AJPCP: Developed a new tool to predict the recurrence of breast cancer

8, 2020 // -- In a recent study published in the international journal American Journal of Physiology-Cell Physiology, scientists from the University of Michigan and others developed a new tool through research. It combines traditional pathology with machine learning to help predict which breast cancer patients really need surgery, and research may hopefully help prevent patients from receiving unnecessary treatment and reduce the cost of treatment, as well as help develop a new generation of drugs to stop breast cancer recurrence.
Breast catheterization (DCIS, Ductal carcinoma in situ) is an early form of disease called stage 0 breast cancer, which is sometimes only a diagnosis of invasive breast cancer, but some patients need surgery, chemotherapy or radiotherapy, others go home to continue to observe the condition, predicting the prognosis of early forms of breast cancer is a scientific challenge scientists have been facing for decades.
picture Source: Dr. In howard R. Petty's study, researchers reported on a way to address this diagnostic dilemma, a technique that tested samples of DCIS patients that were donated to the research structure more than a decade ago and supplemented by the patient's current clinical history. Typically, treatment is more aggressive for pre-immersive breast cancer patients like DCIS, which in the case of DCIS means partial or total mammary tissue removal, but researchers have learned from other studies that more than half of patients do not experience the progression of invasive diseases,
researcher Petty said.
This new approach relies on new findings from researchers that in potentially relapsed DCIS cases and metastasis breast cancer, cells recombine specific enzymes into metabolic platforms under the outer membranes of dangerous tumor cells, which may allow enzymes to function as efficiently as factory assembly lines, making cancer more dangerous.
researchers believe that the enzymes produced by these cell factories promote the invasiveness of tumor cells, while also transferring multiple forms of chemotherapy and radiotherapy.
to predict which DCIS cases could lead to such assembly lines, the researchers tagged biomarkers from patient samples, then photographed them and uploaded them to a cloud computing platform for analysis.
using this method, researchers were able to predict cancer recurrence and non-recurrence with 91% accuracy, with only 4% false negative results, and researchers are constantly improving and optimizing this method.
researcher Petty says the new tool reduces overdiagnosis of life-threatening DCIS, while also allowing scientists to disrupt metabolic platforms in pharmacology, blocking tumor invasion and enhancing chemotherapy and radiotherapy to block tumor recurrence.
tool can also be used to help predict the outcome of other pre-invasion lesions and which patients will respond to specific treatment interventions.
are currently conducting additional retrospective experiments to obtain FDA approval for the new diagnostic testing technology.
original source: Alexandra M. Kraft, Howard R. Petty. Spatial locations of certain enzymes and transporters within preinvasive ductal epithelial cells predict human breast cancer recurrences, American Journal of Physiology-Cell Physiology (2020). DOI: 10.1152/ajpcell.00280.2020。