One of the many challenges in fighting kidney cancer is knowing where it all got started. This research indicates that a gene gone astray, the MET gene, is part of the problem from some patients. It is this type of study that will change the treatment for us, as there will be greater clarity as to the “target” to be reached by “agents of change”. (Nothing political intended, but seems to work here!) Especially of interest to patients who have the variant of papillary RCC, about 10% of us.
MET Variant as a Prognostic Marker in Clear Cell Renal Cell Carcinoma
Dr. Ari Hakimi of Memorial Sloan-Kettering Cancer Ctr.,New York USA
ASCO GU Congress 2014
eCancer reporter Peter Goodwin’s questions are in italics. Where I was not certain of the lecture, I added a (?) to show that. Link below to the actual lecture.
Ari, you have been looking at prognosis or prognostic features, or actually, molecular features of renal cell carcinoma. Can you tell me what you were doing in the study you’ve just been talking about?
There was a great paper that was published by the people at Harvard and Lancet (?) Oncology last year. It found for the first time a prognostic marker that was associated with poor survival in kidney cancer, a molecular marker. This was a variant, a normal variant in the gene, the MET gene. In that study they had several hypotheses they generated from that study, but they didn’t really have enough genetic data to try to figure out what was going on here with this variant in the genes. So what we did was, we took that same concept, that same variant, in the Cancer Genome Atlas Study, which has both patient information and then a host of genomic information. We tried to validate their finding and explore the biology of that marker.
It sounds like a needle in the haystack but you’ve but you become quite familiar with this variant called RS11762213. What you know about it so far?
We know about that it is a variant in the MET oncogene, a very important gene in a lots of different cancers, particularly in papillary renal cell cancer. It’s a gene not thought to be very important in clear cell renal cell carcinoma, but we found that it is, and we explored the variant in an exonic region of the gene–meaning the coding region of the gene. Because the variant is in a coding region of the gene, we thought it might be more than just a prognostic marker. It might also have some biological implications.
Biological implications? What sort of biological implications?
We think the marker may be; we figured out through computational methods, exploring the TCGene data(?) that it might be in the region of enhancement, meaning the variant leads to higher activation of the MET oncogene. In turn, this might explain why these patients have a poorer outcome. It might also have potential therapeutic implications.
So what have you found so far then, about the level of additional risk if you happen to have this variant gene?
Great question. We took about 270 patients from the cohort who had available information. We genotyped them, meaning we determined what percentage of these patients had the risk variant, which is about 10 %, consistent with prior studies. We showed that when these patients had that risk, in addition to the current prognostic features, they had about a 3-4 fold increased risk of cancer-specific death, or tumor recurrence after surgery.
That’s really quite powerful! Am I right that there wasn’t any clear kind of molecular feature to give you some help in the past?
Until this study, which was published last year, there were really only tumor features and patient features that were associated with poor survival in kidney cancer. This is the first study that really showed, that they published last year, to show in two different cohorts that had a molecular feature that added to the prognostic models. We showed, augmented their findings, that if you took the best current prognostic models and stratified patients, added to that model, meaning it improved the predictive accuracy of even the best post-surgical models that are out there.
You are looking actually disease mechanisms–mechanisms of cancer production. You established prognosis, but what about predicting response to therapy?
That’s a great question. Our goal now– that we’ve established that this is a valid biomarker, truly multiple cohorts now showing this marker can stratify patients for aggressive behavior, we can now explore—hopefully–whether this has therapeutic implications because it is in a gene that is a known cancer gene. Because there are multiple drugs that target this gene, and because we think that this variant that is activating this gene, it stands to reason that an inhibitor for these patients with this variant might work. These patients might have another option.
So theoretically a new drug which is an inhibitor for this variant might work. What about existing cancer drugs? Do you have any ideas about if any these do influence that variant?
We don’t know yet. We are trying to find it in cell lines, meaning cell lines that are derived from tumors that are used in the lab, to see MET inhibitors that currently exist and are in phase I or II trials in kidney cancer could potentially be used against patients against this variant. That could be a very powerful tool, and a kind of the precision medicine that were looking for.
This is an amazing achievement, actually going through the Cancer Genome Atlas to find information like this, information about expression. In the realm of the everyday cancer doctor with patients to treat today, tomorrow, what you think the doctor should take home from this development?
The exciting thing about this is to genotype the patient, that is to determine that this patient carries this risk variant, is something you can detect from the patient’s blood or even a swab from the cheek. It’s a very inexpensive. It costs about $10-$70 to get this information for a patient. You can have what is called a liquid biopsy, meaning you need any tissue. You can get it from their own normal cells, because this is germline variant. You can find out this information very affordably and very quickly to determine risk for these patients. Obviously, if we are able to show that it has implications for therapy, that as a whole opens a whole new avenue.
How much hope to have that this it will be possible to manipulate this gene expression by using this kind of drug to target this?
I think that the data there is quite strong for other types of cancers. We know that other genes that are overexpressed or mutated in activated fashion respond quite well to inhibitors. This exists in multiple cancer types, lung cancer, breast cancer, for example. It stands to reason that this would work as well in kidney cancer, and the hope would be that this variant would be actually an activating factor and that we could use that also.
We’re also hearing, and especially at this GU meeting here in San Francisco, about the heterogeneity of the tumors. In fact, you may have tracked down one particular cause of cancer, but there’s another five going to rear their ugly heads at the same time. What you make of that?
Well, that’s definitely a major factor, particularly in kidney cancer, where heterogeneity was really first described in the clinical setting two years ago in the New England Journal. What the nice thing is about this variant is, is that it is germline. It exists in every cell. Thus heterogeneity does not exist in this situation. The variant is present in all cells, including the tumor cells. So if our data does hold up, and it is a therapeutic target, it will not be affected by tumor heterogeneity.
Give me a message to take home for the community cancer doctor very briefly..
The messages that we have truly validated this important finding that was published last year and we truly believe that this is a new prognostic marker and adds to the existing prognostic markers. Time will tell if it will actually help guide treatment of metastatic disease and really change the paradigm for kidney cancer.
Thank you very much.