A multi-institutional team of researchers led by Dr. Laurent Dercle, PhD, of Columbia University Medical Center has developed a machine learning algorithm and radiomics “signature” to estimate overall survival from CT images in patients receiving immunotherapy in two multicenter immunotherapy clinical trials. In testing, the radiomic signature surpassed the Response Evaluation Criteria in Solid Tumors (RECIST) 1.1 guideline for predicting overall survival at six months.
“The results of this prognostic study suggest that the radiomic signature discerned from conventional CT images at baseline and at first follow-up can be used in the clinical setting to provide an accurate early readout of future [overall survival] likelihood in melanoma patients treated with a single programmed cell death blocking agent 1,” the authors wrote.
Although existing criteria rely almost exclusively on tumor size to estimate therapeutic benefit in cancer patients, this approach was not designed to estimate survival benefit and is also challenged by the unique properties immunotherapy, according to the researchers.
“With the primary aim of improving clinical management, we tested the hypothesis that a radiomic signature derived from the size, density and shape of the tumor and its evolution as treatment is administered can estimate [overall survival] in patients with melanoma,” the authors wrote.
They used data provided by the public-private partnership Vol-PACT (Advanced Metrics and Modeling with Volumetric CT for Precision Analysis of Clinical Trial Results) to retrospectively analyze CT scans and clinical data from two landmark randomized clinical trials of pembrolizumab. The analysis included 575 adults with unresectable stage III/IV melanoma.
After evaluating 25 imaging features extracted from segmented tumors on CT images, a random forest machine learning model identified a radiomic signature that best estimates overall survival. This signature included two imaging features related to tumor size and two that reflect changes in tumor imaging phenotype.
The researchers then evaluated the final radiomics machine learning model to assess six-month survival on a CT scan performed three months after treatment on a validation test set of 287 patients treated with pembrolizumab.
|Ability of the CT radiomics signature to predict the survival of patients with melanoma|
|RECIST 1.1||Radiomics machine learning model|
|Area under the curve||0.80||0.92|
“This ability to discriminate between patients with active disease at month 3 could allow physicians to discuss alternative treatments earlier, set appropriate goals of care, or seek clinical trials,” the authors wrote.
Because it only requires the identification and segmentation of target lesions on routine CT scans, the radiomics and machine learning prognosis model could be widely translated into clinical practice using software. publicly available and a clinical decision support software tool, according to the researchers.
They noted, however, that clinical application would currently be limited by the need for manual lesion segmentation, which requires approximately one minute per lesion per scan. However, deep learning methods for automatic lesion detection and segmentation are being developed, which would allow this task to be performed automatically after CT scans are acquired.
In one accompanying commentary, Dr. Michael Farwell and Dr. David Mankoff, PhD, of the University of Pennsylvania said the trial represents an exciting step in applying radiomics to routine CT scans from multicenter clinical trials.
“While there are some hurdles to overcome before this approach becomes part of routine clinical practice, it is only a matter of time before these tools are developed,” Farwell and Mankoff wrote. “As a first step, tools for automatic lesion segmentation and volume change would be a welcome addition to clinical radiology practice. Once these tools are established, it will be quite simple to add radiomics to the analysis. with better prediction of survival for cancer patients who are treated with immune checkpoint blockade.”
Copyright © 2022 AuntMinnie.com