Prospective cohort study of early biosignatures of response to lithium in bipolar-I-disorders: overview of the H2020-funded R-LiNK initiative

Jan Scott, Diego Hidalgo‐Mazzei, Rebecca Strawbridge, Allan H. Young, Matthieu Resche‐Rigon, Bruno Étain, Ole A. Andreassen, Michael Bauer, Djamila Bennabi, Andrew M. Blamire, Fawzi Boumezbeur, Paolo Brambilla, Nadia Cattane, Annamaria Cattaneo, Marie Chupin, Klara Coello, Yann Cointepas, Francesc Colom, David A. Cousins, Caroline Dubertret, Édouard Duchesnay, Adele Ferro, Aitana García-Estela, José Manuel Goikolea, Antoine Grigis, Émmanuel Haffen, Margrethe Collier Høegh, Petter Jakobsen, János Kálmán, Lars Vedel Kessing, Farah Klöhn‐Saghatolislam, Trine Vik Lagerberg, Mikael Landén, Ute Lewitzka, Ashley L. Lütticke, Nicolas Mazer, Monica Mazzelli, Cristina Mora, Thorsten Müller, Estanislao Mur-Milà, Ketil J. Øedegaard, Leif Oltedal, Erik Pålsson, Dimitri Papadopoulos Orfanos, Sergi Papiol, Víctor Pérez, Andreas Reif, Philipp Ritter, Roberta Rossi, Thomas G. Schulze, Fanny Senner, Fiona E. Smith, Letizia Squarcina, Nils Eiel Steen, Pete Thelwall, Cristina Varo, Eduard Vieta, Maj Vinberg, Michèle Wessa, Lars T. Westlye, Frank BellivierInternational Journal of Bipolar Disorders 2019 · DOI

Abstract

The paper outlines the rationale, design and methodology of the first study being undertaken by the newly established R-LiNK collaboration and describes how the project may help to refine the clinical response phenotype and could translate into the personalization of lithium treatment.

Introduction

Lithium remains the gold standard for treating Bipolar I Disorder (BD-I), yet its clinical application is often hindered by a significant "trial-and-error" period. While highly effective for approximately 30% of patients, determining efficacy currently requires an 18-to-24-month observation period. During this time, non-responders face increased risks of relapse, suicide, and unnecessary side effects from a medication that may not be benefiting them.

The R-LiNK initiative, a large-scale European prospective study, aims to address this unmet clinical need. By tracking over 300 patients using a multi-modal array of "biosignatures," the project seeks to identify early predictors of lithium response within weeks of treatment initiation rather than years.

Methodology and Study Design

The R-LiNK study employs an "integrative science" approach to identify biological markers that predict treatment success. Moving away from the search for a single biomarker, the study utilizes a structured protocol that branches into several specialized phenotyping domains.

Overview of the R-LiNK study protocol and patient flow.

Key methodological pillars include:

  • Multi-modal Imaging: Beyond standard MRI, the study uses Proton Magnetic Resonance Spectroscopy ($^1$H-MRS) for brain metabolism and innovative Lithium-7 MRI ($^7$Li-MRI) to visualize lithium distribution in brain tissue 12 weeks into treatment.
  • Functional Omics: Collection of blood samples to analyze high-dimensional data, including genomics, transcriptomics (gene expression), and metabolomics.
  • Digital Phenotyping: Leveraging smartphones for Ecological Momentary Assessment (EMA)—daily self-ratings of mood—combined with wearable actigraphy to monitor sleep-wake cycles and activity levels.
  • Expert Consensus Panel: To ensure accurate classification, an independent panel of five experts reviews two years of longitudinal data (blinded to biomarker data) to determine each patient's response status.

Project Structure and Aims

The initiative is organized into nine distinct work packages (WPs) designed to manage the complexity of multi-site, multi-modal data collection and analysis across 16 European centers.

Table of R-LiNK work packages and their specific research aims.

Key Insights and Strategy

A central hypothesis of R-LiNK is that lithium response is tied to its distribution across specific brain structures, such as the amygdala, rather than just blood serum levels. The $^7$Li-MRI represents a significant technological leap in testing this theory. Furthermore, the study explores whether "digital signatures"—subtle changes in activity or smartphone interaction—can predict relapses before they manifest clinically.

Uniquely for a biomarker study, R-LiNK includes health economics. Simulation models are being built to determine if the cost of these advanced tests is offset by the long-term savings of getting patients on the correct medication faster.

Discussion and Future Directions

While the current cohort of 300+ patients is robust for neuroimaging, the authors acknowledge the statistical challenges of analyzing high-dimensional "omics" data. The 2-year follow-up window is a significant improvement over previous retrospective studies but may still overlook very late responders.

Ultimately, the R-LiNK infrastructure is intended as a permanent platform for precision psychiatry. By replacing trial-and-error with evidence-based biosignatures, the initiative moves clinical practice toward truly personalized medicine for Bipolar I Disorder, potentially extending its methodology to other mood stabilizers and the broader spectrum of bipolar disorders.

Summary updated

Abstract

The paper outlines the rationale, design and methodology of the first study being undertaken by the newly established R-LiNK collaboration and describes how the project may help to refine the clinical response phenotype and could translate into the personalization of lithium treatment.
BibTeX
@article{scott2019prospective,
  author = {Jan Scott and Diego Hidalgo‐Mazzei and Rebecca Strawbridge and Allan H. Young and Matthieu Resche‐Rigon and Bruno Étain and Ole A. Andreassen and Michael Bauer and Djamila Bennabi and Andrew M. Blamire and Fawzi Boumezbeur and Paolo Brambilla and Nadia Cattane and Annamaria Cattaneo and Marie Chupin and Klara Coello and Yann Cointepas and Francesc Colom and David A. Cousins and Caroline Dubertret and Édouard Duchesnay and Adele Ferro and Aitana García-Estela and José Manuel Goikolea and Antoine Grigis and Émmanuel Haffen and Margrethe Collier Høegh and Petter Jakobsen and János Kálmán and Lars Vedel Kessing and Farah Klöhn‐Saghatolislam and Trine Vik Lagerberg and Mikael Landén and Ute Lewitzka and Ashley L. Lütticke and Nicolas Mazer and Monica Mazzelli and Cristina Mora and Thorsten Müller and Estanislao Mur-Milà and Ketil J. Øedegaard and Leif Oltedal and Erik Pålsson and Dimitri Papadopoulos Orfanos and Sergi Papiol and Víctor Pérez and Andreas Reif and Philipp Ritter and Roberta Rossi and Thomas G. Schulze and Fanny Senner and Fiona E. Smith and Letizia Squarcina and Nils Eiel Steen and Pete Thelwall and Cristina Varo and Eduard Vieta and Maj Vinberg and Michèle Wessa and Lars T. Westlye and Frank Bellivier},
  title = {{Prospective cohort study of early biosignatures of response to lithium in bipolar-I-disorders: overview of the H2020-funded R-LiNK initiative}},
  journal = {International Journal of Bipolar Disorders},
  year = {2019},
  doi = {10.1186/s40345-019-0156-x},
}
Figure