Alzheimer’s & Dementia: The Journal of the Alzheimer’s Association

We thank Drs. Di Lorenzo and Koch for sharing their interest and views regarding our recent report on in vivo measurement of synaptic loss in Alzheimer’s disease (AD) using positron emission tomography (PET) to quantify [¹¹C]UCB-J binding to synaptic vesicle glycoprotein 2A (SV2A).¹ We would like to begin by clarifying a few of their comments about our report.

First, we have not provided evidence that synaptic degeneration precedes the occurrence of neurodegeneration and atrophy in early AD. Synaptic loss is presumably one aspect of neurodegeneration and therefore cannot precede it. In our study we observed both synaptic loss and volume loss in participants with early AD, although we strongly emphasized that synaptic PET is measuring more than tissue loss. Most of the reductions we observed in synaptic density were maintained after partial volume correction and thus were not attributable solely to gray matter tissue loss, which appeared to be of lower magnitude and regional extent on structural magnetic resonance imaging (MRI). To determine whether measurable synaptic loss is occurring before any significant atrophy will likely require additional studies beginning at a preclinical stage of disease and extending longitudinally.

Second, our own work does not investigate synaptic plasticity or other forms of synaptic function or efficiency. Although we did not comment on these issues in our article, we do agree with Drs. Di Lorenzo and Koch that the ability to evaluate synaptic function in vivo will likely complement the measurement of synaptic density. In that regard, the authors are to be commended for their pioneering work using transcranial magnetic stimulation (TMS) to investigate in vivo alterations in long-term potentiation (LTP)–like cortical plasticity due to AD.

As suggested by Drs. Di Lorenzo and Koch, the measurement of synaptic density in AD using SV2A PET by our group and others^1-3 may complement the investigation of synaptic function, as exemplified by their own investigation using TMS to evaluate LTP-like cortical plasticity in AD.⁴ Indeed, the reductions in cortical synaptic density measured with SV2A PET presumably reflect a loss of functioning synapses in early AD, leading to impairment in normal neurophysiological processes such as LTP. Furthermore, comparison of the spatial patterns of additional imaging biomarkers for tissue metabolism using fluorodeoxyglucose PET ([¹⁸F]FDG PET) and network connectivity using resting-state functional MRI will enable a better understanding of the relationship between synaptic loss and neuronal and network integrity. Methods to measure LTP-like cortical plasticity in vivo may thus serve as a valuable adjunct to other synaptic biomarkers of AD.

Finally, Drs. Di Lorenzo and Koch highlight the importance of developing valid and sensitive biomarkers of synaptic pathology as outcome measures in therapeutic trials. These will provide particular value for the early-phase testing of agents that target the preservation or restoration of synapses. Such biomarkers are likely to include both structural markers of synaptic density, as well as functional markers of synaptic efficiency, including [¹⁸F]FDG PET, functional MRI network connectivity, and novel measures of LTP-like cortical plasticity. In summary, we agree that these complementary tools provide numerous opportunities, bringing us closer to understanding and ultimately curing AD.