Big Data has created a recent buzz in pharma. As information analytics permeates every aspect of a supply chain, the ability to harness, highlight and discern the most pressing information accessible has the potential to not only increase cost efficiency but also efficacy throughout. From the preclinical phase and on to commercial processing, the promise of harnessing data is not only key to better results, with a direct impact on pipelines, but also on total spend in the industry as a whole. As the McKinsey & Company predicts, applying big data to decision making could work to generate up to $100 billion in added value on an annual basis,1 this is an incredible amount of savings, especially when it is targeted and applied throughout the supply chain.
Outside of the Clinic
The first challenge in big data is collecting information and making it uniform, so it is not only understandable but also accessible. To really engage big data, it is important to think outside the box of the clinical trial environment and focus on how real world situations can be integrated into those in the trial or pending trial, either as an addon or as an entirely separate entity. Rigid clinical trial environments are rarely conducive to situations that would happen outside of the clinic, and moreover, there is a disconnect between the trial setting and the physician’s office that is not only limiting to results but also is not ideal for patients, who need access to potentially life-saving drugs faster.
Unstructured doctor’s notes and pathology reports, scans, images, samples, and various records live in isolated and often times messy files, however, at least they are online. With the normalization of Electronic Health Records (EHR) as standard practice,2 there is now a place in the cloud for records that would have been tossed or lost - this is a big step on its own, and the impact of EHR is nothing to be undermined, however it is time to graduate to the next step, with an eye towards consolidation and analysis in a centralized record bank, where all big data health files can live.
Cutting Out the Noise
These files currently exist as singular entities, however, they should be matched up and sorted accordingly. If electronic records can be sourced and categorized by information type and linked according to searchable criteria, the impact, not only for trials but also healthcare at large will greatly increase. Using key words or algorithmic code data from the physician’s office, data can be incorporated and sent into a central processing station. Data should not only be shared but also standardized. The question then becomes how to extrapolate the necessary material in each situation and cut out the noise.
Data integration is crucial to advancing big data. Data does not work as a tool when it is a disordered entity. It must be integrated, to take it from what is potentially an information overhaul to one single and efficient system that can be sourced by a host of users. However, this brings up the issue of encryption and trusted users, as well as patent protection and consent by patients to have their records integrated into a single system - through a shared data network, that is crucial to progressing data to the next level of usability.
Diluting Data
Firms are already working on how to bring big data down, taking it from an abstraction and incorporating it into the next phase of pharma. The Data Incubator, which was founded in 2012, works to train those with their Ph.D. or Master’s degree on how to work with big data specifically, and ultimately to be equipped with the knowledge of how to deal with massive amounts of data. The company also facilitates the hiring of data scientists and includes an option to train employees.3 Firms that want to capture the potential of big data are encouraged to work with individuals, who have the expertise to handle the data, or to train their staff to be equipped with such knowledge.
One such firm is Genetech - the company has served as a pioneer in big data. It has built the internal capacity for handling big data, including a platform which can impressively analyze billions of patient records in just seconds2 – proving a step closer to leveraging big data than ever before.
Partnerships are further encouraged in order to maximize the potential of big data, especially with CROs and for scale-up. These research organizations can also be consulted with on relevant issues that are intrinsic and specific to clinical trials, such as shared privacy concerns and any legal, intellectual property or regulatory concerns, likely to emerge when dealing with high volume, high stakes data, as well as with increased collaboration.
Limitless Sample
According to Philippe Bishop, M.D. Vice President Clinical Development, Angiogenesis Franchise Head regarding big data and cancer, “only about 2% of people with cancer participate in clinical trials.” This represents a dearth in the population sample sourced for trials, considering, as Bishop points out, “thousands of people are diagnosed and treated every day,” he continues, “which means there is a huge amount of important data that could be obtained to help doctors and patients make better decisions about treatment options and potential outcomes.”4
What is interesting about Bishop’s observations is, just as big data has limitless potential, the size of the sample is increasing exponentially. If an integrated system is implemented and pharma has the ability to gather insight from the hospital, physician’s office, and any other real world setting, then big data breaks the binding nature of the clinical trial scenario. Although drugs will still be monitored and controlled in trials, there is potential to gather insights as they happen.
Real Time Results
This is especially true as the wearable category enters into the conversation. These wearable devices track the user in all instances, 24 hours a day, allowing the person monitored to go about their lives in an ordinary way, all while being studied and having their habits tracked. This essentially adds to the big data conundrum, as more data is generated with this new set of analytics. This also adds in the exciting curve of the promise of real time data, which is a new frontier on all fronts - for pharma, CROs and patients. This sort of access has not existed up until very recently, and although there is tremendous potential, it has yet to be accurately harnessed.
Steve Rosenberg, Oracle’s Senior Vice president and General Manager of its Health Sciences Global Business Unit spoke to PharmaExec regarding the wearable’s phenomenon, though expressing skepticism. “The jury’s still out on wearable’s,” he said, adding, “I would advise taking a watch-and-wait approach. You may be able to measure something 6,500 times a day, but that doesn’t mean that the information is going to be meaningful or useful.”5
Connecting the Dots
Indeed, the addition of wearable seems to complicate the issue of big data, drawing attention to the main thing it is lacking, which is meaning. What good are the numbers without analysis? However, once this is consolidated, wearable’s integrated into a cohesive big data system will literally cipher information of that moment from any and all participants, depending on who is being studied, how and for what, and will take this technology into the next phase of usage. However, especially on a global scale, the possibilities once again seem almost too unlimited.
Taking this into account, Rosenberg comments that wearable’s need to be implemented where it is logical. “You’ll see wearables being used where it makes sense to collect the data on a wearable, but pharma has to be more cognizant of convenience to the patient,” he said. “Every time we put a wearable into a clinical trial, for example, we have to educate the site on how to train the patient to use it,” he added.5
Once this learning curve is overcome and patients, or even those in the general population, come to terms with their devices and even incorporate them into their routine, the potential for real time results integration into big data becomes fantastic. Rosenberg adds, “I think it takes hype to get enough movement, to get pragmatism into the process,” he continued. “There’s not as much data as there is, for example, in banking applications. We are not talking about an unmanageable amount of data; it’s just a question of what it’s going to be used for. There’ll be pockets of success along the way. Two to three years from now you’ll see a pretty large proliferation of data, but until then the success will remain in certain therapeutics areas,” he predicted of the future of this development.5
Data Gold Mining
For big data, this means the next step is closer than ever before, and some might even argue it is in reach. Though big data poses challenges due to the scale and breadth of information that will be available, as is pointed out by Nigel Walker, Managing Director of That’s Nice LLC and Nice Insight. “Some pharma companies are shying away from acting due to incomprehension at the sheer scale of the challenge, wariness of the limitless potential costs, or fear of pioneer disadvantage,” he stated in the Pharma’s Almanac Big Data, first quarter issue.6 However, the road forward looks bright for the companies who can lead the change and much like with serialization, are able to plan in advance.
Organizations looking to be successful in these areas must set up the internal capacity that is needed to be able to handle obvious big data challenges. If this is not an option, then partnering and finding an external resource is optimal.
As further evidenced by Nice Insight, the proof is in the numbers. In the Big Data first quarter issue of the Pharma’s Almanac, data was explored on all fronts after the company conducted seven research surveys.7 These surveys tracked the pharmaceutical landscape of CDMOs, CROs, excipients, intermediates, equipment, private equity and venture capital, as well as the supply chain. These surveys tracked outsourcing, investment and the shape of the market going forward. Using these numbers and subsequent data, the range of the industry can effectively be charted and mapped out to predict what is coming next.
As with all data, big or small, the analysis is where the added value enters the picture. There is little debate that data is intrinsically valuable and has the potential to unlock the way we view ourselves, helping us understand not only what we know about our behavior and how it affects our health, but also getting down to the genetic level and what our code means for unlocking the entire mechanics of our selves and our futures. Again, however, to accurately pull this, one needs to break it down. Data must be treated and organized in a cohesive way, that can shatter all barriers— international or otherwise, so that a central database can begin to make sense of all this information, and so that, before we know it, data is leading the charge. Once this information clearly presents itself, we will be better able to recognize the need, whatever it may be and subsequently treat it. We will then give the data the driver seat in research and development, and allow big data to fully serve its potential to impact not only the healthcare landscape but also the future of pharma and our next major developments.
References
- http://www.mckinsey.com/industries/ pharmaceuticals-and-medical-products/our-insights/ how-big-data-can-revolutionize-pharmaceutical-rand-d
- https://hbr.org/2016/11/the-promise-and-challengeof-big-data-for-pharma
- https://www.thedataincubator.com/
- https://www.gene.com/stories/big-data-and-thebig-c
- http://www.pharmexec.com/big-data-it-crunchtime-pharma?pageID=2
- https://www.pharmasalmanac.com/articles/bigdata-a-big-challenge
- https://www.pharmasalmanac.com/hubfs/ Back%20Issues/PharmasAlmanac_Q1_2017. pdf?t=1492635021555