New biomarker predicts success of immunotherapy in kidney cancer

Immunotherapy increases survival rates in kidney cancer, but does not work for everyone. A Leuven research team has developed a new method to predict which patients will benefit from it. The team of Francesca Finotello (Computational Biomedicine Group) from the University of Innsbruck also contributed.

Their study, published in the journal Nature Medicine, also opens new avenues to even more effective treatments.

Every year, roughly 1,300 people in Austria are diagnosed with kidney cancer. Thanks to immunotherapy, survival rates for metastatic kidney cancer have increased dramatically in recent years: more than half are still alive five years after diagnosis, compared to only 10% in the past. Unfortunately, the innovative treatment does not work for all patients.

To investigate why there is such variation in the efficacy of immunotherapy, and in the hope of better predicting in which patients the treatment will succeed, a Leuven research team set up a large retrospective study. They analyzed many samples from kidney cancer patients treated with immunotherapy at UZ Leuven over the past decade.

Molecular signature

Doctoral researcher and oncologist in training Dr. Lisa Kinget and postdoc Stefan Naulaerts explain, “We examined both tumor biopsies and blood samples with advanced laboratory techniques. Via machine learning, we combined gene expression in the tumor with hereditary characteristics of the patients’ immune system, more specifically the HLA genes, which come in hundreds of variations depending on the individual.

“This approach allowed us to find a ‘molecular signature’ that showed a clear association with clinical response and survival rates. We further confirmed this association on independent samples from more than 1,000 kidney cancer patients from other international studies.”

The lab analyses further showed that a successful response to immunotherapy was tied to a good interaction between two types of immune cells, namely CD8⁺ T cells and macrophages.

Dr. Francesca Finotello from the University of Innsbruck’s Department of Molecular Biology and the Digital Science Center (DiSC) adds, “We integrated and analyzed large-scale multi-omics data from The Cancer Genome Atlas (TCGA) to associate this novel, molecular signature with the mutational landscape of the tumors, demonstrating that it brings orthogonal information with respect to the sole genetic background of cancer cells, capturing efficiently their interaction with the immune system.”

Prof Abhishek D. Garg, KU Leuven says, “Previously, researchers mainly looked at immune cells at the level of individual cell types, which led to oversimplified biomarkers. As a result, we thought macrophages were ‘bad’ for immunotherapy. With this study, we show that the interaction between different types of immune cells in a specific spatial context is more important in fighting kidney cancer.”

Prof Dr. Benoit Beuselinck, medical oncologist at UZ Leuven, says, “In the future, we hope to be able to use our method as a biomarker to predict in which patients immunotherapy will be effective. The new insight that the interaction between certain T cells and macrophages is important for the success of immunotherapy opens up interesting avenues for future treatments.

“We are currently busy setting up new clinical trials of combination therapies to stimulate both cell types and make them work better together, which may be more effective than current treatments.”