Today I presented the challenges in implementation of precision medicine and actually the implementation of precision medicine at MD Anderson with our IMPACT trials, the Initiative for Molecular Profiling in Advanced Cancer Therapy.
The main challenges are access to tissue, fast delivery of the results in order to be used in clinical practice, incomplete understanding of tumour biology as well as lack of effective drugs, and the fact that there is limited access to testing drugs for patients with cancer. To overcome these challenges there are a lot of things that we, the healthcare providers, should try to do. One is we should try to standardise tumour molecular profiling and offer it at the time of diagnosis and during the course of the disease. So far we have been using tumour tissue but moving forward we need to validate the cell free DNA analysis, in other words using analysis of patients’ blood to identify the driver of their disease. We need to optimise bioinformatics analysis and use the existing technology, or advanced technology, to have a complete understanding of tumour biology that causes cancer in individual patients. And to help discover and implement the use of new drugs and new strategies that can inhibit the function of the abnormalities that cause cancer. Therefore we should be able to offer this early in the stage of the disease and during the treatment to prevent progression.
What are the long term implications in terms of treatment?
The long term implications will be we will be able to offer precision medicine to more patients from the time of diagnosis and more effective drugs and at the right time, and to optimise their treatment during the course of the disease and hopefully to cure cancer and offer our patients better quality of life in the process.
We will need to be able to share this data, data sharing is important. The so-called n-of-1 databases are important, in other words to learn from every individual patient and offer this treatment that will learn and the outcomes to subsequent patients who present with similar characteristics and to avoid treating patients, for instance, who will not benefit from specific treatments or who have significant toxicity. In this aspect I would like to point out that there are immunotherapies that are offered to patients and we have seen significant anti-tumour activity for specific tumour types with specific characteristics in many patients but we have also seen rapid progression and severe toxicity in other patients who have been treated with immunotherapy. So we have to be able to develop markers for response and toxicity in those patients and standardise our approach so we can offer them the best treatment possible.
Tell us about the IMPACT studies.
The IMPACT studies, we started them in 2007, the goal was to optimise treatment selection based on patients’ own tumour molecular profiling. We presented the results of 3,700 patients that we analysed from 2007 until 2013. Of those patients 1,307 were treated, 711 were treated with matched therapy and the remaining with non-matched therapy. The overall disease control was 35% in those treated with matched therapy compared to 20% in those treated without matched therapy. Also this improved disease control was associated with improved progression free survival in patients treated with matched therapy which was 4 months for this group compared to 2.8 months for patients treated with non-matched therapy. There was also an improvement in overall survival by approximately 2 months, the median overall survival was 9.7 months in those patients treated with matched therapy. We also performed multivariate analysis and we demonstrated that PI3 kinase pathway alterations were associated with shorter survival and it was an independent factor predicting survival. We found that when we added the intervention which was the type of therapy, matched therapy was also associated with longer survival and it was an independent factor. We developed a model to predict overall survival in patients who were referred to our programme and therefore based on how many risk factors they have we can predict whether they will live less than 3 months, which will disqualify them from participating in trials, versus over 3 months or 6 months – we have a nice algorithm that says that.
What’s your message to clinicians watching this?
The message for clinicians is we need to continue conducting prospective clinical trials, taking into consideration tumour profile, cell free DNA analysis, immune markers and using the best drugs possible, ideally in clinical trials, to try to understand how to better use precision medicine and to identify specific markers and specific drugs that work better for specific patients. In this aspect we are conducting IMPACT2, a randomised study in precision medicine. We perform tumour profiling and we will incorporate cell free DNA analysis while looking at specific markers of DNA analysis as well as immune markers. The goal is to optimise treatment selection; in our revised design patients will have the option to select whether they want to be treated with specific drugs or they want to be randomised to receive the matched therapy, the targeted therapy versus non-targeted therapy, which also includes real therapy, there is no placebo in the programme, and to look at the results and the difference in their progression free survival based on their approach. There are also other efforts in which we participate at MD Anderson such as the TAPUR study relayed by Dr Schilsky as well as NCI MATCH and other trials.