The 21st century world is driven by data, which is at the heart of progress. The pharmaceutical industry is no exception to this. The amount of data available today regarding patient habits, ailments and symptoms is staggering. Gone are the days when companies had to only rely on data obtained from focused clinical trials to gain insights. Today, they can get data from myriad sources and link it together to find patterns that can potentially transform not just the pharma industry, but the quality and longevity of human life.
The sources of data are not just limited to clinical trials; they now stretch to anonymize insurance claims data, genomic data and Electronic Health Record (EHR) data. Tapping into these information streams has become simpler and cheaper than ever, and that helps pharma companies address complex research questions. For instance, real-time EHR data can facilitate better understanding of diseases, treatment patterns and clinical outcomes in real-world scenarios when patients visit hospitals. It also can aid in the inclusion of patients who could previously never enter clinical trials–he elderly, handicapped, or those suffering from rare diseases.
The drawback of previously obtained clinical data was that most insights could never be published. So, many opportunities for gaining knowledge about treatment efficacy and safety were passed up, and at considerable costs. Today, companies source trial data through connected mHealth services, smart wearables and remote monitoring. So, the accuracy and frequency of data is much higher. However, researchers and clinicians now need to access this data in a sustainable way to advance science.
Leveraging Big Data for Clinical Insights
Data obtained by pharma companies runs into terabytes, and is present in complex and unstructured formats. They need an effective mechanism to collect this data, organize it into comparable formats, and then analyze the same. The results must be used to uncover unexpected patterns, or validate existing facts. A lot of companies use this methodology in their research facilities to get insights into the nature and size of sub-population groups who can be served by new treatments.
While it gets harder to find patients who fit into narrow indications for targeted studies, Big Data can come to the rescue. It can find the right patients for trials, and make the process faster and cost-effective. Ultimately, this enables pharma companies to accelerate everything–right from studying symptoms to determining drug compliance. Reducing the time spent on this process can generate substantial cost savings, and it all begins by bringing eligible patients to clinical trials. According to the National Cancer Institute, only 5% of cancer patients partake in trials. One can only imagine the outcome if more were to participate. Other studies have shown that 37% of clinical trials fail to reach their recruitment goals.
Big Data can go beyond patient identification and recruitment, and can even be used to identify targeted therapies based on genetic markers. It can be utilized to evaluate protocol feasibility, identify adverse responses among sub-populations, or assess safety responses during interim evaluations in adaptive trials. What this means is that Big Data is a tool that drug makers cannot afford to neglect for their research efforts.
Nonetheless, very few groups are piloting the use of Big Data. The potential is immense, considering that more than 60% of trials use preliminary online analysis to identify subjects who have actively shared or sought medical information. The most successful examples of Big Data adoption relate to companies leveraging information collected from previous clinical trials to create baseline parameters and control groups. A few notable examples include:
- Yale University Open Data Access (YODA) project – To responsibly share clinical research data between the medical community and society for patient benefits
- $30bn strategic alliance between Columbia University Medical Center and a leading biotechnology company – To create a gene sequencing and analytics facility and a post-doctorate program to further knowledge about disease pathways and genetic mechanisms
- Collaboration between Multiple Myeloma Research Foundation and a data analytics company – To study data of 1000 patients with multiple myeloma; aiming to understand the relationship between genetic profiles and clinical outcomes
- Partnership between a healthcare technology company and National Comprehensive Cancer Network (NCCN) – To develop a cloud-based outcomes database that uses aggregated EHR data to identify patterns in oncology care
It’s clear that partnerships between academic research institutions and Big Data analytics vendors will have a significant impact on future therapies and treatments. Eventually, this will boost personalized medicine as researchers better understand complex diseases and determine treatments.
Barriers to Data-driven Clinical Trials
The challenge in utilizing Big Data comes down to two issues–privacy and ownership. Questions such as who owns the data, and does it invade patient confidentiality, are top of everyone’s mind. Recent medical survey found that while health care industry professionals show higher levels of trust with regard to data usage, the public is skeptical. The masses, overall, prefer that data be used through personal physicians rather than the government or even the health care industry at large. They also wish to separate health care data from daily life data such as social media posts or online shopping/browsing history.
The pharma industry thus needs to determine if individuals give permission to collect some data, can it extend to other aspects? They also need to decide if this process must be opt-in or opt-out by default. This grey area requires some more years of regulatory struggle to find consistent rules even though the benefits of opting for Big Data for clinical trials are pretty evident.