Oncologic Imaging Research Articles

Dual-contrast photon-counting micro-CT using iodine and a novel bismuth-based contrast agent

Objectives. To characterize for the first time in vivo a novel bismuth-based nanoparticular contrast agent developed for preclinical applications. Then, to design and test in vivo a multi-contrast protocol for functional cardiac imaging using the new bismuth nanoparticles and a well-established iodine-based contrast agent. Approach. A micro-computed tomography scanner was assembled and equipped with a photon-counting detector. Five mice were administered with the bismuth-based contrast agent and systematically scanned over 5 h to quantify the contrast enhancement in relevant organs of interest. Subsequently, the multi-contrast agent protocol was tested on three mice. Material decomposition was performed on the acquired spectral data to quantify the concentration of bismuth and iodine in multiple structures, e.g. the myocardium and vasculature. Main results. In the vasculature, the bismuth agent provides a peak enhancement of 1100 HU and a half-life of about 260 min. After the injection, it accumulates in the liver, spleen and intestinal wall reaching a CT value of 440 HU about 5 h post injection. Phantom measurements showed that the bismuth provides more contrast enhancement than iodine for a variety of tube voltages. The multi-contrast protocol for cardiac imaging successfully allowed the simultaneous decomposition of the vasculature, the brown adipose tissue and the myocardium. Significance. The new bismuth-based contrast agent was proven to have a long circulation time suitable for preclinical applications and to provide more contrast than iodine agents. The proposed multi-contrast protocol resulted in a new tool for cardiac functional imaging. Furthermore, thanks to the contrast enhancement provided in the intestinal wall, the novel contrast agent may be used to develop further multi contrast agent protocols for abdominal and oncological imaging.

Standardised Uptake Value in Organ Confined Prostate Cancer in 68-Ga- Prostate-Specific Membrane Antigen Positron Emission Tomography-Computed Tomography Scan and its Correlation with Prostate Specific Antigen Level and Gleason Score.

A positron emission tomography (PET) scan and a computed tomography (CT) scan are an integral part of oncological imaging and other modalities such as magnetic resonance imaging, CT or bone scintigraphy have some limitations in staging the workup of prostate carcinoma. Combined with tissue-specific markers like prostate-specific membrane antigen (PSMA), positron emitter-based functional imaging results have improved. Our study aimed to determine the Standardised Uptake Value (SUVmax) in prostate adenocarcinoma that is confined to the organ in Ga-68-PSMA PET-CT scans and how it correlates with prostate-specific antigen (PSA) levels and Gleason score (GS). This cross-sectional study was conducted at Sindh Institute of Urology and Transplantation (SIUT), Karachi, and includes subjects referred for a Ga68-PSMA PET-CT scan from September 2017 to January 2022. Histopathologic-proven adenocarcinoma prostate patients with organ-confined disease and PSA levels obtained within 6 weeks before the PSMA-PET-CT scan were included in the study. PET-CT images were semi-quantitatively analysed by measuring SUVmax and the result was interpreted using statistical software SPSS version 22.0. A total of 154 patients were analysed. The mean age of patients was 66.57 ± 8.86 years. The GS of all patients ranges from 6 to 10. The mean and median PSA levels were 32.33 ng/mL (range: 0.004-306.00) and 14.20 ng/mL, respectively. The mean SUVmax of all prostatic lesions was 14.67 ± 12.58 and the median value was 10.76. SUVmax was higher in patients with a PSA level of more than ten than those with a <10. The correlation of SUVmax with PSA and GS showed a significant correlation. The SUVmax of organ-confined prostate cancer correlates well with PSA level and GS Median SUVmax and PSA directly relate to GS.

FAPI PET/CT Imaging—An Updated Review

Despite revolutionizing the field of oncological imaging, Positron Emission Tomography (PET) with [18F]Fluorodeoxyglucose (FDG) as its workhorse is limited by a lack of specificity and low sensitivity in certain tumor subtypes. Fibroblast activation protein (FAP), a type II transmembrane glycoprotein, is expressed by cancer-associated fibroblasts (CAFs) that form a major component of the tumor stroma. FAP holds the promise to be a pan-cancer target, owing to its selective over-expression in a vast majority of neoplasms, particularly epithelial cancers. Several radiolabeled FAP inhibitors (FAPI) have been developed for molecular imaging and potential theranostic applications. Preliminary data on FAPI PET/CT remains encouraging, with extensive multi-disciplinary clinical research currently underway. This review summarizes the existing literature on FAPI PET/CT imaging with an emphasis on diagnostic applications, comparison with FDG, pitfalls, and future directions.

Evaluation of the Diagnostic Accuracy of FAPI PET/CT in Oncologic Studies: Systematic Review and Metaanalysis.

Fibroblast-activation protein is a promising target for oncologic molecular imaging. Studies show that fibroblast activation protein inhibitor (FAPI) radiotracers are accurate diagnostics with favorable tumor-to-background ratios across various cancers. Therefore, we performed a systematic review and metaanalysis to assess the diagnostic performance of FAPI PET/CT in comparison with [18F]FDG PET/CT, the most widely used radiotracer in oncology. Methods: We conducted a systematic search in MEDLINE, Embase, Scopus, PubMed, Cochrane Central Register of Controlled Trials, relevant trial registries, and bibliographies. The search consisted of combinations of terms for 3 topics: neoplasia, PET/CT, and FAPI. Two authors independently screened retrieved articles using predefined inclusion and exclusion criteria and extracted the data. Study quality was assessed using the criteria of QUADAS-2 (Quality Assessment of Diagnostic Accuracy Studies 2). For each study, the sensitivity, specificity, and 95% CIs were calculated to determine diagnostic accuracy for primary, nodal, and metastatic lesions. A random-effects metaanalysis was used for pooling the data, and heterogeneity was assessed (I2 index). Results: Thirty-nine studies (1,259 patients) investigating the use of FAPI PET/CT were included. On a patient-based analysis, pooled sensitivity was 0.99 (95% CI, 0.97-1.0) for the detection of primary lesions. Pooled sensitivity for nodal and distant metastases was 0.91 (95% CI, 0.81-0.96) and 0.99 (95% CI, 0.96-1.0), respectively. On a paired analysis between FAPI and [18F]FDG PET/CT, FAPI had a higher sensitivity in the detection of primary, nodal, and metastatic lesions (all P < 0.001). The differences in sensitivities between FAPI and [18F]FDG were statistically significant. In terms of heterogeneity, analyses on primary lesions were moderately affected, distant metastatic lesions were highly affected, and the nodal metastatic analyses had negligible heterogeneity. Conclusion: The diagnostic performance of FAPI PET/CT is superior to that of [18F]FDG in the detection of primary, nodal, and distant metastases. However, further studies are needed to better evaluate its utility and indication in specific cancer types and clinical settings.

Using deep learning–derived image features in radiologic time series to make personalised predictions: proof of concept in colonic transit data

ObjectivesSiamese neural networks (SNN) were used to classify the presence of radiopaque beads as part of a colonic transit time study (CTS). The SNN output was then used as a feature in a time series model to predict progression through a CTS.MethodsThis retrospective study included all patients undergoing a CTS in a single institution from 2010 to 2020. Data were partitioned in an 80/20 Train/Test split. Deep learning models based on a SNN architecture were trained and tested to classify images according to the presence, absence, and number of radiopaque beads and to output the Euclidean distance between the feature representations of the input images. Time series models were used to predict the total duration of the study.ResultsIn total, 568 images of 229 patients (143, 62% female, mean age 57) patients were included. For the classification of the presence of beads, the best performing model (Siamese DenseNET trained with a contrastive loss with unfrozen weights) achieved an accuracy, precision, and recall of 0.988, 0.986, and 1. A Gaussian process regressor (GPR) trained on the outputs of the SNN outperformed both GPR using only the number of beads and basic statistical exponential curve fitting with MAE of 0.9 days compared to 2.3 and 6.3 days (p < 0.05) respectively.ConclusionsSNNs perform well at the identification of radiopaque beads in CTS. For time series prediction our methods were superior at identifying progression through the time series compared to statistical models, enabling more accurate personalised predictions.Clinical relevance statementOur radiologic time series model has potential clinical application in use cases where change assessment is critical (e.g. nodule surveillance, cancer treatment response, and screening programmes) by quantifying change and using it to make more personalised predictions.Key Points• Time series methods have improved but application to radiology lags behind computer vision. Colonic transit studies are a simple radiologic time series measuring function through serial radiographs.• We successfully employed a Siamese neural network (SNN) to compare between radiographs at different points in time and then used the output of SNN as a feature in a Gaussian process regression model to predict progression through the time series.• This novel use of features derived from a neural network on medical imaging data to predict progression has potential clinical application in more complex use cases where change assessment is critical such as in oncologic imaging, monitoring for treatment response, and screening programmes.

Comparative study of the quantitative accuracy of oncological PET imaging based on deep learning methods.

[18F] Fluorodeoxyglucose (FDG) positron emission tomography/computed tomography (PET/CT) is an important tool for tumor assessment. Shortening scanning time and reducing the amount of radioactive tracer remain the most difficult challenges. Deep learning methods have provided powerful solutions, thus making it important to choose an appropriate neural network architecture. A total of 311 tumor patients who underwent 18F-FDG PET/CT were retrospectively collected. The PET collection time was 3 min/bed. The first 15 and 30 s of each bed collection time were selected to simulate low-dose collection, and the pre-90s was used as the clinical standard protocol. Low-dose PET was used as input, convolutional neural network (CNN, 3D Unet as representative) and generative adversarial network (GAN, P2P as representative) were used to predict the full-dose images. The image visual scores, noise levels and quantitative parameters of tumor tissue were compared. There was high consistency in image quality scores among all groups [Kappa =0.719, 95% confidence interval (CI): 0.697-0.741, P<0.001]. There were 264 cases (3D Unet-15s), 311 cases (3D Unet-30s), 89 cases (P2P-15s) and 247 cases (P2P-30s) with image quality score ≥3, respectively. There was significant difference in the score composition among all groups (χ2=1,325.46, P<0.001). Both deep learning models reduced the standard deviation (SD) of background, and increased the signal-to-noise ratio (SNR). When 8%PET images were used as input, P2P and 3D Unet had similar enhancement effect on SNR of tumor lesions, but 3D Unet could significantly improve the contrast-noise ratio (CNR) (P<0.05). There was no significant difference in SUVmean of tumor lesions compared with s-PET group (P>0.05). When 17%PET image was used as input, SNR, CNR and SUVmax of tumor lesion of 3D Unet group had no statistical difference with those of s-PET group (P>0.05). Both GAN and CNN can suppress image noise to varying degrees and improve image quality. However, when 3D Unet reduces the noise of tumor lesions, it can improve the CNR of tumor lesions. Moreover, quantitative parameters of tumor tissue are similar to those under the standard acquisition protocol, which can meet the needs of clinical diagnosis.

Deep learning assessment compared to radiologist reporting for metastatic spinal cord compression on CT.

Metastatic spinal cord compression (MSCC) is a disastrous complication of advanced malignancy. A deep learning (DL) algorithm for MSCC classification on CT could expedite timely diagnosis. In this study, we externally test a DL algorithm for MSCC classification on CT and compare with radiologist assessment. Retrospective collection of CT and corresponding MRI from patients with suspected MSCC was conducted from September 2007 to September 2020. Exclusion criteria were scans with instrumentation, no intravenous contrast, motion artefacts and non-thoracic coverage. Internal CT dataset split was 84% for training/validation and 16% for testing. An external test set was also utilised. Internal training/validation sets were labelled by radiologists with spine imaging specialization (6 and 11-years post-board certification) and were used to further develop a DL algorithm for MSCC classification. The spine imaging specialist (11-years expertise) labelled the test sets (reference standard). For evaluation of DL algorithm performance, internal and external test data were independently reviewed by four radiologists: two spine specialists (Rad1 and Rad2, 7 and 5-years post-board certification, respectively) and two oncological imaging specialists (Rad3 and Rad4, 3 and 5-years post-board certification, respectively). DL model performance was also compared against the CT report issued by the radiologist in a real clinical setting. Inter-rater agreement (Gwet's kappa) and sensitivity/specificity/AUCs were calculated. Overall, 420 CT scans were evaluated (225 patients, mean age=60 ± 11.9[SD]); 354(84%) CTs for training/validation and 66(16%) CTs for internal testing. The DL algorithm showed high inter-rater agreement for three-class MSCC grading with kappas of 0.872 (p<0.001) and 0.844 (p<0.001) on internal and external testing, respectively. On internal testing DL algorithm inter-rater agreement (κ=0.872) was superior to Rad 2 (κ=0.795) and Rad 3 (κ=0.724) (both p<0.001). DL algorithm kappa of 0.844 on external testing was superior to Rad 3 (κ=0.721) (p<0.001). CT report classification of high-grade MSCC disease was poor with only slight inter-rater agreement (κ=0.027) and low sensitivity (44.0), relative to the DL algorithm with almost-perfect inter-rater agreement (κ=0.813) and high sensitivity (94.0) (p<0.001). Deep learning algorithm for metastatic spinal cord compression on CT showed superior performance to the CT report issued by experienced radiologists and could aid earlier diagnosis.

Exploring the Potential of Three-Dimensional Imaging, Printing, and Modeling in Pediatric Surgical Oncology: A New Era of Precision Surgery.

Pediatric surgical oncology is a technically challenging field that relies on CT and MRI as the primary imaging tools for surgical planning. However, recent advances in 3D reconstructions, including Cinematic Rendering, Volume Rendering, 3D modeling, Virtual Reality, Augmented Reality, and 3D printing, are increasingly being used to plan complex cases bringing new insights into pediatric tumors to guide therapeutic decisions and prognosis in different pediatric surgical oncology areas and locations including thoracic, brain, urology, and abdominal surgery. Despite this, challenges to their adoption remain, especially in soft tissue-based specialties such as pediatric surgical oncology. This work explores the main innovative imaging reconstruction techniques, 3D modeling technologies (CAD, VR, AR), and 3D printing applications through the analysis of three real cases of the most common and surgically challenging pediatric tumors: abdominal neuroblastoma, thoracic inlet neuroblastoma, and a bilateral Wilms tumor candidate for nephron-sparing surgery. The results demonstrate that these new imaging and modeling techniques offer a promising alternative for planning complex pediatric oncological cases. A comprehensive analysis of the advantages and limitations of each technique has been carried out to assist in choosing the optimal approach.

The role of ICG-fuorescence imaging in abdominal surgical oncology

The purpose of the study was to summarize and analyze the available data on modern methods of additional imaging using indocyanine green (ICG) used in abdominal surgical oncology.Material and Methods. The search and analysis of publications over the last 7–10 years was carried out using RSCI (Russian Science Citation Index), E-library, PubMed, Scopus, MedLine, Elsevier, Web of Science, ScienceDirect systems. The following keywords were used for the search: “ICG-fluorescent navigation”, “ICG-fluorescence”, “gastric cancer”, “colorectal cancer”, “indocyanine green”.Results. The role of ICG-fluorescent navigation in abdominal oncology, in particular, in surgery for cancer of the stomach, colon and rectum, identification of peritoneal carcinomatosis and liver metastases, as well as in biliary surgery and cholangiography, was shown. Intraoperative visualization of sentinel lymph nodes and mapping of lymph nodes in patients with gastric and colorectal cancers were described. In addition, the information on the diagnostic value of vascular imaging in colorectal surgery, detection of the boundaries of hepatocellular carcinoma and surrounding tissues, as well as real-time liver metastases using indocyanine green, was provided. Special attention was paid to ICG-mediated cholecystectomy and cholangiography. ICG fluorescence imaging in colorectal and gastric cancer surgery makes it possible to individualize the surgical approach and provide functional surgery associated with the performance of a sufficient and prophylactically excessive volume of surgical intervention more effectively. Conclusion. The presented data confirm the relevance of the search for a vector of improving the imaging methods used in abdominal surgical oncology, which requires even more even more multicenter prospective cohort studies to determine the best ways of improving the technique.

Artificial Intelligence in CT and MR Imaging for Oncological Applications.

Cancer care increasingly relies on imaging for patient management. The two most common cross-sectional imaging modalities in oncology are computed tomography (CT) and magnetic resonance imaging (MRI), which provide high-resolution anatomic and physiological imaging. Herewith is a summary of recent applications of rapidly advancing artificial intelligence (AI) in CT and MRI oncological imaging that addresses the benefits and challenges of the resultant opportunities with examples. Major challenges remain, such as how best to integrate AI developments into clinical radiology practice, the vigorous assessment of quantitative CT and MR imaging data accuracy, and reliability for clinical utility and research integrity in oncology. Such challenges necessitate an evaluation of the robustness of imaging biomarkers to be included in AI developments, a culture of data sharing, and the cooperation of knowledgeable academics with vendor scientists and companies operating in radiology and oncology fields. Herein, we will illustrate a few challenges and solutions of these efforts using novel methods for synthesizing different contrast modality images, auto-segmentation, and image reconstruction with examples from lung CT as well as abdome, pelvis, and head and neck MRI. The imaging community must embrace the need for quantitative CT and MRI metrics beyond lesion size measurement. AI methods for the extraction and longitudinal tracking of imaging metrics from registered lesions and understanding the tumor environment will be invaluable for interpreting disease status and treatment efficacy. This is an exciting time to work together to move the imaging field forward with narrow AI-specific tasks. New AI developments using CT and MRI datasets will be used to improve the personalized management of cancer patients.

Comparative Evaluation of [18F]5-Fluoroaminosuberic Acid and (4S)-4-3-[18F]fluoropropyl)-l-Glutamate as System xC--Targeting Radiopharmaceuticals.

System [Formula: see text] is an appealing biomarker for targeting oxidative stress with oncologic PET imaging and can serve as an alternative PET biomarker to other metabolic indicators. In this paper, we report a direct comparison of 2 18F-labeled amino acid radiopharmaceuticals targeting system [Formula: see text], [18F]5-fluoroaminosuberic acid ([18F]FASu) and (4S)-4-(3-[18F]fluoropropyl)-l-glutamate ([18F]FSPG), in terms of their uptake specificity and ability to image glioma and lung cancer xenografts invivo. Methods: Both tracers were synthesized according to previously published procedures. In vitro uptake specificity assays were conducted using prostate (PC-3), glioblastoma (U-87), colorectal (HT-29), ovarian (SKOV3), breast (MDA-MB-231), and lung cancer (A549) cell lines. PET/CT imaging and biodistribution studies were conducted in immunocompromised mice bearing U-87 or A549 xenografts. Results: In vitro cell uptake assays showed that the tracers accumulated in cancer cells in a time-dependent manner and that the uptake of [18F]FASu was blocked by the system [Formula: see text] inhibitor sulfasalazine and rose bengal, but not by system L inhibitor 2-aminobicyclo-(2,2,1)-heptane-2-carboxylic acid, system [Formula: see text] inhibitor L-trans-pyrrolidine-2,4-dicarboxylic acid, or l-serine, which is a substrate for transporter systems A, ACS, B0, and B0,+ Conversely, [18F]FSPG uptake decreased significantly in the presence of an excess of L-trans-pyrrolidine-2,4-dicarboxylic acid in 2 of 3 tested cell lines, indicating some reliance on system [Formula: see text] in these cells. In an invivo setting, [18F]FASu and [18F]FSPG generated good-contrast PET images in U-87 and A549 tumor-bearing mice. Tracer accumulation in A549 tumors was 5.0 ± 0.8 percentage injected dose (%ID)/g ([18F]FASu, n ≥ 5) and 6.3 ± 1.3 %ID/g ([18F]FSPG, n ≥ 6, P = 0.7786), whereas U-87 xenografts demonstrated uptake of 6.1 ± 2.4 %ID/g ([18F]FASu, n ≥ 4) and 11.2 ± 4.1 %ID/g ([18F]FSPG, n ≥ 4, P = 0.0321) at 1 h after injection. Conclusion: [18F]FSPG had greater invitro uptake than [18F]FASu in all cell lines tested; however, our results indicate that residual uptake differences exist between [18F]FSPG and [18F]FASu, suggesting alternative transporter activity in the cell lines tested. In vivo studies demonstrated the ability of both [18F]FASu and [18F]FSPG to image glioblastoma (U-87) and non-small cell lung cancer (A549) xenografts.

PET radiotracers for whole-body in vivo molecular imaging of prostatic neuroendocrine malignancies.

Prostatic neuroendocrine malignancies represent a spectrum of diseases. Treatment-induced neuroendocrine differentiation (tiNED) in hormonally treated adenocarcinoma has been the subject of a large amount of recent research. However, the identification of neuroendocrine features in treatment-naïve prostatic tumor raises a differential diagnosis between prostatic adenocarcinoma with de novo neuroendocrine differentiation (dNED) versus one of the primary prostatic neuroendocrine tumors (P-NETs) and carcinomas (P-NECs). While [18F]FDG is being used as the main PET radiotracer in oncologic imaging and reflects cellular glucose metabolism, other molecules labeled with positron-emitting isotopes, mainly somatostatin-analogues labeled with 68Ga and prostate-specific membrane antigen (PSMA)-ligands labeled with either 18F or 68Ga, are now routinely used in departments of nuclear medicine and molecular imaging, and may be advantageous in imaging prostatic neuroendocrine malignancies. Still, the selection of the preferred PET radiotracer in such cases might be challenging. In the current review, we summarize and discuss published data on these different entities from clinical, biological, and molecular imaging standpoints. Specifically, we review the roles that [18F]FDG, radiolabeled somatostatin-analogues, and radiolabeled PSMA-ligands play in these entities in order to provide the reader with practical recommendations regarding the preferred PET radiotracers for imaging each entity. In cases of tiNED, we conclude that PSMA expression may be low and that [18F]FDG or radiolabeled somatostatin-analogues should be preferred for imaging. In cases of prostatic adenocarcinoma with dNED, we present data that support the superiority of radiolabeled PSMA-ligands. In cases of primary neuroendocrine malignancies, the use of [18F]FDG for imaging high-grade P-NECs and radiolabeled somatostatin-analogues for imaging well-differentiated P-NETs is recommended. KEY POINTS: • The preferred PET radiotracer for imaging prostatic neuroendocrine malignancies depends on the specific clinical scenario and pathologic data. • When neuroendocrine features result from hormonal therapy for prostate cancer, PET-CT should be performed with [18F]FDG or radiolabeled somatostatin-analogue rather than with radiolabeled PSMA-ligand. • When neuroendocrine features are evident in newly diagnosed prostate cancer, differentiating adenocarcinoma from primary neuroendocrine malignancy is challenging but crucial for selection of PET radiotracer and for clinical management.

Role of CT in the detection and management of cancer related complications: a study of 599 patients.

Cancer-related complications (CrC) and any potentially life-threatening findings detected on routine oncological imaging requires urgent intervention and needs proactive management. We conducted a retrospective study to highlight the role of imaging in the detection of CrC on computed tomography (CT)-scan while sharing our experience at a tertiary care cancer hospital. All the reports of the CT scans performed in our department between January 2018 and December 2019 were reviewed and the imaging findings of CrC were recorded. Only the patients who had known malignancy and underwent imaging evaluation at our centre at baseline/follow up/surveillance were included. The clinical details of the patients were recorded and the findings were classified based on the system or organ involved and also on the basis of its impact on clinical management. A total of 14,226 CT scans were performed during the study period, out of which 599 patients had CrC. Most of the CrC were seen involving thorax (265/599, 44.3%) followed by abdomen (229/599, 38.2%) and head and neck (104/599, 17.3%) regions. The commonly encountered CrC were pulmonary infections, superior vena cava obstruction and drug-induced lung changes. CrC have significant impact on the course of management of cancer patients and radiologist plays an important role in early diagnosis and initiation of prompt management of many such patients. CT is an excellent modality for early diagnosis of CrC which guides the oncologist for appropriate treatment.

A dual-transformation with contrastive learning framework for lymph node metastasis prediction in pancreatic cancer.

Pancreatic cancer is a malignant tumor, and its high recurrence rate after surgery is related to the lymph node metastasis status. In clinical practice, a preoperative imaging prediction method is necessary for prognosis assessment and treatment decision; however, there are two major challenges: insufficient data and difficulty in discriminative feature extraction. This paper proposed a deep learning model to predict lymph node metastasis in pancreatic cancer using multiphase CT, where a dual-transformation with contrastive learning framework is developed to overcome the challenges in fine-grained prediction with small sample sizes. Specifically, we designed a novel dynamic surface projection method to transform 3D data into 2D images for effectively using the 3D information, preserving the spatial correlation of the original texture information and reducing computational resources. Then, this dynamic surface projection was combined with the spiral transformation to establish a dual-transformation method for enhancing the diversity and complementarity of the dataset. A dual-transformation-based data augmentation method was also developed to produce numerous 2D-transformed images to alleviate the effect of insufficient samples. Finally, the dual-transformation-guided contrastive learning scheme based on intra-space-transformation consistency and inter-class specificity was designed to mine additional supervised information, thereby extracting more discriminative features. Extensive experiments have shown the promising performance of the proposed model for predicting lymph node metastasis in pancreatic cancer. Our dual-transformation with contrastive learning scheme was further confirmed on an external public dataset, representing a potential paradigm for the fine-grained classification of oncological images with small sample sizes. The code will be released at https://github.com/SJTUBME-QianLab/Dual-transformation.

One-stop [18F]FDG and [68Ga]Ga-DOTA-FAPI-04 total-body PET/CT examination with dual-low activity: a feasibility study.

Positron emission tomography/computed tomography (PET/CT) based on fibroblast activation protein inhibitors (FAPI) has shown complementary values to 2-[18F]-fluoro-2-deoxy-D-glucose ([18F]FDG) in cancer imaging. This study aimed to investigate the feasibility of a one-stop FDG-FAPI dual-tracer imaging protocol with dual-low activity for oncological imaging. Nineteen patients with malignancies underwent one-stop [18F]FDG (0.37MBq/kg) PET (PETFDG) and dual-tracer PET 30-40 and 50-60min (hereafter, PETD30-40 and PETD50-60, respectively) after additional injection of [68Ga]Ga-DOTA-FAPI-04 (0.925MBq/kg), with a single diagnostic CT to generate the PET/CT. The lesion detection rate and tumor-to-normal ratios (TNRs) of tracer uptake were compared between PETFDG/CT and PETD50-60/CT and between PETD50-60/CT and PETD30-40/CT. In addition, a visual scoring system was established to compare the lesion detectability. The dual-tracer PETD50-60 and PETD30-40/CT showed similar performance in detecting primary tumors but presented significantly higher lesion TNRs than PETFDG. Significantly, more metastases with higher TNRs were identified on PETD50-60 than PETFDG (491 vs. 261, P < 0.001). The dual-tracer PETD50-60 received significantly higher visual scores than single PETFDG (111 vs. 10) in demonstrating both primary tumors (12 vs. 2) and metastases (99 vs. 8). However, these differences were not significant between PETD50-60 and PETD30-40. These resulted in tumor upstaging in 44.4% patients taking PET/CT for initial assessment, and more recurrences (68 vs. 7) were identified in patients taking PET/CT for restaging, both on PETD50-60 and PETD30-40, compared to PETFDG. The reduced effective dosimetry per patient (26.2 ± 2.57mSv) was equal to that of a single standard whole-body PET/CT. The one-stop dual-tracer dual-low-activity PET imaging protocol combines the strengths of [18F]FDG and [68Ga]Ga-DOTA-FAPI-04 with shorter duration and lesser radiation and is thus clinically applicable.

Optimisation of scan duration and image quality in oncological 89Zr immunoPET imaging using the Biograph Vision PET/CT

PurposeMonoclonal antibody (mAb)-based PET (immunoPET) imaging can characterise tumour lesions non-invasively. It may be a valuable tool to determine which patients may benefit from treatment with a specific monoclonal antibody (mAb) and evaluate treatment response. For 89Zr immunoPET imaging, higher sensitivity of state-of-the art PET/CT systems equipped with silicon photomultiplier (SiPM)-based detector elements may be beneficial as the low positron abundance of 89Zr causes a low signal-to-noise level. Moreover, the long physical half-life limits the amount of activity that can be administered to the patients leading to poor image quality even when using long scan durations. Here, we investigated the difference in semiquantitative performance between the PMT-based Biograph mCT, our clinical reference system, and the SiPM-based Biograph Vision PET/CT in 89Zr immunoPET imaging. Furthermore, the effects of scan duration reduction using the Vision on semiquantitative imaging parameters and its influence on image quality assessment were evaluated.MethodsData were acquired on day 4 post 37 MBq 89Zr-labelled mAb injection. Five patients underwent a double scan protocol on both systems. Ten patients were scanned only on the Vision. For PET image reconstruction, three protocols were used, i.e. one camera-dependent protocol and European Association of Nuclear Medicine Research Limited (EARL) standards 1 and 2 compliant protocols. Vision data were acquired in listmode and were reprocessed to obtain images at shorter scan durations. Semiquantitative PET image parameters were derived from tumour lesions and healthy tissues to assess differences between systems and scan durations. Differently reconstructed images obtained using the Vision were visually scored regarding image quality by two nuclear medicine physicians.ResultsWhen images were reconstructed using 100% acquisition time on both systems following EARL standard 1 compliant reconstruction protocols, results regarding semiquantification were comparable. For Vision data, reconstructed images that conform to EARL1 standards still resulted in comparable semiquantification at shorter scan durations (75% and 50%) regarding 100% acquisition time.ConclusionScan duration of 89Zr immunoPET imaging using the Vision can be decreased up to 50% compared with using the mCT while maintaining image quality using the EARL1 compliant reconstruction protocol.

Radiolabeled Dendrimer Coated Nanoparticles for Radionuclide Imaging and Therapy: A Systematic Review

Dendrimers are nanoscale-size polymers with a globular structure. They are composed of an internal core and branching dendrons with surface active groups which can be functionalized for medical applications. Different complexes have been developed for imaging and therapeutic purposes. This systematic review aims to summarize the development of newer dendrimers for oncological applications in nuclear medicine. An online literature search was conducted on Pubmed, Scopus, Medline, Cochrane Library, and Web Of Science databases selecting published studies from January 1999 to December 2022. The accepted studies considered the synthesis of dendrimer complexes for oncological nuclear medicine imaging and therapy. 111 articles were identified; 69 articles were excluded because they did not satisfy the selection criteria. Thus, nine duplicate records were removed. The remaining 33 articles were included and selected for quality assessment. Nanomedicine has led researchers to create novel nanocarriers with high affinity for the target. Dendrimers represent feasible imaging probes and therapeutic agents since, through the functionalization of external chemical groups and thanks to the possibility to carry pharmaceuticals, it can be possible to exploit different therapeutic strategies and develop a useful weapon for oncological treatments.

Magnetic Nanoclusters Stabilized with Poly[3,4-Dihydroxybenzhydrazide] as Efficient Therapeutic Agents for Cancer Cells Destruction.

Magnetic structures exhibiting large magnetic moments are sought after in theranostic approaches that combine magnetic hyperthermia treatment (MH) and diagnostic magnetic resonance imaging in oncology, since they offer an enhanced magnetic response to an external magnetic field. We report on the synthesized production of a core-shell magnetic structure using two types of magnetite nanoclusters (MNC) based on a magnetite core and polymer shell. This was achieved through an in situ solvothermal process, using, for the first time, 3,4-dihydroxybenzhydrazide (DHBH) and poly[3,4-dihydroxybenzhydrazide] (PDHBH) as stabilizers. Transmission electron microscopy (TEM) analysis showed the formation of spherical MNC, X-ray photoelectronic spectroscopy (XPS) and Fourier transformed infrared (FT-IR) analysis proved the existence of the polymer shell. Magnetization measurement showed saturation magnetization values of 50 emu/g for PDHBH@MNC and 60 emu/g for DHBH@MNC with very low coercive field and remanence, indicating that the MNC are in a superparamagnetic state at room temperature and are thus suitable for biomedical applications. MNCs were investigated in vitro, on human normal (dermal fibroblasts-BJ) and tumor (colon adenocarcinoma-CACO2, and melanoma-A375) cell lines, in view of toxicity, antitumor effectiveness and selectivity upon magnetic hyperthermia. MNCs exhibited good biocompatibility and were internalized by all cell lines (TEM), with minimal ultrastructural changes. By means of flowcytometry apoptosis detection, fluorimetry, spectrophotometry for mitochondrial membrane potential, oxidative stress, ELISA-caspases, and Western blot-p53 pathway, we show that MH efficiently induced apoptosis mostly via the membrane pathway and to a lower extent by the mitochondrial pathway, the latter mainly observed in melanoma. Contrarily, the apoptosis rate was above the toxicity limit in fibroblasts. Due to its coating, PDHBH@MNC showed selective antitumor efficacy and can be further used in theranostics since the PDHBH polymer provides multiple reaction sites for the attachment of therapeutic molecules.

Comparing Detection Schemes for Adversarial Images against Deep Learning Models for Cancer Imaging.

Deep learning (DL) models have demonstrated state-of-the-art performance in the classification of diagnostic imaging in oncology. However, DL models for medical images can be compromised by adversarial images, where pixel values of input images are manipulated to deceive the DL model. To address this limitation, our study investigates the detectability of adversarial images in oncology using multiple detection schemes. Experiments were conducted on thoracic computed tomography (CT) scans, mammography, and brain magnetic resonance imaging (MRI). For each dataset we trained a convolutional neural network to classify the presence or absence of malignancy. We trained five DL and machine learning (ML)-based detection models and tested their performance in detecting adversarial images. Adversarial images generated using projected gradient descent (PGD) with a perturbation size of 0.004 were detected by the ResNet detection model with an accuracy of 100% for CT, 100% for mammogram, and 90.0% for MRI. Overall, adversarial images were detected with high accuracy in settings where adversarial perturbation was above set thresholds. Adversarial detection should be considered alongside adversarial training as a defense technique to protect DL models for cancer imaging classification from the threat of adversarial images.

Atypical and Typical Distant Sarcoid-like Reactions during Treatment of Metastatic Melanoma.

HomeRadiology: Imaging CancerVol. 5, No. 2 PreviousNext Images in CancerFree AccessAtypical and Typical Distant Sarcoid-like Reactions during Treatment of Metastatic MelanomaLacey J. McIntosh , Evan C. Ruppell, Maria BarileLacey J. McIntosh , Evan C. Ruppell, Maria BarileAuthor AffiliationsFrom the Division of Oncologic and Molecular Imaging (L.J.M., E.C.R.) and Division of Cardiothoracic Imaging (M.B.), UMass Chan Medical School/Memorial Health Care, 55 Lake Avenue North, Worcester, MA 01655.Address correspondence to L.J.M. (email: [email protected]).Lacey J. McIntosh Evan C. RuppellMaria BarilePublished Online:Mar 24 2023https://doi.org/10.1148/rycan.220177MoreSectionsPDF ToolsImage ViewerAdd to favoritesCiteTrack CitationsPermissionsReprints ShareShare onFacebookTwitterLinked In Sarcoid-like reactions are reported in 4%–14% of patients with cancer and are likely driven by T-cell–mediated hypersensitivity reactions forming granulomas (1). They can occur with infection, treatment, or the presence of foreign bodies (2) and have been increasingly reported in the era of precision targeted and immune checkpoint inhibitor therapy (3,4), including the use of BRAF and MEK inhibitors (Figure) (5). The classic pattern mimics sarcoidosis and involves mediastinal and hilar lymph nodes, which may be enlarged and demonstrate increased fluorodeoxyglucose (FDG) uptake at FDG PET/CT; pulmonary nodules and infiltrates may or may not be present (1,2). The incidence of extrathoracic sarcoid-like reactions is unknown but rare (6), with case reports documenting reactions including the liver, spleen, skin, central nervous system, pancreas, bone, and eyes (6,7). Absence of findings at baseline and emergence after treatment initiation can be helpful to identify a sarcoid-like reaction, although it may be indistinguishable from disease and may require tissue sampling. Radiologists should recognize these patterns to avoid confusion with disease progression. If necessary, steroids are the mainstay of treatment; immunotherapy discontinuation may be required in severe cases.Sarcoid-like reactions in a 29-year-old woman with +BRAF V600 mutation metastatic melanoma. (A, B) Baseline fluorodeoxyglucose (FDG) PET/CT maximum intensity projection (MIP) and axial fusion images show FDG-avid left axillary mass (black and white arrows), found to be metastatic melanoma (no primary cutaneous lesion identified). (C, D) Follow-up FDG PET/CT MIP and axial fusion images 8 months after left axillary nodal dissection, while receiving adjuvant dabrafenib and trametinib, show that the left axillary mass has been excised; however, there is a new FDG-avid right axillary lymph node (dashed black and white arrows). Subsequent biopsy showed lymphoid tissue with noncaseating granulomatous inflammation, consistent with atypical presentation of a sarcoid-like reaction. (E, F) Follow-up FDG PET/CT MIP and axial fusion images 2 months later show FDG-avid mediastinal and bilateral hilar lymphadenopathy (short black and white arrows), in a classic pattern of sarcoid-like reaction, which was confirmed with endobronchial US-guided fine needle aspiration cytology.Download as PowerPointOpen in Image Viewer Disclosures of conflicts of interest: L.J.M. Consultant providing clinical trial reads (Bioclinica/Clario, WorldCare Clinical, Imaging Endpoints) and content consultation (Medality) but without conflicts of interest for this article. E.C.R. No relevant relationships. M.B. No relevant relationships.Keywords: Molecular Imaging–Cancer, PET/CTAuthors declared no funding for this work.