Home > Research > Scientific Programmes: Human Cancer Genetics Programme > Hereditary Endocrine Cancer Group

Human Cancer Genetics Programme

Hereditary Endocrine Cancer Group

Group Leader:  Mercedes Robledo
Research highlights
A gain-of-function mutation in DNMT3A causes paraganglioma

The high percentage of patients carrying germline mutations makes pheochromocytomas (PCC) and paragangliomas (PGL) the most heritable of all tumours. However, there are still cases that are not explained by mutations in the known susceptibility genes. We aimed to identify the genetic cause in patients strongly suspected of having hereditary tumours. Whole-exome sequencing was applied to the germline of a parentproband trio (FIGURE). Genome-wide methylome analysis of mutated tissues and targeted deep sequencing of 112 additional samples were also performed. Exome sequencing identified a single, novel de novo mutation in DNMT3A, DNA (Cytosine5-)-Methyltransferase 3 Alpha, affecting a highly conserved residue located close to the aromatic cage responsible for binding the protein to trimethylated histone H3. DNMT3A-mutated tumour and blood tissue from the patient exhibited significant (FDR<0.15) hypermethylation of homeobox-containing genes, providing evidence that the mutation plays an activating role. Targeted deep sequencing revealed the presence of subclonal mutations affecting the same residue in six additional PGLs, all of which exhibited positive staining for H3K9me3. The case described herein not only increases the number of known PCC/ PGL susceptibility genes, but also represents, to the best of our knowledge, the first example of a gain-of-function mutation affecting a DNA methyl transferase gene involved in cancer predisposition.

Multilayer OMIC data in Medullary Thyroid carcinoma identifies the STAT3 pathway as a potential therapeutic target in RETM918T tumours

Medullary thyroid carcinoma (MTC) is a rare disease with few genetic drivers that, when diagnosed at an advanced stage, remains incurable. Due to its rarity, its genomic dissection has not been comprehensively explored. Exploiting multilayer genomic data, considering the transcriptome, miRNome and methylome, it was possible to uncover genes negatively regulated by methylation, such as DKK4, PLCB2, MMP20, miR-10a, miR-30a and miR-200c, using MZ-CRC-1 and TT cell lines. Moreover, hypomethylation may induce activation of key pathways related to the malignant behaviour of RETM918T -related MTCs. Functional annotation enrichment analysis identified the JAK/Stat pathway as a specific hallmark of RETM918T -harbouring MTCs. In vitro studies with MTC cell models pointed to a RETM918T genetic class-specific proliferative dependency on STAT3 activity. Remarkably, the inhibition of STAT3 increased the sensitivity of RETM918T -bearing MTC cells to the FDA-approved RET inhibitor Vandetanib. This combinational treatment could potentially overcome the adverse effects encountered in clinical practice when Vandetanib monotherapy is applied.

Identification of germline genetic variants and tumour microRNAs to predict outcomes in cancer therapies

Personalised cancer treatment is of enormous clinical and social relevance since it can lead to safer and more efficient therapies. This year we focused our efforts on applying next generation sequencing to: i) understand how low frequency genetic variants impact paclitaxel-induced neuropathy, and ii) identify microRNAs predictive of the antiangiogenic drug response in renal cancer patients. Peripheral neuropathy diminishes the quality of life of many cancer patients, sometimes permanently, and limits the dose and efficacy of many cancer drugs. We found that low frequency variants in EPHA6, EPHA5 and EPHA8 genes contribute to the susceptibility to paclitaxelinduced neuropathy. Furthermore, EPHAs neuronal injury repair function suggests that these genes might constitute important neuropathy markers for many neurotoxic drugs. Regarding antiangiogenic therapies, these have drastically improved the survival of kidney cancer patients; however, a fraction of the patients are refractory to these drugs. The first miRNome deepsequencing study on an exceptional series of patients treated with sunitinib revealed microRNAs predictive of sunitinib response. Furthermore, a two microRNA-based classifier discriminated individuals with progressive disease upon sunitinib treatment (P=1.3x10-4) with better predictive value than the commonly used clinicopathological risk factors. Thus, we provide new relevant markers that can help rationalise cancer treatment.