1
|
Huang C, Hecht EM, Soloff EV, Tiwari HA, Bhosale PR, Dasayam A, Galgano SJ, Kambadakone A, Kulkarni NM, Le O, Liau J, Luk L, Rosenthal MH, Sangster GP, Goenka AH. Imaging for Early Detection of Pancreatic Ductal Adenocarcinoma: Updates and Challenges in the Implementation of Screening and Surveillance Programs. AJR Am J Roentgenol 2024; 223:e2431151. [PMID: 38809122 DOI: 10.2214/ajr.24.31151] [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: 05/30/2024]
Abstract
Pancreatic ductal adenocarcinoma (PDA) is one of the most aggressive cancers. It has a poor 5-year survival rate of 12%, partly because most cases are diagnosed at advanced stages, precluding curative surgical resection. Early-stage PDA has significantly better prognoses due to increased potential for curative interventions, making early detection of PDA critically important to improved patient outcomes. We examine current and evolving early detection concepts, screening strategies, diagnostic yields among high-risk individuals, controversies, and limitations of standard-of-care imaging.
Collapse
Affiliation(s)
- Chenchan Huang
- Department of Radiology, NYU Langone Health, 660 First Ave, 3rd Fl, New York, NY 10016
| | | | - Erik V Soloff
- Department of Radiology, University of Washington, Seattle, WA
| | - Hina Arif Tiwari
- Department of Radiology, University of Arizona College of Medicine, Banner University Medicine, Tucson, AZ
| | - Priya R Bhosale
- Department of Radiology, The University of Texas MD Anderson Cancer Center, Bellaire, TX
| | - Anil Dasayam
- Department of Radiology, University of Pittsburgh Medical Center, Pittsburgh, PA
| | - Samuel J Galgano
- Department of Radiology, University of Alabama at Birmingham, Birmingham, AL
| | | | - Naveen M Kulkarni
- Department of Radiology, Medical College of Wisconsin, Milwaukee, WI
| | - Ott Le
- Department of Radiology, The University of Texas MD Anderson Cancer Center, Bellaire, TX
| | - Joy Liau
- Department of Radiology, University of California at San Diego, San Diego, CA
| | - Lyndon Luk
- Department of Radiology, Columbia University Medical Center, New York, NY
| | - Michael H Rosenthal
- Department of Radiology, Dana-Farber Cancer Institute, Brigham and Women's Hospital, Harvard Medical School, Boston, MA
| | | | | |
Collapse
|
2
|
Mukherjee S, Korfiatis P, Patnam NG, Trivedi KH, Karbhari A, Suman G, Fletcher JG, Goenka AH. Assessing the robustness of a machine-learning model for early detection of pancreatic adenocarcinoma (PDA): evaluating resilience to variations in image acquisition and radiomics workflow using image perturbation methods. Abdom Radiol (NY) 2024; 49:964-974. [PMID: 38175255 DOI: 10.1007/s00261-023-04127-1] [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: 06/15/2023] [Revised: 11/08/2023] [Accepted: 11/12/2023] [Indexed: 01/05/2024]
Abstract
PURPOSE To evaluate robustness of a radiomics-based support vector machine (SVM) model for detection of visually occult PDA on pre-diagnostic CTs by simulating common variations in image acquisition and radiomics workflow using image perturbation methods. METHODS Eighteen algorithmically generated-perturbations, which simulated variations in image noise levels (σ, 2σ, 3σ, 5σ), image rotation [both CT image and the corresponding pancreas segmentation mask by 45° and 90° in axial plane], voxel resampling (isotropic and anisotropic), gray-level discretization [bin width (BW) 32 and 64)], and pancreas segmentation (sequential erosions by 3, 4, 6, and 8 pixels and dilations by 3, 4, and 6 pixels from the boundary), were introduced to the original (unperturbed) test subset (n = 128; 45 pre-diagnostic CTs, 83 control CTs with normal pancreas). Radiomic features were extracted from pancreas masks of these additional test subsets, and the model's performance was compared vis-a-vis the unperturbed test subset. RESULTS The model correctly classified 43 out of 45 pre-diagnostic CTs and 75 out of 83 control CTs in the unperturbed test subset, achieving 92.2% accuracy and 0.98 AUC. Model's performance was unaffected by a three-fold increase in noise level except for sensitivity declining to 80% at 3σ (p = 0.02). Performance remained comparable vis-a-vis the unperturbed test subset despite variations in image rotation (p = 0.99), voxel resampling (p = 0.25-0.31), change in gray-level BW to 32 (p = 0.31-0.99), and erosions/dilations up to 4 pixels from the pancreas boundary (p = 0.12-0.34). CONCLUSION The model's high performance for detection of visually occult PDA was robust within a broad range of clinically relevant variations in image acquisition and radiomics workflow.
Collapse
Affiliation(s)
- Sovanlal Mukherjee
- Divisions of Abdominal and Nuclear Imaging, Nuclear Radiology Fellowship, Nuclear Radiology Research Operations, Enterprise PET/MR Research and Development, Department of Radiology, Mayo Clinic, 200 First St SW, Charlton 1, Rochester, MN, 55905, USA
| | - Panagiotis Korfiatis
- Divisions of Abdominal and Nuclear Imaging, Nuclear Radiology Fellowship, Nuclear Radiology Research Operations, Enterprise PET/MR Research and Development, Department of Radiology, Mayo Clinic, 200 First St SW, Charlton 1, Rochester, MN, 55905, USA
| | - Nandakumar G Patnam
- Divisions of Abdominal and Nuclear Imaging, Nuclear Radiology Fellowship, Nuclear Radiology Research Operations, Enterprise PET/MR Research and Development, Department of Radiology, Mayo Clinic, 200 First St SW, Charlton 1, Rochester, MN, 55905, USA
| | - Kamaxi H Trivedi
- Divisions of Abdominal and Nuclear Imaging, Nuclear Radiology Fellowship, Nuclear Radiology Research Operations, Enterprise PET/MR Research and Development, Department of Radiology, Mayo Clinic, 200 First St SW, Charlton 1, Rochester, MN, 55905, USA
| | - Aashna Karbhari
- Divisions of Abdominal and Nuclear Imaging, Nuclear Radiology Fellowship, Nuclear Radiology Research Operations, Enterprise PET/MR Research and Development, Department of Radiology, Mayo Clinic, 200 First St SW, Charlton 1, Rochester, MN, 55905, USA
| | - Garima Suman
- Divisions of Abdominal and Nuclear Imaging, Nuclear Radiology Fellowship, Nuclear Radiology Research Operations, Enterprise PET/MR Research and Development, Department of Radiology, Mayo Clinic, 200 First St SW, Charlton 1, Rochester, MN, 55905, USA
| | - Joel G Fletcher
- Divisions of Abdominal and Nuclear Imaging, Nuclear Radiology Fellowship, Nuclear Radiology Research Operations, Enterprise PET/MR Research and Development, Department of Radiology, Mayo Clinic, 200 First St SW, Charlton 1, Rochester, MN, 55905, USA
| | - Ajit H Goenka
- Divisions of Abdominal and Nuclear Imaging, Nuclear Radiology Fellowship, Nuclear Radiology Research Operations, Enterprise PET/MR Research and Development, Department of Radiology, Mayo Clinic, 200 First St SW, Charlton 1, Rochester, MN, 55905, USA.
| |
Collapse
|
3
|
Korfiatis P, Suman G, Patnam NG, Trivedi KH, Karbhari A, Mukherjee S, Cook C, Klug JR, Patra A, Khasawneh H, Rajamohan N, Fletcher JG, Truty MJ, Majumder S, Bolan CW, Sandrasegaran K, Chari ST, Goenka AH. Automated Artificial Intelligence Model Trained on a Large Data Set Can Detect Pancreas Cancer on Diagnostic Computed Tomography Scans As Well As Visually Occult Preinvasive Cancer on Prediagnostic Computed Tomography Scans. Gastroenterology 2023; 165:1533-1546.e4. [PMID: 37657758 PMCID: PMC10843414 DOI: 10.1053/j.gastro.2023.08.034] [Citation(s) in RCA: 11] [Impact Index Per Article: 11.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 03/24/2023] [Revised: 08/13/2023] [Accepted: 08/17/2023] [Indexed: 09/03/2023]
Abstract
BACKGROUND & AIMS The aims of our case-control study were (1) to develop an automated 3-dimensional (3D) Convolutional Neural Network (CNN) for detection of pancreatic ductal adenocarcinoma (PDA) on diagnostic computed tomography scans (CTs), (2) evaluate its generalizability on multi-institutional public data sets, (3) its utility as a potential screening tool using a simulated cohort with high pretest probability, and (4) its ability to detect visually occult preinvasive cancer on prediagnostic CTs. METHODS A 3D-CNN classification system was trained using algorithmically generated bounding boxes and pancreatic masks on a curated data set of 696 portal phase diagnostic CTs with PDA and 1080 control images with a nonneoplastic pancreas. The model was evaluated on (1) an intramural hold-out test subset (409 CTs with PDA, 829 controls); (2) a simulated cohort with a case-control distribution that matched the risk of PDA in glycemically defined new-onset diabetes, and Enriching New-Onset Diabetes for Pancreatic Cancer score ≥3; (3) multi-institutional public data sets (194 CTs with PDA, 80 controls), and (4) a cohort of 100 prediagnostic CTs (i.e., CTs incidentally acquired 3-36 months before clinical diagnosis of PDA) without a focal mass, and 134 controls. RESULTS Of the CTs in the intramural test subset, 798 (64%) were from other hospitals. The model correctly classified 360 CTs (88%) with PDA and 783 control CTs (94%), with a mean accuracy 0.92 (95% CI, 0.91-0.94), area under the receiver operating characteristic (AUROC) curve of 0.97 (95% CI, 0.96-0.98), sensitivity of 0.88 (95% CI, 0.85-0.91), and specificity of 0.95 (95% CI, 0.93-0.96). Activation areas on heat maps overlapped with the tumor in 350 of 360 CTs (97%). Performance was high across tumor stages (sensitivity of 0.80, 0.87, 0.95, and 1.0 on T1 through T4 stages, respectively), comparable for hypodense vs isodense tumors (sensitivity: 0.90 vs 0.82), different age, sex, CT slice thicknesses, and vendors (all P > .05), and generalizable on both the simulated cohort (accuracy, 0.95 [95% 0.94-0.95]; AUROC curve, 0.97 [95% CI, 0.94-0.99]) and public data sets (accuracy, 0.86 [95% CI, 0.82-0.90]; AUROC curve, 0.90 [95% CI, 0.86-0.95]). Despite being exclusively trained on diagnostic CTs with larger tumors, the model could detect occult PDA on prediagnostic CTs (accuracy, 0.84 [95% CI, 0.79-0.88]; AUROC curve, 0.91 [95% CI, 0.86-0.94]; sensitivity, 0.75 [95% CI, 0.67-0.84]; and specificity, 0.90 [95% CI, 0.85-0.95]) at a median 475 days (range, 93-1082 days) before clinical diagnosis. CONCLUSIONS This automated artificial intelligence model trained on a large and diverse data set shows high accuracy and generalizable performance for detection of PDA on diagnostic CTs as well as for visually occult PDA on prediagnostic CTs. Prospective validation with blood-based biomarkers is warranted to assess the potential for early detection of sporadic PDA in high-risk individuals.
Collapse
Affiliation(s)
| | - Garima Suman
- Department of Radiology, Mayo Clinic, Rochester, Minnesota
| | | | | | | | | | - Cole Cook
- Division of Medical Imaging Technology Services, Mayo Clinic, Rochester, Minnesota
| | - Jason R Klug
- Division of Medical Imaging Technology Services, Mayo Clinic, Rochester, Minnesota
| | - Anurima Patra
- Department of Radiology, Tata Medical Center, Kolkata, India
| | - Hala Khasawneh
- Department of Radiology, Mayo Clinic, Rochester, Minnesota
| | | | | | - Mark J Truty
- Department of Surgery, Mayo Clinic, Rochester, Minnesota
| | - Shounak Majumder
- Department of Gastroenterology, Mayo Clinic, Rochester, Minnesota
| | | | | | - Suresh T Chari
- Department of Gastroenterology, Mayo Clinic, Rochester, Minnesota
| | - Ajit H Goenka
- Department of Radiology, Mayo Clinic, Rochester, Minnesota.
| |
Collapse
|
4
|
Karbhari A, Mosessian S, Trivedi KH, Valla F, Jacobson M, Truty MJ, Patnam NG, Simeone DM, Zan E, Brennan T, Chen H, Kuo PH, Herrmann K, Goenka AH. Gallium-68-labeled fibroblast activation protein inhibitor-46 PET in patients with resectable or borderline resectable pancreatic ductal adenocarcinoma: A phase 2, multicenter, single arm, open label non-randomized study protocol. PLoS One 2023; 18:e0294564. [PMID: 38011131 PMCID: PMC10681241 DOI: 10.1371/journal.pone.0294564] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/13/2023] [Accepted: 10/20/2023] [Indexed: 11/29/2023] Open
Abstract
BACKGROUND Pancreatic ductal adenocarcinoma (PDAC) is a lethal disease prone to widespread metastatic dissemination and characterized by a desmoplastic stroma that contributes to poor outcomes. Fibroblast activation protein (FAP)-expressing Cancer-Associated Fibroblasts (CAFs) are crucial components of the tumor stroma, influencing carcinogenesis, fibrosis, tumor growth, metastases, and treatment resistance. Non-invasive tools to profile CAF identity and function are essential for overcoming CAF-mediated therapy resistance, developing innovative targeted therapies, and improved patient outcomes. We present the design of a multicenter phase 2 study (clinicaltrials.gov identifier NCT05262855) of [68Ga]FAPI-46 PET to image FAP-expressing CAFs in resectable or borderline resectable PDAC. METHODS We will enroll up to 60 adult treatment-naïve patients with confirmed PDAC. These patients will be eligible for curative surgical resection, either without prior treatment (Cohort 1) or after neoadjuvant therapy (NAT) (Cohort 2). A baseline PET scan will be conducted from the vertex to mid-thighs approximately 15 minutes after administering 5 mCi (±2) of [68Ga]FAPI-46 intravenously. Cohort 2 patients will undergo an additional PET after completing NAT but before surgery. Histopathology and FAP immunohistochemistry (IHC) of initial diagnostic biopsy and resected tumor samples will serve as the truth standards. Primary objective is to assess the sensitivity, specificity, and accuracy of [68Ga]FAPI-46 PET for detecting FAP-expressing CAFs. Secondary objectives will assess predictive values and safety profile validation. Exploratory objectives are comparison of diagnostic performance of [68Ga]FAPI-46 PET to standard-of-care imaging, and comparison of pre- versus post-NAT [68Ga]FAPI-46 PET in Cohort 2. CONCLUSION To facilitate the clinical translation of [68Ga]FAPI-46 in PDAC, the current study seeks to implement a coherent strategy to mitigate risks and increase the probability of meeting FDA requirements and stakeholder expectations. The findings from this study could potentially serve as a foundation for a New Drug Application to the FDA. TRIAL REGISTRATION @ClinicalTrials.gov identifier NCT05262855.
Collapse
Affiliation(s)
- Aashna Karbhari
- Department of Radiology, Mayo Clinic, Rochester, Minnesota, United States of America
| | - Sherly Mosessian
- Clinical Development, Sofie Biosciences, Dulles, Virginia, United States of America
| | - Kamaxi H. Trivedi
- Department of Radiology, Mayo Clinic, Rochester, Minnesota, United States of America
| | - Frank Valla
- Radiopharmaceutical and Contract Manufacturing, Sofie Biosciences, Dulles, Virginia, United States of America
| | - Mark Jacobson
- Department of Radiology, Mayo Clinic, Rochester, Minnesota, United States of America
| | - Mark J. Truty
- Department of Surgery, Mayo Clinic, Rochester, Minnesota, United States of America
| | - Nandakumar G. Patnam
- Department of Radiology, Mayo Clinic, Rochester, Minnesota, United States of America
| | - Diane M. Simeone
- Departments of Surgery and Pathology, NYU Langone Health, New York, New York, United States of America
| | - Elcin Zan
- Department of Radiology, Weill Cornell Medicine, New York, New York, United States of America
| | - Tracy Brennan
- Discovery Life Sciences, Newtown, Pennsylvania, United States of America
| | - Hongli Chen
- Discovery Life Sciences, Newtown, Pennsylvania, United States of America
| | - Phillip H. Kuo
- Departments of Medical Imaging, Medicine and Biomedical Engineering, University of Arizona, Tucson, Arizona, United States of America
| | - Ken Herrmann
- Department of Nuclear Medicine, University Hospital Essen, University of Duisburg-Essen, Essen, Germany
| | - Ajit H. Goenka
- Department of Radiology, Mayo Clinic, Rochester, Minnesota, United States of America
| |
Collapse
|
5
|
Mukherjee S, Korfiatis P, Khasawneh H, Rajamohan N, Patra A, Suman G, Singh A, Thakkar J, Patnam NG, Trivedi KH, Karbhari A, Chari ST, Truty MJ, Halfdanarson TR, Bolan CW, Sandrasegaran K, Majumder S, Goenka AH. Bounding box-based 3D AI model for user-guided volumetric segmentation of pancreatic ductal adenocarcinoma on standard-of-care CTs. Pancreatology 2023; 23:522-529. [PMID: 37296006 PMCID: PMC10676442 DOI: 10.1016/j.pan.2023.05.008] [Citation(s) in RCA: 5] [Impact Index Per Article: 5.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 04/26/2023] [Revised: 05/19/2023] [Accepted: 05/20/2023] [Indexed: 06/12/2023]
Abstract
OBJECTIVES To develop a bounding-box-based 3D convolutional neural network (CNN) for user-guided volumetric pancreas ductal adenocarcinoma (PDA) segmentation. METHODS Reference segmentations were obtained on CTs (2006-2020) of treatment-naïve PDA. Images were algorithmically cropped using a tumor-centered bounding box for training a 3D nnUNet-based-CNN. Three radiologists independently segmented tumors on test subset, which were combined with reference segmentations using STAPLE to derive composite segmentations. Generalizability was evaluated on Cancer Imaging Archive (TCIA) (n = 41) and Medical Segmentation Decathlon (MSD) (n = 152) datasets. RESULTS Total 1151 patients [667 males; age:65.3 ± 10.2 years; T1:34, T2:477, T3:237, T4:403; mean (range) tumor diameter:4.34 (1.1-12.6)-cm] were randomly divided between training/validation (n = 921) and test subsets (n = 230; 75% from other institutions). Model had a high DSC (mean ± SD) against reference segmentations (0.84 ± 0.06), which was comparable to its DSC against composite segmentations (0.84 ± 0.11, p = 0.52). Model-predicted versus reference tumor volumes were comparable (mean ± SD) (29.1 ± 42.2-cc versus 27.1 ± 32.9-cc, p = 0.69, CCC = 0.93). Inter-reader variability was high (mean DSC 0.69 ± 0.16), especially for smaller and isodense tumors. Conversely, model's high performance was comparable between tumor stages, volumes and densities (p > 0.05). Model was resilient to different tumor locations, status of pancreatic/biliary ducts, pancreatic atrophy, CT vendors and slice thicknesses, as well as to the epicenter and dimensions of the bounding-box (p > 0.05). Performance was generalizable on MSD (DSC:0.82 ± 0.06) and TCIA datasets (DSC:0.84 ± 0.08). CONCLUSION A computationally efficient bounding box-based AI model developed on a large and diverse dataset shows high accuracy, generalizability, and robustness to clinically encountered variations for user-guided volumetric PDA segmentation including for small and isodense tumors. CLINICAL RELEVANCE AI-driven bounding box-based user-guided PDA segmentation offers a discovery tool for image-based multi-omics models for applications such as risk-stratification, treatment response assessment, and prognostication, which are urgently needed to customize treatment strategies to the unique biological profile of each patient's tumor.
Collapse
Affiliation(s)
- Sovanlal Mukherjee
- Department of Radiology, Mayo Clinic, 200 First Street SW, Rochester, MN, 55905, USA.
| | - Panagiotis Korfiatis
- Department of Radiology, Mayo Clinic, 200 First Street SW, Rochester, MN, 55905, USA.
| | - Hala Khasawneh
- Department of Radiology, Mayo Clinic, 200 First Street SW, Rochester, MN, 55905, USA.
| | - Naveen Rajamohan
- Department of Radiology, Mayo Clinic, 200 First Street SW, Rochester, MN, 55905, USA.
| | - Anurima Patra
- Department of Radiology, Mayo Clinic, 200 First Street SW, Rochester, MN, 55905, USA.
| | - Garima Suman
- Department of Radiology, Mayo Clinic, 200 First Street SW, Rochester, MN, 55905, USA.
| | - Aparna Singh
- Department of Radiology, Mayo Clinic, 200 First Street SW, Rochester, MN, 55905, USA.
| | - Jay Thakkar
- Department of Radiology, Mayo Clinic, 200 First Street SW, Rochester, MN, 55905, USA.
| | - Nandakumar G Patnam
- Department of Radiology, Mayo Clinic, 200 First Street SW, Rochester, MN, 55905, USA.
| | - Kamaxi H Trivedi
- Department of Radiology, Mayo Clinic, 200 First Street SW, Rochester, MN, 55905, USA.
| | - Aashna Karbhari
- Department of Radiology, Mayo Clinic, 200 First Street SW, Rochester, MN, 55905, USA.
| | - Suresh T Chari
- Department of Gastroenterology, Hepatology and Nutrition, The University of Texas MD Anderson Cancer Center, 1515 Holcombe Blvd, Houston, TX, 77030, USA.
| | - Mark J Truty
- Department of Surgery, Mayo Clinic, 200 First Street SW, Rochester, MN, 55905, USA.
| | | | - Candice W Bolan
- Department of Radiology, Mayo Clinic, 4500 San Pablo Rd S, Jacksonville, FL, 32224, USA.
| | - Kumar Sandrasegaran
- Department of Radiology, Mayo Clinic, 13400 E Shea Blvd, Scottsdale, AZ, 85259, USA.
| | - Shounak Majumder
- Department of Gastroenterology, Mayo Clinic, 200 First Street SW, Rochester, MN, 55905, USA.
| | - Ajit H Goenka
- Department of Radiology, Mayo Clinic, 200 First Street SW, Rochester, MN, 55905, USA.
| |
Collapse
|