As a biomedical engineer, one of the reasons I was excited about joining the faculty at the University of Washington in 2014 was the potential to interact with clinicians at one of the strongest medical schools in the country. These hopes were quickly realized when, two weeks after arriving at UW, I received an email from an ambitious first-year resident in the pathology department, Dr. Nick Reder, who wanted to see if we could build a “flat-bed scanner for tissues.” Nick’s request was a pragmatic one — he was frustrated by the large amount of time and labor that was necessary to routinely process the 20 to 30 tissue specimens that resulted from every prostatectomy procedure, especially when many of those specimens did not contain tumor and were not of interest to the pathologists.
“It would be nice to get a sneak peek of the surface of each specimen so that we can focus on the specimens with cancer,” is how Nick pitched the problem to me. At the time, I was not working on imaging large clinical specimens but was focused instead on developing miniature microscopes for in vivo imaging applications on patients. Nick’s ideas and infectious energy, however, inspired me to start brainstorming about an ideal technology to meet his clinical needs. At the time, light-sheet microscopy was becoming popular as a new research tool, particularly amongst developmental biologists and neuroscientists. In thinking through various microscopy strategies, it seemed that many of the well-known advantages of light-sheet microscopy — namely, ultrafast volumetric imaging with efficient generation and collection of fluorescence — would be ideal for slide-free pathology as well, not only for obtaining a “sneak peek” of fresh tissue surfaces, but also, as we would subsequently realize, for volumetric microscopy of surgical specimens and biopsies to improve our ability to treat and diagnose diseases.
The next challenge was finding the right engineer in my lab to build a light-sheet microscope optimized for clinical pathology. Soon after Nick contacted me in the Fall of 2014, I received another timely email, this time from an extremely talented biophotonics engineer, Adam Glaser, who was finishing his doctoral work with Brian Pogue at Dartmouth and looking to relocate to Seattle. Adam was the perfect person for the job and arrived at just the right time as Nick was preparing to spend a year in my lab as a research fellow. By the Spring of 2015, the stars had aligned to bring together what I now consider to be an ideal quartet of characters for an effective collaboration: a clinical mentor in Dr. Larry True, one of the most experienced genitourinary pathologists in the nation, who could partner with myself as an engineering mentor for two highly creative trainees with complementary skills: Adam in the area of optics and Nick in pathology.
Figure 1 | A subset of the co-authors, from left to right: Professor Jonathan Liu, Dr. Adam Glaser, Professor Larry True, Dr. Nick Reder, and Ms. Ye Chen
The work described in Nature Biomedical Engineering is just the tip of the iceberg in terms of what must be accomplished to make slide-free nondestructive 3D pathology a standard of care. As we work to expand our clinical network to translate our technologies, we are also building ties with industry and other researchers to bring these strategies to market and to improve them for real-world clinical use. If there is one thing that this project has taught us, it is that we must always be willing to pivot quickly in our attempts to navigate a path through various obstacles. Working at an institution in which engineers like myself have the opportunity to interact with clinical collaborators on a near-daily basis allows for continuous and rapid course corrections and refinements that are essential to move towards our ultimate goal of improving patient treatments and outcomes.
Video 1 | A volume rendering of a 3D microscopy dataset of a 1-mm diameter prostate biopsy, false colored to mimic conventional H&E histology.
Our paper: Glaser, A. K. et al. Light-sheet microscopy for slide-free non-destructive pathology of large clinical specimens. Nat. Biomed. Eng. 1, 0084 (2017).