Although most diseases are well defined, the patient experience for those who have the same condition can differ dramatically for a multitude of reasons. The physical and psychological impact on patients are significant. Comorbidities, genetics, financial standing, social life, ethnicity, life experiences, environmental factors also play a role. That’s why clinicians and pharma companies need to consider real-world data (RWD) for drug development. Without this crucial information, clinicians and drug developers create “blind spots” that prevent them from fully understanding the condition and what’s needed for effective treatments. A recent webinar hosted by MedCity News and PurpleLab explored the importance of patient-reported outcomes as vital to fixing these blind spots through the lens of non small-cell lung cancer (NSCLC).
RWD are indispensable tools for identifying and highlighting important risk factors that are frequently overlooked. These data, such as social determinants of health and socioeconomic barriers, provide a more holistic view and reveal these “blind spots” that traditional clinical trials, based on their design, simply can’t.
The main objective of clinical trials is to assess a new treatment’s effectiveness and its safety profile, usually in comparison with current, best practice treatments. Clinical trial protocols inform each aspect of a study, from diagnostic tests for biomarkers to the administration of treatment modalities, such as surgery, chemotherapy, targeted therapies or immunotherapy.
Patient-Centered Outcomes
One challenge with clinical trials is that it can take up to one year to recruit participants. Work and family commitments pose potential obstacles to regular, in-person attendance. The clinical nature of a study means that participants’ vital sign data often differ from their home environment. Variations in medication adherence, daily challenges, and the emotional toll of living with a chronic disease manifest differently for each clinical trial participant and can have a significant impact on clinical trial outcomes.
Although randomized clinical trials for people with NSCLC track tumor response and survival, there are many subtle nuances of the condition that significantly impact the quality of life for people with the disease but which clinical studies, broadly speaking, fail to capture. Pain and fatigue are the most common symptoms that can take an emotional toll over a prolonged period of time. They lead to an inability to exercise, make it difficult to walk, contribute to restless sleep, and pose challenges at work. Yet this information is not systematically recorded or assessed in clinical data.
Although clinical trials need a high level of patient adherence to be effective, adherence can be complex to unpack. Unmanaged side effects, comorbidities, such as depression, access to transportation, and the demands of a career can inform a patient’s adherence.
Social Determinants of Health
Defined as non-medical factors that influence a patient’s length and quality of life, social determinants of health (SDoH) span underlying social, economic and environmental conditions faced by individuals and their communities. SDoH are particularly relevant to oncology and are increasingly viewed as a critical factor impacting health and quality of life outcomes. Some studies tie SDoH factors to 75% of cancer occurrences. SDoH are recognized as independent risk factors for poorer health outcomes, exacerbating inequities across the cancer care continuum.
Given that NSCLC is a respiratory condition, it should not be surprising that the environment can play a significant role in the NSCLC patient journey, as Dr. Ben Freiberg, Principal Informatics Systems Lead with Genentech’s gCS Computational Catalysts, pointed out in the webinar.
“The ability to join other data to the data that we have from the patient journey is important. I could say, ‘Look, there’s [a higher number of people with asthma] in this town.’ But what I could also look for are pollution maps that show the types of pollution that exist across the country. Is this an environmental factor that’s weighing in and potentially influencing the development of asthma? The ability to join those other data sources in a way that’s meaningful and really population-based, as opposed to just any individual patient, really empowers looking for those deeper insights, looking for correlations that could be causation, depending on what you come up with for why a certain disease tends to be more prevalent in certain populations of people living in certain places.”
Regional legislation may also dictate the kinds of screening available from state to state.
Steven Emrick, PurpleLab senior vice president of clinical informatics solutions and HealthNexus®, noted that biomarker testing for non-small cell lung cancer for insurance plans that are regulated by the state varies. Sixteen states require this biomarker test.
“PurpleLab actually did a study on this a while ago, and they looked at which patient populations are receiving biomarker testing for non-small cell lung cancer. They found that, per capita, per 100,000 people, the rate of lung cancer between whites and African-Americans was very similar. Biomarker testing was largely skewed towards whites and African-Americans were largely not receiving that. There could be many different reasons for that, but it’s an ongoing issue in this country. There are these advancements in technology, but not everyone has access to those advancements that lead to better outcomes.”
He added: “To me, the North Star of real-world evidence is leveraging that data to influence not only the way drug development, therapy development, diagnostic development are accelerating, but also feeding those insights back into regulators and policymakers to change health outcomes.”
Patient Journey Mapping
One way to help visualize an individual’s complex healthcare experience over time is through patient journey mapping. The goal is to gain a better understanding of the barriers, support, interactions with services, and overall patient outcomes from the point of view of the patient. This approach helps identify points of friction and opportunities for improvement across the continuum of care.
Breaking down the patient journey into distinct stages facilitates comprehensive patient mapping. Among the common frameworks are:
- Pre-diagnosis: Capturing initial symptoms, self-assessment, research, and initial concerns
- Initial Contact: The first direct interaction with the healthcare system (e.g., call center, in-person visit)
- Diagnosis: The process of confirming the condition and its staging
- Treatment: Active management of the disease, including therapies and ongoing care
- Post-treatment/Ongoing Care: Follow-up, symptom management, lifestyle adjustments, and long-term well-being
Leveraging RWD Sources:
Electronic Health Records (EHRs): These provide granular clinical details, including diagnoses (e.g., ICD-10 codes), procedures (e.g., CPT codes), laboratory results (e.g., biomarker tests), prescriptions, and physician notes. While rich in clinical depth, EHRs may lack a comprehensive view of care provided outside a specific health system and often contain unstructured text that requires advanced processing.
Administrative Claims Data: This data captures billing and reimbursement information from payers, offering a longitudinal view of patient encounters across various providers and settings. It is invaluable for understanding healthcare utilization, costs, and tracking patient flow over time.
Patient Registries: These systematically collect specific, often granular, information on patients with a particular disease or receiving a specific treatment. Registries can include data elements not typically found in EHRs or claims, such as detailed biomarker assessments, behavioral factors (e.g., smoking status), and patient-reported outcomes (PROs).
Social Media Data: Public posts and discussions on lung cancer-specific forums and platforms provide unfiltered, real-world insights into patient symptoms, side effects, treatment challenges, and emotional impacts. This offers a raw, real-time understanding of patient sentiment and priorities.
Patient-Reported Outcomes: These are direct reports from patients about their health status, symptoms, functional status, and quality of life, captured without interpretation by a clinician. PROs are crucial for understanding what truly matters to patients, such as symptom relief, quality of life, and treatment satisfaction.
Analytical Techniques: Qualitative data analysis (QDA) is employed for social media posts and interview transcripts to identify themes and patterns. For larger datasets from EHRs and claims, advanced analytics, including AI and machine learning, are used to identify trends and predict patient behaviors. Retrospective and prospective observational studies are common designs for analyzing RWD.
Leveraging the immense, diverse forms of RWD, facilitated by the digitization of patient data, is revolutionizing drug development. It is advancing a more personalized approach to healthcare and helping clinicians to meet patients where they are. It is the only way forward to improve clinical study recruitment and participation and will help us develop more effective drugs that not only improve patient health for diverse patient populations, but also their quality of life. Reducing treatment side effects will also improve medication adherence and lead to a healthier, more robust healthcare industry.
Photo: Natali_Mis, Getty Images