June 10, 2021

Mark Foundation Center at JHU Uses the Tools of Astronomy to Map Tumor Biomarkers

Mapping a cancer tumor in unprecedented detail

An interdisciplinary effort tries to replicate for oncology the successes – and digital tools – of astronomy

Credit: Seyoun Park, Ph.D.
Shown here is a melanoma tissue section. PD-L1 expression is shown in red and PD-1 in blue. The height of the peaks represent the levels of expression. Credit: Seyoun Park, Ph.D.

In the ongoing pursuit of more effective and individualized treatments for cancer, scientists at the Johns Hopkins University, funded by The Mark Foundation for Cancer Research, have started the challenging task of creating an atlas and are seeing promising early results.

The goal of the project is to be able to study cancerous tumor cells, and the human body’s response during immunotherapy, in order to better understand the process. Cancer researchers have never had access to a large database of images – a giant catalogue of the manifold cancer pathologies – and the hope is that the new atlas will lead to a rapid improvement in our knowledge, which should translate into better treatments, according to Dr. Janis Taube and Dr. Alex Szalay of the Johns Hopkins University, the lead researchers. The research team published its first results today in Science.

“The application of advanced mapping techniques from astronomy has the potential to identify predictive biomarkers that will help physicians design precise immunotherapy treatments for individual cancer patients,” says Michele Cleary, Chief Executive Officer, The Mark Foundation for Cancer Research. “These early results are exciting and validate the approach, and we at The Mark Foundation for Cancer Research are proud to support such pathbreaking science.”

Ever since Abraham Ortelius produced the first modern atlas, the Theatrum Orbis Terrarum, in 1570, atlases have opened new horizons. Ortelius’s work unleashed a giant output in cartography – and an increased understanding of physical geography. When two centuries later Carl Linnaeus catalogued plants in his Species Plantarum, it led to a huge understanding of botany. More recently, the Sloan Digital Sky Survey (SDSS) has made available a huge amount of data about the early universe available to researchers. Data collection started some 20 years ago, and more than a billion astronomical objects have been observed. Last year, SDSS published the largest, most detailed three-dimensional map of the universe so far, an enormous atlas for the world’s astrophysicists. The grand goal of the SDSS has been to reveal the large-scale structure of the universe, and thus understand what happened in the first fractions of a second after the Big Bang. Thousands of papers have already been published as a consequence of the huge output of SDSS, and Dr Szalay, who has been a leading figure in the sky survey, hopes to bring the lessons learned over two decades of harvesting and compiling data digitally to this new atlas of cancer cells, which has been named AstroPath.

AstroPath has ambitious goals. It seeks to improve data collection a thousandfold and will make the atlas available in a format similar to Google Maps. It will use automated workflows to process images of tumorous cells, which are being sourced from various hospitals and research labs. Currently, AstroPath already has upward of 226 million detected cells across multiple tumor types. The researchers are wrangling the difficult problem of standardizing different images taken under varied resolutions and imaging techniques. They are hoping to work with a leading microscope provider to create a template for sourcing images (based on the microscope’s optics). They are also trying to build a data set large enough to be able to train machine learning algorithms.

Dr Taube, who is a pathologist, has a near term goal. In recent years, oncologists have made big strides in treating certain advanced metastatic cancers using immunotherapy, she says. Yet not every cancer patient responds to treatment. She is hoping AstroPath will soon help identify biomarkers that will be able to predict if a cancer patient will respond to drug therapy, and, if so, which drugs will be the most effective.

In the paper published in Science, the scientists used the AstroPath platform to understand PD-1 and PD-L1 expression on cancer cells and immune cells in tumor specimens from patients with advanced melanoma who subsequently received anti-PD-1 immunotherapy.

They also looked at three additional proteins expressed by different types of immune cells as well as a marker for the tumor cells themselves, Sox10/S100.

The team found that a particular pattern and intensity of expression of these markers on specific cells in the tumor could strongly predict which patients would respond and survive after anti-PD-1 therapy.

Dr. Szalay says, “We are using the latest technology of mapping the universe to map the tumor microenvironment in the human body. By using this approach, we hope to gain a much better understanding of the structure and vulnerabilities of tumors and their response to immunotherapy.”

Back to all news