Personalizing the Future of Healthcare

The technologies we use on a daily basis – from search engines to streaming music services – recognize patterns in our behaviors and, coupled with vast amounts of data from millions of other individuals, make predictions of what each of us may want in the future. In so doing, the power of data at scale has transformed almost every industry, personalizing our experiences with different platforms in many aspects of our lives. Yet in healthcare we’ve only scratched the surface.

The dream of bringing about this transformation in healthcare has been around for several years, but it wasn’t achievable because the digitization of healthcare has never been at the requisite level, until recently. Depending on who you ask, the future of personalized healthcare can mean many different things, but the unifying theme is that it will be about meaningful data at scale. In our view, that means utilizing technology to acquire more data from each patient than ever before, and bringing together data from more patients to arrive at a deeper understanding of how to treat an individual based on these ultra-high resolution profiles. Only then can we visualize what distinguishes each of us as individuals, and translate that into a clear, personalized path towards improved care for every person.

We know that the use of real-world evidence, molecular information generated from next-generation sequencing, data from wearable devices and mobile apps and results of novel clinical trials can help transform the future of care. However, we need to combine and aggregate this information in a manner that enables answers to questions which meet the needs of patients. And for us to get to those meaningful answers, we need to have a certain breadth and depth – and quality – of that data.

View from the Real World

Real-world evidence – data acquired in everyday clinical practice – can provide valuable insights drawn from information about individuals’ lifestyles, disease biology and treatment outcomes. Thanks to new technologies and data science approaches, as we have been pursuing with our industry partners, we can harness and aggregate real-world data as a powerful complement to traditional clinical trials. We envision applications including new ways to track disease, allowing for optimization of treatment approaches, or insights about patient populations which accelerate clinical development. For example, using data from aggregated medical records, we could potentially replace the control arm in randomized clinical trials to make them more efficient and work with regulatory authorities to increase the speed at which we can bring medicines to patients.

Real-world evidence not only helps us understand existing medicines, but can also inform the development of future treatments. Taking insights from a patient’s bedside back to the research bench is a process we call reverse translation.

A New Era of Diagnostics

For cancer medicines, one of the tools that may inform optimal use are companion diagnostics. Biomarker tests now commonly inform treatment decisions – like PD-L1 for immunotherapies, or alterations in the ALK gene to guide the use of specific targeted therapies.

However, these companion diagnostics are only the beginning of our journey toward the future of personalized healthcare. Technologies like next-generation sequencing can map out an individual’s full genetic makeup, tumor mutations, and other defining molecular features to find the most appropriate treatment. In addition, liquid biopsies may allow us to non-invasively track how a cancer evolves over time and adjust treatment accordingly.

We’re working with others in the industry to advance the development of such next-generation comprehensive diagnostics. With these tools, we can begin to untangle which treatments produce the best outcomes and help healthcare providers decide how to deliver fully customized care plans – the ultimate goal of personalized medicine.

Capturing More with Digital Health

Digital health platforms like wearables and mobile apps on smart phones can help us gather more information and also capture a critical new perspective: the patient voice. People can report detailed information about their symptoms, treatment burden, quality of life and other experiences, actively and passively documenting their health in detail and in real-time, in a way that goes far beyond the standard tests performed episodically at their doctor’s office.

In addition, data monitored in real-time can alert healthcare professionals to problems that might require immediate attention. And looking at longitudinal data may reveal how a person’s disease changes throughout the day or over many months.

Together with more comprehensive next-generation diagnostics, digital health tools can also inform treatment recommendations as the landscape grows more complex. Alongside others in the cancer community, we at Genentech are actively working to support the development of clinical decision support platforms, informed by real-world and clinical trial data as well as molecular and digital data types going into the future.

A Little Help from AI

With meaningful data at scale, we are learning that artificial intelligence (AI) approaches, such as machine learning or deep learning, can help us analyze these vast amounts of healthcare data to derive insights we previously could not have realized.

One area well-suited for AI is medical image analysis. MRIs, CT scans, pathology slides and other images are routinely used in standard care. AI approaches could help streamline processes for image interpretation that currently require hours of labor, allowing for faster decision-making when people with life-threatening illnesses may not have the benefit of time. In addition, AI can potentially enable more consistent interpretation of images, resulting in better-informed treatment decisions.

Moreover, we and others are recognizing that medical imaging data is an incredibly rich data source, harboring additional information and insights into disease which go beyond what human interpretation can capture. We are beginning to use AI algorithms to learn to better distinguish healthy versus diseased tissue, identify early warning signs, and reveal unexpected patterns which enable diagnosis and prediction of clinical outcomes.

The Big Picture

We have traditionally thought about data as a means to an end, but in this new era of meaningful data at scale, we now see that every data point we acquire – while it may initially serve a specific experiment – can continue to serve us in a grander frame. Patient data, across all of these dimensions, actually has enduring additional value when aggregated and accumulated to ever-increasing scale, allowing us to answer new questions we couldn't have before. At the same time, we should also recognize that it’s not as simple as just collecting a large volume of data. We also have to make sure these aggregated data are appropriately de-identified or anonymized to protect patient privacy. To bring this new paradigm to life, the whole healthcare system – from the way we develop therapies to the way care is delivered to patients – has to evolve to accommodate this new kind of patient data and the means to act on the insights that will follow.

By integrating data from all available sources and analyzing it in smart ways, we can finally get a complete picture of an individual’s medical profile and define a truly personalized approach to care. With a greater understanding of how our medicines work in the real world – how they variably work for patients of different ages, ethnicities, genders, family histories, medical conditions, etc. – we can make more informed choices in how we develop new therapies and also deliver more optimal care for each individual person. At the end of the day, that is what personalized medicine is all about.

We’re partnering with leaders in health and technology, as well as physicians and advocates, to realize that vision. It’s an exciting time for everyone working to improve outcomes in cancer and other diseases, and especially for the patients who will ultimately benefit from the modernization and individualization of healthcare. It's a privilege to be part of this evolution, but it’s also a responsibility. The patients are waiting, and we look forward to seeing what’s next.