This is Part 2 of a three-part series. See also Part 1.
28 August 2008. Besides Elan/Wyeth’s and Eli Lilly and Company’s antibodies, two additional anti-amyloid approaches—PBT2 and IVIg—came out with Phase 2 data at ICAD. Jeffrey Cummings of the University of California, Los Angeles, presented results on PBT2. The Alzforum has reported this data this past March, after the drug’s developer, Australian Prana Biotechnology Ltd., first announced results (see ARF related drug news story). The full publication appears in the September Lancet Neurology, now available online (Lannfelt et al., 2008).
At ICAD, some scientists cautioned that the drug’s reported signals in two cognitive tests, i.e., trail making B and category fluency, should not be overinterpreted, as only 29 patients had received the higher, presumably effective, dose. For his part, Cummings said that he saw the main significance of this trial in the drug’s demonstrable ability to lower human CSF Aβ42 levels. This has been difficult to achieve with previous therapeutic agents, including, most recently, Flurizan. Cummings also noted that the trial investigators monitored very closely for any hints of subacute myelo-optic neuropathy, a serious side effect associated with clioquinol in the past, but found no indications of this with PBT2. “We feel the safety and tolerability of PBT2 is comparable to placebo,” Cummings said. Prana at present is looking for funding to support a planned Phase 2b trial.
Norman Relkin of Weill Medical College, Cornell University, New York, presented interim nine-month data from an ongoing 18-month Phase 2 trial of IVIg (aka Gammagard), a preparation of pooled human antibodies made by Baxter Biosciences. While more specific vaccines are being developed, this product drew Relkin’s attention because it has a decades-long safety record in humans for use in other conditions. Relkin and colleagues had previously conducted an open-label, Phase 1 dose-ranging study in people with AD, and reported a symptomatic benefit (Relkin et al., 2008; also ARF related news story). One question that arose in this earlier study was how long this benefit could last. The AD patients in this study had a break in treatment after six months, after which they lost the improvement they had gained during the previous six months on IVIg. Relkin and colleagues address whether this decline was due to the washout of IVIg, or whether this preparation simply does not help longer than six months, in a subsequent placebo-controlled, randomized Phase 2 trial at Weill that is still ongoing. Both trials are small—eight patients in the open-label study, 24 patients in the Phase 2—owing in part to the drug’s cost and availability (see below).
In the Phase 2 trial, four patients each received one of four doses of IVIg for 24 weeks, and eight patients received infusions of saline placebo. Evaluations occurred at 12 weeks and 24 weeks; after that the patients on placebo switched over to one of four doses of IVIg for another 12 months. From this point on, then, the trial had no true placebo group anymore, and the former placebo group was partially blinded in that the rater did not know what dose they were on. The primary outcome measures were the ADAS-Cog, ADCS-CGIC; secondary outcome measures included the ADL and some other cognitive and global measures. (This trial also incorporated CSF biomarker measures, as well as imaging scans using FDG PET to assess brain activity, PIB-PET to measure amyloid levels, and PK11195-PET to measure microglial activation in a subgroup of patients. Those data are not available yet, Relkin said.)
At ICAD, Relkin presented nine-month interim data on the primary and secondary readouts. On all outcomes, the curves between placebo and IVIg separated on the three-, six-, and nine-month time points. The results presented at a press briefing and on a poster were confusing at first glance because the slope graphs indicated a nine-month time point for the placebo group even though those people at this point had been on various doses of drug for the past three months. In an e-mail, Relkin clarified that the main point to take away from this interim analysis is that the symptomatic benefit that was apparent in the Phase 1 study stayed stable for nine months in this second study. “This Phase 2 study was the first time we treated AD patients with IVIg for longer than six months without interruption. The interim analysis presented at ICAD was the first demonstration of the persistence of benefits at nine months when IVIg treatment was uninterrupted,” Relkin wrote. Relkin clearly stated that this small trial was not powered to measure efficacy; hence, numerical comparisons between the groups are less meaningful than the general observation that treated patients maintained their benefit for nine months so far.
The trial confirmed the best dose as being 0.4g/kg infused every two weeks, and this information will help in the conduct of a subsequent Phase 3 trial. Starting this fall, the Alzheimer’s Disease Cooperative Study will enroll 360 patients at some 35 sites into an 18-month trial that incorporates CSF, MRI, FDG PET and PIB-PET. This trial will be pivotal for the prospects of this treatment in AD.
“The NIA-sponsored ADCS is the most appropriate organization to test IVIg as a potential treatment for AD, since IVIg is not patentable and consequently of little interest to the pharmaceutical industry. The partnership between the ADCS, with its expertise in AD clinical trials, and Baxter has made it possible to design a much more comprehensive and exciting Phase 3 trial than would otherwise have been possible,” Relkin said.
If the trial were to be positive, the issue of IVIg’s limited supply would come to the fore. At present, there is a relative shortage of this product, which is made from the blood of several thousand donors. According to Relkin, its sole manufacturer Baxter Bioscience sells all it can prepare within a few weeks for other indications. Scale-up to satisfy demand for a potential new AD indication would require a major investment in manufacturing, and issues related to this have slowed down the planning for the ADCS Phase 3 trial, said the ADCS director Paul Aisen of the University of California, San Diego.
Cost is another issue. A small number of families are working with neurologists besides Relkin to purchase IVIg and treat AD in a loved one outside of formal trials. With supply so limited, this can be expensive. Here is how Relkin explained the issue: “The cost of IVIg is dependent on market factors. There are often considerable mark-ups when it is purchased outside of a hospital setting. Minimum base cost when purchased in bulk by hospitals is about $50 per gram, and we typically give between about 20-30 grams per infusion depending on the subject’s weight. That translates into about $36,000 per year, but with mark-ups that can easily rise to $50,000-$75,000 per year or in extreme cases as high as $100,000 per year. Since supply is limited, for now the high cost helps to discourage premature off-label use that might otherwise deplete the supply available to treat other diseases.”
“If approved, IVIg supply could become a major issue. We are working on strategies to increase the efficiency and ease of delivery, as well as ways to identify best responders predictively (FDG PET looks promising for this purpose). The IVIg manufacturers are now considering plans to increase production. If successful, these initiatives could help extend the available supply to more patients and/or reduce overall expenditures,” Relkin wrote by e-mail.—Gabrielle Strobel.
This is Part 2 of a three-part series. See also Part 1.