Enlisting Big Data in the War Against Cancer

It’s almost impossible to pick up a newspaper without seeing an article describing some potential breakthrough in cancer. If all of the advancements mentioned in the articles lived up to half of their potential, cancer would already be a manageable nuisance, instead of the death sentence it far too often is. Somewhere between the stories we read and the reality we live, there is a disconnect.

The key to more personalized and precise care and eventual cures for diseases such as cancer, diabetes and Alzheimer’s lies largely in our ability to use big data, artificial intelligence and machine learning to better understand the diseases’ genetically based algorithmic patterns. In other words, before we can win the war against cancer, we must first unlock its code. Cancer, like all genomic disorders, is a computation problem.

I’ve worked in technology for the better part of two decades. Yet a few years ago, when a loved one was diagnosed with cancer, I was shocked at how little data and technology had permeated her clinical care. Asking doctors to treat cancer patients without the benefit of modern software is like asking someone to drive at night with no headlights. It’s reckless.

We have mountains of data, algorithms, cutting-edge analytics and software in the palm of our hands every day helping us navigate our world. And while highly regulated industries tend to adopt technology more slowly than nonregulated ones, we can and must bring these tools to health care and to the fight against diseases like cancer.

The time to act is now. Americans spent more than $3.2 trillion on health care in 2015 and The National Institutes of Health projects that by 2020, the total annual cost for cancer treatment in the U.S. will reach $173 billion. Nevertheless, cancer still kills roughly 600,000 people each year in the U.S.; it is the second leading cause of death behind heart disease. And, of course, the toll that cancer takes on individual patients and families is immeasurable.

As consumers of health care, we should demand more. We must empower physicians to make real-time, data-driven decisions. Integrating genomic sequencing with clinical data will allow us to learn from the millions of people diagnosed with cancer every year. Knowing how patients were treated and how they responded will improve the care for everyone that comes after them.

To achieve this, we need to tackle two data challenges. First, we need to generate far more molecular data than currently exists, which starts with driving down the costs of genomic sequencing. Second, we must integrate this data with a patient’s electronic medical record, so that physicians can understand its clinical context: Without marrying genomic data with clinical data, we are assembling only half the information necessary to make personalized medicine a reality.

When the first personal computer was built, it was just a pile of sensors and circuit boards; then someone wrote the first operating system, connecting the keyboard to the screen, running applications and allowing people to enter commands. Eventually that operating system unified many disparate functions into a cohesive problem-solving tool. It’s time to bring these same capabilities to cancer care.

There are reams of data, algorithms, analytics and software available to physicians treating patients. But these valuable information assets exist almost entirely in silos, each disconnected from the other. Without an operating system that brings these tools together, the promise of personalized medicine will remain just a promise.

Cancer has the fortitude of millions of years of evolution on its side. To better treat those who suffer from it – and one day eradicate it entirely – we need to reimagine how we battle the disease. We need the power of big data on our side. Asking doctors to treat patients with anything less is simply unacceptable.

(This piece first appeared at usnews.com)