Computed Tomography Density and β-Amyloid Deposition of Intraorbital Optic Nerve May Assist in Diagnosing Mild Cognitive Impairment and Alzheimer’s Disease: A 18F-Flutemetamol Positron Emission Tomography/Computed Tomography Study

Han Wu, Zhe Lei, Yinghui Ou, Xin Shi, Qian Xu, Keqing Shi, Jing Ding, Qianhua Zhao, Xiuzhe Wang, Xiaolong Cai, Xueyuan Liu, Jingjing Lou, Xingdang LiuFrontiers in Aging Neuroscience 2022 · DOI

TL;DR

This study reveals that the intraorbital optic nerve—an extension of the central nervous system—exhibits measurable amyloid deposition and structural thinning in patients with Alzheimer’s Disease (AD) and Mild Cognitive Impairment (MCI). By combining PET amyloid imaging with standard CT density measurements, researchers achieved high diagnostic accuracy, suggesting the optic nerve could serve as a powerful "ocular window" for early dementia detection.

What Problem Does This Paper Address?

Diagnosing Alzheimer’s Disease in its early stages remains a significant clinical challenge. While brain-based biomarkers like $\beta$-amyloid (Aβ) plaques are the gold standard, PET scans are expensive, and early-stage changes in the brain can sometimes be subtle.

Since the optic nerve is embryologically and anatomically an extension of the brain, researchers have long suspected that it might mirror the neurodegeneration occurring in the cerebral cortex. However, few studies have quantified both the chemical (amyloid) and structural (density) changes of the optic nerve in living patients. This paper investigates whether the intraorbital optic nerve can provide accessible, quantifiable biomarkers to distinguish between healthy aging, MCI, and AD.

Methods

The researchers utilized $^{18}$F-flutemetamol PET/CT imaging, a technique that allows for the simultaneous visualization of amyloid deposits and anatomical structures.

Key technical aspects included:

  • SUVR Calculation: The Standardized Uptake Value Ratio (SUVR) was used to quantify amyloid deposition, with the cerebellum serving as the reference region.
  • CT Density: The researchers measured Hounsfield Units (HU) to determine the physical radio-density of the optic nerve, which reflects its structural integrity.
  • Blinded Analysis: Two independent physicians, blinded to the clinical status of the subjects, performed the measurements using a GE AW 4.5 Workstation.
  • Statistical Modeling: The team used simple linear regression to find correlations and multiple logistic regression to build diagnostic models for classifying patients.

Datasets

The study involved a clinical cohort of 60 subjects recruited from hospitals in Shanghai, China:

  • Normal Controls (NC): 9 subjects.
  • Mild Cognitive Impairment (MCI): 24 subjects.
  • Alzheimer’s Disease (AD): 27 subjects.

The authors noted that recruiting healthy elderly controls was particularly difficult due to local cultural hesitations regarding ionizing radiation, resulting in a smaller control group compared to the patient groups.

Key Insights and Ideas

  • The Optic Nerve as a Proxy: The study confirmed a strong correlation between amyloid levels in the brain's occipital cortex and amyloid levels in the optic nerve. This reinforces the idea that the eye is not just an isolated sensory organ but a direct reflection of brain health.
  • Inverse Relationship of Chemistry and Structure: A critical "aha moment" was the discovery that as amyloid deposition (SUVR) increases, the physical density (HU) of the nerve decreases. This suggests that the accumulation of toxic proteins may be directly linked to the physical "thinning" or degeneration of the nerve fibers.
  • CT Density as a Standalone Tool: Interestingly, the differences in CT density were statistically significant between any two of the three groups. This is promising because CT scans are significantly cheaper and more widely available than PET scans, potentially offering a more accessible screening path.
  • Predictive Power: The researchers found that optic nerve metrics could accurately predict a patient's scores on cognitive tests like the MoCA and MMSE ($R^2$ values of 0.848 and 0.827, respectively).

Key Results

The quantitative findings were robust, particularly for the combined diagnostic models:

  • Diagnostic Accuracy: The Area Under the Curve (AUC) for the multiple logistic regression model was 0.9167 for distinguishing MCI from healthy controls and 0.8951 for distinguishing AD from MCI.
  • Structural Decline: CT density (HU) showed a clear downward trend: NC (~20-24 HU) > MCI (~11-13 HU) > AD (~2-3 HU).
  • Amyloid Increase: Optic nerve SUVR was significantly higher in AD patients compared to healthy controls, mirroring the patterns seen in the brain.

Limitations and Future Work

The authors acknowledge several hurdles:

  • Sample Size: The small number of healthy controls (n=9) limits the generalizability of the findings.
  • Cross-Sectional Design: The study captures a snapshot in time. Longitudinal studies are needed to determine if optic nerve changes occur before or after significant brain amyloidosis.
  • Manual ROI Selection: Measurements were done manually by physicians. Future work could involve automated, AI-driven segmentation to reduce human variability.
  • Mechanistic Clarity: While the correlation is clear, the exact cellular mechanisms—how amyloid leads to decreased CT density in the optic nerve—require further basic science research.

Summary updated

Abstract

The CT density and Aβ deposition of the intraorbital optic nerve were associated with Aβ deposition of the occipital cortex and the severity of cognitive impairment. The intraorbital optic nerve CT density and intraorbital optic nerve Aβ deposition could assist in diagnosing MCI and AD.
BibTeX
@inproceedings{wu2022computed,
  author = {Han Wu and Zhe Lei and Yinghui Ou and Xin Shi and Qian Xu and Keqing Shi and Jing Ding and Qianhua Zhao and Xiuzhe Wang and Xiaolong Cai and Xueyuan Liu and Jingjing Lou and Xingdang Liu and Xingdang Liu and Xingdang Liu},
  title = {{Computed Tomography Density and β-Amyloid Deposition of Intraorbital Optic Nerve May Assist in Diagnosing Mild Cognitive Impairment and Alzheimer’s Disease: A 18F-Flutemetamol Positron Emission Tomography/Computed Tomography Study}},
  booktitle = {Frontiers in Aging Neuroscience},
  year = {2022},
  doi = {10.3389/fnagi.2022.836568},
}
Figure