By combining microscopy with algorithms that can differentiate between healthy and malignant tissues from images, researchers have recently created a novel tool for cancer identification. This technique is different from conventional diagnostic tools, being that it does not rely on the use of contrast dyes or invasive biopsies. This approach has the potential to detect cancer metastases that are currently difficult to identify through standard imaging solutions during operations. These findings were reported in The Optical Society (OSA) journal Biomedical Optics Express.
In evaluating and treating metastatic cancers, it is essential to determine where the cancer has spread and how severely it has done so. This is typically done using cross-sectional radiographic images and white light laparoscopy; however, these techniques can have difficulty finding small cancers deep within healthy tissues.
“Existing techniques are invaluable but suffer from low spatial resolution and often require the use of exogenous contrast agents,” explained research team co-leader Thomas Schnelldorfer, from the Lahey Hospital in Burlington, Massachusetts.
“The method utilized in this work identifies in a completely label-free manner cellular and tissue features at the microscopic level, essentially acting like a biopsy without a knife,” added lead author Dimitra Pouli from Tufts University in Medford, Massachusetts.
How the Cancer Diagnostic Tool Works
Schnelldorfer, Pouli, and colleagues leveraged multiphoton microscopy in this work along with automated image and statistical analysis algorithms to examine biopsies. Multiphoton microscopy involves the delivery of laser light directly to tissue. The laser has the power to potentially cause damage, but it is delivered in a pulsating manner to lower the average intensity and minimize tissue damage.
Signals are released from different components of the tissue as it interacts with this laser. These signals are received by the microscope to form an image. Automatic processing algorithms then analyze this image to identify specific features. These minor details detected by the algorithms are not visible in images acquired by traditional means and are used to identify the tissue as healthy or malignant.i
One of the advantages of this approach is that image acquisition is based on these specific tissue components rather than the contrast from a dye it has been stained with. This noninvasive approach allows for the analysis of component-specific features that relate to form, function, and malignancy.
In their work, the research team applied their unique combination technique to biopsies from the parietal peritoneum of both healthy patients and those with ovarian peritoneal metastases. Lining the walls of the abdominal cavity, this tissue is rich in the structural protein collagen. The researchers’ technique aimed to evaluate the microstructural patterns in the collagen fibers as well as their intermolecular signaling.
It was found that the cancerous tissue displayed patterns that were distinguishable from those of the healthy tissue. The healthy tissue showed more variation in these collagen features, but the metastatic tissues portrayed uniform intensity and smaller collagen fibers. This is to be expected, being that malignancy causes the degradation of surrounding tissues.
Successful Outcomes in Early Research
To evaluate their technique, these researchers used it to analyze biopsies taken from 8 patients with either suspected or confirmed ovarian malignancies. 41 images were taken from the biopsies, from which the algorithm-driven tool was able to correctly identify 40 (97.5% accuracy). In addition, 11 of the samples were properly identified as metastatic and 29 of 30 were correctly classified as healthy, demonstrating 100% sensitivity and 96.6% specificity.
Going forward, the team plans to continue the testing of their method in larger samples of images taken from a wide array of patients. Although this study only aimed to find ovarian cancer that had metastasized in the peritoneum, the researchers believe that this technique could also be used to analyze other tissue and cancer types.
Even though biopsies were analyzed in this initial testing of the technique, the team claims that this method will eventually be used directly on the patient to analyze potentially cancerous tissue. This will hopefully achieve their goal of circumventing dyes and biopsies in analyzing tissues but will require further research before being used during surgery. The scientists must reduce the size of their microscope components and integrate them into surgical instruments to allow for this real-time tissue analysis of images in the operating room.
“This could ultimately help surgeons identify suspicious or diseased areas directly in the operating room in real-time, which in turn would directly affect patient management” #Ovarcome https://t.co/QLY48zDYxP
— Ovarcome (@ovarcome) August 6, 2019
