Several treatments and drugs are available for breast cancer. However, it is not possible to say whether a treatment will help the individual patient or not. Better methods are needed to predict how patients will respond to treatment. Now, researchers at Karolinska Institutet (KI) in Sweden have developed a cell-based method that should be able to predict whether a patient with breast cancer will benefit from a particular treatment or not.
Their findings, “Breast cancer patient-derived whole-tumor cell culture model for efficient drug profiling and treatment response prediction,” are published in Proceedings of the National Academy of Sciences (PNAS).
“Breast cancer is a complex disease comprising multiple distinct subtypes with specific genomic and pathological characteristics,” wrote the researchers. “Although some 30 anti-neoplastic compounds have been approved for clinical use, patient-to-patient variability in drug response is frequently observed. Several patient-derived tumor models have been proposed to serve as therapeutic prediction tools. However, the lack of tumor microenvironment considerations and time-consuming procedures make their clinical utility limited.”
“Today, there are limited possibilities of determining in advance which breast cancer patients benefit from different treatments. This method can predict how patients will respond to certain treatments, which means that unnecessary side effects can be avoided, and costs can be saved. Larger confirmatory studies are needed, but we see that the concept works,” said Johan Hartman, professor at the department of oncology-pathology, Karolinska Institutet, and the study’s corresponding author.
The method the KI researchers developed is based on isolating and cultivating tumor cells as well as supporting cells from patients with breast cancer.
In the current study, the researchers show that it is possible to establish this type of cell-based tumor model from breast tumors and that the cell models are similar to the patients’ tumors of origin.
The tumor models were created from biopsies from 98 patients who had undergone breast cancer surgery. More than 35 existing breast cancer drugs and breast cancer drugs under development were tested on them.
The researchers then examined how accurately the method can predict treatment responses.
The results showed that the treatment responses predicted by the tumor models were consistent with the treatment responses the patient subsequently exhibited. For example, the model predicted the treatment response from the chemotherapy drug epirubicin with 90% accuracy, while four of four patients treated and tested for anti-HER2 monoclonal antibody drugs showed consistency.
“In most cases, we can perform individualized drug testing and get the result within ten days, indicating that this method can work in daily clinical practice. But it can also be used in research and drug development,” explained the study’s first author Xinsong Chen, research specialist at the department of oncology-pathology, Karolinska Institutet.
Looking toward the future, the researchers want to test the method in a larger group of patients and investigate the possibility of combining it with other molecular methods for an even better prediction of treatment responses as well as study resistance mechanisms.
