The Phase III Randomized Double-blind Multi-center HOVON-113 Clinical trial on Treatment of Severe Steroid-refractory Acute GvHD with Mesenchymal Stromal Cells was conducted from January 2015 to January 2020 in the Netherlands, Germany, Spain, Italy, and Belgium. 42 patients were enrolled and randomized to receive either placebo or mesenchymal stromal cells. Unfortunately the trial had to be terminated prematurely due to low accrual. Several mitigation steps were taken to increase patient enrolment in the study but the numbers remained low. The intended number of patients to be enrolled was not reached. This hampered statistical significance for all analysis in the other work packages.
For the drug product, a fully harmonized production protocol was introduced and a detailed IMPD provided. All MSC products were generated, for the reduced number of 42 patients, before recruitment was terminated. To evaluate the purity and identity of the drug product, an 8-color flow cytometry panel based on the Euroflow platform was developed. A potency assay was designed. For both assays a patent was submitted which is currently under consideration at EPO.
To study the immune reconstitution profile, a dedicated sampling collection and handling protocol was developed. Screening of whole blood and biopsy samples for immune markers was performed using CyTOF analysis. Two marker panels (T-cell and non T-cell panels), consisting of lineage-specific markers as well as a set of differentiation, activation and homing markers, allowed for the identification of distinct innate and B cell populations as well as effector T cell subsets with tissue destructive potential which remained persistently high in aGvHD patients who failed to respond to MSC therapy. These cells are equipped with receptors for migration towards aGvHD tissues (GI-tract and skin).
RNA sequencing of PBMC obtained at various time intervals after MSC infusion was performed. This resulted in the identification of a robust immunological neutrophil activation signature in poorly-surviving patients, while an eligibility signature associated with outcome, could not be established. Neutrophil enrichment changed from immature to a granule-rich inflammatory phenotype. Patients who ultimately not survived, were defined by strong neutrophil activation at day 7 post treatment, that persisted throughout the duration of the study. Soluble CD163 was found to be elevated in plasma of the deceased patient subgroup, possibly reflecting the uncontrollable inflammatory responses occurring in these patients. This was supported by the increase of other pro-inflammatory biomarkers such as CXCL9, CD40, altogether indicating the severity of tissue inflammation and the degree of innate and adaptive immune cell activation.
