he National Cancer Institute’s (NCI) Cancer Moonshot program was launched five years ago with the goal of generating new cancer research via an additional $1.8 billion investment into the 21st Century Cures Act. The main goals of the program are “to accelerate scientific discovery in cancer, foster greater collaboration, and improve the sharing of data.” To date, 240 research projects across more than 70 initiatives have been supported by Moonshot funding, with more to come in the next seven years. One such project is Digital Biomarkers for Clinical Impact (DigiBioMarC). This study aims to understand and improve the quality of life for patients receiving cancer therapy, which is critical, because many patients undergoing treatment for serious conditions like cancer often struggle to tolerate their aggressive treatments.

Are clinical studies collecting only a portion of the data that are needed to make rapid and significant advancements in cancer?

The journal Digital Biomarkers defines digital biomarkers as “objective, quantifiable physiological and behavioral data that are collected and measured by means of digital devices such as portables, wearables, implantables, or digestibles.” These clinically significant and objective health data can then be used to “explain, influence, and/or predict health-related outcomes.”

Think of how influenza sets in and changes the body’s physiological aspects before revealing itself in the form of symptoms. For instance, a person’s temperature may rise due to an infection and troponin levels may rise if the flu virus replicates in the heart muscle and causes cardiac damage. What if we could detect early signs of disease before the condition becomes serious?

Researchers from Stanford University have showcased the use of wearable sensors for early symptoms and possible disease detection. This advanced warning or prediction is one of the more potentially exciting benefits offered by digital biomarkers. In one cardiac example, a network of sensors detected clinical signs of decompensated heart failure more than three months before hospitalization. The increased number of toss-and-turns in bed, elevated heart and respiratory rates at night detected by bed sensors, along with decreased physical activity and increased nighttime bathroom visits detected by ambient passive infrared (PIR) motion sensors, combined to serve as a digital biomarker for early detection of changes in heart function before the onset of heart failure symptoms.

Speech is also gaining ground as a biomarker. Physiological, neurological, and even psychological changes can impact how our voice vibrates and sounds. Karger’s Digital Biomarkers journal notes, “Speech offers rich insights into cognition and function and is affected by many psychiatric and neurodegenerative diseases. By requiring the coordination of various cognitive and motor processes, even a short sample of speech may provide a sensitive snapshot of cognition and functioning relevant to many disease areas.” Speech sensors and analytics hold great potential as digital biomarkers for specific symptoms, neurodegenerative and respiratory diseases, and cognitive and memory functions.

It is this promise, combined with the ability to measure patients across multiple time points in the day, that shows new hope for the future of healthcare and clinical trials, especially in areas like oncology.

Biomarkers Get Ahead of Chemotherapy Complications

Scalable and flexible technology solutions that allow passive, continuously measured, analyzed data streams bolstered by connected sensors for direct and indirect measures outside the clinic are being explored in cancer patients. The Digital Biomarkers for Clinical Impact research program aims to understand and improve the quality of life for patients receiving chemotherapy.

A variety of sensors that directly and indirectly detect patient activities helps in the collection and analysis of digital biomarkers before, during, and after chemotherapy to gain a better picture of out-of-clinic physical function and fatigue. This is important for several reasons:

1. Early symptom recognition facilitates earlier and often less complicated supportive care that can also be more effective at addressing the symptom(s). This is better for the patient while reducing care costs and chances of clinical trial dropout.
2. Clinicians and investigators may need to use a patient’s response to treatment as part of any treatment change decision. Thus, longitudinal data can be quite powerful.
3. Patients and providers are aligned in their desire to reduce chemotherapy-related fatigue to help patients enjoy a higher quality of life during treatment and minimize the risk for interruptions in clinical trial participation.

Meaningful to Patients and Clinical Teams

In the Digital Biomarkers for Clinical Impact study, one study partner augments the data from wearable sensors and objective outcomes data that another partner collects. Data are bolstered by smartphone-captured speech and language-based information about motor, cognitive, and respiratory functions. This robustly extends, with very low burden on the patient, the scope of digital measures that can be captured outside the clinic. These data are then combined with patient reports via validated ePROs and electronic diaries that detail how patients are feeling. In this NIH-funded work presented in September 2021 at the American Society of Clinical Oncology (ASCO) Quality Care Symposium, 50 patients on therapy from within the Kaiser Permanente Northern California (KPNC) oncology clinics and their informal caregivers (most often [75%] their spouse) participated in a fully remote decentralized study and used multistakeholder co-created mobile apps for approximately one month. At the end, participants reported that use of a mobile app with specific features improved their own emotional well-being, encouraged patients and caregivers to take better care of themselves (including being more physically active), and increased caregiver awareness of patients’ symptoms and treatment side effects. They also perceived that communication with clinical teams might improve and doctors might be able to act earlier to mitigate symptom or side effect development.

At the same time, in a survey of oncologists at KPNC, findings nicely dovetailed as oncologists reported their belief that patient/caregiver reports of symptoms and physical function can predict adverse events (AEs) and most oncologists would like to have access to physical function/symptom data with many preferring to receive data prior to a visit and/or “critical values.” Oncologists also indicated that they routinely seek caregiver reports of patient symptoms and physical function, and they often use patient/caregiver reports in clinical decision making.

Of course, there are many specifics to work out, such as optimal disease-specific assessment frequency, defining critical values, best-in-class sensor data, and the practical aspects of integration with patient and caregiver lives and existing clinical workflows. However, in many instances, longitudinal changes in remotely captured data will become essential in the future for monitoring therapy-related declines and documenting early improvements, such as quality of life. This will also enable clinical teams to better serve patients, improve health outcomes, and complete clinical trial work at a faster pace. The promise is clear, and testing and implementation are already here.

Reduce Burden and Improve Data

References and sources available upon request.