Decision-support for treatment with 177Lu-PSMA: machine learning predicts response with high accuracy based on PSMA-PET/CT and clinical parameters

Sobhan Moazemi; Annette Erle; Zain Khurshid; Susanne Lütje; Michael Muders; Markus Essler; Thomas Schultz; Ralph A Bundschuh

doi:10.21037/atm-20-6446

Decision-support for treatment with ¹⁷⁷Lu-PSMA: machine learning predicts response with high accuracy based on PSMA-PET/CT and clinical parameters

Ann Transl Med. 2021 May;9(9):818. doi: 10.21037/atm-20-6446.

Authors

Sobhan Moazemi^{1

2}, Annette Erle¹, Zain Khurshid³, Susanne Lütje¹, Michael Muders⁴, Markus Essler¹, Thomas Schultz^{2

5}, Ralph A Bundschuh¹

Affiliations

¹ Department of Nuclear Medicine, University Hospital Bonn, Bonn, Germany.
² Department of Computer Science, University of Bonn, Bonn, Germany.
³ Nuclear Medicine, Oncology and Radiotherapy Institute, Department of Nuclear Medicine, Islamabad, Pakistan.
⁴ Department of Pathology, University Hospital Bonn, Bonn, Germany.
⁵ Bonn-Aachen International Center for Information Technology (B-IT), University of Bonn, Bonn, Germany.

Abstract

Background: Treatment with radiolabeled ligands to prostate-specific membrane antigen (PSMA) is gaining importance in the treatment of patients with advanced prostate carcinoma. Previous imaging with positron emission tomography/computed tomography (PET/CT) is mandatory. The aim of this study was to investigate the role of radiomics features in PSMA-PET/CT scans and clinical parameters to predict response to ¹⁷⁷Lu-PSMA treatment given just baseline PSMA scans using state-of-the-art machine learning (ML) methods.

Methods: A total of 2,070 pathological hotspots annotated in 83 prostate cancer patients undergoing PSMA therapy were analyzed. Two main tasks are performed: (I) analyzing correlation of averaged (per patient) values of radiomics features of individual hotspots and clinical parameters with difference in prostate specific antigen levels (ΔPSA) in pre- and post-therapy as a therapy response indicator. (II) ML-based classification of patients into responders and non-responders based on averaged features values and clinical parameters. To achieve this, machine learning (ML) algorithms and linear regression tests are applied. Grid search, cross validation (CV) and permutation test were performed to assure that the results were significant.

Results: Radiomics features (PET_Min, PET_Correlation, CT_Min, CT_Busyness and CT_Coarseness) and clinical parameters such as Alp1 and Gleason score showed best correlations with ΔPSA. For the treatment response prediction task, 80% area under the curve (AUC), 75% sensitivity (SE), and 75% specificity (SP) were obtained, applying ML support vector machine (SVM) classifier with radial basis function (RBF) kernel on a selection of radiomics features and clinical parameters with strong correlations with ΔPSA.

Conclusions: Machine learning based on ⁶⁸Ga-PSMA PET/CT radiomics features holds promise for the prediction of response to ¹⁷⁷Lu-PSMA treatment, given only base-line ⁶⁸Ga-PSMA scan. In addition, it was shown that, the best correlating set of radiomics features with ΔPSA are superior to clinical parameters for this therapy response prediction task using ML classifiers.

Keywords: Prostate cancer (PC); computed tomography (CT); machine learning (ML); positron emission tomography (PET); prostate specific membrane antigen (PSMA).