Responders vs. non-responders to mesenchymal stromal cells in knee osteoarthritis patients: mechanistic correlates of donor cell attributes and patient features
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Background
Despite recent regulatory approvals for mesenchymal stromal cells (MSCs), including two products for knee and degenerative arthritis globally, MSCs have shown mixed efficacy in advanced clinical trials of knee osteoarthritis (KOA). Analysis of contributing factors to the underlying heterogeneity in patient responsiveness is needed to comprehend these mixed results and to advance clinical translation.
Methods
We provide the basis for a novel algorithm that correlates with responder (defined by stringent international consensus criteria) vs. non-responder classifications using i) the immunomodulatory baseline features of bone m arrow-derived MSCs (MSC(M)) batches, and ii) differential KOA baseline clinical features, local and systemic biomarkers. We use our previous clinical trial with twelve batches of MSC(M) as a test dataset and demonstrate correlations with new longitudinal medium-and long-term clinical data.
Results
We report that KOA clinical responsiveness to MSC treatment is dictated by a combination of the immunomodulatory fitness of the donor MSC(M) batches, clinical phenotype, baseline immune cell profile, and biomarker profile. Specifically, MSC(M) in KOA clinical responders have reduced expression of angiogenic genes, THBS1 , CXCL8 , and ANGPT1 and higher expression of growth factor CCN2 and distinct microRNA signatures, as well as higher in vitro ability to modulate monocytes that significantly distinguish them from MSC(M) from non-responders. Clinical responders had greater OA baseline severity, reduced physical activity; lower local baseline levels of CCL2, VEGFA, HGF; higher local baseline of resistin; and putatively lower frequencies of joint-specific intermediate monocytes/macrophages and activated CD4 T cells. Leave-out-one sensitivity analysis confirmed these findings mitigating small sample size concerns. Importantly, predicted responder status at 12 and 24 months was maintained out to 8-10 years in 5/7 responders and 3/3 non-responders, confirming the utility of our approach.
Conclusions
We provide a systematic, quantitative, evidence-based method for analyzing heterogeneity in KOA patient responsiveness to MSC treatments, dually incorporating batch/donor and patient factors. Our combinatorial analytical method provides a powerful framework for parsing heterogeneity in KOA patient responsiveness to MSC therapies and defining critical quality attributes that can reliably identify potent batches and/or donors of MSCs.
Significance statement
Using our data-rich clinical trial with 12-, 24-months and 8-10-year longitudinal follow-up data, we assess donor/batch cell attributes and patient factors that drive clinical efficacy. Bone Marrow MSC(M) from clinical responders exhibited lower angiogenic gene expression, greater ability to modulate monocytes and differential microRNA expression. Responsive KOA patients had more severe KOA, lower physical activity levels, and lower angiogenic proteins in their joints at baseline. Our quantitative analytical framework can be used prospectively in designing future clinical trials and retrospectively in analyzing MSC effectiveness in KOA patients.
