March 2, 2020

Study reveals roots of leukemia that current chemotherapies can’t reach

John Dick

Researchers find the roots of leukemia relapse are present at diagnosis, uncovering clues to new treatment approaches

Despite significant advances in the treatment of acute lymphoblastic leukemia (ALL), the disease often returns aggressively in many patients after treatment. It is thought that current chemotherapies eliminate most leukemia cells, but groups of resistant cells may survive therapy, progress and eventually cause relapse. Dr. John Dick and collaborators have found these cells.

In a recent study published in Cancer Discovery, Dick and collaborators were able to identify and isolate groups of genetically distinct cells that drive ALL relapse.

The cells, termed diagnosis relapse initiating (dRI) clones were found to have genetic characteristics that differ from the other leukemia cells that are eliminated by treatment.

The study, along with a complementary study published in Blood Cancer Discovery, unraveled the genetic, epigenetic, metabolic and pro-survival molecular pathways driving treatment resistance. Together, these papers provide an integrated genomic and functional approach to describing the underlying genetics and mechanisms of relapse for ALL.

Interestingly, the research group discovered that dRI clones are present at diagnosis, opening opportunities to improve treatment up-front, devise drugs that target these resistant cells and prevent relapse from ever occurring.

Dr. Stephanie Dobson

“Our study has shown that genetic clones that contribute to disease recurrence already possess characteristics such as therapeutic tolerance that distinguish them from other clones at diagnosis,” says Dr. Stephanie Dobson, first author of the study who performed this research as a member of John Dick’s Lab. “Being able to isolate these clones at diagnosis, sometimes years prior to disease recurrence, has enabled us to begin to profile the properties allowing these particular cells to survive and act as reservoirs for relapse. This knowledge can be used to enhance our therapeutic approaches for targeting relapse and relapse-fated cells.”

“Xenografting added considerable new insight into the evolutionary fates and patterns of subclones obtained from diagnosis samples,” says John Dick, who is the co-senior author of the study, Senior Scientist at the Princess Margaret Cancer Centre and leader of OICR’s Acute Leukemia Translational Research Initiative. “We were able to gather extensive information about the genetics of the subclones from our models, which helped us describe the trajectories of each subclone and the order in which they acquired mutations.”

Ordering these mutations relied on the advanced machine learning algorithms designed by Dr. Quaid Morris and Jeff Wintersinger at the University of Toronto.

Research efforts are underway to build on these discoveries and determine how to block dRI clones.

The study was led by researchers at St. Jude Children’s Research Hospital, the Princess Margaret Cancer Centre and the University of Toronto and supported in part by OICR’s Acute Leukemia Translational Research Initiative.

This post has been adapted from the St. Jude Children’s Research Hospital news release.

February 5, 2020

AI algorithm classifies cancer types better than experts

Gurnit Atwal and Wei Jiao

Pan-Cancer Project researchers develop deep learning system that can determine where a cancer originates with better accuracy than human experts

If doctors know where a patient’s cancer started, they can better treat the disease. Unfortunately, this is not always possible, but AI could play a role in solving that.

In a study published today in Nature Communications, a Toronto-based researcher group developed a deep learning system that can accurately classify cancers and identify where they originated based on patterns in their DNA. The system could potentially help clinicians differentiate difficult-to-classify tumours and help recommend the most appropriate treatment option for their patients.

“We reasoned that there was something within the cancer’s DNA that could help us classify these tumours,” says Dr. Quaid Morris, OICR Senior Investigator and co-lead author of the study1. “But I didn’t expect our system to work at well as it does – in some cases, far better than pathologists.”

The team

The initiative began with the dataset: 2,600 whole genomes across 38 tumour types from the Pan-Cancer Analysis of Whole Genomes Project, also known as the Pan-Cancer Project or PCAWG.

Dr. Lincoln Stein, Head, Adaptive Oncology at OICR and member of the Pan-Cancer Project Steering Committee, and his team began to work with these data to identify patterns in a cancer’s genetic material that could help classify these tumours. To them, this was a perfect problem for AI.

When we started to collaborate, We realized we had something amazing.
– Wei Jiao

“Deep learning models excel when they’re trained on large amounts of data,” says Wei Jiao, Research Associate in the Stein Lab and co-first author of the study. “We had an incredibly large dataset to work with, the most comprehensive dataset of whole cancer genomes to date, but we also needed the machine learning expertise.”

The Stein Lab posted their progress on bioRxiv, an open-access repository for biology publications that have not yet been peer-reviewed, which in turn sparked the collaboration between his team and the Morris Lab – a group with deep machine learning expertise.

The system

The development of their deep learning system was not simple. They mined through terabytes of data looking for patterns in the type of mutations, the source of mutations and where mutations occurred in the genome, among other factors.

To their surprise, they found that patterns in driver mutations – the changes in DNA that are thought to ‘drive’ the development of cancer – were not useful in determining where the tumour originated. Instead, they found that patterns in the distribution of mutations and the type of mutation within a patient’s sample could better classify the patient’s disease.

“We knew that we could distinguish between two different types of healthy cells by looking at how the DNA within the cell types are packaged,” says Stein, who is a co-lead author of the study. “We were surprised and gratified that we could do the same using cancer cells.”

“We saw that the tightly-packaged sections – also known as the closed chromatin – would have many more mutations than the loosely wound sections,” says Gurnit Atwal, PhD Candidate in the Morris Lab and co-first author of the study. “It was like the normal cell was casting a shadow on the cancer cell, and we just had to read the shadows.”

To achieve the highest accuracy, the research group developed a deep learning neural network-based system, a type of system that is loosely modeled after the human brain and commonly used to recognize patterns in images, audio and text. Their system achieved an accuracy of 91 per cent – roughly double the accuracy that trained pathologists can achieve using traditional methods when presented with a primary tumour and no clinical information.

Further, they tested their model on an additional 2,000 tumours from patients in the Netherlands who donated their cancer genomic data to the Hartwig Medical Foundation and the system still performed with a remarkably high level of accuracy.

 “As more cancer genomes are sequenced, we can gain the ability to classify rarer cancers,” says Atwal. “Where we are now is great, but there is more work to be done.”

The potential

This study presents a deep learning system that could potentially improve how cancers are classified, enhancing the accuracy of current diagnostic tests and the treatment decisions they inform.

For some patients, this system could tell them where their cancer began, giving them valuable information about which course of treatment to choose. The system also could serve as a tool to help doctors identify whether a tumour in a patient who has been treated for cancer in the past is an entirely new tumour or a recurring tumour that has spread.

“A treatment plan for a cancer that originated in the throat may be very different than one for that originated in the breast, and the treatment for a cancer that has returned is different than for one that has metastasized,” says Atwal. “One day, our tool could help give doctors the power to distinguish these classes of tumours, giving patients valuable information that wouldn’t have been available otherwise.”

The authors of the study suggest that their system could start helping patients soon. They plan to further refine their system for patients with rare cancers before moving towards clinical studies. 

“The potential impact of the system we’ve developed is encouraging,” says Morris. “We look forward to turning this system into a tool that can help clinicians and future cancer patients tackle this disease.”


1Morris is also a Canada CIFAR AI Chair, Faculty Member at the Vector Institute, and Professor at the University of Toronto’s Donnelly Centre for Cellular and Biomolecular Research.


Related links

February 5, 2020

TrackSig: Unlocking the history of cancer

Yulia Rubanova
Yulia Rubanova

Toronto-based machine learning experts map the changes that lead to cancer, revealing opportunities for earlier diagnosis and new approaches to outmaneuver the disease

A tumour is often made up of different cells, some of which have changed – or evolved – over time and gained the ability to grow faster, survive longer and potentially avoid treatment. These cells, which have an ‘evolutionary advantage’, are thought to cause the vast majority of cancer deaths but researchers now have a new tool to tackle tumour evolution: TrackSig.

TrackSig – which was developed by Dr. Quaid Morris and his team at the University of Toronto, the Vector Institute and OICR – is a novel computational method that can map a cancer’s evolutionary history from a single patient sample and in turn help researchers thwart the disease’s next move.

“We combined sequencing with evolutionary theory and mathematical modeling to understand how cancers develop and adapt to resist treatment,” says Yulia Rubanova, PhD Candidate in the Morris Lab and lead author of the study. “This understanding lays the foundation for us to be able to predict – and impede – tumour evolution in future cancer patients.”

This understanding lays the foundation for us to be able to predict – and impede – tumour evolution in future cancer patients
– Yulia Rubanova

TrackSig was published today in Nature Communications alongside nearly two dozen other publications in Nature and its affiliated journals related to the Pan-Cancer Analysis of Whole Genomes Project, also known as the Pan-Cancer Project or PCAWG.

Previous tumour evolution studies focused on identifying the most frequent changes – or mutations – in a patient sample, where the most common mutations represent changes that came earlier in the tumour’s development and less common mutations represent more recent changes. Instead, Morris’ TrackSig charts different types of mutations over time, generating maps of a tumour’s evolutionary history in finer detail and with better accuracy than ever before.

This level of resolution enabled the discovery that many cancer-causing genetic changes occur long before the disease is diagnosed.

“For exceptional cases like in certain ovarian cancers, we were able to see these early events happening 10 to 20 years before the patient has any symptoms,” says Dr. Lincoln Stein, Head of Adaptive Oncology at OICR and member of the Pan-Cancer Project Steering Committee. “This opens up a much larger window of opportunity for earlier detection and treatment than we thought possible.”

The tools and findings from the Pan-Cancer Project are changing the way we think about cancer
Dr. Quaid Morris

With their new detailed maps of tumour evolution, the research group plans to further investigate novel cancer treatment strategies and design new therapies that can better anticipate, prevent and overcome evolution and drug resistance.

“The tools and findings from the Pan-Cancer Project are changing the way we think about cancer,” says Morris. “We’ve uncovered new opportunities to improve diagnosis and treatment, and we’ll continue to strive towards getting the best treatment to patients at the right time.”

TrackSig is freely available for the research community to use at https://github.com/morrislab/TrackSig.


Related links

April 19, 2018

Landmark study links tumour evolution to prostate cancer severity

Largest-ever study of its kind uses a tumour’s past to accurately predict its future

Toronto (April 19, 2018) – Findings from Canadian Prostate Cancer Genome Network (CPC-GENE) researchers and their collaborators, published today in Cell, show that the aggressiveness of an individual prostate cancer can be accurately assessed by looking at how that tumour has evolved. This information can be used to determine what type and how much treatment should be given to each patient, or if any is needed at all.

The researchers analyzed the whole genome sequences of 293 localized prostate cancer tumours, linked to clinical outcome data. These were then further analyzed using machine learning, a type of statistical technique, to infer the evolutionary past of a tumour and to estimate its trajectory. They found that those tumours that had evolved to have multiple types of cancer cells, or subclones, were the most aggressive. Fifty-nine per cent of tumours in the study had this genetic diversity, with 61 per cent of those leading to relapse following standard therapy.

Continue reading – Landmark study links tumour evolution to prostate cancer severity

October 23, 2017

Boost your bioinformatics knowledge at TorBUG

Torbug - Lecture illustration

The Toronto Bioinformatics User Group’s (TorBUG) 2017-2018 season continues this Wednesday, October 23 with two presentations that promise to be of interest to anyone involved in bioinformatics. Dr. Quaid Morris, Associate Professor at the University of Toronto (U of T) will present “The Genetic Archaeology of Individual Cancers”. Brendan Innes, a PhD Candidate in the Bader Lab at U of T will cover “Cell types in single-cell RNAseq.”

Continue reading – Boost your bioinformatics knowledge at TorBUG