Real-World Data & Real-World Impact

In recent years, the rise and demand for real-world data (RWD) has transformed the landscape of healthcare research and drug development. Biopharma has traditionally relied on data from randomized controlled trials (RCTs) to advance drug development pipelines, however, this data is at times limited by sample size, narrow inclusion criteria, and the large time and financial investments required to conduct clinical trials. RWD has garnered increasing interest for its use alongside data from RCTs due to their complementary nature. RWD expands the breadth and depth of available data by leveraging the vast amount of data collected throughout routine patient care such as EHR, claims, and molecular testing, etc. As the volume and variety of health-related data sources expand, the ability to collect, analyze, and apply RWD has become crucial for researchers and drug developers to generate novel insights across patient outcomes, treatment effectiveness, and disease progression.

Today, researchers are using RWD throughout the drug development lifecycle: from identification of novel biomarkers, to enhancing clinical trial design, validating findings, and supporting regulatory submissions. Because RWD enables the analysis of healthcare outcomes across large and diverse patient groups “in the wild”, using RWD alongside traditional RCT data comprises an approach that may ultimately prove more cost-effective and informative than either method alone.

Although the use of RWD has gained traction across biopharma R&D, researchers are often still limited in the types of data they’re able to access and analyze. This is often due to the raw, unstructured nature of data within EHRs, LISs, and beyond, which, without appropriate curation, are extremely difficult to unlock.

Real-World Histopathology Data: The Missing Modality

I am text block. Click edit button to change this text. Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

While clinico-genomic RWD has driven tremendous insights across the drug development lifecycle, there is a critical data modality that has been left untapped due to difficulties with data extraction and analysis – histopathology.

Pathology data plays a pivotal role in precision medicine, elucidating the pathophysiology of diseases, capturing tissue and cellular-level disease manifestations, and quantifying morphological variations influenced by treatment or disease evolution. This essential data is routinely collected at the time of diagnosis and is either stored in physical glass slides or digitally scanned to create whole slide images (WSIs). Pathologists then analyze these slides or images, relying on their expertise and interpretation to guide diagnostic and treatment decisions. Despite the critical information contained within pathology samples and their broad availability, they can be difficult to analyze quantitatively at scale given their inherent data size and complexity.

The integration of pathology into RWD has thus been historically challenging; however, this strategy is advancing rapidly due to the advent of digital pathology and advanced AI-driven image analysis tools, like the natively embedded AI found within PathAI’s Image Management System (IMS), AISight®. By transforming raw, unstructured pathology images into standardized, structured data outputs, pathology data can be seamlessly integrated with clinico-genomic RWD to unlock the full potential of multimodal patient cohorts. These technologies usher in a new era of precision medicine where pathology samples can be rapidly and comprehensively analyzed at the scale necessary to meet global healthcare demands.

Pathology stands as an integral component for precise diagnosis of many different diseases. Despite its complexities and intricacies, advancements in digital pathology and AI-driven technologies are transforming what is possible with routine histopathology samples. Today, researchers can use standard hematoxylin and eosin (H&E)-stained samples to obtain a single-cell, spatially-resolved view of tissue structures and cell populations, as well as examine how patients respond to new therapies, understand prognosis, and even predict molecular biomarkers.

PathAI specializes in developing AI-powered products for the analysis of pathology WSIs, transforming unstructured digital images into structured pathology insights to advance precision medicine. This quantitative histopathology data unlocks completely new possibilities alongside other RWD, enabling a comprehensive understanding of the intricacies of diseases at a molecular, cellular, and now histological and anatomical level.

Transforming WSIs into data-driven insights: H&E images linked with AI-powered insights from PathExplore™ , the world’s first AI solution for rapid, high-resolution analysis of the tumor microenvironment from H&E WSI.

PathExplore delivers >300 histopathology features including:

Total Area of Tumor Tissue
Total Number of Lymphocytes
Density of Lymphocytes in Cancer Stroma
Number of Lymphocytes in Proximity to Cancer Cells
Spatial Distribution of Tumor Infiltrating Lymphocytes, and more

Clinical RWD

Understanding the patient journey is crucial for bolstering RWD as it provides a holistic view of patient healthcare experiences and outcomes. By tracing the sequence of interactions between patients and healthcare systems—from initial symptoms and diagnoses to treatments and follow-ups—RWD captures complexities that controlled clinical trials often cannot replicate. By utilizing clinical data as a modality, this offers significant advantages over traditional clinical trials, which are often expensive, time-consuming, and involve relatively small sample sizes that may not fully represent broader patient populations.

Clinical RWD typically includes data such as:

Demographics
Medical History
Cancer Risk Factors
Physical Assessments
Laboratory Tests
Disease Staging
Medications/Lines of Therapy
Surgeries/Biopsies
Progression/Outcomes

Molecular RWD

Molecular data encompasses a wide array of information, including genome, exome, and transcriptome sequencing data, which are critical components of the multiomics approach for understanding the molecular profiles of patients. Genomics provides insights into the complete DNA sequence and variations within a patient’s genome, while transcriptomics offers a snapshot of gene expression levels, revealing how genes are expressed in different contexts. By integrating these layers of molecular data, multi-omics create a comprehensive picture of molecular processes, enabling a deeper understanding of disease mechanisms and patient-specific responses to treatments.

Molecular RWD may include elements such as:

Whole Genome Sequencing
Whole Exome Sequencing
Whole Transcriptome Sequencing
Single-Cell Transcriptome Sequencing
Microbiome Sequencing

Unlocking the Future of Precision Medicine with Real-World Insights

RWD that includes clinical, molecular, and pathology data has a unique ability to drive innovation throughout the drug development lifecycle. However, pathology data has been historically underutilized due to its lack of resolution and structure. PathAI’s products are uniquely able to transform raw pathology images into structured pathology data. Together with leading RWD providers, PathAI is integrating AI-powered pathology RWD with clinical and molecular RWD to unlock the future of precision medicine. With these multimodal datasets, researchers and drug developers can gain deeper insights into disease mechanisms, biomarkers strategies, and novel therapeutic targets through comprehensive, multi-scale analyses; ultimately making drug development faster and more cost-effective to improve patient outcomes.

How We Can Help

Looking for a real-world dataset?
PathAI has built a network of industry-leading RWD partners, including Aster Insights and ConcertAI, allowing you to access a comprehensive range of curated real-world datasets. These datasets include clinical, molecular, and AI-derived quantitative pathology data, enabling multi-modal analyses across the drug development lifecycle. These datasets are available across oncology indications, including lung, breast, colorectal, melanoma, gastric, kidney, pancreatic, prostate, lymphoma, ovarian, bladder, head & neck, and liver cancers to address your specific use cases.

PathAI also licenses the PathExplore data generated from H&E WSIs from The Cancer Genome Atlas (TCGA), allowing researchers to analyze quantitative pathology features alongside clinical and molecular data from TCGA.

Looking to enrich an existing RWD cohort?
Reach out to us to enrich your RWD with AI-powered pathology. Our solutions transform raw WSIs into standardized and structured tissue and cell insights, enhancing your ability to derive meaningful insights and accelerate discoveries in translational research and clinical development.