Pre-treatment fMRI working memory activation predicts functional remediation outcomes in older adults with bipolar disorder

Introduction: Functional Remediation (FR) is a structured, group-based intervention that combines cognitive training with real-world strategy coaching to improve daily functioning. In older-age bipolar disorder (OABD), FR has been adapted to address age-related cognitive and functional challenges [1]. However, it remains unclear which individuals benefit most from FR, as treatment response shows substantial inter-individual variability and its neural predictors are poorly understood. Identifying neurobiological markers of treatment response may help stratify patients and optimize intervention strategies [2]. We investigated whether pre-treatment fMRI activation during a working memory task predicts functional improvement following FR in OABD.

Methods: Twenty-two OABD patients in full or partial remission, diagnosed according to DSM-5 criteria by experienced psychiatrists, underwent 3T fMRI during an N-back task prior to FR. Additional participants were excluded due to study non-completion or failure to meet fMRI quality control criteria. Psychosocial functioning was assessed at baseline and post-treatment using the Functioning Assessment Short Test (FAST). fMRI preprocessing and first-level analyses followed standard pipelines, including slice timing, motion correction, spatial normalization, smoothing, and quality control. Individual activation estimates during the high-demand condition (2-back vs. rest) were extracted for four regions of interest (ROIs) selected a priori based on prior literature: left and right dorsolateral prefrontal cortex (dlPFC), right ventrolateral PFC (vlPFC), and dorsomedial PFC (dmPFC). Preregistered primary analyses [3] included linear regression models testing whether activation within each of these four ROIs predicted post-treatment FAST scores, adjusting for baseline FAST and correcting for multiple comparisons. We conducted an exploratory analysis to examine a composite index reflecting the relative activation of executive (bilateral dlPFC) versus monitoring/inhibitory regions (vlPFC and dmPFC). This index was entered as a predictor into a regression model predicting post-treatment FAST, controlling baseline FAST.

Results:Individual ROI analyses did not reach statistical significance, although effect directions were consistent with better outcomes being associated with higher dlPFC and lower vlPFC/dmPFC activation. In the exploratory analysis, greater activation in executive versus monitoring/inhibitory regions was associated with better psychosocial functioning after intervention (B = -3.34; p = 0.013).

Discussion: While the preregistered analyses did not reach statistical significance, the findings from the exploratory analysis suggest that the balance of activation across prefrontal control systems—rather than activity within individual regions—may be relevant for FR outcomes in OABD. This pattern is consistent with the notion that effective engagement of executive control networks during cognitively demanding tasks may support the acquisition and implementation of compensatory strategies targeted by FR. In contrast to studies in younger bipolar populations linking vlPFC activity to cognitive outcomes [4,5], our results point to a potential role of preferential engagement of executive control regions during cognitively demanding tasks in predicting functional improvement. This discrepancy may reflect differences between predictors of cognitive versus functional outcomes, or age-related variation in neural mechanisms of treatment response. Given thesmall sample size and exploratory nature of these analyses, findings should be considered preliminary and require replication in larger samples. If replicated, these findings may inform stratification strategies and personalization of FR in OABD.

References

1. Montejo, L., Jiménez, E., Torrent, C., del Mar Bonnín, C., Solé, B., Martínez-Arán, A., Vieta, E., Sánchez-Moreno, J. (2024). Functional Remediation for Older Adults with Bipolar Disorder (FROA-BD): Study protocol for a randomized controlled trial. Spanish Journal of Psychiatry and Mental Health 17, 181–189.
2. Oliva, V., Fico, G., De Prisco, M., Gonda, X., Rosa, A. R., & Vieta, E. (2024). Bipolar disorders: an update on critical aspects. The Lancet regional health. Europe, 48, 101135. https://doi.org/10.1016/j.lanepe.2024.101135
3. Montejo, L., Sole, B., Fortea, L., Jimenez, E., Martinez-Aran, A., Martinez-Heras, E., Sanchez-Moreno, J., Ortuño, M., Pariente, J., Solanes, A., Torrent, C., Vilajosana, E., De Prisco, M., Vieta, E,, & Radua, J. (2024). Study protocol - elucidating the neural correlates of functional remediation for older adults with bipolar disorder. Front Psychiatry, 14:1302255.
4. Miskowiak, K. W., Møller, A. B., & Ott, C. V. (2021). Neuronal and cognitive predictors of improved executive function following action-based cognitive remediation in patients with bipolar disorder. European neuropsychopharmacology: the journal of the European College of Neuropsychopharmacology, 47, 1–10.
5. Ott, C. V., Macoveanu, J., Bowie, C. R., Fisher, P. M., Knudsen, G. M., Kessing, L. V., & Miskowiak, K. W. (2021). Change in prefrontal activity and executive functions after action-based cognitive remediation in bipolar disorder: a randomized controlled trial. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology, 46(6), 1113–1121.