18 July 2008. Two papers in the 19 July Lancet bring both dismay and hope to the ongoing quest for more effective Alzheimer disease treatments. One study—a six-year follow-up of eight patients from Elan’s original AN1792 active immunization trial—confirmed that vaccination with full-length Aβ42 can clear amyloid plaques but showed that this clearance did not slow disease progress. The pages immediately preceding the report of this AD immunotherapy setback carry more promising news: in a six-month Russian trial of 183 patients, Dimebon—a weak inhibitor of cholinesterase and NMDA receptors with neuroprotective properties—improved mild to moderate AD patients in all five of the study’s outcome measures (four cognitive, one global). What’s more, Dimebon’s benefits seemed to hold, and by some measures even increase, through the trial’s six-month blinded extension. The authors note that these persistent benefits distinguish the small-molecule drug from existing approved therapies for mild to moderate AD—none of which have shown increasing improvement past 12 months.
A perplexing mix of evidence from Aβ immunization studies in humans and animal AD models has led researchers in the AD immunotherapy field to a central question: does the immune system help or harm in a person’s fight against AD? The emerging answer is, of course, multi-faceted and situational (for a review of recent evidence, see Boche and Nicoll, 2008).
AN1792’s storied history provides a case in point. Phase 1 trials of this AD vaccine—the first to be tested in people—began in September 2000. In early 2002, Elan Corporation discontinued clinical testing of the vaccine during Phase 2 trials because 6 percent of immunized patients developed cerebral inflammation (see ARF live discussion). Follow-up analysis revealed that among AD patients who got the vaccine, those who made antibodies against the injected Aβ42 preparation seemed to fare better or at least decline less on several measures of cognitive function and daily living (see Hock et al., 2003 and ARF related news story). Another study (Gilman et al., 2005) found no improvement in AN1792-treated subjects using a handful of outcome measures (ADAS-cog, Disability Assessment for Dementia, Clinical Dementia Rating, Mini-Mental State Examination [MMSE], or Clinical Global Impression of Change) but did show some benefits (better neuropsychological test battery (NTB) scores and lower CSF tau levels). More recent studies looking at how AN1792-treated patients were faring four and a half years after the start of the trial have also delivered somewhat hopeful but tentative results (see ARF Madrid news story and ARF Washington news story).
In the new study, James Nicoll at the University of Southampton, UK, with collaborators there and elsewhere, set out to determine how Aβ42 immune response correlated with extent of amyloid plaque removal and longer-term clinical outcomes. Due to patient deaths before and during follow-up, as well as limited numbers offering consent for postmortem analysis, first author Clive Holmes and colleagues were left with eight AN1792 trial participants, all in the treatment group, to analyze. In these patients and an age-matched control group of non-immunized AD patients, the researchers assessed Aβ load by determining the percent of cortical area with Aβ immunoreactivity (clone 6F/3D) and scoring specific histological evidence for plaque removal. For each patient, Aβ load was compared with mean anti-AN1792 antibody titers and longer-term outcomes (survival time and MMSE score before death). A morsel of good news came from the study’s confirmation that the Aβ immunization procedure generally works. Mean Aβ load was lower in the vaccinated patients (2.1 versus 5.1 percent in controls), and though the extent of plaque removal varied greatly, it did correlate to some degree with serum anti-Aβ antibody titers, the researchers found.
The bottom-line news, unfortunately, was grim. In the small cohort they were able to analyze, the researchers saw no measurable clinical gains in the AN1792-treated group. In fact, two patients with near complete clearance of amyloid plaques still succumbed to profound end-stage dementia before they died. Overall, the immunized patients did not live longer, nor did they take longer to reach severe dementia, compared with controls. “Although our findings are based on small numbers of patients, they seem to demonstrate that the presence of plaques is not a prerequisite for progressive cognitive impairment in AD,” wrote Nicoll in an e-mail to ARF.
Donna Wilcock of Duke University in Durham, North Carolina, noted that the authors did not specifically address the possibility that cerebral amyloid angiopathy (CAA) and associated microhemorrhage contributed to the clinical progression of the study patients. She pointed out a recent report (Schroeter et al., 2008) suggesting that even low doses of anti-Aβ antibody, which resulted in virtually no amyloid removal, were associated with subtle increases in microhemorrhage. (See full comment below.)
Based on the new findings, Nicoll expects other immunization strategies (e.g., passive immunization and active immunization with truncated versions of Aβ) to be effective at clearing plaques but predicts that these changes will not correlate well with cognitive improvement. However, he noted that newer protocols may have differential effects on the various forms of Aβ (e.g., plaque, soluble, oligomeric, intraneuronal), which could in turn lead to different cognitive effects. Nicoll expressed cautious hope in using immunization to prevent AD rather than treat it. “A study to determine if Aβ immunization at a young age could prevent the development of AD later in life would be the ultimate test of the Aβ hypothesis,” he wrote. (See full comment below.) Preventive AD vaccines have not yet entered clinical trials, but a recent study (see Movsesyan et al., 2008 and ARF related news story) describes a DNA-based, preventive vaccination approach that seems to work in an AD mouse model.
As the field ponders new avenues for immunotherapy, AD researchers and patients are likely to take heart in the report on Dimebon. Rachelle Smith Doody of Baylor College of Medicine in Houston, Texas, with collaborators elsewhere, led the new study of this small-molecule compound—marketed in Russia as an antihistamine decades ago and repurposed as an AD drug when its anti-cholinesterase and NMDA receptor inhibitory properties were discovered. Studies have shown that Dimebon has neuroprotective effects in rodents, and that this benefit may stem from its ability to stabilize and enhance mitochondrial function (Bachurin et al., 2001; Bachurin et al., 2003). David Hung, CEO of San Francisco-based Medivation, Inc., will discuss evidence for the drug’s novel mitochondrial mechanism of action at ICAD on 30 July. Medivation is developing Dimebon as a treatment for Alzheimer and Huntington diseases.
Much of the newly published data on the Russian trial of Dimebon has been presented previously (see ARF Boston news story, ARF Washington news story, and ARF related news story). In short, the cohort receiving oral Dimebon (20 mg three times a day) showed significant improvement in all cognitive outcome measures (ADAS-cog, MMSE, neuropsychiatric inventory, the Alzheimer’s Disease Cooperative Study—activities of daily living [ADCS-ADL]) as well as in a global measure (Clinician’s Interview-based Impression of Change plus Caregiver Input [CIBIC-plus]). “The interesting thing about those differences is that there was both significant improvement over baseline in all measures and significant decline in the placebo group, and the benefit was driven both by improvement and decline,” noted Doody in an Alzforum interview. Most AD drugs that have shown benefit in clinical trials have only done so compared with placebo groups that declined more sharply. “In the Dimebon case, even if the placebo group was flat, we’d still have a positive study,” Doody said.
Because 86 percent of those who completed the trial chose to enroll in the six-month blinded extension, which retained each person’s original treatment assignment, the researchers essentially got a one-year look at Dimebon, Doody said. “What we saw
was continued benefit above baseline and further decline in the placebo group on all of the outcomes, leading to statistical significance for all the measures.” The Dimebon-placebo difference was greater at week 52 than at week 26 for the ADAS-cog, ADCS-ADL, and CIBIC-plus measures.
In this Russian trial, Dimebon was safe and well tolerated. At week 26, dry mouth and depressed mood were the most common adverse events (14 percent for each symptom, versus 5 percent of placebo patients).
“It's very important that the field be investigating non-amyloid-based interventions as well as amyloid-based interventions,” said Jeff Cummings of the University of California, Los Angeles, who serves on the steering committee to help design Dimebon trials. “This non-amyloidogenic pathway involving unique mechanisms of action appears to result, in this first pivotal trial, in a good effect size, a consistent effect across measurement instruments, and a more persistent effect than we have seen in previous trials.”
In May, Medivation launched a multinational Phase 3 Dimebon trial with an anticipated enrollment of 525 patients in the U.S., Europe, and South America. Like the newly published study, the global trial excludes patients who are taking other anti-dementia drugs, including cholinesterase inhibitors and NMDA receptor antagonists.—Esther Landhuis.
Holmes C, Boche D, Wilkinson D, Yadegarfar G, Hopkins V, Bayer A, Jones RW, Bullock R, Love S, Neal JW, Zotova E, Nicoll JA. Long-term effects of Abeta42 immunisation in Alzheimer’s disease: follow-up of a randomized, placebo-controlled phase I trial. Lancet 2008 July 19;372:216-223. Abstract
Doody RS, Gavrilova SI, Sano M, Thomas RG, Aisen PS, Bachurin SO, Seely L, Hung D on behalf of Dimebon investigators. Effect of Dimebon on cognition, activities of daily living, behaviour, and global function in patients with mild-to-moderate Alzheimer’s disease: a randomized, double-blind, placebo-controlled study. Lancet 2008 July 19;372:207-215. Abstract