Warning! OUTDATED BROWSER DETECTED!   Please update your browser immediately for a better experience on this website. Learn More
21/xsl/MobileMenu.xsltmobileNave880e1541/WorkArea//http://www.rsna.org/RSNANewsDetailWireframe.aspx?pageid=15319&id=22086&ekfxmen_noscript=1&ekfxmensel=falsefalsetruetruetruefalsefalse10-18.0.0.0730truefalse
  •  
     
  • Look Ahead: Imaging’s Critical Role in the Cancer Moonshot Initiative

    C. Carl Jaffe, MD, discusses radiology’s role in the federal initiative aiming to make a decade's worth of progress in cancer prevention, diagnosis and treatment in five years. By C. Carl Jaffe, MD


    June 1, 2017

    Capitalizing on rapid substantial advances in molecular cellular knowledge and empowered by new technologies developed during the Human Genome Initiative (1990–2001) and the National Cancer Institute’s (NCI) subsequent The Cancer Genome Atlas (2004–2012), the cancer scientific community has been emboldened to declare that a cancer knowledge tipping point is in the offing.

    This development encouraged an ambitious federal executive branch announcement in 2016 known as the “Cancer Moonshot,” an initiative led by former Vice President Joe Biden. The original funding initiative stands to be further buttressed by the 21st Century Cures Act. Moving quickly, the NIH formed a visionary Blue Ribbon Panel (BRP) to define achievable goals that could shorten the timeframe of cancer knowledge progress by half — from 10 years to five. Those goals carry exciting potential for the imaging community (see sidebar).

    The APOLLO Consortium

    One of the earliest-defined Moonshot goals that specifically incorporates imaging is the Applied Proteogenomics Organizational Learning and Outcomes (APOLLO) consortium. (read more in the Precision Medicine article on page 8). The departments of Defense (DOD), Veterans Affairs (VA) and NCI have formed a collaboration using state-of-the-art research methods in proteogenomics to more rapidly identify unique targets and pathways for cancer detection and intervention. These methods look for a patient’s genes that may lead to cancer and the expression of these genes in the form of proteins, all of which have potential impact on understanding disease formation and treatment. The full set of medical images, including CT and MRI scans, obtained for each patient before and during treatment will be acquired and curated in NCI’s open-access The Cancer Imaging Archive (TCIA) (see sidebar, Page 7).

    Each individual set of images will be connected to that patient’s clinical, genomic and proteomic data. Big data science techniques can then be applied to understand the relationships amongst these disparate data. Assembling all the data available (analytical, invasive, noninvasive and clinical) may enable researchers to develop predictive and prognostic models to improve patient care. The initial collaborative effort focuses on a cohort of 8,000 patients with lung cancer from the nation’s two largest healthcare systems (VA and DOD) but later will include other tumor types. To broaden the program’s reach, the U.S. has signed memoranda of understanding (MOUs) with eight countries since July 2016 to facilitate collaborations.

    Imaging is Key to Goals of Blue Ribbon Panel

    One BRP goal is to create a national ecosystem for sharing and analyzing cancer data, enabling researchers, clinicians and patients to contribute data that will facilitate efficient data analysis. Concrete steps in that direction have already shown progress. A critical cornerstone resource is the rapidly evolving NCI Genetic Data Commons (GDC) database. As that comprehensive database continues to evolve it will contain internet searchable large-scale genetic, proteomic and clinical outcome data from individual patients linkable to their clinical images in a protected health information (PHI)-compliant but publicly accessible way.

    The NCI’s publicly accessible TCIA has already been operational for the past half-decade. TCIA and similar resources needed by the cross-disciplinary scientific community will likely grow more invaluable to the imaging research community as GDC-TCIA database linkages develop. One of the most common concerns cited as a major impediment to clinical image sharing — PHI leaks in the DICOM headers — has largely been put to rest by TCIA’s time-proven, multi-stage curation process originally based on RSNA’s Clinical Trial Processor (CTP). The software was developed as part of the RSNA-sponsored open source Medical Image Resource Community (MIRC). The national ecosystem will focus on research and discovery. In the future, it is expected that the resulting tools, methods and multi-domain signatures will be integrated into clinically-useful imaging workflows and artificial intelligence systems that will meaningfully improve the overall standard of care in cancer.

    Developing Personalized Cancer Care

    The panel also calls for developing new cancer technologies to characterize tumors and test therapies. The imaging community has made progress by developing quantitatively reproducible radiomic and radiogenomic analytic processes championed by a host of cross-disciplinary radiologist-computer scientist teams. These advances are now making their way into accepted practice within both researcher and clinician imaging communities. The refinement of these technologies will accelerate the transition of tumor volumetrics to replace legacy techniques such as Response Evaluation Criteria in Solid Tumors (RECIST) with more sophisticated imaging signal processing so they can become an essential part of the practice of precision medicine. Instead of being used primarily in clinical imaging trials, the availability of these tools will mean that variations in patients’ responses will be registered and more accurately tracked through imaging as drivers of standard of care therapy. In addition to immediate improvements in care, these quantifiable results can be fed back into research as a resource to further refine such tools and methods.

    Another goal of the BRP is to establish time-sequential 3-D maps of patient-specific tumors that arise from their genetic abnormality and their response to molecularly selected precision therapies. This focus will open new understanding of why some histologically similar cancers respond while others are resistant. Active imaging and computer science teams are demonstrating the value of extracting non-humanly visible texture, shape and boundary properties too subtle to be recognized visually. The mapping of molecular pathways will scale upward to phenotypic properties, and imaging thus will help align these links to tumor phenotypic expression. The growth of cancers from precancerous lesions to advanced metastatic cancer is frequently a growth in volume, which is monitored and measured in large part by conventional clinical imaging.

    Finally, the BRP intends to establish a cancer immunotherapy network to discover why immunotherapy is so effective in some patients but not in others. This particular goal calls for some relevant background. In the 1990s the U.S. Food and Drug Administration (FDA) began to incorporate accelerated approval (AA) as an acceptable drug therapy pathway for approval based on promising tumor objective response revealed in Phase II clinical trials. Though there is an FDA expectation that the pharmaceutical company applicant would still conduct much larger Phase III trials, this flexibility encouraged the pharmaceutical industry as a way to significantly shorten time to market. Imaging’s role in this clinical trial AA framework is critical. Evidence for the importance of imaging as pivotal data in Phase II trials is shown by its now frequent use for defining a solid tumor’s response. This has become especially important for testing immunologically based therapies by the metric known as progression-free survival (PFS). One recent example of PFS’s critical role is evident in the FDA’s favorable review of pembrolizumab for non-small cell lung cancer. This raises the question of how the usual evidence for PFS is established — i.e., it is done by imaging where a non-invasive time sequential measurement of tumor volume is already an accepted methodology.

    The Moonshot has created extensive NIH funding opportunities for the imaging research community (see sidebar below). The Moonshot and APOLLO announcements, accompanied by new research funding opportunities integrated by cross-disciplinary researchers will inevitably advance the march toward precision medicine. Patient imaging, generated by clinical trial researchers and during routine standard of care, will become more widely recognized as essential to validating new therapeutic agents and strategies based on rapid advances in cellular molecular knowledge.

    Access Cancer Imaging Resources

    • For information on the Cancer Moonshot Initiative and APOLLO, including funding opportunities for the imaging research
    community, and the goals of the NIH Blue Ribbon Panel, go to www.cancer.gov.

    • For more information on Cancer Imaging Program FOAs and The Cancer Imaging Archive, go to www.cancerimagingarchive.net.

    • For information on the Genetic Data Commons, go to portal.gdc.cancer.gov.




    Jaffe
    C. Carl Jaffe, MD is a professor of radiology, Boston University School of Medicine, and Professor Emeritus of Medicine (Cardiology), Yale University School of Medicine. He is currently the consultant to the National Cancer Institute’s (NCI) Cancer Imaging Program and previously led The Cancer Imaging Archive (TCIA) initiative which de-identifies and hosts a large, publicly accessible archive of medical cancer images. Dr. Jaffe served as second vice president of RSNA from 1999-2000 and presented the New Horizons Lecture at RSNA 1998.

    1

    2
    The NCI-MATCH Clinical Trial seeks to determine whether treating cancers according to their molecular abnormalities will show evidence of effectiveness. Images courtesy of NCI/National Human Genome Research Institute




    To:
    From:
    Subject:
    Comment:
    Link:
      
  •  
    comments powered by Disqus