1
|
Dhaliwal A, Ma J, Zheng M, Lyu Q, Rajora MA, Ma S, Oliva L, Ku A, Valic M, Wang B, Zheng G. Deep learning for automatic organ and tumor segmentation in nanomedicine pharmacokinetics. Theranostics 2024; 14:973-987. [PMID: 38250039 PMCID: PMC10797295 DOI: 10.7150/thno.90246] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/17/2023] [Accepted: 11/17/2023] [Indexed: 01/23/2024] Open
Abstract
Rationale: Multimodal imaging provides important pharmacokinetic and dosimetry information during nanomedicine development and optimization. However, accurate quantitation is time-consuming, resource intensive, and requires anatomical expertise. Methods: We present NanoMASK: a 3D U-Net adapted deep learning tool capable of rapid, automatic organ segmentation of multimodal imaging data that can output key clinical dosimetry metrics without manual intervention. This model was trained on 355 manually-contoured PET/CT data volumes of mice injected with a variety of nanomaterials and imaged over 48 hours. Results: NanoMASK produced 3-dimensional contours of the heart, lungs, liver, spleen, kidneys, and tumor with high volumetric accuracy (pan-organ average %DSC of 92.5). Pharmacokinetic metrics including %ID/cc, %ID, and SUVmax achieved correlation coefficients exceeding R = 0.987 and relative mean errors below 0.2%. NanoMASK was applied to novel datasets of lipid nanoparticles and antibody-drug conjugates with a minimal drop in accuracy, illustrating its generalizability to different classes of nanomedicines. Furthermore, 20 additional auto-segmentation models were developed using training data subsets based on image modality, experimental imaging timepoint, and tumor status. These were used to explore the fundamental biases and dependencies of auto-segmentation models built on a 3D U-Net architecture, revealing significant differential impacts on organ segmentation accuracy. Conclusions: NanoMASK is an easy-to-use, adaptable tool for improving accuracy and throughput in imaging-based pharmacokinetic studies of nanomedicine. It has been made publicly available to all readers for automatic segmentation and pharmacokinetic analysis across a diverse array of nanoparticles, expediting agent development.
Collapse
Affiliation(s)
- Alex Dhaliwal
- Princess Margaret Cancer Centre, University Health Network, 101 College Street, Toronto, M5G 1L7, Ontario, Canada
- Department of Medical Biophysics, University of Toronto, 101 College Street, Toronto, M5G 1L7, Ontario, Canada
| | - Jun Ma
- Department of Laboratory Medicine and Pathobiology, University of Toronto, 1 King's College Circle, Toronto, M5S 1A8, Ontario, Canada
- Peter Munk Cardiac Centre, University Health Network, 190 Elizabeth St, Toronto, M5G 2C4, Ontario, Canada
- Vector Institute for Artificial Intelligence, 661 University Avenue, Toronto, M4G 1M1, Ontario, Canada
| | - Mark Zheng
- Princess Margaret Cancer Centre, University Health Network, 101 College Street, Toronto, M5G 1L7, Ontario, Canada
| | - Qing Lyu
- Department of Computer Science, University of Toronto, 101 College Street, Toronto, M5G 1L7, Ontario, Canada
| | - Maneesha A. Rajora
- Princess Margaret Cancer Centre, University Health Network, 101 College Street, Toronto, M5G 1L7, Ontario, Canada
- Institute of Biomedical Engineering, University of Toronto, 101 College Street, Toronto, M5G 1L7, Ontario, Canada
| | - Shihao Ma
- Department of Computer Science, University of Toronto, 101 College Street, Toronto, M5G 1L7, Ontario, Canada
- Vector Institute for Artificial Intelligence, 661 University Avenue, Toronto, M4G 1M1, Ontario, Canada
| | - Laura Oliva
- Techna Institute, University Health Network, 190 Elizabeth Street, Toronto, M5G 2C4, Ontario, Canada
| | - Anthony Ku
- Department of Radiology, Stanford University, 1201 Welch Road, Stanford, 94305-5484, California, United States of America
| | - Michael Valic
- Princess Margaret Cancer Centre, University Health Network, 101 College Street, Toronto, M5G 1L7, Ontario, Canada
- Institute of Biomedical Engineering, University of Toronto, 101 College Street, Toronto, M5G 1L7, Ontario, Canada
| | - Bo Wang
- Department of Laboratory Medicine and Pathobiology, University of Toronto, 1 King's College Circle, Toronto, M5S 1A8, Ontario, Canada
- Peter Munk Cardiac Centre, University Health Network, 190 Elizabeth St, Toronto, M5G 2C4, Ontario, Canada
- Department of Computer Science, University of Toronto, 101 College Street, Toronto, M5G 1L7, Ontario, Canada
- Vector Institute for Artificial Intelligence, 661 University Avenue, Toronto, M4G 1M1, Ontario, Canada
| | - Gang Zheng
- Princess Margaret Cancer Centre, University Health Network, 101 College Street, Toronto, M5G 1L7, Ontario, Canada
- Department of Medical Biophysics, University of Toronto, 101 College Street, Toronto, M5G 1L7, Ontario, Canada
- Peter Munk Cardiac Centre, University Health Network, 190 Elizabeth St, Toronto, M5G 2C4, Ontario, Canada
| |
Collapse
|
2
|
Facca VJ, Cai Z, Ku A, Georgiou CJ, Reilly RM. Adjuvant Auger Electron-Emitting Radioimmunotherapy with [ 111In]In-DOTA-Panitumumab in a Mouse Model of Local Recurrence and Metastatic Progression of Human Triple-Negative Breast Cancer. Mol Pharm 2023; 20:6407-6419. [PMID: 37983089 DOI: 10.1021/acs.molpharmaceut.3c00780] [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] [Indexed: 11/21/2023]
Abstract
Triple-negative breast cancer (TNBC) has a high risk for recurrence and metastasis. We studied the effectiveness of Auger electron (AE) radioimmunotherapy (RIT) with antiepidermal growth factor receptor (EGFR) panitumumab conjugated with DOTA complexed to 111In ([111In]In-DOTA-panitumumab) for preventing metastatic progression after local treatment of 231/LM2-4 Luc+ human TNBC tumors in the mammary fat pad of NRG mice. Prior to RIT, the primary tumor was resected, and tumor margins were treated with X-irradiation (XRT; 5 days × 6 Gy/d). RIT was administered 1 day post-XRT by intravenous injection of 26 MBq (15 μg) or 2 × 10 MBq (15 μg each) separated by 7 d. These treatments were compared to tumor resection with or without XRT combined with DOTA-panitumumab (15 μg) or irrelevant [111In]In-DOTA-IgG2 (24 MBq; 15 μg), and efficacy was evaluated by Kaplan-Meier survival curves. The effect of [111In]In-DOTA-panitumumab (23 MBq; 15 μg) after tumor resection without local XRT was also studied. Tumor resection followed by XRT and RIT with 26 MBq [111In]In-DOTA-panitumumab significantly increased the median survival to 35 d compared to tumor resection with or without XRT (23-24 d; P < 0.0001). Local treatment with tumor resection and XRT followed by 2 × 10 MBq of [111In]In-DOTA-panitumumab, DOTA-panitumumab, or [111In]In-DOTA-IgG2 did not significantly improve median survival (26 days for all treatments). RIT alone with [111In]In-DOTA-panitumumab postresection of the tumor without XRT increased median survival to 29 days, though this was not significant. Despite significantly improved survival in mice treated with tumor resection, XRT, and RIT with [111In]In-DOTA-panitumumab, all mice eventually succumbed to advanced metastatic disease by 45 d post-tumor resection. SPECT/CT with [111In]In-DOTA-panitumumab, PET/MRI with [64Cu]Cu-DOTA-panitumumab F(ab')2, and PET/CT with [18F]FDG were used to detect recurrent and metastatic disease. Uptake of [111In]In-DOTA-panitumumab at 4 d p.i. in the MFP tumor was 26.8 ± 9.7% ID/g and in metastatic lymph nodes (LN), lungs, and liver was 34.2 ± 26.9% ID/g, 17.5 ± 6.0% ID/g, and 9.4 ± 2.4%ID/g, respectively, while uptake in the lungs (6.0 ± 0.9% ID/g) and liver (5.2 ± 2.9% ID/g) of non-tumor-bearing NRG was significantly lower (P < 0.05). Radiation-absorbed doses in metastatic LN, lungs, and liver were 9.7 ± 6.1, 6.4 ± 2.1, and 10.9 ± 2.7 Gy, respectively. In conclusion, we demonstrated that RIT with [111In]In-DOTA-panitumumab combined with tumor resection and XRT significantly improved the survival of mice with recurrent TNBC. However, the aggressive nature of 231/LM2-4 Luc+ tumors in NRG mice may have contributed to the tumor recurrence and progression observed.
Collapse
Affiliation(s)
- Valerie J Facca
- Department of Pharmaceutical Sciences, Leslie Dan Faculty of Pharmacy, University of Toronto, 144 College St., Toronto, Ontario M5S 3M2, Canada
| | - Zhongli Cai
- Department of Pharmaceutical Sciences, Leslie Dan Faculty of Pharmacy, University of Toronto, 144 College St., Toronto, Ontario M5S 3M2, Canada
| | - Anthony Ku
- Department of Pharmaceutical Sciences, Leslie Dan Faculty of Pharmacy, University of Toronto, 144 College St., Toronto, Ontario M5S 3M2, Canada
| | - Constantine J Georgiou
- Department of Pharmaceutical Sciences, Leslie Dan Faculty of Pharmacy, University of Toronto, 144 College St., Toronto, Ontario M5S 3M2, Canada
| | - Raymond M Reilly
- Department of Pharmaceutical Sciences, Leslie Dan Faculty of Pharmacy, University of Toronto, 144 College St., Toronto, Ontario M5S 3M2, Canada
- Department of Medical Imaging, Temerty Faculty of Medicine, University of Toronto, Toronto, Ontario M5S 1A8, Canada
- Joint Department of Medical Imaging and Princess Margaret Cancer Centre, University Health Network, Toronto, Ontario M5G 2C1, Canada
| |
Collapse
|
4
|
Li D, Li X, Zhao J, Tan F. Advances in nuclear medicine-based molecular imaging in head and neck squamous cell carcinoma. J Transl Med 2022; 20:358. [PMID: 35962347 PMCID: PMC9373390 DOI: 10.1186/s12967-022-03559-5] [Citation(s) in RCA: 7] [Impact Index Per Article: 3.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/13/2022] [Accepted: 07/25/2022] [Indexed: 12/24/2022] Open
Abstract
Head and neck squamous cell carcinomas (HNSCCs) are often aggressive, making advanced disease very difficult to treat using contemporary modalities, such as surgery, radiation therapy, and chemotherapy. However, targeted therapy, e.g., cetuximab, an epidermal growth factor receptor inhibitor, has demonstrated survival benefit in HNSCC patients with locoregional failure or distant metastasis. Molecular imaging aims at various biomarkers used in targeted therapy, and nuclear medicine-based molecular imaging is a real-time and non-invasive modality with the potential to identify tumor in an earlier and more treatable stage, before anatomic-based imaging reveals diseases. The objective of this comprehensive review is to summarize recent advances in nuclear medicine-based molecular imaging for HNSCC focusing on several commonly radiolabeled biomarkers. The preclinical and clinical applications of these candidate imaging strategies are divided into three categories: those targeting tumor cells, tumor microenvironment, and tumor angiogenesis. This review endeavors to expand the knowledge of molecular biology of HNSCC and help realizing diagnostic potential of molecular imaging in clinical nuclear medicine.
Collapse
Affiliation(s)
- Danni Li
- Shanghai Fourth People's Hospital, and School of Medicine, Tongji University, Shanghai, China
| | - Xuran Li
- Shanghai Fourth People's Hospital, and School of Medicine, Tongji University, Shanghai, China
| | - Jun Zhao
- Department of Nuclear Medicine, Shanghai East Hospital, School of Medicine, Tongji University, Shanghai, China
| | - Fei Tan
- Shanghai Fourth People's Hospital, and School of Medicine, Tongji University, Shanghai, China. .,The Royal College of Surgeons in Ireland, Dublin, Ireland. .,The Royal College of Surgeons of England, London, UK.
| |
Collapse
|
5
|
Manafi-Farid R, Ataeinia B, Ranjbar S, Jamshidi Araghi Z, Moradi MM, Pirich C, Beheshti M. ImmunoPET: Antibody-Based PET Imaging in Solid Tumors. Front Med (Lausanne) 2022; 9:916693. [PMID: 35836956 PMCID: PMC9273828 DOI: 10.3389/fmed.2022.916693] [Citation(s) in RCA: 13] [Impact Index Per Article: 6.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/09/2022] [Accepted: 05/24/2022] [Indexed: 12/13/2022] Open
Abstract
Immuno-positron emission tomography (immunoPET) is a molecular imaging modality combining the high sensitivity of PET with the specific targeting ability of monoclonal antibodies. Various radioimmunotracers have been successfully developed to target a broad spectrum of molecules expressed by malignant cells or tumor microenvironments. Only a few are translated into clinical studies and barely into clinical practices. Some drawbacks include slow radioimmunotracer kinetics, high physiologic uptake in lymphoid organs, and heterogeneous activity in tumoral lesions. Measures are taken to overcome the disadvantages, and new tracers are being developed. In this review, we aim to mention the fundamental components of immunoPET imaging, explore the groundbreaking success achieved using this new technique, and review different radioimmunotracers employed in various solid tumors to elaborate on this relatively new imaging modality.
Collapse
Affiliation(s)
- Reyhaneh Manafi-Farid
- Research Center for Nuclear Medicine, Shariati Hospital, Tehran University of Medical Sciences, Tehran, Iran
| | - Bahar Ataeinia
- Department of Radiology, Martinos Center for Biomedical Imaging, Massachusetts General Hospital and Harvard Medical School, Boston, MA, United States
| | - Shaghayegh Ranjbar
- Division of Molecular Imaging and Theranostics, Department of Nuclear Medicine, University Hospital Salzburg, Paracelsus Medical University, Salzburg, Austria
| | - Zahra Jamshidi Araghi
- Rajaie Cardiovascular Medical and Research Center, Iran University of Medical Sciences, Tehran, Iran
| | - Mohammad Mobin Moradi
- Research Center for Nuclear Medicine, Shariati Hospital, Tehran University of Medical Sciences, Tehran, Iran
| | - Christian Pirich
- Division of Molecular Imaging and Theranostics, Department of Nuclear Medicine, University Hospital Salzburg, Paracelsus Medical University, Salzburg, Austria
| | - Mohsen Beheshti
- Division of Molecular Imaging and Theranostics, Department of Nuclear Medicine, University Hospital Salzburg, Paracelsus Medical University, Salzburg, Austria
| |
Collapse
|