Genomic analysis of childhood tumours; biological and therapeutic implications

Bookmark and Share
Published: 4 Jul 2017
Views: 1959
Rating:
Save
Dr Peter Lichter - German Cancer Research Center (DKFZ), Heidelberg, Germany

Dr Lichter speaks with ecancer at WIN 2017 about germline mutations and heritable predispositions behind in paediatric brain cancer.

He outlines the Sonic Hedgehog family of genes as common driver in medulloblastoma, and discusses how translocation of genes to enhancer regions can trigger disease onset.

Dr Lichter notes that guidelines for treatments must reflect mutational subtypes, with patients whose tumours show enhanced activation of Wnt-Catenin signalling among the best prognosis compared to APC and Sonic subtypes.

The study I presented is the first part of a story that the second part will be presented later today by Stefan Pfister. This is that we look at paediatric tumours and we look for options that we find to do treatment based on genomic alterations in the tumour genomes of these children. In particular the focus that I presented here was on paediatric brain tumours so I talked about novel findings that we have in terms of pathomechanisms that are possibly targeted, findings that there are a few alterations that I presented for which already drugs exist and what we learned when we treat patients. One particular aspect was that we also look into the detail of possible hereditary predisposition of these tumours and we found that in some subgroups of these paediatric brain tumours the hereditary predisposition is much, much higher, as expected, and this has immediately consequences for genetic counselling, for the modalities of treatment and so on.

Which hereditary genes are you looking at?

This was particularly in medulloblastoma which is the most common highly malignant brain tumour in children. There is a subgroup that is called sonic hedgehog subgroup because this particular biomedical pathway is activated. In this subgroup one in five children has a hereditary predisposition. This is basically affecting six different genes, some of the well-known ones like TP53 but also other ones.

What is the prognosis for medulloblastoma?

It’s very different depending on the subgroups that I briefly addressed. There are four main subgroups, the first one has a very, very good prognosis and a very good cure rate, almost everyone is cured. However, unfortunately it’s the smallest subgroup and there are two other subgroups for which I presented today a number of new pathomechanisms that could possibly be targeted. For those the five year overall survival is not as good and therefore we hope that we can put these new targets over some new alternatives in the future.

Very interestingly, we find rearrangement so that a gene is basically translocated onto a regulatory element such as an enhancer very often which comes from a very different gene from a very different genomic area. Because of this the gene is regulated in a way that it should not be regulated and therefore this contributes, basically, to the malignancy of this particular cell to progression and sometimes even dissemination to the cell. So this is a phenomenon that was known years ago from a few lymphomas but now we also find it in solid tumours and we find it specifically in some of these subgroups of the medulloblastoma, some specific genes with specific enhancers.

What does the genomic screening mean for the development of treatments and treatment guidelines?

If we now get all the findings together with the guidelines we have to keep in mind that, for instance for the Wnt subgroup there is a very good successful treatment, as I said. But there is a very, very high percentage of mutation of a particular gene, catenin. If this mutation is not there then we see we have this inherited situation with a different gene, APC. So this would be specific guidelines even though it’s a good subgroup - screen for this mutation, if it’s not present go for the other gene and consider counselling. In other cases where we have this, as I just said, enhancer hijacking in group 3 and group 4 tumours where we need to develop a therapeutic option. So we need to do more preclinical work, that’s what we are doing in animal models at the moment to see whether we can achieve site specific enhancer inhibition. That’s in principle doable but we have to show it.