As of: February 20, 2024, 5:23 a.m
Comments
Press
Split
Using artificial intelligence, Swiss researchers have succeeded in detecting tumors based on altered blood cells.
The approach could simplify cancer diagnosis in the future.
An early diagnosis can significantly increase the chances of a cure for cancer.
For this reason, many health insurance companies cover various cancer screening examinations from a certain age.
However, many of them are complex and cancers are often only discovered when they already cause symptoms.
Researchers are therefore looking for processes that can be implemented quickly and deliver reliable results.
Researchers from the Swiss
Paul Scherrer Institute (PSI)
have now achieved an important breakthrough.
Cancer diagnosis: Blood cells indicate a tumor
The Paul Scherrer Institute: Important medical progress was achieved here.
© Andreas Haas/IMAGO
The research group led by GV Shivashankar, head of mechano-genomics at PSI and professor at ETH Zurich, succeeded in recognizing tumors based on changes in the organization of the nucleus of certain blood cells.
They received support from artificial intelligence (AI).
With their method, the experts were able to distinguish between healthy patients and those suffering from cancer with an accuracy of around 85 percent.
The test also provided information about what type of tumor it is.
Including:
Melanoma (skin cancer)
Gliomas (tumors of the central nervous system)
Head and neck tumors
“This is the first time in the world that someone has achieved this,” explains Shivashankar.
The researchers published their findings in the journal
“npj Precision Oncology”
.
Early cancer detection: This is how the new blood test works
The basis of the new procedure is a simple blood sample.
The researchers concentrated on lymphocytes and monocytes.
These white blood cells can recognize and eliminate altered cells in the body's own cells (tumor cells).
They have a round nucleus that is easily visible under the microscope.
And this is exactly what the researchers set their sights on: because, according to the experts' assumption, the genetic material of the white blood cells in the cell nucleus reacts to substances in the blood that tumors secrete (so-called “secretome”).
These substances activate the chromatin in the cell nucleus, the material from which chromoses are made, and change the organization of the genetic material.
The researchers use this change as a biomarker for cancer.
The blood cells were a kind of detector.
My news
Do you always wake up at the same time?
The Chinese organ clock could provide the explanation
Reduce belly fat: If you know one rule, you are guaranteed to read your stomach
Possible sign of a heart attack?
What you need to know if your heart is racing at night
Awake at the “wolf hour” between two and three o’clock: How to sleep read on
Clear your nose in just one minute - without nasal spray or other medication
No appointment in the practice necessary: Read which specialists you can see directly without a referral
AI-supported work of researchers
The researchers received help from artificial intelligence (AI): the chromatin of the blood cells of healthy and sick patients was first examined for around 200 features under the microscope and the images were fed into an AI.
The software learned to recognize the differences using “supervised learning” and was then able to distinguish between diseased and healthy cells using an algorithm.
These differences are not visible to the human eye.
The accuracy of the AI was 85 percent.
In order to distinguish between different tumors, the researchers then fed in chromatin data from blood cells from sick patients: ten with a glioma tumor, ten with a meningeal tumor and ten from patients with an ear and nose tumor.
Here too, the AI was able to distinguish between tumor types with an accuracy of 85 percent.
Software also detects the effectiveness of radiation therapy
The researchers were also interested in whether the procedure could be used to monitor the success of cancer therapy.
To do this, they asked cancer patients who were receiving radiation therapy at the ZPT Center for Proton Therapy to provide blood samples.
Before, during and after radiation therapy.
As expected, there was a reduced amount of tumor signals in the blood after treatment.
The DNA of the blood cells returned to its normal form.
“It was astonishing to observe how the structure of the chromatin moved closer to the healthy pattern over the course of the treatment,” summarizes Shivashankar.
Procedure could also be used to diagnose other types of cancer
In the future, the method based on blood cell chromatin could also help detect other types of cancer and track the progress of radiation therapy, surgery and chemotherapy.
However, further research with a larger number of participants and under clinical conditions is necessary, say the PSI scientists.
But: “The method is there,” Shivashankar is certain.
According to
the German Cancer Society
, various imaging procedures such as MRI, CT, X-rays, but also endoscopies (e.g. colonoscopy) and biopsies (removal of cell tissue and subsequent laboratory examination) have been used to diagnose cancer (depending on the type of cancer).
Blood and urine samples can also provide clues.