Home
Perspectives
- Assessing Tumor Burden, Automatically
Assessing Tumor Burden, Automatically
Elba Etchebehere, MD, and Mariana Camacho collaborate and review patient data.
With the help of syngo.via and its Multi-foci Segmentation tool, Elba Etchebehere, MD, PhD, is able to accurately assess tumor burden by identifying individual lesions, which increases the prognostic power for her patients in Campinas, Brazil.
Photos: Pisco Del Gaiso
Download your print version here.
Etchebehere is an assistant professor at UNICAMP (Campinas State University), one of the most important research universities in Brazil. She also directs a major nuclear medicine private practice. Yet, even with all of her clinical experience, she says Siemens Healthineers’ syngo.via and its Multi-foci Segmentation (MFS) tool make a huge difference when she’s assessing tumor burden in a patient. “I would go insane if I had to do this manually in a patient with multiple lesions”, says Etchebehere. “It would take me all day, and unfortunately I don’t have an entire hour to spend with each prostate cancer patient—with so many exams to take into account, it would not be feasible.”
Prognostic power
The MFS tool allows doctors to calculate tumor burden in a semi-automatic fashion by using predefined parameters to identify areas with high tracer uptake and draw volumes of interest with great precision. Another major advantage of MFS is that it is able to provide estimates of tumor burden that consider only the naturally irregular contours of each lesion. According to Etchebehere, “when I try to calculate tumor burden manually, especially the skeletal tumor burden, I end up defining spherical volumes of interest around the lesion. However, the actual volume of interest is not spherical, of course, and that could lead me to overestimate a patient’s tumor load and design an inadequate response to his or her disease.”
There is a reliable body of research showing that an accurate assessment of tumor load has great prognostic power.
Elba Etchebehere, MD, PhD
That is the main clinical importance of the precise data the MFS tool provides. “There is a reliable body of research showing that an accurate assessment of tumor load has great prognostic power,” Etchebehere explains. “If you measure tumor burden efficiently, that gives you a wealth of information about what you can expect for that patient in the future, in terms of likelihood of survival. It’s an objective tool and that’s crucial when you, or the referring physician, need to discuss and make decisions with the patient about the next steps in his or her treatment: whether to adopt a more aggressive and costly therapy or even to decide whether to stop the treatment altogether.”
Enhanced workflow
MFS makes a huge difference, in terms of workflow, for assessing patients with widespread tumors, Etchebehere says. “Without it, I simply wouldn’t know what to do with patients who have 10 or more lesions. In those cases, it saves up to 80% or 90% of my time. With MFS, you have the data you need like this,” she says, snapping her fingers. syngo.via’s proprietary ALPHA technology employs automatic organ recognition to align data from previous studies. Beyond this alignment, syngo.via’s Cross-Timepoint Evaluation aides in the visualization of how a patient’s case progresses through time, which enhances Etchebehere‘s clinical work. “It’s extremely important, even essential, to have this visual timeline in front of you when you’re treating a cancer patient,” she emphasizes. “With this visual timeline, you have a pretty clear idea of what has changed between the initial diagnosis and the different rounds of chemotherapy and radiotherapy, such as when something is not working, when you have to change course, and how much tumor load has changed.”
As good as gold—or better
Etchebehere and her colleagues in Campinas are beginning to publish the results of their work with syngo.via and its MFS tool in medical journals. To date, one article has been published while another article has been accepted for publication.
The difference was just phenomenal, both in terms of the time we saved with the semi-automatic approach and in the reliability of the prognosis.
Elba Etchebehere, MD, PhD
“Although manual measurements are still considered the gold standard, we compared manual quantification of tumor load with the use of the MFS tool in about 150 breast cancer patients. The difference was just phenomenal, both in terms of the time we saved with the semi-automatic approach and in the reliability of the prognosis,” she states. Etchebehere presented some of these results during the 2017 meeting of the Society of Nuclear Medicine and Molecular Imaging (SNMMI) in Denver, Colorado, USA. Other papers analyzing patients with lung and prostate tumors are also in the works.
“When you’re collaborating with other doctors, the best thing about the MFS tool is its potential for reproducibility,” Etchebehere argues. “If you have several centers working on the same research project, it’s easy to transfer your data to your colleague’s workstation, no matter where in the world he or she is working. With robust and reproducible results, you have the chance to use a ‘Big Data’ approach in cancer research, which has huge potential benefits for patients.”
Etchebehere says she regularly uses the Deauville criteria, which is a five-point scoring system for lymphomas based on the tumor’s relative avidity for glucose-analog uptake as compared with the uptake at the mediastinum and the liver. Scores higher than 3 (meaning a higher avidity for glucose-analog uptake in the tumor compared to the liver) are usually a sign that therapeutic strategies may not be working and should be reassessed. In an upcoming release, syngo.via will incorporate automation of the Deauville score. “I am intrigued to learn if Siemens Healthineers will improve the qualitative assessment, especially in equivocal cases, which can ultimately alter patient management. Let’s see what they can come up with,” she quips.
Share this page
- The statements by Siemens Healthineers customers described herein are based on results that were achieved in the customer’s unique setting. Since there is no “typical” hospital and many variables exist (e.g., hospital size, case mix, level of IT adoption) there can be no guarantee that other customers will achieve the same results.