Antagonizing the Hedgehog Pathway with Vismodegib Impairs Malignant Pleural Mesothelioma Growth In Vivo by Affecting Stroma.

An autocrine-driven upregulation of the Hedgehog (Hh) signaling pathway has been described in malignant pleural mesothelioma (MPM), in which the ligand, desert Hh (DHH), was produced from tumor cells. However, our investigation revealed that the Hh pathway is activated in both tumor and stroma of MPM tumor specimens and an orthotopic immunocompetent rat MPM model. This was demonstrated by positive immunohistochemical staining of Glioma-associated oncogene 1 (GLI1) and Patched1 (PTCH1) in both tumor and stromal fractions. DHH was predominantly expressed in the tumor fractions. To further investigate the role of the Hh pathway in MPM stroma, we antagonized Hh signaling in the rat model of MPM using a Hh antagonist, vismodegib, (100 mg/kg orally). Daily treatment with vismodegib efficiently downregulated Hh target genes Gli1, Hedgehog Interacting Protein (Hhip), and Ptch1, and caused a significant reduction of tumor volume and tumor growth delay. Immunohistochemical analyses revealed that vismodegib treatment primarily downregulated GLI1 and HHIP in the stromal compartment along with a reduced expression of previously described fibroblast Hh-responsive genes such as Fibronectin (Fn1) and Vegfa Primary cells isolated from the rat model cultured in 3% O2 continued to express Dhh but did not respond to vismodegib in vitro However, culture supernatant from these cells stimulated Gli1, Ptch1, and Fn1 expression in mouse embryonic fibroblasts, which was suppressed by vismodegib. Our study provides new evidence regarding the role of Hh signaling in MPM stroma in the maintenance of tumor growth, emphasizing Hh signaling as a treatment target for MPM. Mol Cancer Ther; 15(5); 1095-105. ©2016 AACR.

Low Merlin expression and high Survivin labeling index are indicators for poor prognosis in patients with malignant pleural mesothelioma.

INTRODUCTION: Alterations of the tumor suppressor Neurofibromatosis type II (NF2) have been reported in about 40% of Malignant pleural mesothelioma (MPM) patients. NF2 (Merlin) deficiency leads to alterations of the Hippo pathway; resulting in activation of the oncogenic Yes Associated Protein-1 (YAP1). Our aim was to investigate the association between these alterations and clinical outcomes. MATERIAL AND METHODS: Tissue microarrays composed of MPM tumors derived from 2 independent MPM cohorts were employed for this study. Immunohistochemical expression of Merlin, YAP1 and its target genes, Survivin and connective tissue growth factor (CTGF) were assessed in nuclear and cytoplasmic fractions. Cohort 1 was comprised of 145 patients intended to be treated with chemotherapy (CTX) followed by extrapleural pneumonectomy (EPP), thus both pre- and post-CTX tissues were available. Cohort 2 was comprised of 59 patients treated with EPP followed by intraoperative hyperthermic cisplatin and/or adjuvant CTX and/or radiotherapy. Marker expression was quantified by means of labeling index (%) for nuclear Survivin and by H-score for the other markers. The dichotomized marker expression was tested for the association with overall survival (OS) and freedom from recurrence (FFR). RESULTS: Kaplan-Meier survival curves revealed a significant association between low cytoplasmic Merlin expression in pre-induction CTX tissues of cohort 1 with shorter FFR (p = 0.02) and OS (p = 0.03). The same tendency was observed in the chemotherapy naïve tissues obtained during EPP of cohort 2. Low nuclear Merlin expression in post-CTX tissues (available from cohort 1 only) was associated with shorter FFR (p = 0.04) and OS (p = 0.05). High nuclear Survivin labeling indices in both pre- and post-CTX tissues of cohort 1 was associated with shorter FFR (p = 0.02). In cohort 2, this was associated with both FFR and OS (p = 0.046 and p = 0.002, respectively). In multivariate analysis, low expression of cytoplasmic Merlin remained an independent prognosticator for shorter FFR of cohort 1 [hazard ratio (HR) = 0.5, 95% confidence interval (CI) = 0.3-0.9, p = 0.001] and OS [HR = 0.5, 95% CI = 0.3-1, p = 0.04]. High Survivin labeling index was an independent prognostic factor for shorter FFR in patients from cohort 1 [HR = 3.4, 95% CI = 1.7-6.8, p = 0.006] and shorter OS in patients from cohort 2 [HR = 2.35, 95% CI = 1.27-4.33, p = 0.006]. CONCLUSIONS: Our findings uncover the significance of Merlin protein expression and Survivin labeling index as prognosticators for poor clinical outcome in two independent MPM cohorts. If confirmed, these markers may be used to identify subgroups of patients benefitting from additional treatment.

Evaluation of imaging techniques for the assessment of tumour progression in an orthotopic rat model of malignant pleural mesothelioma†.

OBJECTIVES: An orthotopic rat tumour recurrence model for malignant pleural mesothelioma (MPM) provides clinical similarity to patients and is useful for drug testing combined with surgical intervention. Importantly, a reliable imaging method is required allowing for noninvasive and repetitive evaluation of the tumour load. We compared the tumour load assessed by bioluminescence and magnetic resonance imaging (MRI) to the macroscopic tumour volume as a reference standard. METHODS: A total of 500,000 syngeneic rat MPM cells transfected with luciferase were implanted underneath the parietal pleura of immunocompetent rats (n=13). From the second day after implantation, bioluminescence measurements of the tumour load expressed as the maximum bioluminescent intensity (photon/second) were performed daily after intraperitoneal injection of the luciferase substrate, d-luciferin, to observe the first occurrence of tumour. Six days after the first detection of tumour, bioluminescence, MRI and macroscopic tumour volume measurement were conducted. For MRI, a 4.7-Tesla small animal imager equipped with a 1H whole-body rat coil was employed using T2-weighted fast spin-echo sequences. Tumour burden (mm3) was quantified from magnetic resonance transverse images by two independent readers by manual segmentation. Finally, the tumour burden assessed by bioluminescence and MRI was correlated (Pearson's correlation) with the macroscopic measurement of tumour (ellipsoid) volume. RESULTS: In all rats, a single tumour nodule was found at the inoculation site with a median macroscopic volume of 46 mm3 (18-377 mm3). For tumour burden quantification of MRIs, we observed good interobserver correlation (R2=0.81, P<0.0001) as well as significant association with the macroscopic tumour volume (R2=0.59, P=0.002). However, the signal intensity of bioluminescence did not correspond to the macroscopic tumour volume (R2=0.01, P=0.76). CONCLUSIONS: MRI is a reliable and reproducible noninvasive in vivo imaging method for MPM tumour burden assessment for the present MPM model.