Data Use Projects

The MOHCCN Gold Cohort is Canada’s largest and most comprehensive cancer case resource, comprising paired clinical and genomic data from a diverse cohort of 15,000 Canadian cancer patients.

Network researchers may request access to Gold Cohort data for analysis projects.

Learn more about how to submit a data access request.

Approved Gold Cohort Data Access Projects

Precision Targeting of ecDNA-Driven Tumors through Bioinformatic Identification of Synthetic Lethal Vulnerabilities

Lead Investigator: David Cescon
Lead Institution: Princess Margaret Cancer Centre

Summary:

Cancer cells sometimes have extra loops of DNA called extrachromosomal DNA (ecDNA), which are separate from the normally packaged chromosomes. EcDNA often carries genes that make cancers grow faster, become aggressive, and resist treatment. Patients whose tumors contain ecDNA seem to have poorer outcomes. At the Princess Margaret Cancer Centre, we found that early-stage breast cancer patients with ecDNA-containing tumors responded poorly to curative therapy (chemotherapy and endocrine therapy) and had a higher chance of their cancer coming back. Knowing this, we need new treatments targeting ecDNA-containing cancers to improve outcomes for patients. Our study aims to find these new treatments by identifying weaknesses in ecDNA-driven cancer cells using a concept called synthetic lethality (SL). SL occurs when cancer cells survive losing function in one gene but die when a second function is also disrupted. We think that we can apply this approach for the presence of ecDNA and the genes that are contained in them. Using genomic data from thousands of Canadian cancer patients (the Marathon of Hope Cancer Centres Network Gold Cohort), we will find tumors containing ecDNA and identify gene weaknesses unique to these tumors. First, we will detect ecDNA and the genes it carries in this unique data resource using specialized computer software. Next, we'll find gene pairs where disrupting one gene kills only ecDNA-positive cancer cells, leaving healthy cells unharmed. We will share our findings and methods openly, helping scientists develop new treatments to improve outcomes for patients with ecDNA-positive cancers as quickly as possible.

Date of Request Submission: July 13, 2025
Date of Approval: August 11, 2025
Duration: 2 years (to August 11, 2027)

Improving assessment of immune cell infiltration in solid tumors (VDJump software)

Lead Investigator: Jennifer Chan
Lead Institution: University of Calgary

Summary:

Tissue samples for bulk tumor genome sequencing often contain non-tumor cells, including tumor infiltrating lymphocytes (TILs) acting as part of the body's immune response to the tumor. TILs are important prognostic and treatment indicators in many cancers. Existing methods for estimating the proportion of TILs and myeloid white blood cells in tumors are based on having an RNA sequence dataset for the tumor. Uniquely, VDJump makes TIL estimates (B cell and T cell fractions) from DNA data, which can be more readily generated or already available for many cancer cases.

Date of Request Submission: July 13, 2025
Date of Approval: August 11, 2025
Duration: 2 years (to August 11, 2027)

Detection of Clonal Hematopoiesis Driven by Structural Variants

Lead Investigator: Aly Karsan
Lead Institution: BC Cancer

Summary:

As we get older, our blood cells can develop DNA changes, or mutations, that help some cells grow faster than others. When these mutated cells start to multiply more than usual, it is  called  clonal  hematopoiesis  (CH). CH is fairly common — about 1 in 10 people over age 65 are estimated to have it. While CH doesn’t always lead to illness, CH raises the risk for blood cancers, as well as conditions like heart disease and diabetes.

Evidence indicates that CH may be even more common than currently thought, because  some types of genetic  changes  are  harder  to  detect  with  standard methods. We  want  to explore  those  harder‐to‐find  changes, especially  a  kind called chromosomal translocations. These happen when parts of  two different chromosomes swap places, sometimes creating new, abnormal genes that can lead to cancer. Surprisingly, versions of these cancer‐related changes have been found in the blood of healthy people, which suggests they might also be involved in CH, even in those without cancer. However, this idea has not been studied on a large scale. 

This project will look at genetic data from normal control blood samples collected in  the  MOHCCN  consortium.  The  goal  is  to  see  whether  these  chromosomal translocations  are  present  in  older  adults  without  disease,  and  whether  they become more common with age. If so, that would support the idea that these changes are part of CH. 

We will use advanced computer tools and powerful data‐sharing systems to scan for these large‐scale DNA changes, which are usually missed in routine genetic analyses. This pilot study will also test whether researchers across different sites can collaborate effectively using shared resources. 

If successful, the study could change how physicians and scientists detect and understand CH. Instead of focusing only on small mutations, they may also need to look for bigger structural changes in DNA. In the long run, this could improve early detection of blood cancers and help prevent other health problems linked to CH.

 

Date of Request Submission: July 13, 2025
Date of Approval: August 11, 2025
Duration: 2 years (to August 11, 2027)

Computational synthetic lethal analysis of MOHCCN genomic data to expand precision cancer genomic medicine options for glioblastoma multiforme patients

Lead Investigator: Marco Marra
Lead Institutions: Canada’s Michael Smith Genome Sciences Centre, University of British Columbia

Summary:

Today, only a small number of genetic changes found in cancer patients can be matched to drugs. This is especially true for aggressive cancers like glioblastoma, where most patients have no treatment options based on their tumour’s unique genetics.

Our project aims to improve this by finding new ways to match existing drugs to more patients. We use a strategy called synthetic lethality—a situation where two genes, when altered together, can cause cancer cells to die, while healthy cells remain unharmed. To do this, we use a computational tool called GRETTA, developed by our team, to search for these gene pairs using large datasets. GRETTA has already been used to match existing drugs to several cancers based on their unique genetic features. Now, we plan to apply this method to ~120 glioblastoma cases from across Canada, using genetic data from the MOHCCN gold cohort. We’ll computationally search for genetic patterns that may enable an existing drug to kill glioblastoma cells, even if those drugs weren’t originally designed for brain cancer.

In addition to potentially helping patients who currently have few or no treatment options, this project will also test how well different MOHCCN members across Canada can share and use genetic data together to accelerate research. This will strengthen national collaboration in precision medicine and help bring the benefits of this research and others to more Canadians.

Date of Request Submission: July 13, 2025
Date of Approval: August 11, 2025
Duration: 2 years (to August 11, 2027)

Combined analysis of MOHCCN Gold Cohort and TCGA/ICGC data to Improve Detection of Low-Frequency Cancer Driver Genes

Lead Investigator: Ian Watson
Lead Institution: McGill University Goodman Cancer Institute

Summary:

Cancer is caused by changes in our DNA that allow cells to grow out of control. Some of these changes, called driver mutations, directly contribute to cancer development. Finding these driver mutations helps researchers understand how cancers form and grow and can also guide treatment decisions or lead to new therapies.

Over the past decade, large international research efforts, like The Cancer Genome Atlas (TCGA) and the International Cancer Genome Consortium (ICGC), have studied thousands of tumors and successfully identified many common driver mutations in cancer. However, many less common, but still important, driver mutations remain undiscovered. This is especially true in cancers that have many DNA changes, where it becomes harder to tell which mutations are important and which are just “background noise.” A previous study showed that thousands of patient samples are needed to reliably detect these rare drivers.

This project will utilize the MOHCCN Gold Cohort, a growing national resource that includes detailed genetic and clinical data from cancer patients across Canada, to look for these rare driver mutations. We will analyze the complete DNA sequence (whole-genome sequencing) of many tumors to identify genes that are mutated more often than expected by chance. We will also look beyond the parts of DNA that code for proteins, and search for driver mutations in non-coding regions—areas that help regulate genes, but where mutations are harder to study and researchers still don’t have a clear understanding of their function from smaller studies.

Importantly, we will combine different types of data to improve the accuracy of our results. For example, we will use RNA data (which shows which genes are turned on or off in the tumor) to check whether a mutation actually affects gene activity. We will also examine whether certain mutations are paired with losses of other parts of the genome, which is a pattern often seen in important genes that play a role in suppressing tumor formation. Finally, we will compare our findings to large experimental datasets that show which genes are essential for cancer cell survival, helping us focus on mutations that are most likely to matter.

The goal of this work is twofold: first, to discover new cancer driver genes and mutations that could lead to better diagnostics or therapies in the future, and second, to demonstrate that sharing and analyzing data across cancer centers in Canada through the MOHCCN platform is both feasible and scientifically valuable.

In the long run, this project could help identify new targets for therapy, improve how we diagnose and classify cancers, and contribute to more personalized care for cancer patients. It will also help build Canada’s capacity to do large-scale, collaborative cancer research that benefits patients nationwide.

Date of Request Submission: July 13, 2025
Date of Approval: August 11, 2025
Duration: 2 years (to August 11, 2027)

The Personalized OncoGenomics Program (POG)

Lead Investigator: Steven Jones
Lead Institution: BC Genome Sciences Centre

Summary:

The Personalized OncoGenomics (POG) Program, which is now funded as part of the Marathon of Hope Cancer Centres Network (MOHCCN), is a clinical research initiative dedicated to improving outcomes for patients facing advanced and metastatic cancers. The program has successfully sequenced tumors from over 1,900 adult and pediatric patients to date. The rimary goal of POG is to comprehensively analyze the genetic makeup of these difficult-to-treat cancers to determine which specific therapies are most likely to effectively combat the disease.

To achieve this, the program utilizes advanced whole-genome and transcriptome sequencing. In practical terms, this means researchers look at the complete DNA sequence of a patient's tumor—alongside their normal, healthy DNA—to identify specific genetic mutations that are driving the cancer's growth. Furthermore, by analyzing gene activity, the team can understand exactly how the tumour is behaving and how its genetic landscape has been altered by any therapies the patient has already received. These crucial insights are actively integrated into clinical care; they are relayed directly back to the patient's clinical management team through detailed reports and collaborative molecular tumour boards, directly informing future clinical decision-making.

While the POG program has already revealed vital predictive markers regarding how well patients will respond to certain therapies, cancer is an incredibly complex disease. To confidently confirm these retrospective findings and discover even more about how advanced cancers behave, researchers need to compare local data against a much larger, diverse pool of patients. This is the core purpose of this data access proposal. The MOHCCN has amassed a highly valuable, nationwide resource of clinical and genomic data from thousands of cancer patients across Canada, known as the Gold Cohort.

In this project, our team intends to utilize the sequence data available through the MOHCCN to validate and deeply interpret the novel genetic and transcriptomic observations we have made within the POG program. By comparing POG patients to the broader MOHCCN population, we can minimize technical biases and more easily identify distinct genomic features that contribute to cancer development, progression, and drug resistance. We will specifically look at a variety of tumor characteristics across these datasets, including small mutations, structural changes in the DNA, and how genes are expressed in the tumor microenvironment.

Ultimately, merging the focused insights of the POG initiative with the massive scale of the MOHCCN database will significantly enhance our ability to interpret complex patient data. It will allow us to validate new findings regarding why some advanced cancers resist therapy while others respond and help pinpoint specific genetic variants that have true biological and clinical importance. By communicating these validated findings through clinical reports and sharing them with the broader scientific community, this project will directly contribute to the nationwide advancement of precision medicine, ensuring that cancer patients receive the most individualized and effective treatments possible.

Date of Request Submission: April 20, 2026
Date of Approval: May 28, 2026
Duration: 2 years (to May 28, 2028)

Characterization of prostate cancer tumor tropism

Lead Investigator: Frédéric Pouliot
Lead Institution: Université Laval

Summary:

In Canada, prostate cancer (PCa) is the second leading cause of cancer death among men. During carcinogenesis, tumor cells remain dependent on hormones notably androgens, and current treatments aim to block androgen activity. However, despite these pharmacological approaches, 30% of patients will eventually develop metastatic cancer that is resistant to these treatments. Cancer cells can then colonize new organs distant from the prostate, such as the lungs, liver, bones, and others. At this advanced stage of the disease, there is no curative or organ-specific anti-metastatic treatment. It therefore becomes urgent to better understand the mechanisms underlying the establishment of these metastatic sites. A better understanding of the biology of the metastatic process is essential to improving clinical management of patients. To develop new therapeutic targets specific to each metastatic site, our objective is to characterize DNA profiling and gene expression in metastases to develop Bayesian models. Thus, bioinformatics tools will enable the identification of somatic and germline mutations (DNA) specific to each metastatic site such as bone, liver, and lung. By integrating patients’ clinical parameters and radiomic features, this profiling will identify the specific cellular mechanisms that enable metastatic development toward distinct niches. Ultimately, we will gain a much better understanding of the biology of advanced prostate cancer to improve patient management based on the biological reality of their disease.

 

Date of Request Submission: June 9, 2026
Date of Approval: July 13, 2026
Duration: 2 years (to July 13, 2028)

Defining the Clinical Utility of Whole-Genome and Transcriptome Sequencing in Primary Brain Tumours

Lead Investigator: Kevin Wang
Lead Institution: Princess Margaret Cancer Centre

Summary:

Patients with primary brain tumours still have limited access to precision oncology approaches compared with many other cancers. This project will analyze whole-genome and transcriptome sequencing data from approximately 331 MOHCCN brain tumour cases and compare them with a Hartwig Medical Foundation brain tumour cohort processed using comparable WGTS workflows. We will study tumour DNA, matched normal DNA, RNA sequencing, and linked clinical data to determine how often comprehensive sequencing identifies diagnostically relevant, prognostic, or potentially actionable findings beyond standard testing. The goal is to clarify when WGTS can improve classification and clinical management of primary brain tumours in real-world practice.

 

Date of Request Submission: June 16, 2026
Date of Approval: July 20, 2026
Duration: 2 years (to July 20, 2028)

Optimized gene expression-based computational inference of tumour mutational burden to enable first-in-class immunotherapy response prediction

Lead Investigator: Benjamin Haibe-Kains
Lead Institution: Princess Margaret Cancer Centre

Summary:

Immunotherapy has improved outcomes for many patients with cancer, but not all patients benefit. One of the strongest predictors of benefit is tumour mutational burden (TMB), which measures the number of genetic mutations in a tumour. However, current methods of measuring TMB rely on large-scale sequencing, which is expensive, resource-intensive, and not widely available in clinical practice.

This project will develop a way to measure TMB using standard RNA-based data. By applying advanced computational models to large patient cohorts with paired RNA and genomic sequencing, we will create and validate predictors of TMB that can also reveal how immune cells interact with tumours. We will then adapt this method into a practical test that can be run directly on tumour tissue, together with other important immune biomarkers. In addition, we will explore whether TMB can be inferred and associated with routine H&E-stained pathology slides, as these slides are part of standard care and offer rich spatial cellular and tissue architectures.

If successful, this research will make precision immunotherapy more affordable, scalable, and accessible, improving patient selection and supporting the design of future clinical trials.

 

Date of Request Submission: April 27, 2026
Date of Approval: August 12, 2026
Duration: 2 years (to August 12, 2028)