Today we’re talking about precision oncology which, I have to say, is my favourite conversation topic. We were really still strategizing about how to best implement biomarker-driven care, focussing on genomically informed medicine. This is the area that’s made the most advances although clearly now there’s a lot of excitement about adding additional analytes, including transcriptomics and immunoprofiling. But at least from day-to-day routine care more and more genomic testing is being utilised for advanced disease.
There are a lot of different platforms for genomic testing and the increasing challenge is not the testing per se but the interpretation of the genomic test results. Equally important, of course, is having access to the therapeutic options. So we talked today a little bit about how do we optimise interpretation of the genomic test results. This is an area that we really, at MD Anderson, have worked on extensively. We have put together a decisions programme; we have a team of oncologists interested in genomic medicine, research scientists that have a molecular oncology background and really a passion for impacting patient outcomes and then computational scientists and informaticians really very thoughtful about how could we optimally select care.
The increasing challenge has been, of course, that genomic testing is getting broader and with that we’re not only seeing common recurrent mutations but we’re more and more challenged by seeing variants of unknown significance. So we’re putting together a platform for decision support so when a treating physician gets a genomic test result we can immediately help them by interpreting – is it an alteration in an actionable gene? Is this specific alteration predicted to affect the function of that gene and if so, what do we know about it? Is there any clinical data that affects therapeutic sensitivity, whether it’s resistance or sensitivity to agents, approved or investigational? Further, if that data is lacking is there any preclinical data suggesting that this would be impactful? And if this is an actionable gene with a functional alteration then what is the best option, what is the best investigational option, what is the best standard option? The genomic match option may or may not be the best option so what is the best option taking all the information at hand and all therapeutic options? So it’s very important to create decision support tools that really factor in the whole big picture.
Further, today we talked about how do we transition from our current focus which has been very much monotherapy or single gene-based information now that we increasingly realise patients have complex genomic reports, they have multiple drivers or multiple alterations that may affect the sensitivity to single targeted therapy. How do we build combination therapies that can leverage the information? We know more and more about mechanisms of resistance, mechanisms of acquired resistance and there is already data emerging about targeting acquired resistance mechanisms, targeting adaptive responses.
Today our focus is to optimise the therapies we have while building new generation clinical trials that really incorporate all this information about resistance mechanisms so we can combine therapies to be more effective so we can treat patients better from the first time around.
Will these tools be shared with other institutes?
Yes. Some of these tools that we are creating are publically available. We have a website, personalisedcancertherapy.org, that has information on some of the more commonly altered genes and variants that are known to affect function with clinical trial matching efforts for those. In addition, many of our new tools computationally as we generate them we try and make them available in the public domain. But also we’re really trying to create a network of clinical trials that are platform-based or looking at unique opportunities that are better at targeting genomic alterations with rational combinations to have more durable responses when they do recur or objective responses when currently agents are cytostatic.
