The most common cause of disease-related death in children is cancer. However, diagnosis and treatment options for pediatric brain tumors have remained unchanged for decades, signaling an urgent need for new therapies. Dr. Waszak explains that progress is imminent: “within the last 10 years, we’ve advanced our understanding of pediatric brain tumors, and how both somatic and germline mutations contribute to tumor development, including what the key genetic drivers and molecular mechanisms are.”

On the heels of these scientific developments, Dr. Waszak and his Computational Oncology research group at the Centre for Molecular Medicine Norway (NCMM) study highly aggressive brain tumors, such as medulloblastoma, diffuse intrinsic pontine glioma (DIPG), and diffuse midline glioma (DMG), that primarily occur in children and young adults. 

Specifically, Dr. Waszak’s research program, which got underway at NCMM in March 2020, has ambitious goals to understand the cellular origin and genetic evolution of childhood brain tumors, to develop rapid, cost-efficient, and routinely available methods for brain tumor diagnostics, and to study cancer genomes in the context of clinical studies.

A novel classification system: from cell type of origin to genetic drivers and tumor evolution

Genomic instability in aggressive brain tumors is often very high. Every tumor has a different set of alterations, including gain and loss of chromosomes, genetic rearrangements, and somatic mutations. With the high amount of heterogeneity in tumors, every patient is unique. However, the critical changes that drive tumor formation are very often the same. Dr. Waszak sees the opportunity here: 

By focusing on brain tumor and patient genomes, we can map the most recurrent drivers and derive the chain of somatic mutations during tumor evolution. We observed that many pediatric tumors must have followed very precise routes of evolution such that only certain cell types are capable of transforming into a malignant tumor.

Take childhood medulloblastoma, for example. Children with a genetic syndrome very often develop the exact same type, even subtype, of brain tumor, typically along the same path of genome evolution, within the same time period. Dr. Waszak explains, “there’s a lot of heterogeneity, but once you identify a particular tumor subtype, you can actually predict the evolution based on the combination of somatic mutations.” 

Dr. Waszak and his group use these key genetic drivers to figure out the original cell state: 

We are very interested in understanding the actual developmental cell type from which a combination of somatic mutations was capable of triggering tumor development. We develop new computational methods that try to infer the lineage and cell of origin, and we are building a brain tumor classification system that places the physiological cell type, not the mutation, first. 

Computational efforts toward improved diagnosis and disease monitoring

Molecular diagnostics is rapidly developing, but many hospitals lack the resources for its application, e.g., clinical genome sequencing in pediatric neuro-oncology. Thanks to support from an Ian’s Friends Foundation grant, Dr. Waszak’s group is developing new computational methods that read out the molecular and genetic subtypes of a disease simply from routinely acquired  histopathological images. 

This information could enable more patients to be eligible for clinical trials. Dr. Waszak explains:

Many clinical trials, especially for DMG, require not only a radiographic but also a genetic diagnosis. For example, an ongoing immunotherapy trial is testing a peptide-based vaccination strategy that is tailored for patients with Histone 3 gene (H3.3K27M)-mutant DMGs. 

And, if a clinician lacks evidence that a patient's tumor is driven by such a mutation, then they are not able to enroll them into such a study.

With a computational platform the group is on its way to developing new, cost-effective, and tissue-sparing genetic diagnostics, just from standard clinical imaging data. Clinicians could then use this information for ordering follow-up tests and determining eligibility for clinical trial enrollment. 

But, what about the situation where surgery or biopsy is too great of a risk? 

In the majority of cases, a spinal tap is almost always possible. There is good reason to believe that cell-free material in spinal fluid could be used to provide valuable information for brain tumor diagnosis and treatment monitoring. Dr. Waszak explains: 

Clinicians rely on magnetic resonance imaging to assess treatment responses of brain tumors, but assessing a radiological response is challenging, especially for infiltrating tumors such as DIPG and DMG. So we want to study if cell-free material in liquid biopsies might be used as a source to monitor the behavior of a brain tumor.

For example:

The majority of DMGs have a hotspot driver mutation in a Histone 3 gene. Researchers have designed a ddPCR (Digital Droplet PCR) assay for this particular mutation. Just based on this mutation, it's possible to quantify tumor presence in spinal fluid. 

Building on this, he continues:

We are developing new assays and computational methods to help determine if cell-free material might allow genome-wide prediction of tumor responses to a drug. Our technical development is based on patient-derived brain tumor cell lines, and we use cell culture media as a surrogate for spinal fluid. For the next stage, we have a collaboration with Dr. Carl Koschmann at the University of Michigan to potentially test clinical samples.

If successful, this approach could be especially beneficial for monitoring tumor behavior before and during treatment. In addition, material in the spinal fluid might provide an untapped opportunity. While a small needle biopsy may only represent one area of a tumor, information from spinal fluid could widen the view. 

Robust science from interdisciplinary, international collaboration

Each childhood brain tumor subtype can be considered a rare disease. Thus, researchers around the world must work together. The global research community in the Pacific Pediatric Neuro-Oncology Consortium (PNOC) provides a network and platform aimed to advance new therapeutic strategies for childhood brain cancer. Supported by progressive and dynamic family-run foundations, this international community works together to better understand the biology and develop new clinical trials. 

In October 2021, Dr. Waszak joined the Department of Neurology at the University of California, San Francisco (UCSF) as an Associate Adjunct Professor. This affiliation allows his group to become more involved in preclinical PNOC working groups and trials and in multidisciplinary neuro-onology tumor boards at UCSF.

Dr. Waszak provides an example of how the consortium enriches his and others’ work: 

When I started my group leader position I joined DMG-ACT, the PNOC working group on diffuse midline glioma. There are 30 members from Australia, the United States, and Europe with different skill sets and resources, from CRISPR screening to proteomics, and cell line models to computational biology. It is a vibrant community with regular results-oriented meetings connecting the member labs. 

This platform for interdisciplinary research collaboration has diversified research teams to include clinicians, molecular biologists, computer scientists, pathologists, and biostatisticians. To Dr. Waszak, “these projects are robust because different labs join forces for one particular problem.” For example, “diffuse intrinsic pontine glioma (DIPG) is among the deadliest of pediatric brain tumors. Median survival ranges between 8-11 months and 90% of diagnosed children die from this disease within two years.” 

He continues:

We recently finished a manuscript reporting the results of a precision medicine trial for children with newly diagnosed DIPGs. Comprehensive molecular profiling was performed, and a multidisciplinary molecular tumor board assigned a therapy plan with up to four FDA-approved drugs based on the findings. Such an extensive interventional clinical trial for children with DIPG has not been done previously.

While PNOC collaborations are integral to Dr. Waszak’s research program, local and Nordic connections provide additional, essential expertise. The team collaborates with the Chemical Neuroscience Group at NCMM, led by Dr. Camila Esguerra, to develop novel pediatric brain tumor models, and with the Machine Learning in Biomedicine Group at FIMM, led by Dr. Esa Pitkänen, to study cancer genomes. FIMM, the Institute for Molecular Medicine Finland, is the Finnish node of the Nordic EMBL Partnership for Molecular Medicine.

Contribution to WHO international standards

As a result of the past five years , Dr. Waszak was invited to contribute, along with 200 other authors and editors, to the 5^th edition of the World Health Organisation (WHO) Classification of Tumors of the Central Nervous System. This WHO book is the international standard for oncologists and pathologists and serves as a guide for use in the design of studies monitoring response to therapy and clinical outcome. Dr. Waszak reflects on the significant publication:

The new 2021 edition presents a fundamental change in the way pediatric central nervous system tumors are classified. For example, pediatric tumors are now described separately from adult tumors; integrated diagnosis is based on histology with molecular diagnostics; and many new pediatric brain tumor entities are included. 

He continues:

I am really excited that our research [1,2,3] has contributed to updated sections, diagnoses, and pathogenesis mechanisms on embryonal tumors, and that, together with Prof. Stefan Pfister at DKFZ, I was able to write a new section to introduce ELP1-medulloblastoma as a new genetic tumor predisposition syndrome.

Hurdles in rare disease research

In 2018 the European Union (EU) implemented the General Data Protection Regulation (GDPR) in the EU and European Economic Area. This EU law regulates the international transfer and usage of personal data.

While GDPR was created for the protection of personal data, the regulation causes a major hurdle in rare disease research. Dr. Waszak explains:

For rare diseases, such as pediatric brain tumors, Norway, Finland and Denmark might see just a handful of patients each year. It's tricky to perform any preclinical research on so few samples. Data sharing across borders and continents is essential. But, GDPR makes it very challenging. Even if we find a solution to data sharing in Northern Europe, there remains an issue between the legislation in the United States and in Europe.

PNOC helps. Dr. Waszak describes an example of how data sharing is solved within the United States:

The Children’s Brain Tumor Network has acquired about 1000 samples from hospitals around the United States and made the data derived from them accessible. A comprehensive genomic data sharing platform like this doesn't exist in Europe. There's no centralized European genomic data bank that allows one to query the same tumor across different countries. In Europe there are resources to make this happen, but it will be challenging. 

Seeding and motivating a career in research

Stepping back from the science, Dr. Waszak reflects on the early seeds of his career and the path to computational (neuro-)oncology: “I originally wanted to study computer science. But then I spent over a year at the Red Cross, which made me want to connect my interests in computer science with life sciences.”

As a result, he enrolled in a computer science program with a strong focus on biotechnology in Freising, near Munich, in Germany. The program included the engineering and computer science he was looking for, but also molecular biology. Intrigued by the Gene Center Munich, Dr. Waszak contacted Dr. Helmut Blum who took him into the lab. Dr. Waszak recalls how his initial interest in genomics was sparked:

Dr. Blum is a very passionate teacher. I started to work on Sanger sequencing, operating the machine and purifying samples. I studied diffusion kinetics of nucleic acids through Millipore membranes with a diffractometer he recovered from the 1970s. His enthusiasm sparked my interest to study DNA. He also taught me functional genomics and how to analyze gene expression microarrays. And then I discovered my love for data science.

Around that time the 1000 Genomes Project was beginning to ramp up, and Dr. Waszak had the chance to develop a new computational method to study copy number variation of olfactory receptor genes at the Weizmann Institute of Science in Rehovot, Israel with Prof. Doron Lancet. 

Looking to move into gene regulation and transcriptomics, he then followed an opportunity with Prof. Bart Deplancke at the École Polytechnique Fédérale de Lausanne, Switzerland in systems genetics and gene regulation, which eventually led to his PhD degree. 

The deep dive into brain tumor research followed during his postdoc with Prof. Jan Korbel at EMBL Heidelberg in cancer genomics and Prof. Stefan Pfister at the German Cancer Research Center in pediatric neuro-oncology. He recalls that the research questions were intriguing and there were so many unknowns:

My entry point into this topic was during my postdoc, to answer the question of if medulloblastoma, a rare pediatric brain tumor, always occurs sporadically. The field knew about genetic syndromes that predispose to brain tumors, but it's never been systematically analyzed across all genes. I found it very challenging and interesting to study patient genomes and to try to answer the question of how many of these patients have an underlying genetic predisposition.

Each step of Dr. Waszak’s career has been led by curiosity. And, that is no different for his research program as a group leader at NCMM. When asked what motivates him now, he responds: 

To make discoveries. Now there are more resources and long-term visions to develop new approaches and to participate in clinical studies. It also motivates me now to use data science to actually aid and help clinicians. I am always on the lookout for how I can reuse clinical data. One of my goals is to not reinvent the wheel, but to make it spin faster.

Current collaborations with Prof. Sabine Mueller (UCSF), Dr. Javad Nazarian (University of Zurich), and Dr. Carl Koschmann (University of Michigan) have fueled his research program to better understand the biology of DMGs and DIPGs, develop new diagnostic assays, and identify new therapeutic targets for these aggressive brain tumors.

Recognizing the role of his host institute and the University of Oslo, he explains, “NCMM has given me the freedom and full support to develop an independent line of research and explore new directions and opportunities as a junior group leader.”

Dr. Waszak warmly explains that every mentor in his career so far has changed him, pushing him in new directions. He is extremely grateful and humble for their influence.

Listening to Dr. Waszak talk about his research and mentors through the years is inspiring. His motivation for science comes from being able to connect with colleagues around the world to discuss tough problems and initiate new lines of collaborative work. 

And from the purest of motivators in science - the curiosity to discover something new.
 

The “Behind the Science'' profile series takes an in-depth look at a scientist or group within the Nordic EMBL Partnership. 

Brief career summary

Dr. Sebastian Waszak was initially trained in bioinformatics. He completed his PhD with special distinction in biotechnology and bioengineering at the École Polytechnique Fédérale de Lausanne (EPFL) in 2014. Dr. Waszak then moved to the European Molecular Biology Laboratory (EMBL) as an EMBO and SNSF Postdoctoral Fellow and extended his research into cancer genomics and personalized medicine. In early 2020, Dr. Waszak began his career as a group leader at NCMM, heading the Computational Oncology research group with a commitment to childhood brain cancer. In October 2021, he signed a Memorandum of Understanding with the Pacific Pediatric Neuro-Oncology Consortium (PNOC) and became an Associate Adjunct Professor at the Department of Neurology at the University of California, San Francisco (UCSF) enabling closer collaboration with the pediatric neuro-oncology community within the PNOC trial network and at UCSF.