Table 3.

Available evidence of validity of disease markers by clinical aim

Marker	Tools for measurement	Criteria
		Specific*	Validated‡	Precise§	Reliable	Noninvasive	Simple to perform	Inexpensive
Diagnosing incipient Alzheimer disease
Medial temporal atrophy on visual rating	Visual rating scales	No	Yes	Moderately sensitive to early disease but poorly specific	Yes	Yes	Yes	Yes
Hippocampal atrophy on volumetry	Manual tracing, automated tools	No	Yes	Moderately sensitive to early disease but poorly specific	Yes	Yes	Y/N	No
Entorhinal and parahippocampal atrophy on volumetry	Manual tracing, thickness measurement	No	No	Moderately sensitive to early disease but poorly specific	Y/N	Yes	No	No
Three-dimensional atrophy patterns	Support vector machines, STAND score	Yes	Yes	Yes	Y/N	Yes	Yes	Y/N
Tracking progression in clinical trials
Hippocampal atrophy rate	Manual tracing, automated tools	No	NA	NA	Yes	Yes	Y/N	NA
Cortical thinning pattern	Freesurfer, cortical pattern matching, CIVET algorithm	Yes	NA	NA	Yes	Yes	Y/N	NA
Ventricular dilation rate	Threshold-based semi-automated measure, boundary shift integral	No	NA	NA	Yes	Yes	Yes	Yes
Whole-brain atrophy rate	Boundary shift integral, SIENAX software	No	NA	NA	Yes	Yes	No	NA

Diagnosis is considered as a one-time assessment. Criteria for validity were originally developed for diagnosis¹⁴² and adapted here to track disease progression.

^*Able to detect a fundamental feature of Alzheimer disease neuropathology.

^‡Validated in neuropathologically confirmed Alzheimer disease cases.

^§Able to detect Alzheimer disease early in its course and distinguish it from other dementias.

Abbreviations: NA, not applicable; STAND, structural abnormality index; Y/N, moderately or uncertain.