Mutations in PTF1A cause pancreatic and cerebellar agenesis.

Individuals with permanent neonatal diabetes mellitus usually present within the first three months of life and require insulin treatment. We recently identified a locus on chromosome 10p13-p12.1 involved in permanent neonatal diabetes mellitus associated with pancreatic and cerebellar agenesis in a genome-wide linkage search of a consanguineous Pakistani family. Here we report the further linkage analysis of this family and a second family of Northern European descent segregating an identical phenotype. Positional cloning identified the mutations 705insG and C886T in the gene PTF1A, encoding pancreas transcription factor 1alpha, as disease-causing sequence changes. Both mutations cause truncation of the expressed PTF1A protein C-terminal to the basic-helix-loop-helix domain. Reporter-gene studies using a minimal PTF1A deletion mutant indicate that the deleted region defines a new domain that is crucial for the function of this protein. PTF1A is known to have a role in mammalian pancreatic development, and the clinical phenotype of the affected individuals implicated the protein as a key regulator of cerebellar neurogenesis. The essential role of PTF1A in normal cerebellar development was confirmed by detailed neuropathological analysis of Ptf1a(-/-) mice.

Establishment of In Vivo Acquired Resistance to Chemotherapy Via Individual Dose Escalation Treatment Regime.

The vast majority of cancer deaths are the result of drug resistance. The lack of superior preclinical models that better reflect the complexity of relapsed disease hinders the development of novel therapeutics. 2D and 3D in vitro cell-based assays have provided some information, but this is limited and does not consider the role of the tumor microenvironment. The development of an in vivo assay can allow to generate resistance, while taking into account the role of the tumor microenvironment and the tumor structure. To achieve this, we have developed an in vivo dose-escalation protocol that models the acquisition of resistance. This model of chemo-resistant neuroblastoma presented with metastases and a genetic signature characteristic of clinical relapsed tumors (Yogev et al. Cancer Res. 79:5382-5393, 2019). We believe that this protocol can be used to generate faithful models for other types of relapse disease; these could serve as reliable tools while developing novel therapies.

In Vivo Modeling of Chemoresistant Neuroblastoma Provides New Insights into Chemorefractory Disease and Metastasis.

Neuroblastoma is a pediatric cancer that is frequently metastatic and resistant to conventional treatment. In part, a lack of natively metastatic, chemoresistant in vivo models has limited our insight into the development of aggressive disease. The Th-MYCN genetically engineered mouse model develops rapidly progressive chemosensitive neuroblastoma and lacks clinically relevant metastases. To study tumor progression in a context more reflective of clinical therapy, we delivered multicycle treatment with cyclophosphamide to Th-MYCN mice, individualizing therapy using MRI, to generate the Th-MYCN CPM32 model. These mice developed chemoresistance and spontaneous bone marrow metastases. Tumors exhibited an altered immune microenvironment with increased stroma and tumor-associated fibroblasts. Analysis of copy number aberrations revealed genomic changes characteristic of human MYCN-amplified neuroblastoma, specifically copy number gains at mouse chromosome 11, syntenic with gains on human chromosome 17q. RNA sequencing revealed enriched expression of genes associated with 17q gain and upregulation of genes associated with high-risk neuroblastoma, such as the cell-cycle regulator cyclin B1-interacting protein 1 (Ccnb1ip1) and thymidine kinase (TK1). The antiapoptotic, prometastatic JAK-STAT3 pathway was activated in chemoresistant tumors, and treatment with the JAK1/JAK2 inhibitor CYT387 reduced progression of chemoresistant tumors and increased survival. Our results highlight that under treatment conditions that mimic chemotherapy in human patients, Th-MYCN mice develop genomic, microenvironmental, and clinical features reminiscent of human chemorefractory disease. The Th-MYCN CPM32 model therefore is a useful tool to dissect in detail mechanisms that drive metastasis and chemoresistance, and highlights dysregulation of signaling pathways such as JAK-STAT3 that could be targeted to improve treatment of aggressive disease. SIGNIFICANCE: An in vivo mouse model of high-risk treatment-resistant neuroblastoma exhibits changes in the tumor microenvironment, widespread metastases, and sensitivity to JAK1/2 inhibition.

No evidence for epigenetic inactivation of fumarate hydratase in leiomyomas and leiomyosarcomas.

Germline mutations in Fumarate Hydratase (FH) cause the development of leiomyomas and leiomyosarcomas in the syndromes Multiple Cutaneous and Uterine Leiomyomata (MCUL1) and Hereditary Leiomyomatosis and Renal Cell Cancer (HLRCC). There is little evidence, however, that FH mutation plays a role in the development of sporadic leiomyomas or leiomyosarcomas. Such observations do not, however, exclude a role for FH in tumour development outside the context of MCUL1/HLRCC, as it is possible that FH expression could be silenced by epigenetic mechanisms. To explore this possibility we have developed a highly specific antibody to FH and analysed a series of forty-five fresh-frozen uterine leiomyomas and nine leiomyosarcomas for FH expression.

Overgrowth syndromes: is dysfunctional PI3-kinase signalling a unifying mechanism?

Studies in drosophila and animal models have shown that the phosphoinositide-3-kinase (PI3-kinase) axis plays a central role in normal development, defining the number and size of cells in tissues. Dysfunction of this pathway leads to growth anomalies and has been established to play a key role in the pathogenesis of Cowden syndrome and tuberous sclerosis. It is probable that dysfunction of this pathway is the basis of other disorders especially those typified by asymmetric overgrowth.

BRAF mutations are detectable in conjunctival but not uveal melanomas.

Activating mutations in exon 15 of BRAF have been detected in a high proportion of cutaneous melanomas. To determine whether such mutations are a feature of conjunctival or uveal melanomas, we screened DNA from these tumours. Twenty-one conjunctival and 88 uveal tumours were included in the study. Mutation analysis of BRAF exons 11 and 15 was undertaken using a combination of conformationally sensitive gel electrophoresis and direct sequencing. Mutations in exon 15 were detected in three of the conjunctival tumours (two V599E and one E585 K). None of the uveal tumours possessed a BRAF mutation in either exon 15 or 11. We conclude that uveal melanomas arise independently of oncogenic BRAF mutations, but the development of a proportion of conjunctival tumours involves mutation of this gene.

A new familial cancer syndrome including predisposition to Wilms tumor and neuroblastoma.

Wilms tumor and neuroblastoma are childhood tumors of the kidney and undifferentiated neural crest cells, respectively. Both disorders are primarily sporadic, but familial Wilms tumor pedigrees and familial neuroblastoma pedigrees are each well recognized and account for approximately 1-3% of each tumor type. Families with Wilms tumor and neuroblastoma in the same, or related individuals, have not been reported. Here, we present nine families with two or more individuals with Wilms tumor and/or neuroblastoma. The affected individuals were otherwise well, without syndromic features. Although this co-occurrence might be due to chance in some families, the coexistence of two rare embryonal tumors in related individuals of multiple families suggests an underlying genetic susceptibility to both tumors. We undertook mutational analysis of the genes known to predispose to non-syndromic familial Wilms tumor (WT1) or neuroblastoma (PHOX2B, ALK) which excluded these as the underlying predisposition genes in the nine families. We also excluded epigenetic and copy-number abnormalities at 11p15 which are known to predispose to embryonal tumors including Wilms tumor and neuroblastoma. Overall, these data suggest that families with both Wilms tumor and neuroblastoma represent a previously unrecognized familial cancer syndrome in which the underlying predisposition gene(s) remain to be determined.