1. The FDA Decides the Fate of Two Drugs for AIDS-Related Malignancies
2. Three New Antivirals Trickle From the Pipeline
3. Hamburg Highlights

On June 23rd, the FDA's Oncologic Drug Advisory Committee held a hearing to determine the fate of Bristol Myers-Squibb's supplemental new drug application for paclitaxol (Taxol) as second-line treatment for AIDS-related Kaposi's sarcoma. Taxol is a popular, potent chemotherapeutic agent which had originally been approved by the FDA for use in breast cancer, and is also used off-label for ovarian cancer.

Bristol's application included 85 patients with progressive KS from two Phase II studies: #174 conducted at the National Cancer Institute and #281 conducted at University of Southern California and Harvard. Because Bristol was seeking an indication for Taxol as second-line KS therapy (that is, patients who have failed or are intolerant to first-line treatment), the main focus was on the 59 patients (69%) who had received prior systemic therapy. The vast majority of the 59 patients were considered "poor prognosis" candidates for therapy because they had a median CD4 count of 16, other prior opportunistic infections (86%), and flu-like symptoms (54%) typical of abnormal activation of one's B cells. Their KS involvement was pronounced: only 7% had KS limited to their skin, 42% also had oral involvement and 30% had KS in their lungs.

According to the FDA's assessment, an overall response was documented in 35 of the 59 patients, 2 with a "complete response" and 33 with a "partial response." The median time to the first response was 8.1 weeks (range 2.9-36.0 weeks, and the median time to progression for all 59 patients was 6.2 months (95% range: 4.6-8.7 months). Median survival for all patients (including the ones who had neither a complete nor a partial response to treatment) was 13.7 months. For those patients with prior systemic therapy, the median survival was 14.1 months, but of those, patients with pulmonary KS had a median survival of about half that time (7.4 months). Overall, 84% of patients developed secondary opportunistic infections during the treatment for KS. At the time of the analysis, over half of the total patients (forty-six of eighty-five) were dead.

Neutropenia (a dangerously low number of bacteria-fighting white blood cells) was the major toxicity associated with Taxol, with 100% of the patients in study 174 and 62% of those in study 281 developing grade 3 or 4 neutropenia. (Co-administration of G-CSF was more common in #281 than in #174.) Fever associated with neutropenia developed in 55% and 9% of the #174 and #281 patients, respectively. Other toxicities included, alopecia (hair loss) in 91% of patients, asthenia (weakness or loss of strength) in 84%, nausea/vomiting in 69%, diarrhea in 79%, arthralgia or myalgia (joint or muscle pain) in 64%, peripheral neuropathy in 58%, alterations in SGOT liver enzymes in 49%, and renal (kidney) toxicity in 24%. Even with such high rates of toxicity, only 7 (8%) patients discontinued Taxol due to adverse events.

After all these data were presented to the FDA's Oncologic Drug Committee, a vote of 8 to 4 was rendered in favor of approving Taxol as second-line treatment for patients with KS. Bristol got its KS indication for Taxol, but the FDA stipulated that when used to treat KS in HIV-infected patients, Taxol must always be administered with G-CSF.

Two Taxol post-marketing studies generated by the AIDS Committee of the NCI's Eastern Cooperative Oncology Group (ECOG) are soon to open: a Phase II study which will test the.pharmacological drug interactions of Taxol with all approved protease inhibitors, and a Phase III study of Taxol versus Doxil for first-line treatment of progressive KS. The endpoints for the Phase III study are efficacy, toxicity, and clinical benefit.

On the same day as the Taxol hearing, Ilex Pharmaceuticals presented its data on mitoguazone (MGBG) for the treatment of patients with HIV-related non-Hodgkin's lymphoma (NHL) who have previously failed at least one "potentially curative" regimen. Currently, there are no approved drugs (or a standard of care) for patients with HIV-related lymphoma who have failed or are intolerant to standard, first-line chemotherapy.

The MGBG application included a total of 90 relapsed or refractory NHL patients who were enrolled in two open-label, single-arm Phase II studies. Of the 90 patients, 65 patients (72%) had more than 3 "poor prognostic" factors; only one patient had none.

According to Ilex, six patients achieved a "complete response" and 7 achieved a "partial response." Median time-to-response was 53 days, and the median duration of response was 144+ days. Overall, time-to-tumor progression was 40 days, time-to-treatment failure was 30 days, and the median survival time was 84+ days. For the responders, time-to-tumor progression was 163+ days, time-to-treatment failure was 151 days, and the median survival time was 269 days.

After closely examining Ilex's methods for assessing treatment responses, the FDA medical officer adjusted Ilex's reported 14.4% overall response rate down to 6% due to either patient ineligibility, protocol violation or simple disbelief. Even at 14.4%, the Oncologic Drug Committee members were nowhere near bowled over by MGBG's purported success rate, but when even that unimpressive response rate was whittled down to a single digit, they wanted nothing more to do with the drug-period! They promptly voted 0 to 12 against approving MGBG. Since the FDA usually follows its advisory committee's decisions, it seems highly unlikely that MGBG will be approved.

While it may have been rational for the Committee to vote against approval of MGBG in light of the FDA's data analysis, some activists and lookers-on are concerned that the FDA did not give Ilex adequate time to formally respond to the response rate discrepancy. So too, they are, while the Oncologic Drug Committee is honorably staffed with a phalanx of distinguished breast, prostate, lung and ovarian cancer doctors, representation of AIDS physicians is sorely lacking. And the differences between HIV-related and non-HIV lymphoma are striking. For the future review of new AIDS malignancy agents, the FDA must insure that its oncologic drug advisory committee has adequate representation of AIDS oncologists so that it may professionally, logically and sensibly advise.

A group of new anti-HIV therapies are about to enter expanded access programs, but heavily pre-treated patients aren't out of the woods yet. Due to cross resistance, at least one of the new products isn't expected to work in people with extensive prior nucleoside experience, and it may be difficult to cobble together a regimen from the two remaining new agents that can reasonably be expected to produce sufficient virologic suppression.

1592 -- late for nearly everyone

Thanks to Glaxo-Welcome's crack marketing team (and a cynically delayed release designed so as not to cannibalize the recent launch of "Combivir"), perhaps the most hotly anticipated of the new therapies is Wellcome's 1592U89 (now known by the name "abacavir"). Early study results suggested as much as a 2 log reduction in plasma HIV RNA when treatment naïve patients were treated with 1592. Data recently presented by Dr. David Hardy to an investigator's meeting sponsored by Glaxo-Wellcome, however, suggests that heavily pre-treated patients may not benefit from 1592. While patients pre-treated with ddI or d4T experienced relatively normal reductions in plasma HIV RNA (about 1.5 log) after twelve weeks of 1592 treatment, patients with extensive AZT or AZT+3TC double nucleoside therapy were much less responsive: after twelve weeks of therapy, patients with more than a year of prior AZT treatment had only a .05 log reduction while patients with extensive AZT+3TC therapy had only a 0.05 log reduction. Since the lion's share of the traditional expanded access population will have received loads of prior AZT (and invariably 3TC too), it is unlikely that they will receive much benefit from 1592 -- at least based on the results from this study.

Due to community pressure, plans for expanded access to 1592 keep changing. The most recent plan was for 1592 to become available beginning September 15, with participating sites allowed to enroll 2 patients per week until the end of the year, some 16 weeks. Eligible patients would have a CD4+ T-cell count < 100 and a plasma viral load > 30,000. The program plans to enroll approximately 2,400 patients in North America and another 900 patients in Europe. Separate (and smaller) expanded access programs will make 1592 available for pediatrics as well as for adults with HIV dementia.

Sustiva, sustava, sustain me

In contrast to the disappointing cross-resistance news with 1592, DuPont-Merck's new non-nucleoside reverse transcriptase inhibitor (NNRTI) DMP-266 (now with the generic name "efavirenz" and trade name "Sustiva"), seems to be active against the most common resistance mutation (Y181) associated with use of nevirapine and delavirdine, the two commercially available NNRTIs. As long as only one mutation (181) is present, current dosing of DMP should be enough to suppress virus, however when two NNRTI resistance mutations (Y181 and K103) are present, DMP may not be enough to suppress replication. Early studies suggest that DMP may be taken once a day, and can reduce viral RNA counts by about 1.5 logs. The most common side effects are lightheadedness and rash.

DuPont-Merck recently announced an expanded access program for DMP that will be available to a limited number of patients (approximately 2,000) through the end of 1997, and will then broaden to include additional patients in early 1998. Patients with < 50 T-cells who are failing current therapeutic regimens are eligible for the program as it is currently designed. Participants, though, will be required to take DMP with a second new drug, such as 1592 or adefovir (see below). DuPont Merck expects to file its NDA for efavirenz in March of next year. For more information on the program, patients and physicians can call 1-800-998-6854.

Adefovir itself

Gilead Science's adefovir dipivoxil (formerly known as bis-POM PMEA) is a nucleotide analogue which will also be made available through an expanded access program said to be very similar to DuPont Merck's program for DMP-266. Adefovir, to be marketed under the trade name Preveon, is active against the hepatitis B virus and CMV as well as HIV. The drug has been reported to deplete the body's store of L-carnitine levels (which is made by the body to help supply energy to muscles), so supplementation (500 mg once daily) with this amino acid is recommended for people taking adefovir. Gilead's expanded access for adefovir is slated to begin mid-November. Eligible patients will be required to 1) have failed a regimen containing at least two commercially available nucleoside RTIs and one commercially available protease inhibitor; 2) have a CD4 cell count < 50 cells/mm³ 3) a plasma HIV RNA copy number > 30,000 within the past 2 months; and 4) not qualify for other adefovir clinical trials. Participants in the adefovir program will also be advised, as in the DMP expanded access program, to add at least one new antiretroviral agent to the mix. Adefovir is administered as a 120 mg tablet once daily. For more information about the adefovir expanded access program, call Gilead at (800) GILEAD-5.

The bottom line...

While two new protease inhibitors, Glaxo-Welcome's GW-141 and Abbott Laboratories' ABT-378 are in the pipeline, there are not yet estimates on when these drugs will become available. Early reports suggest that GW-141 will be cross-resistant with indinavir and ritonavir, however in vitro studies suggest that ABT-378 may be effective against viral strains that are resistant to other protease inhibitors. (More critical on-lookers warn that the 84 mutation knocks ABT-378 on its back.)

Cross-resistance between anti-HIV therapies is continuing to baffle design of expanded access programs. Those patients for whom expanded access is intended -- patients with few other therapeutic options -- will often be the patients least likely to benefit from new therapies. This creates powerful disincentives for companies to make drug available -- just witness the incredibly shrinking expanded access programs over recent years.

On the flip side, companies such as Glaxo-Wellcome have also learned that expanded access schemes represent a powerful promotional tool for manipulating the HIV/AIDS communities. When Glaxo-Wellcome limited access to 3TC just prior to approval, it created strong pressure from AIDS activists to approve the drug despite serious limitations in the available knowledge regarding use of the product. That the company seems to be repeating this pattern with 1592 (Witness, for example, the relentless headline-making paeans from patients with advanced disease who dutifully propose that, "This product could save my life."), is testament to the manufacturer's marketing skills and the community's collective amnesia.

Programs like expanded access and accelerated approval were designed to make available therapies that were desperately needed by patients with few other treatment options. When new products are unlikely to be effective in these patients, AIDS activists will have to ask themselves if programs such as expanded access and accelerated approval are serving their needs, or are simply highly effective pre-approval marketing efforts by pharmaceutical manufacturers. These difficult questions are likely to haunt the AIDS communities for the foreseeable future.

• David Ho proposes immunotherapeutic vaccination to help the immune system clear the long-lived "3rd" compartment of HIV-infected cells. These latently infected CD4+ T-cells with integrated HIV DNA currently represent the greatest obstacle to the potential eradication of HIV infection. And a British study may just beat him to the punch: Frances Gotch (London) reported that her group will randomize patients to HAART plus 1) an immunomodulator, 2) the Salk (Immunogen) vaccine, 3) both or 4) placebo.

• On the other side of the Atlantic, Dr. Luc Montaigner proposes the use of an HIV-nef vaccine in order to immunize uninfected persons from productive infection with HIV. In an open plenary lecture at the Sixth European Conference on Clinical Aspects and Treatment of HIV Infection in Hamburg, Germany (October 11-15), the official discoverer of HIV explained that HIV's regulatory proteins nef and tat are responsible for the continual propagation of what would ordinarily be a containable infection with HIV. Montaigner's goal: by inactivating nef early on, the pool of activated CD4+ cells susceptible to infection would be small enough for -- perhaps -- the body's own immune response to clear the initial infection.

• Jean-Pierre Sommadossi (UAB) presents compelling evidence of what is being called "pharmacologic resistance" to the entire nucleoside class of drugs after long-term AZT use. Changes in the cellular metabolism of AZT resulted in reduced efficacy of the other nucleoside analogue agents. Whether or not -- or to what extent -- this sort of pharmacologic resistance develops after prolonged use of the other nukes (and other classes of antiretroviral agents) is not known.

• On behalf of the Delta coordinating committee, London's A. Babiker showed that changes in CD4+ T-cells and HIV plasma RNA levels break down as surrogate markers over the long term. "Similar RNA responses with different treatments do not have similar associations with clinical outcome," Babiker explained. He speculates that a large part of this dis-association is likely due to the increased toxicity of combination regimens -- ill-effects not directly captured by changes in viral load or T-cell counts.

• Scott Eastman, speaking at one of the many commercial "satellite" symposia during the two days prior to the Hamburg meeting, presented disturbing data about the variation in blood levels of different antiretroviral drugs. It's not just indinavir where you see significant variation in plasma drug levels, Eastman explained. While plasma levels of indinavir varied up to 8-fold, plasma levels of 3TC and nevirapine were also found to vary significantly (3-5 fold).

• Two-year follow-up of the high-profile Merck 035 study provides both good news and disconcerting news, depending on how carefully you look at the data. While the number that made it into the newspapers was "80% still undetectable at 2 years," that's using the bDNA assay with a lower limit of detection of 500 copies/mL. Using the more sensitive Roche UltraDirect assay (with a lower limit of 20 copies/mL), the proportion of 035 patients with "undetectable" plasma HIV RNA fall to 60%. (Which is beginning to look more like INCAS!) While experts disagree on how significant the difference between <500 and <20 copies/mL is, emerging clinical data are becoming more and more convincing that a nadir of <20 assures a significantly more durable response to treatment.

• In a French study of similar design called AVANTI 2, which treated drug naïve patients with the same triple combo (AZT+3TC+indinavir), the proportion of study participants with plasma HIV RNA < 20 copies/mL at 52 weeks was also 60% -- the same proportion seen at 2 years in 035's AZT-experienced population.

• A study of indinavir in combination with DMP-266 found that 82% of the IDV+DMP recipients whose RNA fell to between 1-400 copies/mL experienced viral rebound (defined as any RNA measurement > 400 copies/mL) within 100 days of initiation of treatment compared to only 25% of patients whose plasma viral load fell to < 1 copy/mL after initiation of the double combination treatment. The ultra, ultra sensitive assay used in this study was the new (experimental) Amplicor assay.