Re: Netherlands Cancer Institute : ‘Tumor avatars’ predict patients’ response to immunotherapy

lyftv-sciences Date d'inscription : 12/03/2019

AI can help improve precision radiotherapy

The Netherlands Cancer Institute, University of Amsterdam (UvA), and Elekta will collaborate on the development of new AI strategies for the further improvement of precision radiotherapy. This concerns the personalization of treatment by improving the quality of imaging used during treatment, predicting and accounting for changes in the patient’s anatomy over time, and automatically adapting radiation delivery each time a patient is treated.

The collaborating parties have recently been awarded a grant of the Rijksdienst voor Ondernemend Nederland RVO, making the Partnership for Online Personalized AI-driven Adaptive RT (POP-AART) lab a reality. The lab, part of the Innovation Center for Artificial Intelligence, aims to use state-of-the-art AI methods to optimize online adaptive radiotherapy. The goal of adaptive radiotherapy is to account for changes that can occur in patients over the course of treatment and to ‘adapt’ the original prescription for every day of treatment to ensure the right dose is delivered at the right time to the right location, with minimal damage to healthy tissue. This a great challenge that should be a part of every treatment session, because the patient and the tumor are continuously in motion and change shape over time. In the past this has been impossible to achieve due to computational and workflow complexities, but AI has the potential to make that achievable in the near future.

Jan-Jakob Sonke, group leader adaptive radiotherapy, professor by special appointment at the UvA, and one of the two lab directors, greatly looks forward to the start of the lab. “Not only is it our ambition to create incredibly innovative AI algorithms – we also want to make sure that they can be applied to clinical practice so patients can receive even better treatment.”

Efstratios Gavves, associate professor on computer vision and deep learning at the UvA, and director of the lab: “The time is ripe to combine the power of Artificial Intelligence and big data with the cutting-edge expertise of the University of Amsterdam and the Netherlands Cancer institute, for an application with tremendous impact: battling cancer.”

The lab will pioneer the uses of novel deep learning approaches to address challenges in modern radiation oncology including personalization, speed and precision. Multiple work streams will run in parallel over the course of the partnership, and each will address various key elements in the radiotherapy process including the quality of daily treatment images, and strategies to detect and adapt for changes a patient’s anatomy and methods to continually personalize the treatment plan for patients.

Maurits Walleswinkel, President Linac Solutions & Chief Product Officer at Elekta, says “We believe AI has an critical role to play in assisting clinicians to provide increasingly personalized and complex cancer treatment to patients. Elekta’s participation in POP-AART underlines our commitment to precision radiation medicine.”

https://icai.ai/pop-aart-lab/

lyftv-sciences Date d'inscription : 12/03/2019

Tumor fragments in the lab are able to predict whether the corresponding real-life patients will benefit from immunotherapy. “We’ve solved a major problem many scientists had been facing: preserving a tumors original composition and structure outside of the patient in the lab”, says cancer researcher Daniela Thommen from the Netherlands Cancer Institute. On 8 July, the results of her study are published in Nature Medicine.

While some cancer patients experience incredible results from immunotherapy, many others do not benefit from this treatment, which puts patients’ own immune systems to work. By treating tiny fragments of tumor tissue from real patients in the lab, Daniela Thommen aims to improve this situation. With this new platform, she tries to match the right treatment with the right patient. "We first cut patient tumor samples into small pieces and then treat these 'tumor avatars' outside the patient’s body with different therapies, to see which one works".

The big question is: does such a ‘tumor avatar’ in the lab really reflect how a patient responds to a treatment? The latest research from the Thommen and Schumacher research groups together with many NKI-clinicians, confirms that indeed these tumor avatars’ response to treatment in the lab predicts whether the former owner of this tissue (the patient) will respond to the treatment in real life. The researchers analyzed the reaction of the tumor avatars in the lab to the type of immunotherapy called PD-1 blockade, and linked this information to treatment responses from 38 patients with various cancer types.

“These results confirm that we have now a very powerful model system in place which we can use to develop new diagnostics, and in this way personalize immunotherapy”, says Thommen. “We also found some unknown predictors for response or resistance to immunotherapy. We identified three different subgroups of tumors that do not respond, for example. And we saw that the tumors that did respond had been infiltrated by a specific type of immune cells and formed more immune cell niches in their tumor, the so-called tertiary lymphoid structures. These different markers can now be further tested as predictive markers for treatment response, separately or in combination.”

Learn more about:

Daniela Thommen and her tumor-avatars
The scientific paper in Nature Medicine
The story behind the paper on Nature’s website

Note to the editor

For more information please contact Hilje Papma, Science Communication Advisor at the Netherlands Cancer Institute (h.papma@nki.nl, +31643468357/+31 20 512 28 50).

About the Netherlands Cancer Institute

The Netherlands Cancer Institute is among the world’s best comprehensive cancer centers, combining innovative fundamental, translational, and clinical research with dedicated patient care. Our research institute gratefully acknowledges funding from the Dutch Ministry of Health, Welfare and Sport, the Dutch Cancer Society, and individual donors. For more information please visit our websites www.nki.nl and www.avl.nl.

<== Retour à LYFtv.com

Nouveautés : les derniers articles ==>