Introduction: The classical model of blood stem cell differentiation is branch-like linear HSPC progression from progenitor multi/oligo potent stem cells to differentiated mature cell types. An alternative model, referred to as CLOUD-HSPC, proposes bone marrow (BM) HSPCs continuous and direct differentiation into unipotent cells of myeloid lineage (neutrophils (N), basophils (B), eosinophils (E), monocytes (M), mast cells, megakaryocytes (platelets) and erythrocytes without linear stem cell commitment (Velten Nat Cell Biol 2017). Pegfilgrastim (Peg), and the non-colony stimulating agent Plinabulin (Plin) prevent chemotherapy (chemo) induced neutropenia (CIN) in cancer patients (pts), and both exert BM effects leading to CD34+ cell mobilization (Blayney ASH 2017, 2018). In mice, Plin enhances LSK (Lin-Sca+cKit3+) cell differentiation in BM (Ghosh, AACR 2018). Here we clinically validated the CLOUD-HSPC concept by analyzing peripheral blood mononuclear cell (PBMC) counts derived from two bifurcating progenitor cells 1. Granulocyte-Monocyte Progenitor (GMP) producing N,M,E and B and 2. Common Lymphoid Progenitor (CLP) producing lymphocytes (L), after exposure to Peg or Plin in a CIN setting. With linear stem cell commitment, increase in the differentiated mature cell counts would positively inter-correlate, and with CLOUD-HSPC, they would not.
Methods: In the Phase 3 portion of study BPI-2358-105 (NCT03102606), pts with NSCLC, BC or HRPC received pre-medication with dexamethasone (Dex) on day (D) -1,0, and 1 and docetaxel (Doc) on day (D) 1. Pts were randomized 1:1 to either Plin 40 mg (n=52), given 30 min after Doc infusion D1, or Peg 6 mg (n=53), D2. Central laboratory (Covance Laboratory) PBMC counts were obtained at screening, D 1,2, 6,7,8,9,10, 15 in Cycle 1, and correlated with each other at a pre-planned interim analysis. D1 and D2 data points were omitted due to confounding effects of Dex and demargination. Data of Plin and Peg were combined, since both drugs exert BM effects leading to increased N cell counts and showed similar correlation trends.
Results: Maximum increases in % from screening value on D5-D15 of each N,M,E,B and L counts correlated with each other. Pearson correlation coefficients (r) after linear regression and corresponding p-value are summarized below.
Conclusions: There is a linear and positive inter-correlation between differentiated mature cells derived from the GMP and CLP lineage, proving strong clinical evidence against CLOUD-HSPC, but in favor of the classical linear commitment model.
Mohanlal:BeyondSpring Pharmaceuticals: Employment. Blayney:BeyondSpring Pharmaceuticals: Research Funding. Huang:BeyondSpring Pharmaceuticals: Employment.
