6 Ways Big Data Is Driving Personalized Medicine Revolution
Researchers are poised to make huge advances in medicine, particularly in how we treat cancer and arthritis. See how big data and IT are contributing.
![](https://eu-images.contentstack.com/v3/assets/blt69509c9116440be8/blt7fe50846341bf756/64cb4cfd2177aed1a90ce9c2/intro.jpg?width=700&auto=webp&quality=80&disable=upscale)
Drugs can be expensive, difficult to research, hard to get approved, and, according to a recent report, don't work on large parts of the population. These factors likely put a great deal of pressure on pharmaceutical companies to research drugs that have the highest probability of turning a profit rather than those that could help the most people.
But this paradigm may be shifting with the help of IT and big data.
The industry has found new ways IT and big data are making a major impact on the way drugs are being researched by helping create more effective trials.
Before we examine the benefits IT is bringing to this arena, let's try to understand what's wrong with the traditional (and ongoing) way most drugs enter trial.
Most drugs don't work. Statistics published in the journal Nature show that among the 10 highest-grossing drugs prescribed in the US, even the best work in only one in four patients. Some work in only one of twenty-five. Statins, commonly prescribed cholesterol drugs, work correctly in only one in fifty patients, according to the article.
Nature cites multiple reasons for this, but the basic is that our different genetic makeups (genome), proteins in our body (proteome), and body flora (the bacteria and other stuff that grows inside of us that we don't like to think about) affect how drugs work. Over the past few decades, medicine has often (and sometimes legitimately) been accused of focusing too heavily on Western patient pools (cohorts), excluding minorities and people from other countries, and therefore of doing a bad job of creating drugs that work for all ethnicities.
[See 10 iPhone Healthcare, Fitness Apps That Actually Work.]
As drug companies try to address those criticisms and create more accurate drug trials, the cost and time of developing drugs can increase due to delays in creating patient pools, difficulty finding pools with the right genetic makeup, and the need to increase the length of trials to find drugs that will work for greater portions of the population.
Add it up, and the cost to produce a new drug is now $2.6 billion and rising 8.5% every year, according to the Tufts Center for the Study of Drug Development. This is an untenable situation.
Changing the way we design and administer treatment trials, using big data to bring "personalized" or "precision" medicine to drug trials and research, can potentially reduce costs, allow the right drugs to be prescribed faster, and improve outcomes at lower costs. It also may mean faster drug development with easier margins. It is one of those rare win-wins for pharmaceutical companies and patients.
Take a look at six new ways IT and Big Data are helping drug trials that are currently going on.
What if we could study the ongoing health records of more than 1 million people to learn which individuals respond to certain types of drugs, are at risk for a certain disease, maintain health and fitness, age, and die? That's the goal of the $130 million Precision Medicine Initiative Cohort Program (part of a precision health initiative, with an overall $215 million annual budget) being done by the National Institute of Health. The goal is to enroll over 1 million Americans in the cohort in the next three to four years. The data (anonymous, of course) from all 1 million individuals will be furnished to any interested researcher who wants to study one of the largest cohorts ever made available.
All members of the cohort will have their genomes sequenced, and their health history, lifestyle habits, and environmental exposures tracked. By doing so, the study will yield a treasure trove of big data. It will allow people to track the effectiveness of medicine based on genetic markers and identify certain biomarkers that signify that people might be at risk for a given disease.
The cohort could also serve as a platform for smaller trials. For instance, if you needed 100 people with a specific genetic trait who are also taking a certain drug, it could take years to develop a cohort for a study. The Precision Medicine Initiative may be able to identify the people you need in minutes. The NIH also aims to use mobile apps and information from the trial to help the people in the trial lead healthier lives. The agency hopes that by their example they will encourage healthier lifestyles in the general population. IT's advances in mobile devices, cheaper genome sequencing, databases, big data, and electronic health records make knowledge and health goals possible.
Cancer research program NCI-Match is intended to be a major part of the National Institute of Health's Precision Medicine Initiative. It will enroll about 1,000 people in an effort to match specific types of tumors with specific medicines.
The program will seek out people with tumors that have failed to respond to standard cancer treatments and match those tumors with drugs known to have better outcomes based on certain genetic markers. The hope is to build a database of drugs that have positive effects on different types of tumors in order to get the best treatment to patients the fastest.
The study should be an ongoing ode to personalized medicine and big data, as it will continue to grow new "arms" to study new types of tumors and track new medicines and their effect on certain types of tumors. This has the potential to be the most important cancer trial in history, as it hopefully will unlock the secret to curing multiple rare and fatal types of cancer by matching the individual's genome to the right drug.
This would not have been possible until recently, because genome sequencing would have taken too long. Faster computers, better software, and better databases are at the heart of the success of trials like these.
One problem in the treatment of cancer, especially breast cancer, is the false positive. Ten percent of the time, women who get a screening mammogram are called back for further study, but only 5% of the women who get called back actually have cancer. The Wisdom Study is designed to enroll 100,000 women to see if mammograms are really the best way to detect breast cancer. The interesting thing about the study from an IT perspective is that it isn't using an electronic health record provider or a specialized database to manage the study. It is using Salesforce, the cloud-based CRM.
According to the director of the research, Dr. Laura Esserman, Salesforce offers more flexibility, better data management, and the ability to more easily treat patients like people. When something as ubiquitous as Salesforce is being used for clinical trials on a grand scale, you know IT is having a major impact. The cloud, consumerization of IT, and cloud analytics are all responsible.
Another new concept in drug and health trials is the "N-of-1" concept. Most trials require hundreds or thousands of "N's" (people enrolled in the study) to make sure the project has statistical significance. But we're starting to realize that, with some trials, larger cohorts mean you have X's, Y's, and Z's mixed in with your N's.
In other words, on a genetic level we're comparing apples to oranges and we're adding static to the data. The Tanner Project is specifically looking for N's-of-1, people with variations from the statistical norm. The study looks for "stage zero" signs of hereditary diseases. In other words, it is looking for signs of potential diseases before people officially have the disease.
The hope is that in finding those markers before the disease begins, treatments can be started sooner. For instance, doctors often look for a specific blood protein as a sign of ovarian cancer. In most people, the count of that protein has to be 30 to 35 to be significant. In some people, a much lower number can be a sign of the beginnings of cancer. If it can be discovered why the number is lower in some people (the N-of-1), the information can be extended to the hundreds or thousands of similar traits, turning an N-of-1 into a statistically useful study and helping people with that genetic and physical makeup.
Half of those who suffer from metastatic melanoma, skin cancer that spreads to other parts of the body, have a genetic alteration called BRAF. But there are promising treatments for those with that particular genetic alteration. For everyone else, there is no known treatment. The Dream Team assembled for this study hopes to find other potential genetic mutations, either in the tumors or in the patients, that could lead to better treatments. Like most of the other cancer research teams reviewed here, the Dream Team will be using faster computers to sequence more than 3.1 billion base pairs of DNA to look essentially for needles in a haystack -- a genetic marker like BRAF that can help explain why the cancer is hard to treat and could possibly be translate into a cure.
You probably noticed that most of these trials I have mentioned so far have centered on cancer. That disease is a chosen focus of precision and personalized medicine because of the huge variation of success in cancer treatment drugs based on genetic factors. However, cancer isn't the only disease where trials are being conducted. Another common area is arthritis treatment.
Multiple trials have been conducted to determine the best drugs and dosage for those suffering from the disorder. Specifically, rheumatoid arthritis, which has several different biomarkers, has been shown to have significant variance in treatment response. Ultimately, there's no reason -- with faster DNA sequencing, cloud analytics, and cheaper storage -- why large-scale precision medicine trials couldn't be used for any disease.
Drug trials and large-scale gene studies aren't the only ways IT is advancing personalized medicine. Simple things like the Fitbit and other consumer devices may eventually lead to breakthroughs in health based on big data. Companies such as UBC, among others, are shaving costs simply by using large databases to do a better job of finding patients to enter studies.
While the NIH's Precision Medicine Cohort is open to all, many drug companies are starting similar (or larger) databases. In some cases, those databases are a result of pooling data within the pharmaceutical industry.
All of this is powered by major breakthroughs in IT and analytics. If you've ever helped develop technology such as a database, cloud architecture, a supercomputer, big data, mobile, or networking, you've probably had a hand in inspiring these studies that could one day help conquer some horrible diseases.
Drug trials and large-scale gene studies aren't the only ways IT is advancing personalized medicine. Simple things like the Fitbit and other consumer devices may eventually lead to breakthroughs in health based on big data. Companies such as UBC, among others, are shaving costs simply by using large databases to do a better job of finding patients to enter studies.
While the NIH's Precision Medicine Cohort is open to all, many drug companies are starting similar (or larger) databases. In some cases, those databases are a result of pooling data within the pharmaceutical industry.
All of this is powered by major breakthroughs in IT and analytics. If you've ever helped develop technology such as a database, cloud architecture, a supercomputer, big data, mobile, or networking, you've probably had a hand in inspiring these studies that could one day help conquer some horrible diseases.
-
About the Author(s)
You May Also Like