Identifying changes in the physical properties of cells is essential for the diagnosis and treatment of some diseases such as cancer. However, diagnosis requires the expertise of a pathologist. A team of scientists has developed a quick and easy method to analyze tissue biopsies for cancer that may make pathologists a thing of the past.
If a cancerous tumor is biopsied, a professional pathologist should analyze the sample and assess the health of the tissue. This is a process that needs to be quick and precise as it is usually done while the patient is on the operating table.
Taking a biopsy of a solid tumor is the most common way to assess cancer aggressiveness and guide surgeons in managing patients during and after surgery. However, while it seems like a simple process, it is not. A pathologist should slice, stain, and examine the sample under a microscope before giving an expert opinion. Very labor and resource consuming.
But waiting for a pathologist’s report may soon be a thing of the past. They tested the method using mouse tissue.
The new device uses a “tissue grinder” to reduce samples down to single cells. This process takes less than 5 minutes. Cells are then analyzed using real-time fluorescence and deformability cytometry (RT-FDC). It uses a syringe pump to push a stream of single cells through microscopic contractions where cells are subjected to hydrodynamic shear stress and pressure.
Cell deformability is a useful method for diagnosing disease, especially the invasiveness of cancer cells. The ability of cancer cells to transform means that they become more likely to invade other cells, leading to tumor metastasis.
RT-FDC can analyze up to 1,000 cells per second. This is 36,000 times faster than older conventional methods of analyzing cell deformability. Images are taken of each cell as it passes through, its physical attributes are evaluated, and the image alone identifies tissue cell subtypes.
“Traditional methods of analyzing biopsy samples only allow pathologists to see cells,” said Markéta Kubánková, co-lead author of the study. “We can do a physical examination of individual cells, which allows us to work with a lot more information.”
However, observation of the physical structure of cells alone is not sufficient for diagnosis. Researchers then added AI in the form of a machine learning model that evaluates the data obtained by RT-FDC and rapidly assesses whether a biopsy sample contains tumor cells.
The entire process from sample processing to data analysis took less than 30 minutes. This means it can be performed while the patient is still on the operating table and the results can be provided to the surgeon without referral to a pathologist.
Researchers say this is an advantage. This is because it is not always possible to catch a pathologist while an operation is being performed, which in some cases means that samples cannot be analyzed until after the operation is over.
“Some results often mean that patients have to return to the hospital for another operation in a few days,” the researchers said.
In addition to examining cancerous tumors, this device was also used to detect tissue inflammation in models of inflammatory bowel disease. Researchers hope that new methods of cell analysis will one day be used in clinical settings.
“This was a proof-of-concept study. This method allowed us to determine the presence of tumor tissue in a sample very quickly and accurately,” said Despina Soteriou, co-first author of the study. “The next step is to work very closely with clinicians to determine how this method can be applied clinically.”
The study was published in a journal Nature Biomedical Engineering.
Source: Max Planck Center for Physics and Medicine (PDF)
