Hallmark discoveries in the biology of Wilms tumour.

The modern study of Wilms tumour was prompted nearly 50 years ago, when Alfred Knudson proposed the 'two-hit' model of tumour development. Since then, the efforts of researchers worldwide have substantially expanded our knowledge of Wilms tumour biology, including major advances in genetics - from cloning the first Wilms tumour gene to high-throughput studies that have revealed the genetic landscape of this tumour. These discoveries improve understanding of the embryonal origin of Wilms tumour, familial occurrences and associated syndromic conditions. Many efforts have been made to find and clinically apply prognostic biomarkers to Wilms tumour, for which outcomes are generally favourable, but treatment of some affected individuals remains challenging. Challenges are also posed by the intratumoural heterogeneity of biomarkers. Furthermore, preclinical models of Wilms tumour, from cell lines to organoid cultures, have evolved. Despite these many achievements, much still remains to be discovered: further molecular understanding of relapse in Wilms tumour and of the multiple origins of bilateral Wilms tumour are two examples of areas under active investigation. International collaboration, especially when large tumour series are required to obtain robust data, will help to answer some of the remaining unresolved questions.

A Gradual Transition Toward Anaplasia in Wilms Tumor Through Tolerance to Genetic Damage.

Patients with Wilms tumor (WT) in general have excellent survival, but the prognosis of patients belonging to the subgroup of WT with diffuse anaplasia (DA) is poor due to frequent resistance to chemotherapy. We hypothesized that DA WT cells might undergo changes, such as acquiring a persistent tolerance to DNA damage and copy number aberrations (CNAs), which could eventually lead to their resistance to chemotherapy treatment. Tissue sections from chemotherapy-treated DA WTs (n = 12) were compared with chemotherapy-treated nonanaplastic WTs (n = 15) in a tissue microarray system, enabling analysis of 769 tumor regions. All regions were scored for anaplastic features and immunohistochemistry was used to quantify p53 expression, proliferation index (Ki67), and DNA double-strand breaks (γH2AX). CNAs were assessed by array-based genotyping and TP53 mutations using targeted sequencing. Proliferation index and the frequency of DNA double-strand breaks (γH2AX dot expression) increased with higher anaplasia scores. Almost all (95.6%) areas with full-scale anaplasia had TP53 mutations or loss of heterozygosity, along with an increased amount of CNAs. Interestingly, areas with wild-type TP53 with loss of heterozygosity and only one feature of anaplasia (anaplasia score 1) also had significantly higher proliferation indices, more DNA double-strand breaks, and more CNAs than regions without any anaplastic features (score 0); such areas may be preanaplastic cell populations under selective pressure for TP53 mutations. In conclusion, we suggest that chemoresistance of DA WTs may be partly explained by a high proliferative capability of anaplastic cells, which also have a high burden of double-stranded DNA breaks and CNAs, and that there is a gradual emergence of anaplasia in WT.

Gene expression-based dissection of inter-histotypes, intra-histotype and intra-tumor heterogeneity in pediatric tumors.

Intra-tumor heterogeneity (ITH) fosters tumor evolution, resistance to therapy, and relapse. Recently, many evidence have been accumulated on the occurrence of genetic ITH in pediatric cancers. With this study we aimed to address the downstream effects that genetic and epigenetic ITH, and tumor-microenvironment interactions may produce within a tumor mass. To this aim, we investigated by high-throughput gene expression multiple samples of 5 hepatoblastomas, 5 neuroblastomas, 5 rhabdomyosarcomas, and 5 Wilms tumors. Principal component analysis, single sample hallmark gene sets analysis, and weighted gene co-expression network analysis were performed on gene expression data. We observed that the different tumors clustered by histotype, and then by case, and in addition, a variable degree of ITH was visible in all the investigated cases. The ITH highlighted in this study can represent a challenge in tumor treatment since we demonstrated that different druggable hallmarks and targets may be heterogeneously present within the same tumor mass, and this can potentially lead to therapeutic failure. Despite this heterogeneity, we could highlight some commonalities among the different histotypes investigated, supporting the feasibility to move in the clinic from a histotype-driven to a target-driven, sometimes agnostic, approach at least in some cases.

Finding the way to Wilms tumor by comparing the primary and relapse tumor samples.

In this issue of Cell Reports Medicine, Gadd and colleagues presented on behalf of the Children's Oncology Group their comprehensive analysis of genetic changes associated with relapse in children with favorable histology Wilms tumor.

Analysis of the mutational status of SIX1/2 and microRNA processing genes in paired primary and relapsed Wilms tumors and association with relapse.

Whereas 90% of patients with Wilms tumor (WT) reach cure, approximately half of patients developing a recurrent tumor die of the disease. Therefore, to disclose events leading to recurrence represents a clinical need. To study paired primary/recurrent tumor samples, being aware of the intra-tumoral heterogeneity, might help finding these answers. We previously suggested that mutations in SIX1 and DROSHA underlie WT recurrence. With the aim to better investigate this scenario, we collected 19 paired primary/recurrent tumors and 10 primary tumors from relapsing patients and searched for mutations in the SIX1/2 genes and microRNA processing genes (miRNAPGs). We found SIX1 mutation in one case, miRNAPGs mutations in seven cases, and the co-occurrence of SIX1 and miRNAPG mutations in one case. We could observe that, whereas in primary tumors the mutations could be heterogeneously present, in all cases they were positively selected and homogeneously present in the recurrent disease, as also indicated by a "moderate" and "almost perfect" agreement (according to the Landis and Koch classification criteria) between paired samples. Analysis of SIX1/2 genes and miRNAPGs in 50 non-relapsing WTs disclosed SIX2 mutation in one case and miRNAPGs mutations in seven. A borderline statistically significant association was observed between miRNAPGs mutations and the occurrence of relapse (p value: 0.05). These data suggest that SIX1 and miRNAPGs mutations may provide an advantage during tumor progression to recurrence and can represent oncogenic drivers in WT development.

Genetic and epigenetic analyses guided by high resolution whole-genome SNP array reveals a possible role of CHEK2 in Wilms tumour susceptibility.

Wilms tumour (WT), the most frequent malignant childhood renal tumour, shows a high degree of genetic and epigenetic heterogeneity. Loss of imprinting on chromosome 11p15 is found in a large fraction of cases and mutations in a few genes, including WT1, CTNNB1, WTX, TP53 and, more recently, SIX1, SIX2 and micro RNA processing genes (miRNAPGs), have been observed. However, these alterations are not sufficient to describe the entire spectrum of genetic defects underlying WT development. We inspected data obtained from a previously performed genome-wide single nucleotide polymorphism (SNP) array analysis on 96 WT samples. By selecting focal regions commonly involved in chromosomal anomalies, we identified genes with a possible role in WT development, based on the prior knowledge of their biological relevance, including MYCN, DIS3L2, MIR562, HACE1, GLI3, CDKN2A and CDKN2B, PALB2, and CHEK2. The MYCN hotspot mutation c.131C>T was detected in seven cases (7.3%). Full sequencing of the remaining genes disclosed 16 rare missense variants and a splicing mutation. Most of these were present at the germline level. Promoter analysis of HACE1, CDKN2A and CDKN2B disclosed partial methylation affecting HACE1 in a consistent fraction of cases (85%). Interestingly, of the four missense variants identified in CHEK2, three were predicted to be deleterious by in silico analyses, while an additional variant was observed to alter mRNA splicing, generating a functionally defective protein. Our study adds additional information on putative WT genes, and adds evidences involving CHEK2 in WT susceptibility.

Retina-derived POU domain factor 1 coordinates expression of genes relevant to renal and neuronal development.

Retina-derived POU domain Factor 1 (RPF-1), a member of POU transcription factor family, is encoded by POU6F2 gene, addressed by interstitial deletions at chromosome 7p14 in Wilms tumor (WT). Its expression has been detected in developing kidney and nervous system, suggesting an early role for this gene in regulating development of these organs. To investigate into its functions and determine its role in transcriptional regulation, we generated an inducible stable transfectant from HEK293 cells. RPF-1 showed nuclear localization, elevated stability, and transactivation of promoters featuring POU consensus sites, and led to reduced cell proliferation and in vivo tumor growth. By addressing the whole transcriptome regulated by its induction, we could detect a gross alteration of gene expression that is consistent with promoter occupancy predicted by genome-wide Chip-chip analysis. Comparison of bound regulatory regions with differentially expressed genes allowed identification of 217 candidate targets. Enrichment of divergent octamers in predicted regulatory regions revealed promiscuous binding to bipartite POUS and POUH consensus half-sites with intervening spacers. Gel-shift competition assay confirmed the specificity of RPF-1 binding to consensus motifs, and demonstrated that the Ser-rich region upstream of the POU domain is indispensable to achieve DNA-binding. Promoter-reporter activity addressing a few target genes indicated a dependence by RPF-1 on transcriptional response. In agreement with its expression in developing kidney and nervous system, the induced transcriptome appears to indicate a function for this protein in early renal differentiation and neuronal cell fate, providing a resource for understanding its role in the processes thereby regulated.

Chromosomal anomalies at 1q, 3, 16q, and mutations of SIX1 and DROSHA genes underlie Wilms tumor recurrences.

Approximately half of children suffering from recurrent Wilms tumor (WT) develop resistance to salvage therapies. Hence the importance to disclose events driving tumor progression/recurrence. Future therapeutic trials, conducted in the setting of relapsing patients, will need to prioritize targets present in the recurrent lesions. Different studies identified primary tumor-specific signatures associated with poor prognosis. However, given the difficulty in recruiting specimens from recurrent WTs, little work has been done to compare the molecular profile of paired primary/recurrent diseases. We studied the genomic profile of a cohort of eight pairs of primary/recurrent WTs through whole-genome SNP arrays, and investigated known WT-associated genes, including SIX1, SIX2 and micro RNA processor genes, whose mutations have been recently proposed as associated with worse outcome. Through this approach, we sought to uncover anomalies characterizing tumor recurrence, either acquired de novo or already present in the primary disease, and to investigate whether they overlapped with known molecular prognostic signatures. Among the aberrations that we disclosed as potentially acquired de novo in recurrences, some had been already recognized in primary tumors as associated with a higher risk of relapse. These included allelic imbalances of chromosome 1q and of chromosome 3, and CN losses on chromosome 16q. In addition, we found that SIX1 and DROSHA mutations can be heterogeneous events (both spatially and temporally) within primary tumors, and that their co-occurrence might be positively selected in the progression to recurrent disease. Overall, these results provide new insights into genomic and genetic events underlying WT progression/recurrence.

Constitutional de novo deletion of the FBXW7 gene in a patient with focal segmental glomerulosclerosis and multiple primitive tumors.

Multiple primary malignant neoplasms are rare entities in the clinical setting, but represent an important issue in the clinical management of patients since they could be expression of a genetic predisposition to malignancy. A high resolution genome wide array CGH led us to identify the first case of a de novo constitutional deletion confined to the FBXW7 gene, a well known tumor suppressor, in a patient with a syndromic phenotype characterized by focal segmental glomerulosclerosis and multiple primary early/atypical onset tumors, including Hodgkin's lymphoma, Wilms tumor and breast cancer. Other genetic defects may be associated with patient's phenotype. In this light, constitutional mutations at BRCA1, BRCA2, TP53, PALB2 and WT1 genes were excluded by performing sequencing and MLPA analysis; similarly, we ruled out constitutional abnormalities at the imprinted 11p15 region by methylation specific -MLPA assay. Our observations sustain the role of FBXW7 as cancer predisposition gene and expand the spectrum of its possible associated diseases.

PALB2 sequencing in Italian familial breast cancer cases reveals a high-risk mutation recurrent in the province of Bergamo.

PURPOSE: Monoallelic germ-line deleterious mutations of PALB2 (partner and localizer of BRCA2) are associated with breast cancer risk and have been found in several populations, with carrier frequencies of ~1-2%. Initially, these mutations were considered to have moderate penetrance, but accumulating evidence now indicates that they are associated with much higher risk. METHODS: In this study, we sequenced the PALB2 coding regions unlinked to BRCA (breast cancer) genes in 575 probands from Italian breast cancer families recruited in Milan. RESULTS: We found 12 carriers (2.1%) of deleterious mutations, and none of the mutations was found in 784 controls collected in Milan. One of these mutations, the c.1027C>T (p.Gln343X), was found to be recurrent in the province of Bergamo in northern Italy, being detected in 6/113 (5.3%) familial breast cancer cases and 2/477 (0.4%) controls recruited in this area (Fisher's exact test: P < 0.01). CONCLUSIONS: Our data provide confirmatory findings that, in the Italian population also, deleterious mutations of PALB2 are relatively frequent predisposing factors for breast cancer and may be associated with high risk of the disease.