At this time for biliary tract cancers there’s a great interest in precision medicine. It’s estimated that probably 40-50% of bile duct cancers have actionable mutations that can be accessed with approved drugs or in the context of clinical trials. One area that’s specifically interesting is HER2 alterations in biliary tract cancers. The more common is amplification or overexpression but a subset, barely 1-2% of these cancers, have somatic mutations in HER2 and these mutations are known to be oncogenic, they drive cancer growth and represent a potential therapeutic target.
So the data that we presented was the biliary cohort of the SUMMIT study. SUMMIT is a global multicentre phase II study looking at the drug neratinib, which is a tyrosine kinase inhibitor that inhibits HER2, with patients that have HER2 somatic mutations. The biliary cohort had 20 patients enrol and what we were able to show was that a subset of these patients had clinical benefit. Of the 20 patients two patients had a partial response, one had an unconfirmed partial response and three patients had stable disease that exceeded 16 weeks. So 30% of patients had clinical benefit, 10% of patients had a confirmed objective response.
Another finding from the study was in one excellent responder we were able to use next generation sequencing, both from the tumour and liquid biopsies, to demonstrate a potential resistance mechanism. Finally, we found that this treatment is generally safe in patients with bile duct cancers.
Can you go into more detail about the patient where next generation sequencing was used to demonstrate a potential resistance mechanism?
Sure. That patient was a 71 year old woman with gall bladder adenocarcinoma. The tumour expressed both HER2 amplified but also had a missense mutation in HER2. That patient had received three prior lines of chemotherapy and had progressed on all three. The patient initiated neratinib at 240mg orally daily. Very rapidly we saw a drop in the tumour marker CA99 and we saw a significant reduction into tumour at both an eight week scan and then at a twelve week confirmation of response scan. Unfortunately at 24 weeks the patient progressed and we had in our possession both a biopsy of the original tumour as well as a new biopsy of the escape lesion. We performed MSK-IMPACT which is our next generation programme where we look at nearly 500 oncogenes and tumour suppressors. What we found was in the pre-treatment sample, of course, we saw the erbB2 alterations, a TP53 alteration and several others. On the post-progression sample the erbB2 alterations were lost, TP53 were maintained and then there were several other new genes that were not previously seen. So this indicates that there’s a sub-clone that doesn’t express erbB2 that essentially emerged with selective pressure.
We also were able to do a liquid biopsy which takes cell free DNA and apply a similar type of next generation sequencing. In this we could see the erbB2 alterations as well as the TP53 at all sampling time points. However, at the emergence of resistance we saw several other new mutations that weren’t present in the original tissue and that included a BRCA alteration, an erbB2 missense mutation and an erbB3 missense mutation of unclear significance. But this illustrates a more broader point in biliary tract cancers which is that there is sub-clonal heterogeneity and we’re just now trying to explore that and to understand that better with the hope that that might lead to better prognostication for patients as well as better prediction and treatment alignment.
How much time was taken to identify some of these mutations and how feasible is that for general practice?
The idea behind this, as you noted, requires genotyping patients at clinical grade in real time. This leads to both feasibility issues and cost issues. I would say that across the world there are different time frames by which we are able to do this presently. At our institution we do perform next generation sequencing for biliary tract cancers, namely because there are FDA approved drugs that require having knowledge of that. For other parts of the world that may not be standard of care at the present time.
As you could have seen from the presentation before mine, actually several patients had genotyping but weren’t able to make it on to a clinical trial. But I think that illustrates the need to perform sequencing faster, better, cheaper and make it widely accessible. But over time we’ll see how that evolves.
What are the next steps for the main body of the trial?
At this point the study continues to accrue because it hasn’t reached its specified number of evaluable patients. Once we see the totality of the data there would be additional plans for further drug development in this space.
I would note that outside of this study there are several other HER2 targeted compounds that are being proposed or ongoing in biliary tract cancers. Some of those data have already been presented and over the next few years we’ll still see continued interest in HER2 in biliary tract cancers as well as several other genomic targets.