On the Choice of Longitudinal Models for the Analysis of Antitumor Efficacy in Mouse Clinical Trials of Patient-derived Xenograft Models.

In translational oncology research, the patient-derived xenograft (PDX) model and its use in mouse clinical trials (MCT) are increasingly described. This involves transplanting a human tumor into a mouse and studying its evolution during follow-up or until death. A MCT contains several PDXs in which several mice are randomized to different treatment arms. Our aim was to compare longitudinal modeling of tumor growth using mixed and joint models. Mixed and joint models were compared in a real MCT (N = 225 mice) to estimate the effect of a chemotherapy and a simulation study. Mixed models assume that death is predictable by observed tumor volumes (data missing at random, MAR) while the joint models assume that death depends on nonobserved tumor volumes (data missing not at random, MNAR). In the real dataset, of 103 deaths, 97 mice were sacrificed when reaching a predetermined tumor size (MAR data). Joint and mixed model estimates of tumor growth slopes differed significantly [0.24 (0.13;0.36)log(mm3)/week for mixed model vs. -0.02 [-0.16;0.11] for joint model]. By disrupting the MAR process of mice deaths (inducing MNAR process), the estimate of the joint model was 0.24 [0.04;0.45], close to mixed model estimation for the original dataset. The simulation results confirmed the bias in the slope estimate from the joint model. Using a MCT example, we show that joint model can provide biased estimates under MAR mechanisms of dropout. We thus recommend to carefully choose the statistical model according to nature of mice deaths. Significance: This work brings new arguments to a controversy on the correct choice of statistical modeling methods for the analysis of MCTs. We conclude that mixed models are more robust than joint models.

Liposomal Irinotecan Shows a Larger Therapeutic Index than Non-liposomal Irinotecan in Patient-Derived Xenograft Models of Pancreatic Cancer.

INTRODUCTION: Liposomal irinotecan promotes controlled sustained release of irinotecan (CPT-11), therefore, we hypothesize that the therapeutic index (quantitative measurement of the relative efficacy/safety ratio of a drug) will be higher for liposomal than non-liposomal irinotecan. METHODS: We compared the therapeutic indexes of liposomal and non-liposomal irinotecan in mice bearing subcutaneous patient-derived xenograft (PDX) pancreatic tumors under dosing regimens approximating the clinical setting. Following preliminary drug sensitivity/antitumor activity analyses on three PDX tumor models, one model was selected for analyses of efficacy, biomarker, toxicology, pharmacokinetics in mice receiving liposomal irinotecan (2.5, 10, 50 mg/kg/week) or non-liposomal irinotecan (10, 25, 50 mg/kg/week). The maximum tolerated dose (MTD) for each treatment was 50 mg/kg/week. RESULTS: Using the selected IM-PAN-001 model at the MTD (both treatments, 50 mg/kg/week), antitumor activity, phospho-histone gamma-H2AX protein staining in cancer cell nuclei, histological tumor regression, and plasma levels of CPT-11 and its active metabolite SN-38 after 24 h were greater with liposomal than non-liposomal irinotecan, but tumor SN-38 levels were similar. At the lowest doses assessed, antitumor activity, histological tumor regression, and jejunum and bone marrow toxicity were similar. Based on these findings, liposomal and non-liposomal irinotecan had therapeutic indexes of 20 and 5, respectively. CONCLUSION: This non-clinical study showed a fourfold broader therapeutic index with liposomal than non-liposomal irinotecan in mice bearing IM-PAN-001 PDX pancreatic tumors, even at optimal dosing for the two drugs. These findings support the clinical benefit observed with liposomal irinotecan in patients with pancreatic cancer.

Exploring the Complementarity of Pancreatic Ductal Adenocarcinoma Preclinical Models.

Purpose: Compare pancreatic ductal adenocarcinoma (PDAC), preclinical models, by their transcriptome and drug response landscapes to evaluate their complementarity. Experimental Design: Three paired PDAC preclinical models-patient-derived xenografts (PDX), xenograft-derived pancreatic organoids (XDPO) and xenograft-derived primary cell cultures (XDPCC)-were derived from 20 patients and analyzed at the transcriptomic and chemosensitivity level. Transcriptomic characterization was performed using the basal-like/classical subtyping and the PDAC molecular gradient (PAMG). Chemosensitivity for gemcitabine, irinotecan, 5-fluorouracil and oxaliplatin was established and the associated biological pathways were determined using independent component analysis (ICA) on the transcriptome of each model. The selection criteria used to identify the different components was the chemosensitivity score (CSS) found for each drug in each model. Results: PDX was the most dispersed model whereas XDPO and XDPCC were mainly classical and basal-like, respectively. Chemosensitivity scoring determines that PDX and XDPO display a positive correlation for three out of four drugs tested, whereas PDX and XDPCC did not correlate. No match was observed for each tumor chemosensitivity in the different models. Finally, pathway analysis shows a significant association between PDX and XDPO for the chemosensitivity-associated pathways and PDX and XDPCC for the chemoresistance-associated pathways. Conclusions: Each PDAC preclinical model possesses a unique basal-like/classical transcriptomic phenotype that strongly influences their global chemosensitivity. Each preclinical model is imperfect but complementary, suggesting that a more representative approach of the clinical reality could be obtained by combining them. Translational Relevance: The identification of molecular signatures that underpin drug sensitivity to chemotherapy in PDAC remains clinically challenging. Importantly, the vast majority of studies using preclinical in vivo and in vitro models fail when transferred to patients in a clinical setting despite initially promising results. This study presents for the first time a comparison between three preclinical models directly derived from the same patients. We show that their applicability to preclinical studies should be considered with a complementary focus, avoiding tumor-based direct extrapolations, which might generate misleading conclusions and consequently the overlook of clinically relevant features.

Mechanical induction of the tumorigenic ß-catenin pathway by tumour growth pressure.

The tumour microenvironment may contribute to tumorigenesis owing to mechanical forces such as fibrotic stiffness or mechanical pressure caused by the expansion of hyper-proliferative cells. Here we explore the contribution of the mechanical pressure exerted by tumour growth onto non-tumorous adjacent epithelium. In the early stage of mouse colon tumour development in the Notch(+)Apc(+/1638N) mouse model, we observed mechanistic pressure stress in the non-tumorous epithelial cells caused by hyper-proliferative adjacent crypts overexpressing active Notch, which is associated with increased Ret and ß-catenin signalling. We thus developed a method that allows the delivery of a defined mechanical pressure in vivo, by subcutaneously inserting a magnet close to the mouse colon. The implanted magnet generated a magnetic force on ultra-magnetic liposomes, stabilized in the mesenchymal cells of the connective tissue surrounding colonic crypts after intravenous injection. The magnetically induced pressure quantitatively mimicked the endogenous early tumour growth stress in the order of 1,200 Pa, without affecting tissue stiffness, as monitored by ultrasound strain imaging and shear wave elastography. The exertion of pressure mimicking that of tumour growth led to rapid Ret activation and downstream phosphorylation of ß-catenin on Tyr654, imparing its interaction with the E-cadherin in adherens junctions, and which was followed by ß-catenin nuclear translocation after 15 days. As a consequence, increased expression of ß-catenin-target genes was observed at 1 month, together with crypt enlargement accompanying the formation of early tumorous aberrant crypt foci. Mechanical activation of the tumorigenic ß-catenin pathway suggests unexplored modes of tumour propagation based on mechanical signalling pathways in healthy epithelial cells surrounding the tumour, which may contribute to tumour heterogeneity.

CD47 agonist peptides induce programmed cell death in refractory chronic lymphocytic leukemia B cells via PLCγ1 activation: evidence from mice and humans.

BACKGROUND: Chronic lymphocytic leukemia (CLL), the most common adulthood leukemia, is characterized by the accumulation of abnormal CD5+ B lymphocytes, which results in a progressive failure of the immune system. Despite intense research efforts, drug resistance remains a major cause of treatment failure in CLL, particularly in patients with dysfunctional TP53. The objective of our work was to identify potential approaches that might overcome CLL drug refractoriness by examining the pro-apoptotic potential of targeting the cell surface receptor CD47 with serum-stable agonist peptides. METHODS AND FINDINGS: In peripheral blood samples collected from 80 patients with CLL with positive and adverse prognostic features, we performed in vitro genetic and molecular analyses that demonstrate that the targeting of CD47 with peptides derived from the C-terminal domain of thrombospondin-1 efficiently kills the malignant CLL B cells, including those from high-risk individuals with a dysfunctional TP53 gene, while sparing the normal T and B lymphocytes from the CLL patients. Further studies reveal that the differential response of normal B lymphocytes, collected from 20 healthy donors, and leukemic B cells to CD47 peptide targeting results from the sustained activation in CLL B cells of phospholipase C gamma-1 (PLCγ1), a protein that is significantly over-expressed in CLL. Once phosphorylated at tyrosine 783, PLCγ1 enables a Ca2+-mediated, caspase-independent programmed cell death (PCD) pathway that is not down-modulated by the lymphocyte microenvironment. Accordingly, down-regulation of PLCγ1 or pharmacological inhibition of PLCγ1 phosphorylation abolishes CD47-mediated killing. Additionally, in a CLL-xenograft model developed in NOD/scid gamma mice, we demonstrate that the injection of CD47 agonist peptides reduces tumor burden without inducing anemia or toxicity in blood, liver, or kidney. The limitations of our study are mainly linked to the affinity of the peptides targeting CD47, which might be improved to reach the standard requirements in drug development, and the lack of a CLL animal model that fully mimics the human disease. CONCLUSIONS: Our work provides substantial progress in (i) the development of serum-stable CD47 agonist peptides that are highly effective at inducing PCD in CLL, (ii) the understanding of the molecular events regulating a novel PCD pathway that overcomes CLL apoptotic avoidance, (iii) the identification of PLCγ1 as an over-expressed protein in CLL B cells, and (iv) the description of a novel peptide-based strategy against CLL.

Caspase-independent type III programmed cell death in chronic lymphocytic leukemia: the key role of the F-actin cytoskeleton.

BACKGROUND: Programmed cell death has been traditionally related with caspase activation. However, it is now accepted that caspase-independent forms of programmed cell death also regulate cell death. In chronic lymphocytic leukemia, CD47 ligation induces one of these alternative forms of cell death: type III programmed cell death. This poorly understood process is characterized by cytoplasmic hallmarks, such as mitochondrial damage. To gain insights into the molecular pathways regulating type III programmed cell death in chronic lymphocytic leukemia, we performed extensive biochemical and cell biology assessments. DESIGN AND METHODS: After CD47 triggering, purified B-cells from 20 patients with chronic lymphocytic leukemia were studied by flow cytometry, immunofluorescence and three-dimensional imaging, immunoblotting, electron microscopy, and fibrillar/globular actin measurements. Finally, we subjected CD47-treated chronic lymphocytic leukemia cells to a phagocytosis assay. RESULTS: We first confirmed that induction of type III programmed cell death is an efficient means of triggering cell death in chronic lymphocytic leukemia. Further, we demonstrated that the signaling events induced by CD47 ligation provoked a reduction in cell size. This alteration is related to F-actin disruption, as the two other cytoskeleton networks, microtubules and intermediate filaments, remain undisturbed in type III programmed cell death. Strikingly, we revealed that the pharmacological modulation of F-actin dynamics regulated this type of death. Finally, our data delineated a new programmed cell death pathway in chronic lymphocytic leukemia initiated by CD47 triggering, and followed by serine protease activation, F-actin rearrangement, mitochondrial damage, phosphatidylserine exposure, and cell clearance. CONCLUSIONS: Our work reveals a key molecular tool in the modulation of cell death in chronic lymphocytic leukemia: F-actin. By assessing the regulation of F-actin and type III programmed cell death, this analysis provides new options for destroying chronic lymphocytic leukemia cells, such as a combination of therapies based on apoptosis regulators (e.g., caspases, Bcl-2, Bax) along with alternative therapies based on type III death effectors (e.g., F-actin).

Drp1 mediates caspase-independent type III cell death in normal and leukemic cells.

Ligation of CD47 triggers caspase-independent programmed cell death (PCD) in normal and leukemic cells. Here, we characterize the morphological and biochemical features of this type of death and show that it displays the hallmarks of type III PCD. A molecular and biochemical approach has led us to identify a key mediator of this type of death, dynamin-related protein 1 (Drp1). CD47 ligation induces Drp1 translocation from cytosol to mitochondria, a process controlled by chymotrypsin-like serine proteases. Once in mitochondria, Drp1 provokes an impairment of the mitochondrial electron transport chain, which results in dissipation of mitochondrial transmembrane potential, reactive oxygen species generation, and a drop in ATP levels. Surprisingly, neither the activation of the most representative proapoptotic members of the Bcl-2 family, such as Bax or Bak, nor the release of apoptogenic proteins AIF (apoptosis-inducing factor), cytochrome c, endonuclease G (EndoG), Omi/HtrA2, or Smac/DIABLO from mitochondria to cytosol is observed. Responsiveness of cells to CD47 ligation increases following Drp1 overexpression, while Drp1 downregulation confers resistance to CD47-mediated death. Importantly, in B-cell chronic lymphocytic leukemia cells, mRNA levels of Drp1 strongly correlate with death sensitivity. Thus, this previously unknown mechanism controlling caspase-independent type III PCD may provide the basis for novel therapeutic approaches to overcome apoptotic avoidance in malignant cells.