Why do we need the Precision Medicine Calculator?
Medicine and especially oncology has entered a new era referred as personalized and precision medicine. Instead of using scientific data generated on large number of average patients physicians have to combine classical evidence with molecular evidence linked to the genetic background of the patient and his/her cancer cells.
We have clinical proof that if we can identify the molecular cause of cancer, the “driver” gene in a patient, targeted drugs targeting this alteration are very effective, sometimes can even lead to cure or at least years of overall survival benefit (e.g. CML – imatinib, lung cancer – gefitinib).
The effect is very similar to the efficacy of antibiotics in infectious diseases, when the pathogenic cause, the bacteria are eliminated and not only the consequence, the inflammation is suppressed. Similarly, now we want to antibiotics of cancer. We have the tools to explore these molecular causes and we have hundreds of targeted drugs in pipelines to choose from. Now the last barrier is that we do not have a decoding machine, which would translate the molecular information into treatment decisions.
We are living in an exciting time. After sequencing the human genome in 2003, the majority genetic alterations have been explored in cancer by the Cancer Genome Atlas project and other large international research projects by 2014.
The Catalogue of Somatic Mutations of Cancer contains 2.1 million different variants in 550 cancer genes. Each tumor can contain a combination of up to 8 cancer genes. This means that more than 50% of cancer patients have cancer mutations less frequent than 1%. There are hundreds of targeted therapies in development. Physicians face the great challenge to make the best decision based on complex molecular information and evidence (24 million in PUBMED) and limited experience with rare alterations.
This needs computational support. Oncologists cannot spend hours to search for relevant information at each patient. This important step in medicine is acknowledged by the American Association of Cancer Research (ASCO) which is indicated by the official title of their upcoming annual meeting: “transforming data into learning” with computer chip as a logo.
Newest technologies enable personal genomics to become a commodity. The greatest bottleneck we face today is that most genetic alterations are “variants of unknown significance” or “VUS”. The most frequent alterations have been annotated already and linked to targeted therapies, which have shown great clinical activity. Therefore, the majority of scientific community believe, that therapies targeting the “molecular cause” of cancer are the future, but there is also a strong scepticism about time we need to understand the clinical significance of the large number of rare alterations. The other great challenge is the intra-tumour heterogeneity and continuous mutagenesis, which are responsible for primary and secondary resistance to targeted therapies used as a mono-therapy. Therefore we will need combination of therapies, but the number of potential combinations, associated cost and side effects pose a great challenge to cancer research.
Recently, there is a great enthusiasm around immunotherapies, which have achieved unprecedented long progression free survivals in several patients. But not all patients respond to the same immunotherapy and the cost of these treatments is well over 100,000 USD per treatment. Therefore, we will need molecular predictors of efficacy to personalize immunotherapies as well.
In this new era personalized medicine, there is a need to adapt treatment to each patient based on each patient’s “story”. There is an advantage to learning and sharing these stories in development of treatments across patient populations. A system for automated sharing and adapting treatments accordingly for each individual patient would be advantageous, but does not exist.
The large number of genetic variations is a great challenge, but the analytical reproducibility of genetic data is a unique opportunity to share experience. It is easy to see that only collective and collaborative intelligence of thousands of doctors and millions of patients can accumulate enough knowledge to conquer cancer.