1
|
Jones AK. Known Unknowns: Gaps in Dose Distribution in Radioembolization, and in Our Understanding of Them. J Vasc Interv Radiol 2024; 35:1613-1615. [PMID: 39053847 DOI: 10.1016/j.jvir.2024.07.011] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/19/2024] [Revised: 07/14/2024] [Accepted: 07/16/2024] [Indexed: 07/27/2024] Open
Affiliation(s)
- A Kyle Jones
- Departments of Imaging Physics and Interventional Radiology, University of Texas MD Anderson Cancer Center, Houston, Texas.
| |
Collapse
|
2
|
Mansouri Z, Salimi Y, Hajianfar G, Wolf NB, Knappe L, Xhepa G, Gleyzolle A, Ricoeur A, Garibotto V, Mainta I, Zaidi H. The role of biomarkers and dosimetry parameters in overall and progression free survival prediction for patients treated with personalized 90Y glass microspheres SIRT: a preliminary machine learning study. Eur J Nucl Med Mol Imaging 2024; 51:4111-4126. [PMID: 38981950 DOI: 10.1007/s00259-024-06805-8] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/13/2024] [Accepted: 06/17/2024] [Indexed: 07/11/2024]
Abstract
BACKGROUND Overall Survival (OS) and Progression-Free Survival (PFS) analyses are crucial metrics for evaluating the efficacy and impact of treatment. This study evaluated the role of clinical biomarkers and dosimetry parameters on survival outcomes of patients undergoing 90Y selective internal radiation therapy (SIRT). MATERIALS/METHODS This preliminary and retrospective analysis included 17 patients with hepatocellular carcinoma (HCC) treated with 90Y SIRT. The patients underwent personalized treatment planning and voxel-wise dosimetry. After the procedure, the OS and PFS were evaluated. Three structures were delineated including tumoral liver (TL), normal perfused liver (NPL), and whole normal liver (WNL). 289 dose-volume constraints (DVCs) were extracted from dose-volume histograms of physical and biological effective dose (BED) maps calculated on 99mTc-MAA and 90Y SPECT/CT images. Subsequently, the DVCs and 16 clinical biomarkers were used as features for univariate and multivariate analysis. Cox proportional hazard ratio (HR) was employed for univariate analysis. HR and the concordance index (C-Index) were calculated for each feature. Using eight different strategies, a cross-combination of various models and feature selection (FS) methods was applied for multivariate analysis. The performance of each model was assessed using an averaged C-Index on a three-fold nested cross-validation framework. The Kaplan-Meier (KM) curve was employed for univariate and machine learning (ML) model performance assessment. RESULTS The median OS was 11 months [95% CI: 8.5, 13.09], whereas the PFS was seven months [95% CI: 5.6, 10.98]. Univariate analysis demonstrated the presence of Ascites (HR: 9.2[1.8,47]) and the aim of SIRT (segmentectomy, lobectomy, palliative) (HR: 0.066 [0.0057, 0.78]), Aspartate aminotransferase (AST) level (HR:0.1 [0.012-0.86]), and MAA-Dose-V205(%)-TL (HR:8.5[1,72]) as predictors for OS. 90Y-derived parameters were associated with PFS but not with OS. MAA-Dose-V205(%)-WNL, MAA-BED-V400(%)-WNL with (HR:13 [1.5-120]) and 90Y-Dose-mean-TL, 90Y-D50-TL-Gy, 90Y-Dose-V205(%)-TL, 90Y-Dose- D50-TL-Gy, and 90Y-BED-V400(%)-TL (HR:15 [1.8-120]) were highly associated with PFS among dosimetry parameters. The highest C-index observed in multivariate analysis using ML was 0.94 ± 0.13 obtained from Variable Hunting-variable-importance (VH.VIMP) FS and Cox Proportional Hazard model predicting OS, using clinical features. However, the combination of VH. VIMP FS method with a Generalized Linear Model Network model predicting OS using Therapy strategy features outperformed the other models in terms of both C-index and stratification of KM curves (C-Index: 0.93 ± 0.14 and log-rank p-value of 0.023 for KM curve stratification). CONCLUSION This preliminary study confirmed the role played by baseline clinical biomarkers and dosimetry parameters in predicting the treatment outcome, paving the way for the establishment of a dose-effect relationship. In addition, the feasibility of using ML along with these features was demonstrated as a helpful tool in the clinical management of patients, both prior to and following 90Y-SIRT.
Collapse
Affiliation(s)
- Zahra Mansouri
- Division of Nuclear Medicine and Molecular Imaging, Diagnostic Department, Geneva University Hospital, CH-1211, Geneva, Switzerland
| | - Yazdan Salimi
- Division of Nuclear Medicine and Molecular Imaging, Diagnostic Department, Geneva University Hospital, CH-1211, Geneva, Switzerland
| | - Ghasem Hajianfar
- Division of Nuclear Medicine and Molecular Imaging, Diagnostic Department, Geneva University Hospital, CH-1211, Geneva, Switzerland
| | - Nicola Bianchetto Wolf
- Division of Nuclear Medicine and Molecular Imaging, Diagnostic Department, Geneva University Hospital, CH-1211, Geneva, Switzerland
| | - Luisa Knappe
- Division of Nuclear Medicine and Molecular Imaging, Diagnostic Department, Geneva University Hospital, CH-1211, Geneva, Switzerland
| | - Genti Xhepa
- Service of Radiology, Geneva University Hospital, CH-1211, Geneva, Switzerland
| | - Adrien Gleyzolle
- Service of Radiology, Geneva University Hospital, CH-1211, Geneva, Switzerland
| | - Alexis Ricoeur
- Service of Radiology, Geneva University Hospital, CH-1211, Geneva, Switzerland
| | - Valentina Garibotto
- Division of Nuclear Medicine and Molecular Imaging, Diagnostic Department, Geneva University Hospital, CH-1211, Geneva, Switzerland
- Centre for Biomedical Imaging (CIBM), Geneva, Switzerland
| | - Ismini Mainta
- Division of Nuclear Medicine and Molecular Imaging, Diagnostic Department, Geneva University Hospital, CH-1211, Geneva, Switzerland
| | - Habib Zaidi
- Division of Nuclear Medicine and Molecular Imaging, Diagnostic Department, Geneva University Hospital, CH-1211, Geneva, Switzerland.
- Department of Nuclear Medicine and Molecular Imaging, University of Groningen, University Medical Center Groningen, Groningen, Netherlands.
- Department of Nuclear Medicine, University of Southern Denmark, Odense, Denmark.
- University Research and Innovation Center, Óbuda University, Budapest, Hungary.
| |
Collapse
|
3
|
Topcuoglu OM, Orhan T, Gormez A, Alan N. Are survival outcomes dependent on the tumour dose threshold of 139 Gy in patients with chemorefractory metastatic colorectal cancer treated with yttrium-90 radioembolization using glass particles? A real-world single-centre study. Br J Radiol 2024; 97:1255-1260. [PMID: 38730551 PMCID: PMC11186554 DOI: 10.1093/bjr/tqae096] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/17/2024] [Revised: 04/10/2024] [Accepted: 05/05/2024] [Indexed: 05/13/2024] Open
Abstract
OBJECTIVES To compare the survival and objective response rate (ORR) of the patients receiving estimated tumour absorbed dose (ETAD) <140 Gy versus ETAD ≥140 Gy in patients with advanced chemorefractory colorectal carcinoma liver metastases (CRCLM) treated with yttrium-90 transarterial radioembolization (90Y TARE). METHODS Between August 2016 and August 2023 adult patients with unresectable, chemorefractory CRCLM treated with 90Y TARE using glass particles, were retrospectively enrolled. Primary outcomes were overall survival (OS) and hepatic progression free survival (hPFS). Secondary outcome was ORR. RESULTS A total of 40 patients with a mean age of 66.2 ± 7.8 years met the inclusion criteria. Mean ETAD for group 1 (ETAD <140 Gy) and group 2 (ETAD ≥140) were 131.2 ± 17.4 Gy versus 195 ± 45.6 Gy, respectively. The mean OS and hPFS for group 1 versus group 2 were 12 ± 10.3 months and 8.1 ± 9.3 months versus 9.3 ± 3 months and 7.1 ± 8.4 months, respectively and there were no significant differences (P = .181 and P = .366, respectively). ORR did not show significant difference between the groups (P = .432). CONCLUSION In real-world practice, no significant difference was found in OS, hPFS, and ORR between patients who received ETAD <140 Gy versus ETAD ≥140 Gy in patients with CRCLM, in this series. ADVANCES IN KNOWLEDGE This study demonstrated that increased tumour absorbed doses in radioembolization may not provide additional significant advantage for OS and hPFS for patients with CRCLM.
Collapse
Affiliation(s)
| | - Tolga Orhan
- Department of Radiology, Yeditepe University Hospitals, Kosuyolu 34718, Turkey
| | - Ayşegul Gormez
- Department of Radiology, Yeditepe University Hospitals, Kosuyolu 34718, Turkey
| | - Nalan Alan
- Department of Nuclear Medicine, Yeditepe University Hospitals, Kosuyolu 34718, Turkey
| |
Collapse
|
4
|
Salimi Y, Mansouri Z, Hajianfar G, Sanaat A, Shiri I, Zaidi H. Fully automated explainable abdominal CT contrast media phase classification using organ segmentation and machine learning. Med Phys 2024; 51:4095-4104. [PMID: 38629779 DOI: 10.1002/mp.17076] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/28/2023] [Revised: 03/19/2024] [Accepted: 04/02/2024] [Indexed: 06/05/2024] Open
Abstract
BACKGROUND Contrast-enhanced computed tomography (CECT) provides much more information compared to non-enhanced CT images, especially for the differentiation of malignancies, such as liver carcinomas. Contrast media injection phase information is usually missing on public datasets and not standardized in the clinic even in the same region and language. This is a barrier to effective use of available CECT images in clinical research. PURPOSE The aim of this study is to detect contrast media injection phase from CT images by means of organ segmentation and machine learning algorithms. METHODS A total number of 2509 CT images split into four subsets of non-contrast (class #0), arterial (class #1), venous (class #2), and delayed (class #3) after contrast media injection were collected from two CT scanners. Seven organs including the liver, spleen, heart, kidneys, lungs, urinary bladder, and aorta along with body contour masks were generated by pre-trained deep learning algorithms. Subsequently, five first-order statistical features including average, standard deviation, 10, 50, and 90 percentiles extracted from the above-mentioned masks were fed to machine learning models after feature selection and reduction to classify the CT images in one of four above mentioned classes. A 10-fold data split strategy was followed. The performance of our methodology was evaluated in terms of classification accuracy metrics. RESULTS The best performance was achieved by Boruta feature selection and RF model with average area under the curve of more than 0.999 and accuracy of 0.9936 averaged over four classes and 10 folds. Boruta feature selection selected all predictor features. The lowest classification was observed for class #2 (0.9888), which is already an excellent result. In the 10-fold strategy, only 33 cases from 2509 cases (∼1.4%) were misclassified. The performance over all folds was consistent. CONCLUSIONS We developed a fast, accurate, reliable, and explainable methodology to classify contrast media phases which may be useful in data curation and annotation in big online datasets or local datasets with non-standard or no series description. Our model containing two steps of deep learning and machine learning may help to exploit available datasets more effectively.
Collapse
Affiliation(s)
- Yazdan Salimi
- Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, Geneva, Switzerland
| | - Zahra Mansouri
- Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, Geneva, Switzerland
| | - Ghasem Hajianfar
- Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, Geneva, Switzerland
| | - Amirhossein Sanaat
- Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, Geneva, Switzerland
| | - Isaac Shiri
- Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, Geneva, Switzerland
- Department of Cardiology, Inselspital, Bern University Hospital, University of Bern, Bern, Switzerland
| | - Habib Zaidi
- Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, Geneva, Switzerland
- Department of Nuclear Medicine and Molecular Imaging, University of Groningen, University Medical Center Groningen, Groningen, Netherlands
- Department of Nuclear Medicine, University of Southern Denmark, Odense, Denmark
- University Research and Innovation Center, Óbuda University, Budapest, Hungary
| |
Collapse
|
5
|
Noipinit N, Sukprakun C, Siricharoen P, Khamwan K. Comparison of absorbed doses to the tumoral and non-tumoral liver in HCC patients undergoing 99mTc-MAA and 90Y-microspheres radioembolization. Ann Nucl Med 2024; 38:210-218. [PMID: 38142421 DOI: 10.1007/s12149-023-01890-5] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/02/2023] [Accepted: 11/26/2023] [Indexed: 12/26/2023]
Abstract
PURPOSE This study aimed to determine the absorbed doses in the tumoral-liver and non-tumoral liver of hepatocellular carcinoma (HCC) patients undergoing radioembolization with Yttrium-90 (90Y) resin microspheres, and compared with those derived from 99mTc-MAA using the partition model. METHODS A total of 42 HCC patients (28 males and 14 females, mean age 65 ± 11.51 years) who received 45 treatment sessions with 90Y-microspheres between 2016 and 2021 were included. Pre-treatment 99mTc-MAA and post-treatment 90Y-bremsstrahlung SPECT/CT were acquired for each patient. Semi-automated segmentation of regions of interest (ROIs) was performed using MIM Encore software to determine the tumor-liver ratio (TLR) encompassing the liver volume, tumoral-liver, and lungs, and verified by both nuclear medicine physician and interventional radiologist. A partition dosimetry model was used to estimate the administered activity of 90Y-microspheres and the absorbed doses to the tumoral-liver and non-tumoral liver. The student's paired t test and Bland-Altman plot were used for the statistical analysis. RESULTS The mean TLR values obtained from 99mTc-MAA SPECT/CT and 90Y-bremsstrahlung SPECT/CT were 4.78 ± 3.51 and 2.73 ± 1.18, respectively. The mean planning administered activity of 90Y-microspheres based on 99mTc-MAA SPECT/CT was 1.56 ± 0.80 GBq, while the implanted administered activity was 2.53 ± 1.23 GBq (p value < 0.001). The mean absorbed doses in the tumoral-liver estimated from 99mTc-MAA and 90Y-bremsstrahlung SPECT/CT were 127.44 ± 4.36 Gy and 135.98 ± 6.30 Gy, respectively. The corresponding mean absorbed doses in the non-tumoral liver were 34.61 ± 13.93 Gy and 55.04 ± 16.36 Gy. CONCLUSION This study provides evidence that the administered activity of 90Y-microspheres, as estimated from 90Y-bremsstrahlung SPECT/CT, was significantly higher than that estimated from 99mTc-MAA SPECT/CT resulted in increased absorbed doses in both the tumoral-liver and non-tumoral liver. However, 99mTc-MAA SPECT/CT remains a valuable planning tool for predicting the distribution of 90Y-microspheres in liver cancer treatment.
Collapse
Affiliation(s)
- Nut Noipinit
- Medical Physics Program, Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand
- Division of Nuclear Medicine, Department of Radiology, King Chulalongkorn Memorial Hospital, The Thai Red Cross Society, Bangkok, Thailand
- Chulalongkorn University Biomedical Imaging Group, Department of Radiology, Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand
| | - Chanan Sukprakun
- Division of Nuclear Medicine, Department of Radiology, Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand
| | - Punnarai Siricharoen
- Department of Computer Engineering, Faculty of Engineering, Chulalongkorn University, Bangkok, Thailand
| | - Kitiwat Khamwan
- Medical Physics Program, Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand.
- Division of Nuclear Medicine, Department of Radiology, Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand.
- Chulalongkorn University Biomedical Imaging Group, Department of Radiology, Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand.
| |
Collapse
|
6
|
Marquis H, Ocampo Ramos JC, Carter LM, Zanzonico P, Bolch WE, Laforest R, Kesner AL. MIRD Pamphlet No. 29: MIRDy90-A 90Y Research Microsphere Dosimetry Tool. J Nucl Med 2024; 65:jnumed.123.266743. [PMID: 38388514 PMCID: PMC11064830 DOI: 10.2967/jnumed.123.266743] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/03/2023] [Accepted: 01/23/2024] [Indexed: 02/24/2024] Open
Abstract
90Y-microsphere radioembolization has become a well-established treatment option for liver malignancies and is one of the first U.S. Food and Drug Administration-approved unsealed radionuclide brachytherapy devices to incorporate dosimetry-based treatment planning. Several different mathematical models are used to calculate the patient-specific prescribed activity of 90Y, namely, body surface area (SIR-Spheres only), MIRD single compartment, and MIRD dual compartment (partition). Under the auspices of the MIRDsoft initiative to develop community dosimetry software and tools, the body surface area, MIRD single-compartment, MIRD dual-compartment, and MIRD multicompartment models have been integrated into a MIRDy90 software worksheet. The worksheet was built in MS Excel to estimate and compare prescribed activities calculated via these respective models. The MIRDy90 software was validated against available tools for calculating 90Y prescribed activity. The results of MIRDy90 calculations were compared with those obtained from vendor and community-developed tools, and the calculations agreed well. The MIRDy90 worksheet was developed to provide a vetted tool to better evaluate patient-specific prescribed activities calculated via different models, as well as model influences with respect to varying input parameters. MIRDy90 allows users to interact and visualize the results of various parameter combinations. Variables, equations, and calculations are described in the MIRDy90 documentation and articulated in the MIRDy90 worksheet. The worksheet is distributed as a free tool to build expertise within the medical physics community and create a vetted standard for model and variable management.
Collapse
Affiliation(s)
- Harry Marquis
- Department of Medical Physics, Memorial Sloan Kettering Cancer Center, New York, New York
| | - Juan C Ocampo Ramos
- Department of Medical Physics, Memorial Sloan Kettering Cancer Center, New York, New York
| | - Lukas M Carter
- Department of Medical Physics, Memorial Sloan Kettering Cancer Center, New York, New York
| | - Pat Zanzonico
- Department of Medical Physics, Memorial Sloan Kettering Cancer Center, New York, New York
| | - Wesley E Bolch
- J. Crayton Pruitt Family Department of Biomedical Engineering, University of Florida, Gainesville, Florida; and
| | - Richard Laforest
- Mallinckrodt Institute of Radiology, Washington University School of Medicine, St. Louis, Missouri
| | - Adam L Kesner
- Department of Medical Physics, Memorial Sloan Kettering Cancer Center, New York, New York;
| |
Collapse
|
7
|
Wendler T, Kreissl MC, Schemmer B, Rogasch JMM, De Benetti F. Artificial Intelligence-powered automatic volume calculation in medical images - available tools, performance and challenges for nuclear medicine. Nuklearmedizin 2023; 62:343-353. [PMID: 37995707 PMCID: PMC10667065 DOI: 10.1055/a-2200-2145] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/16/2023] [Accepted: 10/26/2023] [Indexed: 11/25/2023]
Abstract
Volumetry is crucial in oncology and endocrinology, for diagnosis, treatment planning, and evaluating response to therapy for several diseases. The integration of Artificial Intelligence (AI) and Deep Learning (DL) has significantly accelerated the automatization of volumetric calculations, enhancing accuracy and reducing variability and labor. In this review, we show that a high correlation has been observed between Machine Learning (ML) methods and expert assessments in tumor volumetry; Yet, it is recognized as more challenging than organ volumetry. Liver volumetry has shown progression in accuracy with a decrease in error. If a relative error below 10 % is acceptable, ML-based liver volumetry can be considered reliable for standardized imaging protocols if used in patients without major anomalies. Similarly, ML-supported automatic kidney volumetry has also shown consistency and reliability in volumetric calculations. In contrast, AI-supported thyroid volumetry has not been extensively developed, despite initial works in 3D ultrasound showing promising results in terms of accuracy and reproducibility. Despite the advancements presented in the reviewed literature, the lack of standardization limits the generalizability of ML methods across diverse scenarios. The domain gap, i. e., the difference in probability distribution of training and inference data, is of paramount importance before clinical deployment of AI, to maintain accuracy and reliability in patient care. The increasing availability of improved segmentation tools is expected to further incorporate AI methods into routine workflows where volumetry will play a more prominent role in radionuclide therapy planning and quantitative follow-up of disease evolution.
Collapse
Affiliation(s)
- Thomas Wendler
- Clinical Computational Medical Imaging Research, Department of Diagnostic and Interventional Radiology and Neuroradiology, University Hospital Augsburg, Germany
- Institute of Digital Medicine, Universitätsklinikum Augsburg, Germany
- Computer-Aided Medical Procedures and Augmented Reality School of Computation, Information and Technology, Technical University of Munich, Munich, Germany
| | | | | | - Julian Manuel Michael Rogasch
- Department of Nuclear Medicine, Charité – Universitätsmedizin Berlin, corporate member of Freie Universität Berlin and Humboldt-Universität zu Berlin,Germany
| | - Francesca De Benetti
- Computer-Aided Medical Procedures and Augmented Reality School of Computation, Information and Technology, Technical University of Munich, Munich, Germany
| |
Collapse
|