While researchers pursuing passive immunotherapy approaches struggle to refine trial protocols (see Part 4 of this series), active immunotherapies presented setbacks at the 7th Clinical Trials on Alzheimer’s Disease (CTAD) conference, held November 20 to 22 in Philadelphia. Speakers clarified baffling data from AFFiRiS’ Phase 2 trial of AD02, in which the placebo was said to beat the treatment. The treatment did not work and is not being pursued any more. Another active vaccine, Pfizer/Janssen Alzheimer Immunotherapy’s ACC-001, nudged biomarkers, yet the company has shelved development. There were no presentations at CTAD on CAD106, currently the most advanced active immunotherapy in AD. Passive approaches were the hotter topic here, with a new candidate— Eli Lilly’s LY3002813—making its debut.
AFFiRiS AG, Vienna, unfurled perhaps the most curious story of the meeting. At an earlier press conference, the company had reported that its placebo worked better than its drug, and announced plans to further develop the placebo, which it declined to identify (see Jun 2014 news story). In Philadelphia, the company’s Achim Schneeberger cleared up the mystery, revealing that the placebo was alum, an aluminum salt (typically potassium aluminum sulfate) frequently used as an adjuvant for vaccines. The researchers injected either 1 mg or 2 mg alum into the control cohorts, while the three treatment groups received either 1 or 2 mg of alum along with varying amounts of the active vaccine, AD02. The Phase 2 trial enrolled 332 people with early AD from six European countries.
Those who received 2 mg alum but no vaccine had the best outcomes. Huh? Yes, you read it correctly. This group reportedly declined less than the others, and did so consistently across several measures, including the ADAS-Cog, Activities of Daily Living (ADL), and Clinical Dementia Rating Sum of Boxes (CDR-SB). In fact, almost half of this control cohort remained cognitively stable over the 18 months of the trial, Schneeberger reported. Their hippocampi shrank less compared to the other groups’, correlating with higher ADAS-Cog scores. This cohort also stabilized on measures of neuropsychiatric symptoms and quality of life. Dropout rates were similar between the cohorts.
Several clinicians at CTAD privately suggested this may be a chance finding, but Schneeberger disagreed. “It is unlikely that the effect is by chance. We believe there is a disease-modifying effect,” he said. Some attendees questioned why, if alum protects, didn't 2 mg alum given in conjunction with AD02 slow decline? Schneeberger suggested that the AD02 peptide might interfere with the effect of alum by coating it.
Could the 2 mg alum group have declined as expected, while AD02 had a deleterious effect that made those taking it deteriorate faster than normal? Probably not, said Suzanne Hendrix, a statistician at Pentara Corporation, Salt Lake City, who works on contract with AFFiRiS. Hendrix presented analyses that compared the decline in this trial with decline in historical datasets including the Alzheimer’s Disease Neuroimaging Initiative (ADNI), the Alzheimer’s Disease Cooperative Study (ADCS), and a trial simulation tool developed by the Coalition Against Major Diseases (CAMD) (see Jul 2013 news story). All cohorts in the AFFiRiS trial declined cognitively and functionally at the expected rates—except the 2 mg alum group, Hendrix said. “The 2 mg alum group was the outlier,” she added. She subdivided the cohort by disease severity and found that only those with mild AD appeared to benefit from alum; moderate patients worsened at expected rates.
AFFiRiS’ next goal is to discern how alum might slow decline, Schneeberger said. Hendrix noted that the 2 mg dose was chosen to provoke an immune reaction at the injection site. Perhaps the adjuvant somehow slows Alzheimer’s disease by stimulating the immune system, she suggested. To explore this, AFFiRiS will look for changes in cytokine levels in patient samples. The company plans to launch a new confirmatory trial in Europe, Hendrix told Alzforum. A U.S. trial would face additional safety hurdles, because 2 mg alum exceeds the 1.25 mg limit set by the Food and Drug Administration. In Europe, alum is approved at higher doses and frequently used in immunotherapy for a variety of conditions. Adding to the puzzle, epidemiological studies have linked aluminum exposure to Alzheimer’s risk (see, e.g., Bondy, 2014). In fact, there has been longstanding debate about aluminum’s role in Alzheimer’s, and families have been discarding their aluminum pots and pans for years.
Meanwhile, AD02 itself provided no statistically significant benefit, despite earlier results (see Jun 2012 conference story). AD02 was designed to stimulate the immune system to make antibodies against Aβ. With active vaccines, the dose of injected peptide itself matters less than the titer of the antibodies the patient’s B cells generate in response. That antibody concentration, in effect, represents the exposure of the body to “active drug.” Schneeberger presented no antibody titer data at CTAD. He told Alzforum that most patients did mount an immune response, which varied with the dosage of AD02 and adjuvant. The company has terminated an extension study of AD02. Appearing puzzled by the data, the audience asked no questions after Schneeberger’s talk.
Another active vaccine has also reached the end of its development life. Janssen’s ACC-001, originally developed by Elan/Wyeth, is a peptide of seven amino acids from the N-terminus of Aβ (see Apr 2008 news story; Jun 2008 news story). At CTAD, Nzeera Ketter of Janssen presented results from the company’s Phase 2 trial in 125 volunteers with mild to moderate AD. Participants received six intramuscular injections of either placebo, 3 mg, or 10 mg ACC-001 with the adjuvant QS-21. More than 90 percent of participants in the active groups responded to the vaccine by making antibodies, with titers modestly higher at 10 mg than 3, Ketter noted. Safety was as expected, with 5.8 percent of the cohort developing the leaky brain blood vessels known as ARIA-E.
Also as expected, no significant difference in cognitive decline appeared in this small study. However, those on the drug did accumulate less brain amyloid as assessed by florbetapir PET imaging. This effect increased at the higher vaccine dose, but missed statistical significance. Plasma Aβ levels rose along with antibody titers, suggesting enhanced clearance from brain. In the treatment group, phosphorylated tau in the cerebrospinal fluid (CSF) dropped slightly, again hinting at a treatment benefit. The only statistically significant difference between the cohorts was more shrinkage in brain volume in the high-dose treatment group. Why this happens is unclear, but it mirrors what has been seen in other antibody trials, Ketter noted (see Jul 2004 conference story; Nov 2012 conference story). Though small, these biomarker effects demonstrate that the generated antibodies hit their target, Ketter said.
Janssen also tested the same doses of ACC-001 in an amyloid imaging and safety Phase 2 trial of 63 early AD patients. Presented on a poster at CTAD, those data were likewise negative overall, with subtle hints of biomarker movement. In response to an audience question, Ketter noted that Janssen has shelved this vaccine.
Even as some vaccines reach the end of the road, new hopefuls appear. Eli Lilly has an antibody against a modified form of Aβ in Phase 1. LY3002813, also known as N3pG, targets an N-truncated, pyroglutamate form of Aβ that aggregates readily and forms the core of many plaques. LY3002813 cleared deposits in mouse studies without producing microhemorrhages (see Dec 2012 news story). John Sims at Lilly described the trial. At four U.S. and two Japanese sites, the single-ascending-dose study will test intravenous injections of 0.1, 0.3, 1, 3, and 10 mg/kg, as well as subcutaneous injections of 3 mg/kg. Participants have a positive florbetapir scan, and range from MCI due to AD or mild to moderate AD as diagnosed with the AA-NIA revised criteria. In the first month, the most common adverse events were diarrhea, dizziness, and falls, but no changes in microhemorrhages (ARIA-H) and no ARIA-E, Sims reported. So far so good, but there are many more hurdles to jump.—Madolyn Bowman Rogers
- Immunotherapy I: Baby Steps, but No Breakthroughs
- In Surprise, Placebo, not Aβ Vaccine, Said to Slow Alzheimer’s
- AD Trial Simulation Tool Receives Regulators’ Blessings
- Stockholm: New Strategies for Immunotherapy
- Trial Troika—Immunotherapy Interrupted, Lipitor Lags, Dimebon Delivers
- Research Brief: Elan/Wyeth Vaccine Back on Track
- Philadelphia: Can a Shrinking Brain Be Good for You?
- CTAD: New Data on Sola, Bapi, Spark Theragnostics Debate
- Could Antibodies Against Pyroglutamate Safely Break Down Plaques?
- Bondy SC. Prolonged exposure to low levels of aluminum leads to changes associated with brain aging and neurodegeneration. Toxicology. 2014 Jan 6;315:1-7. Epub 2013 Nov 1 PubMed.