ODYSSEY stands for Oncology Data Science. What we do is to correlate the clinical parameters, the features, of patients that we treat at the Vall d’Hebron Institute of Oncology with the molecular profiling that is done to either enrich patients in clinical trials and also understand the biology, the prognosis and the predictive value of these markers. So it means understanding whether an alteration in the tumour confers a poor outcome, worse survival for the patients or if it’s a predictive marker which means whether it sensitises to a particular chemotherapy or a targeted therapy or an immunotherapy. So what we do is correlative analysis between the clinical parameters and the molecular features.

That leads neatly on to the presentation you’re giving at this year’s conference.

Exactly. We focus mainly in colorectal cancer. There have been many advances, technological advances, in, for example, circulating tumour DNA. So with liquid biopsies we are able to identify minimal residual disease after a potentially curative surgery of colorectal cancer. So we detect in the blood if there is residual disease in the body that is microscopic that you don’t detect with other technologies, that’s supersensitive. Then you understand by profiling the tumour what are the drivers of this potential metastatic behaviour and then what will be the best therapeutic approach, what will be the best chemotherapy and the combinations with immunotherapies that you give after surgery to increase the chances of response.

So it’s more a conceptual view of all these new advances, understanding of the biology of the disease, what are the drivers of a metastasis and how to tackle them with circulating DNA analysis and profiling of the tumour.

Beyond personalising to tumour genotype you can also personalise with the timing and the staging with neoadjuvant therapy, for example.

Neoadjuvant therapy, exactly. In breast cancer it’s another interesting example. Before starting neoadjuvant therapy, during the course of neoadjuvant therapy and at the time of surgery you get biopsies at multiple time points, you correlate with circulating tumour DNA analysis and you identify what are the residual markers of more aggressive disease that you are not able to completely eradicate with your therapy. Then you give your adjuvant therapy accordingly. So you personalise the therapy by adding multiple factors. So it’s not a single marker that you use any more, or only clinical parameters, you have to combine multiple molecular markers to understand what is the driver of the resistance to therapy or the best therapy that you give in the adjuvant setting.

There are many markers which have been associated with better or worse outcomes and some very famous, very wide, very broad biomarkers, is there not the concern that by increasingly narrowing in on one tumour type, one mutation, a very personalised therapy, then you risk almost missing out on the broader targets?

That’s a good point but at the same time as we are personalising therapy the diseases are right now not only breast cancer, every patient has a different breast cancer. So by personalising, really understanding that individual case, the mutations that you detect, the alterations that you detect in that particular case are very important for the evolution of the disease. So there is no way to generalise in breast cancer or in colorectal cancer anymore. You have bigger global patterns of what is the driver of resistance or sensitivity and prognosis but at the same time in that particular patient we have to combine multiple approaches to understand what is really making a difference, if you want to make a difference in the evolution of the disease with therapies.

Speaking of combination therapies, we’ve done a lot of work with people working on CDK inhibition and we’ve got the results from abiraterone today over the hormonal therapy. Where do you see the future for either triplet or quadruple therapies?

Exactly.  With targeted therapies we more or less reach a plateau with the number of therapies that you can give because of the toxicity. By combining multiple targeted therapies it’s hard to go beyond two targeted therapies. You can combine probably with a hormonal therapy or now with the immunotherapy. So I see the future with combination therapy definitely and the beauty of immunotherapy is that it’s not completely depending on the molecular fingerprint of the particular alteration that you identify, so it’s not only one bottleneck. So the residual disease that you cannot target with a specific therapy you may be able to just break with immunotherapy and they are much less toxic to combine as compared to multiple targeted therapies or chemotherapy. So the idea for the future is definitely a combination of immunotherapies with targeted therapies or hormonal therapies in some diseases.