Cancer Moonshot

Big Data against Cancer

Sergey Syrov, XXII CENTURY

The US government plans to invest $1 billion in the Cancer Moonshot project. This should speed up research in the field of cancer detection, treatment and prevention.

The curator of the project is Vice President Joe Biden. His son, Beau Biden, died of brain cancer in 2015 at the age of 46. The name of the project appeared after Joe Biden's speech, in which he called for a breakthrough in the fight against cancer, for a new national achievement, "flight to the moon". This is an ambitious and important task – about 600,000 Americans die from cancer every year, and 1.6 million US residents are diagnosed with the disease.

Cancer Moonshot's goal is to double the current pace of innovation in the medical industry. This is not an easy task, taking into account the fact that many research teams are engaged in the fight against cancer. There is a lot of money for this, so $ 5.2 billion has been allocated only to the National Cancer Institute of the USA (National Cancer Institute, NCI – part of the National Institutes of Health (NIH)), and more than $ 100 billion is spent annually in the world to fight cancer.

But the action plan announced by Biden does not involve large expenditures, the construction of cancer treatment centers or the start of new scientific developments. Instead, it is proposed to simplify bureaucratic procedures and find a way to combine the efforts of regulatory authorities, industry, researchers, patient groups and charities. It is assumed that joint efforts will achieve results exceeding the capabilities of individual participants in the process. Thus, the main part of the work is data collection and organization of their exchange.

"From my personal experience, I know that research and therapies are on the verge of incredible discoveries," Biden wrote in a post on medium.com in January. – Only in the last four years we have seen amazing progress. And this is a turning point. But scientific discoveries, data and research results are stored in the bins, preventing rapid progress and greater accessibility of innovations for patients."

Data and technological innovations can play an important role in the "medical revolution" if they become publicly available, Biden believes. If the project manages to open up to researchers those vast arrays of cancer-related data that various organizations have collected and stored for years, then the success of the "flight to the moon" is very likely.

"Almost every cancer center holds a database–genetic history, medical records, and tissue banks–that may hold the key to improving certain cancer treatments," Biden writes. – Enabling researchers and oncologists to use this storehouse of information is vital to accelerate the pace of progress towards a cure. If we make this data compatible and accessible to scientists, researchers and doctors, we will accelerate scientific advances, improve patient care and move closer to a cure."

Nidhi Aggarwal, Director of Promotion and Strategic Development at Tamr, talks about the difficulties Cancer Moonshot faces when working with big data.

"There are thousands of possible sources, public and private, that could potentially be used to advance research – everything from data obtained during clinical trials consisting of several thousand parameters to genomic datasets that may contain millions of elements," Aggarwal notes. "There are many problems and they are diverse."

Tamr announced this week that it will provide all researchers associated with Cancer Moonshot with its big data software. Tamr programs use machine learning algorithms designed to organize, prepare and integrate weakly structured and unstructured data for analysis. The company also employs a lot of experts to monitor the work and improve algorithms.

Tamr software is used by pharmaceutical companies such as Novartis, GlaxoSmithKline and Merck.

In addition to the problem of processing a large amount of data, there are other difficulties, including the lack of uniform data standards. Aggarwal notes that there have been attempts to standardize data, so the FDA issued a requirement that all electronic data on clinical trials should be presented in a form that complies with data exchange standards (CDISC), and the Office of the National Coordinator for Health Information Technology (ONC) plans to build an "interactive, a confidential, secure nationwide healthcare information system." But so far neither the FDA nor the ONC have been able to achieve the stated goals.

Instead of relying on the efforts of Federal Standardization Agencies, Aggarwal recommends using the approach used in the SpaceX cargo spacecraft project to find a way to partner with the public and private sectors. She also says that standardization and unification should be observed from the very beginning of work on the project.

Finally, she notes that the scale of data and the challenges of interaction are so great that people and teams should not try to solve the problem of fighting cancer alone. Machine learning and artificial intelligence are developing rapidly, and we need to make full use of these tools.

Portal "Eternal youth" http://vechnayamolodost.ru  25.03.2016