Credit: Gpointstudio.

Precision medicine is revolutionising cancer treatments, with whole genome sequencing and transcriptomics enabling tailored treatments for individual patients. However, as most drugs act on proteins not genes, studying proteins directly might reveal otherwise missed treatments.

Dr Philipp Lange and his research group at the University of British Columbia and BC Children’s Hospital Research Institute, Vancouver, Canada, specialise in proteomics, the study of proteins, and how these are altered in childhood cancers.

Together with scientists across Canada, they recently published a groundbreaking study where proteomics found a unique treatment for a young cancer patient. The patient’s tumour was then grown in a chicken egg allowing quick testing of the identified drug. This study was a team effort of PROFYLE (Precision Oncology For Young people), a Canadian initiative to find personalised treatments for young people with difficult-to-treat cancers.

ScienceVega spoke to Philipp and his current team, which includes research associate Dr Kirsty Ferguson, senior research technician Dr Verena Goebeler, and bioinformaticians Agustina Conrrero and Dr Enes Kemal Ergin, about their research and plans for the future of precision medicine for childhood cancers.

Lead authors of the case study (left to right) Dr Georgina Barnabas, Dr Verena Goebeler, Dr Philipp Lange, Dr James Lim, Tariq Bhat Credit: © UBC/ Philipp Lange.

ScienceVega: Why are cancers in young people different to those in older people?

PL: Childhood cancers have 14 times fewer somatic mutations than adult tumours, that is, fewer mutations acquired throughout life. This is because children have had a shorter exposure time to environmental risks than adults and their cancers often happen when faults occur during development. This means analysing proteins directly could be particularly helpful in identifying treatments for children. Proteomics also brings us closer to the actual molecules driving cancers. There are many layers of regulation between DNA, RNA and the final protein which carries out functions in a cell. For example, proteins can have chemical changes after they have been produced called post-translational modifications. Because of this, measurements of DNA or RNA may not correspond directly to the proteins that most anti-cancer drugs target.

Compared to adult cancers, research into childhood cancers is also significantly underfunded, with many fewer treatments developed specifically for children. One reason for this is that many childhood cancers are considered rare, by adult cancer definitions. Children have unique physiologies and their ongoing development must be considered. This makes finding personalised treatments even more pressing for children and young people, to provide a lifeline when few options exist.

ScienceVega: In the case study, what did proteomics show that genomics didn’t?

PL: When the patient’s tumour first regrew, genomic analysis suggested treatment with a receptor tyrosine kinase inhibitor. Unfortunately, the patient’s tumour became resistant to this drug and returned for a third time. At this stage, DNA and RNA analyses did not reveal any more options. However, led by Dr Georgina Barnabas in my lab, proteomics found the tumour had high levels of a mitochondrial enzyme compared to the surrounding normal tissue. This suggested a way to specifically target the metabolism of the tumour. We also did proteomics analysis on earlier tumour samples from the patient and this enzyme was found to be at consistently high levels, giving us even more confidence in it as a ‘target protein’.

Members of Dr Lange’s research team at BC Children’s Hospital Research Institute (left to right) Farah Jayousi, Agustina Conrrero, Dr Kirsty Ferguson, Dr Verena Goebeler, Dr Philipp Lange. Credit: © UBC/ Philipp Lange.

ScienceVega: How did this lead to a new treatment for the patient?

PL: Importantly, we were able to identify this target protein within two weeks of receiving the tumour biopsy, giving us time to test how we could ‘drug’ it. Interestingly, we read that the target protein can be inhibited by an anti-depressant already known to be safe for children. We then tested this drug in a chicken egg model of the patient’s cancer in the lab to confirm it slowed tumour growth. As a result, the patient was treated with this drug, and although the tumour continued to grow, the speed of growth was decreased. This showed for the first time that proteomics could guide real-time treatment for a patient.

ScienceVega: Can you explain the chicken egg process? Is this a new method, how exactly is it done and why did you choose it?

VG: The chicken egg, or chick chorioallantoic membrane (CAM) model, was established by PhD student Tariq Bhat in James Lim’s lab at BC Children’s Hospital. Together with James’ lab, we are part of the BRAvE Initiative (Better Responses through Avatars and Evidence) which uses different lab methods to study patient tumour cells to pinpoint which drugs work best. The chicken egg model is already used in cancer research, but not widely for children’s cancers. It involves grafting small pieces of tumour onto the chicken egg membrane, which provides a network of blood vessels needed for the tumour to grow. This is a low-cost way of growing a patient’s own tumour outside of their body and allows drugs to be tested very quickly, even in just one to two weeks. This speed is vital when generating data that might help find a new treatment.

Glass slide with a tumour slice stained with Hematoxylin and Eosin (H&E). Credit: © UBC/ Kirsty Ferguson.

ScienceVega: How do you see the proteomics route progressing? Will it be used in conjunction with genomics for a better overarching diagnostic process?

KF: Yes, this is exactly our aim. As was done in the first case study, we aim to integrate proteomics with genomics and transcriptomics, which are already performed routinely here at BC Cancer by the Personalized OncoGenomics Program. We know that each technique adds a complementary layer of information and hope adding proteomics will increase the chances of finding potential treatments for children whose options are limited. There are many steps to our process and at the moment we are working on optimising these so we can take on more cases. There is a team of us in the Lange lab working together and we each have a unique role.

ScienceVega: Can you each explain a bit about your role in the precision oncology process?

VG: When the hospital collects a biopsy from a patient, this is preserved and embedded in a small block of wax. This allows very thin slices of the tumour to be cut and attached to glass slides for analysis. The slides are then stained with dyes called hematoxylin and eosin (H&E), which stain nuclei purple and cytoplasm pink. Dr Jonathan Bush, paediatric pathologist at BC Children’s Hospital, then marks the tumour and normal tissue under a microscope. Here’s where my work really kicks in. I use a very precise laser beam to cut out these specific areas, a technique called laser capture microdissection. I then isolate the proteins and use an enzyme to digest the protein molecules into smaller peptides (like cutting a long chain of beads into smaller strings of beads with scissors). I then analyse the peptides on a liquid chromatography-tandem mass spectrometer (LC-MS/MS), so we can find the protein differences in the tumour compared to non-cancerous normal tissue.

AC: LC-MS/MS is an analytical chemistry technique where peptides are first separated (by liquid chromatography), before being ionised and injected into the mass spectrometer. Tandem mass spectrometry (MS/MS) measures the mass-to-charge ratio of the ions, then splits specific ions into smaller fragments, before measuring their mass-to-charge ratio again. This allows us to identify the abundance of peptides in the sample, as well as their amino acid sequences. Enes and I are the bioinformaticians who then analyse this data. We use algorithms to see which proteins are present, based on the peptides identified, and to what level – this helps us to find any differences between the tumour and the surrounding non-cancerous tissue. I am also making a database of drugs and the proteins they target, to help us then identify possible treatments.

KF: Next I read about the proteins identified in the context of the wider cancer literature to help us understand what molecules might be driving the tumour. I also read about potential drugs and past and current clinical trials, before generating a report to present to the genomics analysts and clinicians. Together, we will discuss our results and the implications for possible treatment options for the patient. In some cases, the BRAvE initiative may have patient tumour cells growing in the lab so drug testing can also be done to validate any proteins we find, or findings from genomics analysis.

The mass spectrometry facility in the Lange laboratory at BC Children’s Hospital Research Institute. Credit: UBC/ Philipp Lange.

ScienceVega: What are some of the challenges in expanding this to other patients, and how will you solve these?

VG: In some cases, tumours are not uniform and contain many different cell types (what we call ‘heterogeneity’). To make sure we have a representative view of the patient’s cancer, Dr Bush will mark extra areas which I cut out with the laser capture microdissection microscope. We are also interested to know if the protein changes we find are specific to the patient, or found in all cancers of that type. To do this, I’m also analysing tumours from ‘tissue microarrays’ which contain many small pieces of tumour from different patients.

EKE: In terms of the data analysis, expanding to other patients can be challenging because of differences in sample quality and biological variability. To overcome this, we develop analysis pipelines, which allow us to easily analyse new cases in a reproducible way. This is important to make sure we have consistent and reliable results, and will also mean we can perform the analysis quick enough to feedback to the hospital.

KF: To use proteomics clinically, we also need to link up with other information available for each patient, such as genomics analysis. This means we need to have data within a few weeks of receiving the tumour biopsy, so we can discuss and compare our findings with the hospital. As Philipp mentioned, a target protein was identified within two weeks for the recent case study, so we believe this speed is achievable! Communicating our research to other scientists, as well as the patients and their families, will also become more and more important. Whilst we have big hopes for the future, it is important to understand we are doing research and are continually learning.

ScienceVega: What is next in your research?

PL: Moving forward, we plan to analyse a number of childhood cancer cases and gradually optimise our process. Once we have reached a sufficient number, we will be able to accurately assess the benefits of proteomics in guiding clinical care. Our work is unique in going all the way from the hospital to the lab bench and back to the patient, and moving research to the clinic like this has many considerations. Ultimately, our goal is to introduce proteomics into hospitals so that it can be used routinely for precision medicine of childhood, and adult, cancers.

As the group have highlighted, teamwork is really essential in science. We are part of a large paediatric cancer network in Canada called ACCESS (Advancing Childhood Cancer Experience, Science and Survivorship), where scientists, clinicians and people with lived experience are all working together to improve the lives of children with cancer across the country. As a group, we hope to gradually change the lives of patients and their families through access to the best treatments possible.

Verena Goebeler in the Lange Lab at BC Children’s Hospital Research Institute. Image © BCCHR, Moisés Schini 2023.

Dr Philipp Lange, Canada Research Chair in Translational Proteomics of Childhood Malignancies and
Associate Professor of Pathology and Laboratory Medicine;

Dr Kirsty Ferguson, Research Associate;

Dr Verena Goebeler, Senior Research Technician;

Agustina Conrrero, Bioinformatician; and

Dr Enes Kemal Ergin, Bioinformatician.

BC Children’s Hospital Research Institute

and

The University of British Columbia, Canada

Reference: Barnabas GD, Bhat TA, Goebeler V, et al. Proteomics and personalized PDX models identify treatment for a progressive malignancy within an actionable timeframe. EMBO Mol Med (2025)17:625-64. https://www.embopress.org/doi/full/10.1038/s44321-025-00212-8