Researchers can predict effect of anti–cancer drug on cancer cells in the lab.
As part of a preclinical study, members of the OncoTrack consortium (which includes researchers from Charité – Universitätsmedizin Berlin and Max Planck Institute for Molecular Genetics) analyzed tumor samples from patients with colorectal cancer. The researchers were looking for biomarkers – molecules that are typically associated with different tumor subtypes, and which provide valuable information for diagnosis and treatment. They identified molecules capable of predicting the efficacy of two commonly used drugs: cetuximab, an inhibitor of the epidermal growth–factor receptor (EGFR), and the chemotherapy drug 5FU.

Results from this study were published in the current issue of the journal Nature Communications.

“Colorectal cancer represents a heterogeneous group of conditions, which is why currently available drugs only work with variable success,” explains Prof. Marie–Laure Yaspo, the study's principal investigator and a researcher at the Max Planck Institute for Molecular Genetics in Berlin. “We are of course aware of the existence of molecular subtypes. What remains mostly unknown, however, is how these subtypes impact treatment.” In order to improve our ability to predict how a patient will respond to a specific drug, scientists must obtain detailed molecular profiles of both the cancer and the patient.

Charité–based researchers collaborated with scientists from University Hospital Graz to collect tumor samples from more than 100 bowel–cancer patients at different stages of disease. Samples were then grown either in a petri dish or in specially–reared mice before being exposed to the drugs. The researchers were able to gain a better understanding of how molecular changes affect the way in which tumors respond to drug treatment. “As a next step, we will need to determine which of the newly–identified molecular predictors have clinical relevance when treating patients with bowel cancer,” says Prof. Ulrich Keilholz, Acting Director of the Charité Comprehensive Cancer Center.

The researchers started by determining the genetic composition of the tumors and the 'transcriptome' profile, i. e. the sum total of all RNA molecules produced inside a cell. Doing so allowed the researchers to produce 'molecular fingerprints' for all tumors sampled. The different tumors were then tested in terms of their response to different drug treatments, which allowed the researchers to link a particular tumor's molecular fingerprint with its response to various drugs.

If a group of tumors was shown to respond to a particular drug, the researchers tried to identify which typical biomarkers were associated with this type of cancer. They were able to identify two biomarker signatures capable of predicting whether a bowel cancer can be successfully treated using the EGFR inhibitor cetuxmiab and the chemotherapy agent 5FU. “These results will allow us to develop diagnostic devices that will improve the prediction of drug efficacy. In future, it might become possible to devise individualized treatments based on the type of cancer involved,” explains Prof. Yaspo.