Research Article Biomarkers of Rehabilitation Therapy Vary according to Stroke Severity Erin Burke Quinlan, 1 Lucy Dodakian, 2 Jill See, 2 Alison McKenzie, 3 Jill Campbell Stewart, 4 and Steven C. Cramer 1,2 1 Department of Anatomy & Neurobiology, University of California, Irvine, CA, USA 2 Department of Neurology, University of California, Irvine, CA, USA 3 Department of Physical Therapy, Chapman University, Orange, CA, USA 4 Department of Exercise Science, University of South Carolina, Columbia, SC, USA Correspondence should be addressed to Steven C. Cramer; scramer@uci.edu Received 1 September 2017; Revised 10 January 2018; Accepted 23 January 2018; Published 12 March 2018 Academic Editor: Malgorzata Kossut Copyright © 2018 Erin Burke Quinlan et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Biomarkers that capture treatment effects could improve the precision of clinical decision making for restorative therapies. We examined the performance of candidate structural, functional, and angiogenesis-related MRI biomarkers before and after a 3-week course of standardized robotic therapy in 18 patients with chronic stroke and hypothesized that results vary significantly according to stroke severity. Patients were 4.1 ± 1 months poststroke, with baseline arm Fugl-Meyer scores of 20–60. When all patients were examined together, no imaging measure changed over time in a manner that correlated with treatment-induced motor gains. However, when also considering the interaction with baseline motor status, treatment-induced motor gains were significantly related to change in three functional connectivity measures: ipsilesional motor cortex connectivity with (1) contralesional motor cortex (p =0 003), (2) contralesional dorsal premotor cortex (p =0 005), and (3) ipsilesional dorsal premotor cortex (p =0 004). In more impaired patients, larger treatment gains were associated with greater increases in functional connectivity, whereas in less impaired patients larger treatment gains were associated with greater decreases in functional connectivity. Functional connectivity measures performed best as biomarkers of treatment effects after stroke. The relationship between changes in functional connectivity and treatment gains varied according to baseline stroke severity. Biomarkers of restorative therapy effects are not one-size-fits-all after stroke. 1. Introduction Restorative therapies that promote plasticity within surviving neural tissue [1] can improve recovery, but patient responses are highly variable. Identifying biomarkers that provide information about neural events underlying treatment- related gains could improve individualization of rehabilita- tion therapy [2], as stroke rehabilitative care clinical decision making is still primarily based on behavioral assessments [3]. A biomarker can be defined as a laboratory measurement reflecting the activity of a disease process [4] that changes in parallel with clinical status [5]. Imaging techniques have been identified that are related to neural events underlying brain plasticity [6, 7] and are candidate biomarkers of treatment-induced motor gains after stroke [8–10], including functional MRI (fMRI) [11] and diffusion tensor imaging (DTI) [7, 12, 13]. Restorative therapies in stroke animal models have been associated with induction of angiogenesis derived from T2 * -weighted susceptibility-weighted imaging (SWI) [14], but this MRI-based measure has been little stud- ied in humans as a potential biomarker related to plasticity. The current study compared the utility of these imaging- based biomarker candidates based on the extent to which each changed in parallel with motor gains across a course of therapy in patients early in the chronic phase of stroke. The first study hypothesis was that motor gains across three Hindawi Neural Plasticity Volume 2018, Article ID 9867196, 8 pages https://doi.org/10.1155/2018/9867196