Digital pathology review equally efficacious as glass slides in Blueprint phase II
Digital images and glass slides reviews of immunostained programmed death-ligand 1 (PD-L1) assays are comparable in reliability, says a researcher.
Speaking at the recent International Association for the Study of Lung Cancer (IASLC) 18th World Conference on Lung Cancer (WCLC) in Yokohama, Japan, Dr Ming Tsao, chairman of the Blueprint 2 committee and professor of Laboratory Medicine and Pathobiology at University of Toronto, Canada, said one of the conclusions reached by the committee was that PD-L1 tumour cell scoring on digital images and glass slides demonstrate comparable reliability. In other words, the review done on digital slides closely paralleled those made on glass slides.
The materials for the Blueprint Phase 2A study were 81 samples collected from routine clinical pathology practice (to reflect different sample and histological types). In the study, 25 pathologists from the International Association for the Study of Lung Cancer (IASLC) Pathology Committee independently scored the five sets of slides which were individually stained by the five PD-L1
(22C3, 28-8, SP142, SP263 and 73-10) tests.
“The results of Blueprint 2 involving routine clinical lung cancer samples and 25 pathologists largely reaffirms the Blueprint 1 results,” said Tsao. Of the five assays, three (22C3, 28-8 and SP263) are comparable for scoring tumour cells and can thus be interchanged in their use as immunohistochemistry tests—as long as the clinical algorithm is matched with the individual drug in patient selection.
About the Blueprint Project
The Blueprint project came about in 2015 when a group of four pharmaceutical and two diagnostic companies released a Blueprint proposal to analytically compare and characterize all their immunohistochemistry (IHC)-based PD-L1 companion diagnostics for NSCLC. The stakeholders were part of a workshop (jointly organized by the American Association for Cancer Research, US FDA and American Society of Clinical Oncology) to examine whether multiple companion diagnostics intended for the same class of therapeutics could be harmonized.
The rationale for Blueprint was that the information generated by the project would serve as the groundwork for further studies that will be educate patients, physicians and pathologists on how to utilize the test results to determine the best treatment pathway.
Blueprint phase 2 aimed to validate the comparability results obtained from Blueprint phase 1 for five assays with a larger cohort of lung cancer specimens and a larger pool of pathologists from all over the world. Additionally, phase 2 of Blueprint also tried to assess concordance of PD-L1 scoring using light microscopy compared with digital reads done through a web system.