Quality Processes in Clinical Digital Pathology Operations at Memorial Sloan Kettering Cancer Center

Background: Pathologists are becoming increasingly familiar with digital pathology for primary diagnostic use, and the field is witnessing a growth in pathologist buy-in and comfort using the technology. However, in order to enable successful clinical implementation, a suitable, multi-component digital pathology infrastructure must be in place. Each laboratory and organization will require foundational components, including hardware, software, and network components to support the digital pathology system. The exact infrastructure blueprint chosen for institutional deployment will depend on the laboratory use cases, resources, and overall strategy. An often-overlooked component of the clinical digital pathology operation is the need to ensure quality whole slide imaging workflows that will control image quality for the downstream pathology use cases. These quality considerations need to be applied in all phases of the digital pathology process: preanalytical, analytical and post analytical phases and get updated with new technologies and workflows that are being adopted and as scanning volumes increase. The objective of this study is to identify quality processes required for clinical grade digital pathology operations.Methods: Our team at Memorial Sloan Kettering Cancer Center (MSK) started developing a quality management system (QMS) shortly after the integration of digital pathology into our clinical systems in 2020. With the expansion of our operation in recent years, a departmental wide effort to identify additional potential quality needs for the different digitization phases was identified. The effort included multiple teams and a systemic review of all digital pathology related workflows from specimen accessioning to slide storage that led to changes in some of these workflows. The financial implication of these new workflows was calculated to ensure long term sustainability.Results: Identified new workflows in different phases of the pathology operation resulted in changes in glass slide delivery, the addition of automated tracking capabilities, added quality dashboards, turnaround time monitoring and overall lean operations that were integrated in our daily workflows. New centralized scanning laboratory workflows resulted in a decrease in the downtime of the digital scanners. A major effort was given to finding an automated solution for the labor-intensive image quality review process, which was previously found to be essential for our large-scale clinical pathology effort.Conclusions: The transition to fully digital pathology operations requires adjustments to current workflows. A systemic approach to identify quality improvements should be ongoing as technologies and scanning operations expand. The talk will describe these integration efforts and the financial implications of these improvements on our overall pathology operations.

Learning Objectives

1. Understand quality management system requirement for digital pathology.
2. Identify potential quality risks in different digital pathology workflows.
3. Develop an ongoing approach for a sustainable quality management system for their operations.