Latest Research News
ProCan Progress Update
By CMRI | June 05, 2019
Since the ACRF International Centre for the Proteome of Human Cancer (ProCan®) launched at Children’s Medical Research Institute (CMRI) in the last quarter of 2016, it has made excellent progress.
ProCan was made possible by ACRF’s 30th Anniversary $10 million grant, with further substantial support from Cancer Institute NSW and other funders. In April 2018, the Federal government committed $20.4 million of MRFF funding over four years, and is now close to completing the contracts for that support.
This international cancer research program depends on a purpose-built facility featuring a series of specialised instruments (barocyclers and mass spectrometers) that can process and generate proteomes (i.e., quantitation of thousands of proteins) from very small cancer samples and at high throughput. Two years after launch, ProCan is still one of only two high-throughput proteomics facilities worldwide and the only one focused on cancer.
This ambitious world-first research program aims to analyse the human cancer proteome on an unprecedented scale. ProCan’s primary plans over its expected seven-year initial phase include proteome generation from up to 70,000 cancer samples of all types, from paediatric and adult cancer patients whose treatment outcomes are known. Well-annotated cancer samples are being sourced from collaborators and tissue banks worldwide.
The ProCan team is also creating a secure and robust data management platform to enable storage, handling, analysis and dissemination of the very large amounts (petabytes) of data generated. A team of computational scientists work to analyse ProCan’s pipeline of cancer big data, to help answer key clinical questions about human cancers and determine protein signatures/biomarkers that accurately identify each cancer type and sub-type and predict an individual cancer’s response to treatments. The predictive power of the ProCan database is expected to increase steadily as more samples are analysed.
ProCan’s ultimate goal (which extends beyond the initial seven-year data-generation and analysis horizon) is to develop clinically-validated technologies and tools to transform cancer diagnosis and treatment decision-making. These clinical tools would greatly improve the personalised management of cancer by predicting with far greater accuracy than is possible today, the treatments most likely to be successful for an individual patient’s cancer and those that are very likely to fail and should therefore be avoided. The ability to use the most effective treatment regimen first-tine will improve cancer treatment success rates (including the proportion of patients cured) and will result in fewer side effects from futile therapies.
The ProCan project relies on global collaboration and involves multidisciplinary teams across diverse fields, such as proteomics, cancer cell biology, oncology, cancer pathology, computer science, software engineering, data science and mathematics.
The present report outlines ProCan’s progress in its second full year of operation. Activities in 2018 have primarily focused on establishing the protocols required for consistent, high-throughput operation of the proteomics facility, and creating the software and analytic processes required to handle proteomic data on this unprecedented scale. At the same time, we have been establishing local, national and international collaborations, including an EU Horizon 2020 consortium on paediatric cancer, and processing large numbers of samples.
A snapshot shows that scientific projects carried out by ProCan involved some significantly large numbers of samples, with the projects completed or underway in 2018 much larger than 2017. The aim of 2019 is to ramp up the facility into continuous high-throughput using the technologies and processes we have been developing.
• Improved standard operating procedures to support moving to continuous high-throughput by mid-2019.
• Sample handling and processing improvements which increase the output of each barocycler, possibly by a factor of 2.
• A ProCan Consumer forum was held at CMRI on 17 October 2018.
• Began building a metadata and workflow management engine (Swon) that ensures reproducibility of the computational proteomics pipeline. Although still under development, this has demonstrably improved the robustness of analyses.
• Protocols for frozen tissue sectioning and digital histopathology review of matched sections have been established.
• Made excellent progress developing infrastructure for data management, searching, and archiving. The data include metadata covering all processing steps within ProCan: in the pathology and proteomics labs, and all computational processing of the digital data.
• Two peer-reviewed ProCan publications demonstrating improvements in crucial, underlying techniques (Barocycler and Mass Spectrometry).
• Assembling a series of computer programs (a “software pipeline”) to convert raw mass spectrometry data into a set of peptide data points containing tens of thousands of items – the “peptide matrix”.
• In collaboration with Amazon Web Services, we demonstrated our software pipeline can scale to meet the unprecedented requirements of ProCan (accepted for publication in Journal of Proteome Research).
• Made good progress towards another state-ofthe- art pipeline for generating “spectral reference libraries” (i.e., information required to search the raw data and identify peptides); this work was presented at the Australian Proteomics conference in February 2019.