Aims

Longitudinal assessments of brain morphometry in aging and neurodegeneration are needed as an important reference to be used by clinical trials by Alzheimer’s disease drug candidates.
However, to detect changes at local brain region level, processing pipelines, brain atlases and statistical assessments need to be optimally designed and match to each other.

Methods

We used longitudinal surface based morphometry (SBM) in combination with longitudinal voxel based morphometry (VBM) together with the latest high resolution brain atlas HCP MMP 1.0 for surface-based image parcellation as suggested by the Human Connectome Project. Moreover, we corrected for variations in the observed time interval of longitudinal assessments by an Annual Percent Change (APC) formular using a multiplicative model for observed changes. Finally, we describe criteria for an adequate statistical multiple testing correction for high resolution imaging results. We test our approach at the example of 25 Alzheimer’s disease patients and 25 cognitively normal controls from the ADNI3 study.

Results

We found 3 ROI-based measurements from longitudinal VBM and 22 ROI-based measurements from longitudinal SBM, which showed significantly different brain morphometry alterations in patients compared to controls after multiple testing correction. Moreover, the distribution of APC values for cortical volume, area and thickness allowed for a robust separation of the two groups of AD and CN subjects (Fig. 1).

Conclusions

The example demonstrated that using precise longitudinal morphometry processing pipelines together with an adequate statistical assessment can reveal significant group differences with a high confidence level even for relatively small study sizes and an increased separability between the groups compared to longitudinal VBM alone. The same imaging and statistics approach is well applicable to clinical trials of Alzheimer’s disease drug candidates and may reveal significant results even with smaller and more economic study sizes.

*Simon Rechberger and Yong Li contributed equally to this work.

Acknowledgements:
This work was supported by Eurostars project E! 113682 MS-CONNECT through the German Federal Ministry of Education and Research (BMBF) under grant number 01QE2025A and by the Human Brain Project of the European Union (GA 945539).

Data used in preparation of this article were obtained from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database (adni.loni.usc.edu). As such, the investigators within the ADNI contributed to the design and implementation of ADNI and/or provided data but did not participate in analysis or writing of this report. A complete listing of ADNI investigators can be found at: http://adni.loni.usc.edu/wp-content/uploads/how_to_apply/ADNI_Acknowledgement_List.pdf