Benefits of Digitized Pathology for Clinical Trial Sponsors and Pathology Teams

Advancements in ultra-speed scanners, high-resolution imaging tools, and artificial intelligence-driven data management strategies are reshaping age-old lab imaging processes within clinical trials. These transformative solutions and workflows bring many benefits, including streamlining operational workloads and enhancing the accuracy and speed of data analysis. 

However, some sponsors and lab teams hesitate to integrate these pathology solutions into clinical trials. This hesitation has made it necessary to better understand how digitized pathology can help address long-term challenges in imaging analysis, especially when trial timelines and budgets matter more than ever.

Manual vs. digital imaging 

Digitized pathology offers tangible benefits for clinical trial efficiency by enhancing traditional workflows with tech-enabled ones: 

• Efficiency in time, cost, and collaboration 

Slow-moving and disjointed, one-by-one glass slide image analysis is being replaced by digitized tissue slides and scanners, which are more easily shared for teaching/training purposes and second opinions on complex cases. Replacing fragile glass slides that have higher shipping costs and damage risk with automated digital slides stored on cloud sites that are accessible via comprehensive platforms means experts across the world can review a case within seconds. 

• Increased resolution and dimensionality

With AI-based solutions, pathologists can view tissue sections on a molecular level in 3D. 3D solutions can simultaneously display tissue at 2X and 200X magnification, providing clinicians with critical measurements (e.g., comparisons of tumor nuclear features between cell and tissue types) and deeper insights. 

• Quantity and quality

Manually analyzing distances between key elements in tissues can be time-consuming for overloaded pathologists. With computational systems, experts can easily obtain volume measurements to review and compare, enhancing their throughput. Relational contexts can also be analyzed using spatial analytic tools.

The shortage of pathologists

The industry-wide shortage of experienced pathologists adds reason to consider adopting tech-enabled diagnostic processes—not to replace experts but rather to reduce the burden on the existing workforce. There are several ways digital solutions can help: 

• During COVID-19, pathologist shortages led to a significant backlog of patient cases needing diagnoses. The Centers for Medicare & Medicaid Services encouraged remote review of pathology slides to ensure workforce safety while maintaining workflow. With increased use, the cost of imaging scanners has rapidly declined and they have also become more portable. Scanner integration allows pathologists to review and provide analyses from any location without time lags. As patients resume visiting sites in person for elective surgeries and other delayed procedures, pathologists can carry out remote review and analysis when convenient   
• AI-driven in vitro diagnostics tools are increasingly being validated by regulators for patient use (e.g., validation via CE marking for IVD use in Europe). As such, time-consuming, monotonous tasks (e.g., counting mitosis) can be reduced by digitizing image slides, freeing up experts’ time to focus on quality of care. 

Only the beginning 

Tech-driven advancements are already helping accelerate the process of image review and diagnosis. With new specialized cameras, photographic technologies, computational approaches, and comprehensive data management platforms, we are creating opportunities for higher throughput and a more holistic approach to care.

Taking in increasingly complex processes for review along with demographic data, lab results, and more, pathologists can analyze patients from various levels, providing a diagnosis with subtle yet deeper insights. 

However, as the capabilities of these solutions improve, discussions around updating regulatory guidance, increasing patient data privacy, and reducing algorithmic gender/race bias are required.