NEW YORK – Hopp Children's Cancer Center Heidelberg in Germany has launched a consortium to expand access to a methylation-based tumor classifier that could help children in low-income countries receive better diagnoses.
Led by Stefan Pfister, a pediatric oncologist who led the development of a central nervous system tumor classifier using data from methylation arrays, Hopp has joined with the German Cancer Research Center (DKFZ), Heidelberg University Hospital (UKHD), and Illumina to provide training for labs around the world and funding to carry out this testing in their home countries. German nonprofit Ein Herz für Kinder (A Heart for Children) has provided financial support to help train healthcare providers and lab staff.
So far, the program has trained 37 people from labs in Argentina, Brazil, Chile, Egypt, India, Indonesia, Jordan, Pakistan, Qatar, and South Africa.
As part of the program, Illumina is donating reagents to run the tests back at participants' home institutions and providing expertise and technical support to run them. How long the initiative will run and how much the reagents are worth is unclear.
"Cancers in children are very diverse and therefore respond quite differently to radiation and chemotherapy," Pfister said. "Classifying the tumors as precisely as possible is therefore quite crucial for effective treatment." The classifier works with sarcomas as well as CNS tumors, he noted.
The program may also provide new insights on methylation patterns in populations that have been underrepresented in research.
Pfister's team first published on the assay in Nature in 2018. It uses Illumina methylation arrays to profile samples and a machine-learning approach to create classifications. In a prospective study, the researchers applied it to 1,155 CNS tumors, obtaining profiles for all but 51 cases and methylation-based classifications for 977 of them.
Methylation patterns have now been adopted by the World Health Organization for classification of pediatric tumors. To date, the Heidelberg method has analyzed more than 120,000 tumors from over 40 countries for research purposes.
"It has kind of become essential if you want to make a specific diagnosis," said Ahmed Gilani, a neuropathologist who has gone through the training program and is bringing the assay and array-based methylation testing to Pakistan. Gilani splits his time between Children's Hospital Colorado and Aga Khan University Hospital in Karachi. "If you don't have access to methylation data, then you end up lumping tumors into an umbrella diagnosis," which is associated with worse prognosis and fewer appropriate treatments, he said.
Gilani attended a weeklong training in Germany, along with a colleague who runs the research lab in Pakistan and a bioinformatician. The course included lectures as well as hands-on wet and dry lab training.
"One of the strengths of this initiative is that it's funding our university for reagents and not the instrument," he said. "A lot of other projects funded infrastructure, but when it comes to [operating] costs, we're not able to generate that. That has led to many instruments not being used. In this project, we'll get all the reagents needed for individual patients for the life of the project."
Gilani estimated that he'll be able to do approximately 500 cases per year. "We've talked to other large institutions in Pakistan," he said. "They'll be sending us their complicated cases, as well."
Access to this type of testing is especially important in Pakistan, which has a population that skews younger, he said. "We have about 3,000 cases of brain tumors in a year, and essentially half come to the university in Karachi where I practice," he said. Up to two-thirds can be diagnosed with microscopy-based methods already available, "but about one-third to one-half do not get adequate diagnoses," Gilani added.
He's interested in addressing the assay's utility in diagnosing pediatric sarcomas, another disease area that requires advanced molecular diagnostics currently not available in Pakistan.
In addition to providing critical healthcare to children around the world, Pfister said there are "scientific reasons to be excited about this."
"If you look at all the genome data published, it's basically biased towards Europeans," he said. "There are a lot of discoveries to be made in populations that are massively underrepresented."
His lab has already seen differences in tumor profiles from populations in the Middle East. "You have a completely different spectrum of tumors," he said.
The consortium is also helping to interpret these new profiles by offering virtual molecular tumor boards led by experts from the Heidelberg-based institutions and Washington University in St. Louis.
"From our experience offering this to the world, we get a lot of these requests every week," Pfister said.