Electronic ‘nose’ can predict response to lung cancer immunotherapy

GlobalData Healthcare 1 October 2019 (Last Updated October 1st, 2019 15:54)

Researchers at the Amsterdam University Medical Centers have developed an electronic nose (eNose) that detects chemicals in the breath of lung cancer patients, distinguishing those who will and will not respond to therapy.

Electronic ‘nose’ can predict response to lung cancer immunotherapy

Patients diagnosed with lung cancer typically benefit from immunotherapy, a form of treatment that uses the body’s immune system to induce an anti-tumour response, thus defeating cancer.

However, the response rate to immunotherapy varies and secondary resistance with disease progression is very common. To address this, researchers at the Amsterdam University Medical Centers developed an electronic nose (eNose) that detects chemicals in the breath of lung cancer patients, distinguishing those who will and will not respond to therapy.

The eNose contains sensors that detect volatile organic compounds (VOCs); chemicals present in about one per cent of exhaled breath. These chemicals have been proven to be a useful source of information for early lung cancer diagnosis. When using the eNose, patients take a deep breath, hold it for five seconds, and slowly exhale into the device. The eNose sensors respond to the VOCs in the exhaled breath. The sensor readings are then sent to an online server and compared to an online database where machine learning algorithms distinguish whether or not the patient will respond to immunotherapy.

The study consisted of 143 patients with advanced non-small cell lung carcinoma (NSCLC), a type of cancer that constitutes approximately 85% of lung cancer patients. The patients used eNose two weeks before they began the immunotherapy treatment and they were once again examined three months later to assess the response to treatment. The study showed that exhaled breath analysis by eNose allows discrimination between responders and non-responders to immunotherapy with an accuracy of 85%.

Patients are currently tested by immunohistochemistry (IHC), which is considered the gold standard for selecting patients who will respond to immunotherapy. The technique, which involves testing tissue samples for the programmed death-ligand 1 (PD-L1), a target for immunotherapy, is invasive and time-consuming. The eNose technology, on the other hand, is non-invasive and provides feedback within seconds.

The researchers indicate that the use of eNose can potentially avoid application of ineffective treatment and unnecessary delays in patients who do not respond to immunotherapy. They plan to conduct a large prospective multi-centre study on NSCLC outcomes as the next step.