Patients with advanced cancer are typically treated using a combination of methods. If possible, surgical removal is attempted first, followed by chemotherapy to eradicate any remaining disease. Each step increases the chances of patients to survive the disease, but also increases their risk of severe side effects that may result in poor quality of life or even death.

To strike an appropriate balance between cure and quality of life, it is important to know in advance which individual patients will respond to chemotherapy and who will not. Physicians therefore sometimes switch the order of treatments, performing chemotherapy first, and surgery second. This is called “neoadjuvant chemotherapy”.  It allows the tumor to be monitored on medical images during the chemotherapy in order to intervene quickly if the medication is ineffective.

Unfortunately, current methods to measure tumor response have limited accuracy. This is caused in part because medical images cannot visualize microscopic disease and because not all parts of the tumor respond equally well to the same medication.

A consortium of hospitals and industry has joined forces in the LIMA project to tackle this problem.  They will develop new methods to increase the chances of cure without compromising quality of life. To allow results to be translated to patients as soon as possible, the study has been designed to follow the current clinical routine of doctors as closely as possible.

In patients with breast cancer and rectal cancer, magnetic resonance imaging (MRI) will be employed to study the biology of the tumor before and during the treatment using state-of-the-art technology.  MRI uses magnetic fields and radio waves to create very detailed pictures of the tumor, its functional behaviour and chemical composition. Secondly, blood samples of the same patients will be analysed before and during treatment to automatically look for microscopic tumor cells and their characteristics. Finally, all available information will be combined with clinical information using machine learning techniques to predict as soon as possible after the start of the treatment whether the treatment will eradicate all cancer.