This article is more than 5 years old. Information may no longer be current.
Novel microscope shows potential for scanning tumors during surgery, assessing biopsies
Researchers at University of Washington have invented a microscope that can potentially image the margins of large tissue specimens in a rapid and nondestructive manner, with the same level of detail as traditional pathology.
The microscope conserves tumor tissue for genetic testing and diagnosis. It also allows experts to assess biopsy samples in 3D view.
“Slide-based pathology is still an analog technique, much like radiology was several decades ago when X-rays were obtained on film,” Jonathan T.C. Liu, PhD, associate professor and Bryan T. McMinn endowed professorship director of the Molecular Biophotonics Laboratory at University of Washington, said in a press release. “By imaging tissues in 3D without having to mount thin tissue sections on glass slides, we are trying to transform pathology much like 3D X-ray CT has transformed radiology. [Although] it is possible to scan microscope slides for digital pathology, we digitally image the intact tissues and bypass the need to prepare slides, which is simpler, faster and potentially less expensive.”
HemOnc Today spoke with Liu about how this microscope works, the efficacy it has demonstrated so far, how much additional research must be done to confirm its benefits and the potential impact it could have for patients.
Question: How does the microscope work?
Answer: Conventional histology technology has been around for decades and has not fundamentally changed for more than 100 years. The technology takes tissues, whether surgically excised or biopsied, and chemically and mechanically processes the tissues first with fixatives and embedding them in paraffin wax blocks so they can be cut with what amounts to a fancy deli-meat slicer. This is necessary because light does not penetrate tissues very well. To get a clear image with a traditional microscope, one must cut the tissue into extremely thin sections and mount it on a glass slide. This process is very involved, usually requiring a few days, and is not something one can do in real time during a surgical operation. Frozen sectioning is a faster alternative to gold-standard histology of paraffin-embedded tissues, but is less reliable for many tissue types. A common problem for all forms of slide-based histology is that pathologists can typically only visualize a small number of thin 2D sections from a large 3D tissue. In other words, there is a major sampling problem where experts are often imaging much less than 1% of the whole specimen. To try to address these limitations, our microscope enables fresh intact tissues to be imaged without having to cut them with a knife or embed them in paraffin wax or a freezing medium. It is a much simpler approach through which a tissue specimen freshly obtained from a patient can be stained for just a few seconds to a few minutes in a fluorescent staining mixture and then placed directly on our microscope. Our microscope looks like a flat-bed scanner for tissues. There is a glass plate at the top of the microscope and the tissues are simply placed on top of that plate and scanned from below. We are able to obtain images that are almost identical to gold-standard histology images of tissue sections on glass slides. Except now, we can nondestructively image the tissues and provide 3D information, which we believe will help pathologists to diagnose many lesions more accurately and effectively. This, of course, is important for guiding treatment decisions and improving the quality of care for patients.
Q: What type of efficacy has the microscope demonstrated so far?
A: So far, we have anecdotal demonstrations and have not conducted a large clinical study. This is our next plan, to show that our technology can improve patient outcomes by improving the diagnosis and grading of tumors. One example of a potential clinical application that we described in our recent paper is examining prostate biopsies. We have shown that comprehensively imaging the biopsies in 3D may allow pathologists to more accurately determine the grade of a tumor because they can see the complex glandular structures more clearly. This is important, for example, because patients with Gleason pattern 3 tumors usually do not require treatment, but Gleason pattern 4 patients are often recommended to undergo surgery or radiation therapy. These differences in tumor grade have dramatic consequences for the patient, especially because prostate cancer treatments can have serious side effects for men. Another application is imaging the margins of breast cancer surgical specimens to guide surgical procedures, for which imaging speed is important.
Q: How much additional research must be done to confirm its benefits?
A: For the surgical-guidance application of our technology, our next steps will be to prove that we can look at the surfaces of the excised tissue within approximately 20 minutes, while the patient is still in the operating room, and that we can accurately guide the surgeon on whether they need to continue cutting or if they are done. In particular, for breast cancer lumpectomy procedures, between 20% and 40% of patients require multiple surgeries because it is difficult for surgeons to know if they have successfully removed all of the tumor. Because the current histopathology procedure is so slow, the success of the surgery is not known until several days later. For the other application of our technology, which is to examine biopsies in 3D, extensive studies will eventually need to be done — with large cohorts of patients — to show that our technologies are superior to conventional histology for predicting patient outcomes and for guiding treatment decisions.
Q: What is the potential impact it can have for patients?
A: For patients with prostate cancer, the impact of using our technology for examining biopsy specimens would be to determine which patients require treatments such as surgery, and which patients should just be monitored through active surveillance. All treatments have side effects. For prostate cancer, they are particularly severe, including incontinence and impotence. Prostate cancer tends to be a slow-growing cancer, but it can sometimes be aggressive. Clinicians need to determine which patients require aggressive treatments and which do not. There is an acknowledged problem with overtreatment, not only in prostate cancer but also for breast cancer. Part of this is because pathology is not perfect. The way that we look at biopsy specimens right now is not as accurate as it could be because we are not looking at the entire biopsy and we are not looking at it in 3D. In addition, conventional histology is destructive of tissue, and reduces the quality and quantity of tissue that is available for modern molecular assays such as DNA and RNA sequencing. One of the biggest impacts we can have with our technology is to improve the ability of pathologists to accurately examine biopsy specimens. Also, for the application of surgical guidance, a major impact would be to reduce the rate at which patients need to undergo multiple surgeries to remove their tumors. By having a device with which we can perform rapid 3D pathology of biopsies, and image tumor margins during surgery, there is the potential to reduce health care costs and improve the quality of life and outcomes for patients with cancer.
Q: Is there anything else that you would like to mention?
A: I am fortunate to work in a department and university in which collaborations between engineers, scientists and clinicians is not only encouraged, but is facilitated by the geographic proximity of the engineering and medical schools, and also by the quality and creativeness of the investigators. This project would not have started without pathologists such as Nicholas P. Reder, MD, MPH, and Lawrence D. True, MD, reaching out to me for solutions to their unmet clinical needs. Also, Adam Glaser, a postdoctoral research fellow in my lab, whose talents and creativity were essential for developing an ideal solution to the problems that Nick and Larry face on a daily basis. Having the right team is essential to solve most significant real-world problems, and I feel lucky to have a ‘dream team’ of collaborators to work with on these exciting medical problems. – by Jennifer Southall
Reference:
Glaser AK, et al. Nat Biomed Eng. 2017;doi:10.1038/s41551-017-0084.
For more information:
Jonathan T.C. Liu, PhD, can be reached at University of Washington, Department of Mechanical Engineering, 901 12th Ave., Seattle, WA 98122; email: jonliu@uw.edu.
Disclosures : Liu and colleagues report they are in the process of patenting their technologies.