A robust and reproducible connectome fingerprint of ketamine is highly associated with the connectomic signature of antidepressants

Chadi G. Abdallah, Kyung-Heup Ahn, Lynnette A. Averill, Samaneh Nemati, Christopher L. Averill, Samar Fouda, Mohini Ranganathan, Peter Morgan, Deepak Cyril D’Souza, Daniel H. Mathalon, John H. Krystal, Naomi Driesen

Abstract

Over the past decade, various N-Methyl-D-Aspartate (NMDA) modulators have failed in clinical trials, underscoring the challenges of developing novel rapid-acting antidepressants based solely on receptor or regional targets. Identifying the effect of ketamine on brain circuitry and networks is becoming increasingly critical. In this longitudinal fMRI study of 265 participants, we used a validated predictive model to assess brain functional connectivity without seed selection. We identified a robust and reproducible "connectome fingerprint" (CFP) in healthy participants during ketamine infusion, which was consistent with reduced connectivity within primary cortices and increased connectivity between the executive network and the rest of the brain. This Ketamine CFP was highly associated with a previously identified Antidepressant CFP and predicted clinical response to sertraline at 8 weeks in patients with major depressive disorder (MDD).

The Challenge of Rapid-Acting Antidepressants

Major Depressive Disorder (MDD) is a leading cause of disability worldwide. While ketamine has emerged as a revolutionary rapid-acting antidepressant (RAAD), developing new drugs with similar benefits but fewer side effects has proven difficult. Traditional research often focuses narrowly on molecular receptors or specific brain regions. This study argues for a broader perspective: understanding how these drugs reorganize the brain's entire "wiring diagram"—its functional connectome. The goal is to find a biomarker that is robust, reproducible across different groups, and clinically relevant for predicting patient outcomes.

Methods and Datasets

The researchers employed functional connectome fingerprinting, analyzing the strength of connections between 424 different nodes across the whole brain using the Akiki-Abdallah (AA) hierarchical connectivity atlas. This approach utilized Network Restricted Strength (NRS) to manage data complexity while retaining network affiliations like the Default Mode or Central Executive networks.

The study utilized data from 265 participants across four independent trials:

  • Cohort A (n=25): Healthy participants used to identify the initial Ketamine CFP during infusion.
  • Cohorts B (n=22) & C (n=18): Independent healthy samples used to test reproducibility.
  • Cohort D (n=200): MDD patients randomized to sertraline or placebo, used to test clinical relevance.

Key Insights: Reorganizing the Connectome

Ketamine does not simply increase or decrease connectivity globally; it causes a specific reorganization of brain networks. The study identified a "top-down" shift characterized by reduced connectivity within primary sensory areas (visual and sensorimotor) and increased connectivity between the central executive network and the rest of the brain. This suggests that ketamine may enhance top-down emotional regulation by strengthening the executive network's influence over other brain systems.

Connectome maps and brain surface visualizations (Sum of Edges) illustrating the spatial distribution of the Ketamine CFP, highlighting the shift from primary sensory to executive network connectivity.

Key Results: Reproducibility and Clinical Prediction

The Ketamine CFP proved to be remarkably stable and predictive across all cohorts. Key findings include:

  • Generalizability: The fingerprint successfully generalized from the initial training set to Cohorts B and C, proving it was not a statistical fluke.
  • A Common Antidepressant Signature: There was a massive overlap between the Ketamine CFP and a previously identified "Antidepressant CFP," with a correlation of r = 0.95. This suggests that despite different molecular targets, ketamine and traditional antidepressants may converge on a final common pathway in brain circuitry.
  • Early Prediction of Success: In the MDD cohort (Cohort D), changes in the Ketamine CFP measured just one week into treatment significantly predicted clinical improvement by week eight ($r = 0.30, p < 0.001$).
Graphical Abstract: The Ketamine Connectome Fingerprint (CFP) and its high correlation (r=0.95) with the Antidepressant Fingerprint across four independent cohorts.

Limitations and Future Work

While the results are robust, the authors note that the Ketamine CFP was derived during task performance, which may differ from resting-state connectivity. Additionally, while the network shifts are clear, the exact cellular mechanisms—such as changes in synaptic density—remain hypothetical. Future research will focus on whether this fingerprint can serve as a screening tool for new rapid-acting antidepressant candidates in early-phase clinical trials, potentially streamlining drug development.

Summary updated

BibTeX
@inproceedings{abdallah2020robust,
  author = {Chadi G. Abdallah and Kyung-Heup Ahn and Lynnette A. Averill and Samaneh Nemati and Christopher L. Averill and Samar Fouda and Mohini Ranganathan and Peter Morgan and Deepak Cyril D’Souza and Daniel H. Mathalon and John H. Krystal and Naomi Driesen},
  title = {{A robust and reproducible connectome fingerprint of ketamine is highly associated with the connectomic signature of antidepressants}},
  booktitle = {Neuropsychopharmacology},
  year = {2020},
  doi = {10.1038/s41386-020-00864-9},
}
Figure