combination therapy, both with other protease inhibitors and with other anti-HIV drugs.

At the recent Retroviruses conference in Washington D.C., nelfinavir developer Agouron Pharmaceuticals, Inc. released six-month safety, viral load, and CD4 results from three major clinical trials. All three tested the same two doses of nelfinavir, 500 mg three times a day vs. 750 mg three times a day, in patients with fairly advanced HIV infection (median CD4 counts of approximately 276, 279, and 288, but with relatively high viral loads, averages of approximately 164,000, 142,000, and 151,000 copies, in the three trials respectively). A minimum viral load of 15,000 copies was required to qualify for this trial, so that large decreases in viral load could be seen before reaching the lower cut-off of what the viral load test could measure. [Note: The numbers above, the latest from Agouron on February 19, differ from those in the Retroviruses conference abstract, mainly because average copy numbers are given here, while averages of the logs of the numbers were used in the abstract. Also, some patients are included or excluded differently, based on further analysis of the data in the several months since the abstracts went to press. None of the conclusions have changed.]

(1) The first trial tested nelfinavir alone, with 91 HIV-positive volunteers (both antiretroviral naive and experienced) randomly assigned to one of the two doses. To obtain comparison data with no drug, some of the volunteers were given a placebo for the first four weeks.

(2) Another trial compared d4T alone, vs. d4T in combination with the low dose or the high dose of nelfinavir, in 297 volunteers, most of whom were antiretroviral experienced.

(3) The third trial compared AZT+3TC, vs. the triple combination of AZT+3TC+low dose nelfinavir, vs. AZT+3TC+high dose nelfinavir, in 308 volunteers -- who had little or no prior antiretroviral use (no more than one month of AZT).

All of the trials showed statistically significant improvements in CD4 count and viral load in the volunteers who received nelfinavir, compared to those who did not. The best and most sustained results were with the triple combination -- the one which the company identified as most relevant to prospective clinical use of the drug. After six months, average viral load reductions were 2.3 log (200 fold) and 2.5 log (315 fold) for the triple combination with the low and high doses of nelfinavir, vs. 1.4 log (25 fold) for AZT plus 3TC only. The average CD4 increase was 160 for the nelfinavir-containing arms, vs. 105 for AZT plus 3TC only. Probably most importantly, viral load became undetectable (under 100 copies in the test used) in 65% and 81% of those receiving triple combination with low and high dose nelfinavir, vs. only 18% with AZT plus 3TC only -- despite the fact that all these groups of volunteers had started with a relatively high viral load.

No "clinical endpoint" results are yet available from these trials.

Comment: Dosage Issues

We do not know which dose (500 mg or 750 mg, three times a day) the FDA is likely to approve. At the time of the Retroviruses conference, it was widely feared that the company would seek approval for the lower dose, for most patients at least. A number of doctors who had worked with the drug asked for approval for the higher dose. (And, months earlier, treatment activists had asked that doses higher than 750 mg be investigated.)

The issue seems to be that the trials so far have not shown a statistically significant difference between the doses (except for volunteers with viral load over 100,000 copies, in the triple-combination trial, for whom the 750 mg dose was significantly better than 500 mg). For most patients, the data are equivalent for the two doses -- so far. But only six months of data are now available, and people will of course be using the drug for longer. The fear of many experts is that over time, the virus will escape control of the drug by developing resistance to it -- a process which is facilitated if the dose is too low. This may already be happening at six months in those with the highest viral load, who would be expected to encounter resistance problems first -- as suggested by the statistically significant superiority of the higher dose in the triple combination trial, for those with over 100,000 copies only. It is possible that the reason the doses appeared equivalent for other patients is that at lower starting viral loads, the problem may often take more than six months to develop.

Physicians also want the higher dose approved for reimbursement reasons. If the lower dose becomes official, and then doctors find that they should be prescribing more, there will be serious problems getting payers to reimburse for more drug than the FDA has officially approved. But if the higher dose is approved, doctors will of course have no complaints from payers if they prescribe less.

The company might have the opposite incentive. When a drug is being priced to cover investment, etc., then a lower dose will result in a higher price per gram. Later, moving from the lower dose to a higher dose would mean an automatic revenue increase for the company. Similarly, a move downward from the high dose would result in a revenue decrease outside the company's control.

Of course a lower dose could have the advantage of minimizing side effects -- not all of which are likely to be found in relatively short-term clinical trials. Balancing an additional long-term safety margin against the need for long-term control of the virus will be a difficult, continuing issue. Dosing for all HIV treatments could be improved if tests for blood levels were available to physicians, so that doses could be adjusted for an individual's absorption and metabolism of the drug.

Note: See AIDS TREATMENT NEWS #263 for information about new nelfinavir expanded-access programs for children and for adults, and about a new clinical trial comparing triple combination with nelfinavir to triple combination with ritonavir.