WIN 2016
Preclinical strategies for precision medicine in colorectal cancer
Dr Livio Trusolino - Candiolo Cancer Institute, Turin, Italy
We are interested in identifying novel oncogenic drivers in colorectal cancer and validating them as viable therapeutic targets. The targeted therapies that are currently done in colorectal cancer are two antibodies, cetuximab and panitumumab, which inactivate the epidermal growth factor receptor EGFR. When given to a selected population of patients their activity is limited to a very small fraction of cases, 10-12%. Now there is a validated biomarker that correlates with resistance which is mutant KRAS which accounts for approximately 50% of all colorectal cancers. So by excluding all patients with KRAS mutant tumours from treatment we can arithmetically increase the prevalence of responders from 10% to 20%. But there are plenty of cases that do not respond at all and we do want to understand why. To do so we have started collecting systematically viable surgical samples from cancer patients that on one side are extensively annotated at a molecular level for discovery purposes and on the other are implanted and propagated in mice to generate treatment cohorts that receive therapies on a rational basis according to the information that emerges from the discovery pipeline.
For example, we have been able to identify on top of KRAS mutations other genetic alterations in other oncogenes such as HER2 amplifications, MEK1 activating mutations, EGFR amplifications and PDGFR mutations that specifically correlate with resistance to anti-EGFR antibodies that are causatively involved in standing resistance and that are drugable targets because when challenged in the mice tumours shrink. The most mature results that we have gathered so far are those on HER2 amplification because these are the ones that historically have been identified first. The preclinical results have been translated into a clinical trial in which the very same treatment modalities that were used in mice and that were successful in mice were transferred into patients. This trial is now completed with very successful results also in patients.
Has there been any resistance to the treatment documented?
Yes, we have some initial results about the occurrence of acquired resistance to HER2 blockade in HER2 amplified tumours. It looks like one potential mechanism is the emergence of activating mutations in PIK3CA which encodes for PI3 kinase. This is something that has already been shown for HER2 positive breast cancer and gastric cancer.
The idea is, of course, sooner or later to come up with a more granular certification of patients based on genetically defined small subsets. In the future what is likely to become a reality is the identification of very low frequency genetic alterations that occur in 0.5%, 1%, 2% of all patients. So we will need to reason differently and to try to change the way we face with the logistics because we will not be able to identify such mutations by screening for one mutation in one thousand patients. We will have to screen one thousand patients for many mutations and then allocate all the different patients to the pertinent treatment modality.
