New results from the AMBAR trial show that plasma exchange leads to statistically significant functional and cognitive improvements in patients with mild to moderate Alzheimer’s .
As our readers surely know, the world is spending much more resources on fighting the diseases of aging than on solving aging itself. In particular, tens of billions of dollars have been poured into looking for a cure for the deadly Alzheimer’s disease (AD), with few to no results. One analysis puts the rate of success of AD clinical trials at 0.4% . So, when a human AD trial yields encouraging results, it’s big news.
The underlying hypothesis
AMBAR (Alzheimer’s Management by Albumin Replacement) was a clinical trial conducted by the Spanish pharmaceutical company Grifols to study the effects of plasma exchange (PE) on AD. With more than 400 participants, it was registered as a Stage IIb trial in the US and a Stage III trial in the EU and ended a while ago. Plasma exchange involves replacing a part of a patient’s blood plasma with donor plasma or a plasma substitute. In AMBAR, various groups received various concentrations (either 5% or 20%) of medicinal albumin in their plasma transfers.
AMBAR was based on the “peripheral sink” hypothesis. Alzheimer’s disease is linked to the accumulation of the misfolded protein amyloid beta (Aß) in the brain. The hypothesis evolved from an experiment where administration of anti-Aß antibodies lowered brain Aß levels in mice even though the antibodies had not penetrated the blood-brain barrier. This suggested that Aß is cleared by moving it from the brain to the cerebrospinal fluid (CSF), then to the peripheral blood, and out of the body. Therefore, if you clear Aß from the blood, this makes room for more Aß to trickle down.
In the blood, more than 90% of Aß is tied to molecules of another protein, albumin. According to this hypothesis, if we replace albumin from the patient’s blood with fresh albumin via PE, this should also remove a lot of Aß and increase the trickle down effect.
AMBAR’s primary endpoint results caused some furor back in the day. They were not clear-cut but also not as disappointing as the results of the vast majority of other AD trials. Statistically significant slowing in AD progression was observed in the range of 42%-70%, depending on the test, the amount of albumin in the fresh plasma, and the severity of the disease.
Now, the results on the secondary endpoints are in, and they are mostly positive as well. Those secondary endpoints addressed various aspects of cognitive function via a battery of tests, while the primary endpoints were less specific. Importantly, in some cases, there was an upward trend – that is, patients’ scores actually improved during the study period. This is unusual in AD research, where even the most successful interventions merely slow the progression of the disease.
Once again, the results were dependent on the dosage of albumin and on the disease stage. For instance, only the mild-AD / high-dosage group, but not patients with moderate AD, showed significant improvement in language fluency tests. Same goes for mental processing speed tests. The results regarding verbal learning and memory were less affected by the severity of AD but also less pronounced.
Lots of caveats
Unfortunately, deciphering the results might not be easy. After the results on primary endpoints were published, a paper interpreting them and analyzing the AMBAR design appeared. As it points out, despite the results being positive, the mechanism of action remains unclear.
Since plasma may contain many ingredients, it is not easy to determine which are responsible for the positive effect. For instance, mechanisms like inflammation and oxidative stress may contribute to AD’s neuropathology, while PE removes inflammatory cytokines from peripheral blood. Albumin itself is also known to have neuroprotective action unrelated to Aß . Unfortunately, tests that could have ruled out or confirm some of those hypotheses were not performed in AMBAR.
Moreover, there are some discrepancies in the primary endpoints’ and. secondary endpoints’ results. For instance, for the primary endpoints, the effect on patients with moderate AD was more significant, while for the secondary it was the opposite: patients with mild AD had more pronounced benefits.
Importantly, most participants, in addition to albumin, also received immunoglobulin (antibodies) in their plasma transfers to correct for a possible immunological deficit, which is a known side effect of PE. In this new paper, the researchers themselves point out that in some tests, immunoglobulin supplementation accounted for a good deal of the change.
Last but not least, the variance was much higher in the results from the treatment groups compared to the control group. This might mean that while the progression of AD is roughly the same in most people, the PE treatment, for unknown reasons, does not benefit everyone equally.
All those caveats notwithstanding, it is hard not to be optimistic about the results, especially considering the abysmal record of previous AD clinical trials. While AMBAR was designed less than perfectly, and some of its results were not clear-cut, they are good enough to provide a strong incentive for a closer look. Plasma exchange remains a hot topic in longevity research, and AMBAR results strengthen the link between it and aging.
 Boada, M., López, O. L., Olazarán, J., Núñez, L., Pfeffer, M., Puente, O., … & AMBAR Trial Group. (2021). Neuropsychological, neuropsychiatric, and quality-of-life assessments in Alzheimer’s disease patients treated with plasma exchange with albumin replacement from the randomized AMBAR study. Alzheimer’s & Dementia.
 Cummings, J. L., Morstorf, T., & Zhong, K. (2014). Alzheimer’s disease drug-development pipeline: few candidates, frequent failures. Alzheimer’s research & therapy, 6(4), 1-7.
 Khatri, R., Afzal, M. R., Rodriguez, G. J., Maud, A., Miran, M. S., Qureshi, M. A., … & Qureshi, A. I. (2018). Albumin-induced neuroprotection in focal cerebral ischemia in the ALIAS trial: does severity, mechanism, and time of infusion matter?. Neurocritical care, 28(1), 60-64.