Researchers have created a method to identify circulating tumor cells found in the blood of patients with pancreatic and lung cancer.
Researchers from the University of Michigan have developed a way of detecting circulating tumor cells in the bloodstream of pancreatic cancer and lung cancer patients.
The findings were published in Biosensors and Bioelectronics.
As tumors grow, they release cells into the bloodstream.
Even though these circulating tumor cells are significantly fewer in number compared to millions of other blood cells, early detection can potentially lead to better treatment results.
For example, pancreatic cancer often has a grim outlook because it is usually discovered too late for effective treatment.
Similarly, lung cancer detection rates, particularly after recurrence, are also disappointing.
Currently, methods for detecting circulating tumor cells involve tagging certain proteins present on tumor cells with fluorescent dyes.
These dyed cells can then be easily identified in blood samples.
However, there are drawbacks.
Some cells might not express these proteins on their surface, leading to missed detections.
The existing methods also overlook crucial information concerning the internal state of the cancer cells.
“The current techniques typically employ methods that end up destroying the cancer cells, which prevents us from studying them further,” explained Sunitha Nagrath, a professor of chemical engineering.
“We recognized the need for an alternative approach to identify circulating tumor cells while keeping them alive.”
To achieve this, the researchers utilized biolasers.
While this method still includes staining cancer cells with dyes, it does not harm them. Instead of relying on surface proteins, the researchers were able to stain a component vital to all cells — their nucleus.
Using blood samples from pancreatic cancer patients, the researchers first directed all cells through a circular structure called Labyrinth, which pre-sorted the circulating tumor cells since they are slightly larger than other white blood cells.
“Think of it this way: riding a bicycle around a curve is different than driving a truck — the forces you experience vary significantly. This difference causes the larger tumor cells to gather in different areas compared to smaller white blood cells,” Nagrath said.
Next, they placed the tumor cells between two mirrors and directed a laser at them one by one.
When the laser’s power was sufficient, the cells emitted light and could be referred to as cell lasers.
“The light emitted from a cell laser is considerably stronger than what we obtain from standard fluorescent methods,” stated Xudong (Sherman) Fan, a professor of biomedical engineering.
“The images from laser emissions also differ; in fluorescence, the cells appear as glowing spheres. However, with a laser, different shapes can be observed, offering insights into the organization of DNA within cancer cells.”
Nonetheless, these distinctions are subtle.
The researchers thus sought assistance from machine learning.
By utilizing the Deep Cell-Laser Classifier model, they accurately identified pancreatic cancer cells 99% of the time.
Remarkably, the model was so proficient that it could also recognize lung cancer cells without needing any extra training, even though it was initially trained on pancreatic cancer cells.
“While there are a few research teams exploring biolasers, we are the first to apply it in clinical studies related to cancers and circulating tumor cells,” Fan noted.
Looking ahead, the team aims to create a device capable of isolating cancer cells after they’ve been identified.
“With our current setup, if you want to collect circulating tumor cells, you must remove the top mirror, which risks causing the cell to move, and then you lose track of it,” Fan explained.
“We aspire to design a system where cells pass through the laser excitation spot individually and subsequently enter a sorting apparatus to help separate and gather cells for further analysis.”
The researchers also plan to analyze the light patterns produced by the cells to gain insights into which tumors may be more aggressive or resistant to treatment.
“Each of these circulating cells can vary widely,” Nagrath mentioned. “Understanding how aggressive cells evolve during treatment cycles would be invaluable.”
“This project wouldn’t have been possible without such a collaborative team effort,” Nagrath concluded.
