K Bernstein¹*, W Shi²*, D Schlesinger³*, H Lin⁴*, X Gu⁵*, T Li⁶*, (1) NYU Langone Medical Center, New York, NY, (2) Thomas Jefferson University, Philadelphia, PA, (3) University of Virginia Health Systems, Charlottesville, VA, (4) University of California San Francisco, San Francisco, CA, (5) Stanford University, Dallas, TX, (6) University of Pennsylvania, Philadelphia, PA

• 
  K Bernstein
• 
  W Shi
• 
  D Schlesinger
• 
  H Lin
• 
  X Gu
• 
  T Li

• 
  
• 
  
• 
  
• 
  
• 
  
• 
  
• 
  

TU-GH-BRC-0 (Tuesday, 7/12/2022) 1:45 PM - 3:45 PM [Eastern Time (GMT-4)]

Ballroom C

With the improvement of management and survival of patients with advanced stage cancer, radiation therapy teams in today’s clinics are seeing an increasing number of patients receiving stereotactic radiosurgery (SRS) to treat multiple and small brain metastasis (BM). Many of these patients undergo multiple and/or repeated SRS to longitudinally manage new and/or recurrent BM for years. This prolonged course of disease presents substantial challenges for the treatment team because:
1) The onset of BM is predominantly small and multiple, thus manual delineation is challenging and labor-intensive;
2) Tracking multiple lesions across multiple MRIs over years is difficult, particularly in differentiating recurrence from radiation necrosis, which could result in unintentional misses or re-treatments for a particular lesion;
3) The quality of manual multiple and/or repeated SRS planning is highly variable among planners and delivery platforms;
4) Clinical data is often segregated in different databases, and is difficult to mine;
5) Outcome and toxicity are not well characterized for this cohort of patients

Recently, new tools and technologies have been developed to address the above challenges in BM management using SRS:
1) Automation and AI-assisted tools have been developed to assist with BM identification and contouring
2) Integration of hospital EMRs and other department-specific standalone databases creates more robust datastores of clinical and SRS-specific data
3) Prospective registries for structured data have been developed for cross-institution data integration.
These new developments provide great opportunity for improved data-mining to be conducted, particularly for outcomes analytics for both single- and multi-institutional studies. As more and more patients with brain metastasis are managed in both academic clinics and community hospitals, it is important to understand how these tools work, and how to best use them to achieve efficient and effective management of multiple and/or repeated BMs.

In this session we invite speakers with diverse expertise and background (physicians, clinical physicists, translational researchers) to discuss their unique perspectives on this topic, and to provide the community with an overview of current applications and future directions of automation and AI in longitudinal management of BMs using SRS.

Learning objectives:
1. Understand the clinical landscape of radiation therapy with regard to longitudinal management of brain metastasis: what are physicians concerned about?
2. Learn about the development, design, functionality, and organization of a national, prospective stereotactic radiosurgery registry.
3. Understand techniques and issues regarding the integration of the hospital EMR with other information systems and techniques for clinical record data mining for cross-departmental brain metastasis management.
4. Understand AI-enabled tools that can automated BMs delineation and dosimetry tracking.
5. Detail particular concerns in treatment planning for multi-course brain metastasis treatment, as well as important uncertainty parameters.

Funding Support, Disclosures, and Conflict of Interest: TL received research grant, travel expenses, and honoraria from Varian Medical Systems unrelated to this work. DS reports that some content for his presentation was provided by Brainlab and Neuropoint Alliance. He has no financial conflicts of interest. XG reports research supported by NIH R01 CA235723 and NIH SBIR 75N91021C00031.

Handouts

• 175-63512-16291659-183116-1591726624.pdf (Kenneth Bernstein)
• 175-63515-16291659-183569-735621344.pdf (Hui Lin)
• 175-63514-16291659-183863-773833228.pdf (David Schlesinger)

Keywords

Stereotactic Radiosurgery, Segmentation, Data Interpolation

Taxonomy

TH- External Beam- Photons: intracranial stereotactic/SBRT

Contact Email