In recent years, the world has become much more aware of the promise of Big Data, machine learning, and artificial intelligence. Indeed, the ability to extract unprecedented insights from vast troves of unstructured data, using smart algorithms and game-changing computing power, is expected to be tremendously valuable to those industries that are bold enough to apply the technique.

In drug development, it’s a particularly seductive idea. Drug development is a complex and expensive business that involves looking for a molecule that can affect a biologic process in a very specific way, developing that molecule into a prototype drug, and then driving it through three phases of clinical trials – big data benefits are hugely powerful. After all, the failure rate of this approach is so high that it can take decades and costs hundreds of millions of dollars to get a drug to market. There’s got to be a better way, and for some, big data holds out the greatest promise of making the drug-development process less arbitrary and more effective.

Big data strategies are already being used to improve the way in which potential drug molecules are chosen for further analysis, to speed up pre-clinical testing, to track the progress of those tests more effectively, and to correlate test effects with particular variables. And there’s promising research underway about using in-silico simulations to explore how an organism will react to a drug, rather than having to do the work in vitro, on a lab bench.

A Research and Markets report on Artificial Intelligence (AI) in healthcare and pharmaceutical Industries predicts the spend on AI to increase from $463 million in 2019, to $2,452.7 million by 2025. This includes money being pumped into big data approaches to drug development. If they work, then perhaps we could quickly create more effective versions of current drugs with fewer side effects. Perhaps, for example, we could use big data to analyze whether existing drugs could be repurposed to work on so-called ‘orphan’ illnesses, which affect such small (or poor) populations that it is currently not cost-effective to develop therapies for them. And eventually, perhaps big data techniques will enable personalized medicine, in which individual drugs are tuned to each user’s unique needs and circumstances at the point of use.

How do we capitalize upon this promise? One of the most important steps we can take is to ensure that the data which underpins these big data drug-development strategies is well-founded. Tools that capture real-world evidence will improve the quality and timeliness of the data to which big data strategies are applied, or accelerate the speed with which testing and clinical trial insights are fed back to researchers. This can only improve the process.

How do we achieve this? Connected drug-delivery systems may provide part of the answer. These systems combine traditional therapeutics with computer-mediated delivery to help ensure that patients get the right dose of the right drug at the right time. The approach has several advantages. It can help patients get the full benefits of their prescribed drugs, by ensuring they comply with the dosage regime. It can tune drug delivery to the patient’s situation by monitoring the relevant vital signs and then adjusting drug doses to match the current need. It can enable patients to be treated at home rather than in a hospital. And it can create a tight feedback loop between the patient and physician for improved monitoring and oversight – while also providing well-validated patient data for analysis in other contexts.

Connected drug-delivery systems are an attractive option for improving patient care now and for enabling more effective big data drug-development strategies in the future. Developing such systems is not straightforward, though. It demands deep experience in medical-grade electronics design, firmware and software development. It also requires experienced human-factors oriented design teams to create simple user interfaces and intuitive user experiences. All this has to be done while ensuring high levels of patient safety, privacy, and data security – and meeting the requirements of a robust regulatory environment.

Molex and Phillips-Medisize, a Molex company are well positioned to meet the challenges of developing connected drug-delivery systems through our experience in hardware, software, and user-experience design. We have worked with pharma companies to facilitate their connected drug-delivery projects, supporting the process from concept development through to clinical supply and commercial launch of CE-marked, FDA-approved devices.

We’re playing other roles in the development of connected drug-delivery systems. At one end of the process, we provide interconnect solutions that enable the efficient operation of the kind of cloud data centers in which big data analytics is carried out. At the other end of the process, Molex is enabling the roll-out of next-generation 5G mobile networks and 5G enabled handsets that will connect these devices and create new use cases.

5G has several critical enabling characteristics for connected medicine. The 5G standard is designed to support 1000 times more devices per unit area than 4G, so if connected drug-delivery systems become as commonplace as fitness trackers, 5G will be well equipped to provide the required connectivity. Additionally, low-energy, low-data-rate communication standards may enable future drug-delivery or related medical devices to connect via simple IoT protocols. And 5G may play a critical role in bringing high-performance connectivity to rural areas, enabling telemedicine in remote areas and reducing barriers to connected drug-delivery devices where they may be most needed.

As the old adage says: garbage in, garbage out. Using connected drug-delivery systems to provide well-validated patient data should help improve patient care and therapeutic effectiveness in the short term. In the longer term, real-world evidence collected by connected drug-delivery systems will complement the drug-development process. If the two strategies can work together effectively – and we hope they can – these solutions will help pharma companies to develop better therapies more quickly to the benefit, first and foremost, of the patient.