Paired 18F-Fluorodeoxyglucose (18F-FDG), and 64Cu-Copper(II)-diacetyl-bis(N(4)-methylthiosemicarbazone) (64Cu-ATSM) PET Scans in Dogs with Spontaneous Tumors and Evaluation for Hypoxia-Directed Therapy

Hypoxia is associated with neoplastic tissue, protecting cancer cells from death by irradiation and chemotherapy. Identification of hypoxic volume of tumors could optimize patient selection for hypoxia-directed medical, immunological, and radiation therapies. Clostridium novyi-NT (CNV-NT) is an oncolytic bacterium derived from attenuated wild-type Clostridium novyi spores, which germinates exclusively in the anaerobic core of tumors with low-oxygen content. The hypothesis was that 64Cu-ATSM would localize to regions of hypoxia, and that greater hypoxic volume would result in greater germination of Clostridium novyi-NT (CNV-NT). Tumor-bearing companion dogs were recruited to a veterinary clinical trial. Dogs received a CT scan, 18F-FDG PET scan (74 MBq) and 64Cu-ATSM PET scan (74 MBq). Scan regions of interest were defined as the highest 20% of counts/voxel for each PET scan, and regions with voxels overlapping between the two scans. Maximum standardized uptake value (MaxSUV) and threshold volume were calculated. Direct oximetry was performed in select tumors. Tumor types evaluated included nerve sheath tumor (10), apocrine carcinoma (1), melanoma (3) and oral sarcoma (6). MaxSUVATSM ranged from 0.3–6.6. Measured oxygen tension ranged from 0.05–89.9 mmHg. Inverse of MaxSUVATSM had a linear relationship with oxygen tension (R2 = 0.53, P = 0.0048). Hypoxia <8 mmHg was associated with an SUVATSM > 1.0. Hypoxic volume ranged from 0 to 100% of gross tumor volume (GTV) and MaxSUVATSM was positively correlated with hypoxic volume (R = 0.674; P = 0.0001), but not GTV (P = 0.182). Tumor hypoxic volume was heterogeneous in location and distribution. 64Cu-ATSM-avid regions were associated with differential CT attenuation. Hypoxic volume did not predict CNV-NT germination. 64Cu-ATSM PET scanning predicts hypoxia patterns within spontaneously occurring tumors of dogs as measured by direct oxymetry. Total tumor volume does not accurately predict degree or proportion of tumor hypoxia.

Primary Parotid Non-Hodgkin’s B-Cell Lymphoma in an Elderly Woman on Pembrolizumab for Metastatic Invasive Papillary Urothelial Cell Carcinoma of the Bladder

Primary lymphomas of the parotid are rare (4-5%) and seldom appear in patients with pre-existing metastatic cancer from other primary sources. We present a primary marginal zone B-cell lymphoma of the mucosaassociated lymphoid tissue (MALT) in an 84-year-old female with preexisting metastatic bladder cancer. A PET scan that identified positive pelvic/cervical lymphadenopathy and bilateral parotid masses. She underwent transurethral resection of a bladder tumor and started on pembrolizumab chemotherapy. After two years, the left parotid mass decreased in size but the right increased to 3.9cm. Right superficial parotidectomy diagnosed B-Cell Marginal zone lymphoma, staining positive for CD20, PAX5, and Bcl2. pembrolizumab was held and the patient was started on 4 weekly rituximab infusions. A PET scan done 3 months after completion of rituximab showed a good response to chemotherapy. This is the first reported case of a primary parotid gland lymphoma in a patient with active metastatic bladder cancer.

Acute ischemia of the lower limb due to non-bacterial thrombotic endocarditis with recent venous thromboembolic disease as the initial manifestation of lung adenocarcinoma: A case report

Background Non-bacterial thrombotic endocarditis (NBTE) is a rare condition. Optimal management is based on early diagnosis which remains difficult. Case summary A 75-year-old male patient was admitted to the hospital with acute ischemia of the left lower limb due to popliteal artery occlusion despite anticoagulation with rivaroxaban for pulmonary embolism diagnosed two weeks earlier. Transoesophageal echocardiography (TEE) showed a mobile vegetation with mild mitral valve regurgitation. Positron emission tomography (PET) scan did not show hyperfixation at the mitral valve but rather lymphadenopathy hyperfixation at different sites. Biopsy of a lymph node from Barety’s space identified a bronchopulmonary adenocarcinoma. The outcome was favorable after popliteal artery thrombectomy and low-molecular-weight heparin treatment. The patient was referred to the department of onco-pneumology for further care. Discussion Upon clinical presentation, the combination of an arterial and prior venous thrombotic event suggested that the origin could be either a patent foramen ovale (PFO) or a thrombosis from an underlying cancer. A transthoracic echocardiography and TEE excluded a PFO and demonstrated a mobile echogenic mass at the mitral valve site together with a mild regurgitation. The diagnosis of non-bacterial thrombotic endocarditis was suggested given the absence of clinical and biological infectious signs, negative blood cultures and serology for endocarditis, the presence of both arterial and venous thrombosis, as well as the presence of intra-thoracic lymphadenopathy hyperfixation on the PET scan for which a biopsy demonstrated lung adenocarcinoma.

Performance evaluation in [18F]Florbetaben brain PET images classification using 3D Convolutional Neural Network

High accuracy has been reported in deep learning classification for amyloid brain scans, an important factor in Alzheimer’s disease diagnosis. However, the possibility of overfitting should be considered, as this model is fitted with sample data. Therefore, we created and evaluated an [18F]Florbetaben amyloid brain positron emission tomography (PET) scan classification model with a Dong-A University Hospital (DAUH) dataset based on a convolutional neural network (CNN), and performed external validation with the Alzheimer’s Disease Neuroimaging Initiative dataset. Spatial normalization, count normalization, and skull stripping preprocessing were performed on the DAUH and external datasets. However, smoothing was only performed on the external dataset. Three types of models were used, depending on their structure: Inception3D, ResNet3D, and VGG3D. After training with 80% of the DAUH dataset, an appropriate model was selected, and the rest of the DAUH dataset was used for model evaluation. The generalization potential of the selected model was then validated using the external dataset. The accuracy of the model evaluation for Inception3D, ResNet3D, and VGG3D was 95.4%, 92.0%, and 97.7%, and the accuracy of the external validation was 76.7%, 67.1%, and 85.3%, respectively. Inception3D and ResNet3D were retrained with the external dataset; then, the area under the curve was compared to determine the binary classification performance with a significance level of less than 0.05. When external validation was performed again after fine tuning, the performance improved to 15.3%p for Inception3D and 16.9%p for ResNet3D. In [18F]Florbetaben amyloid brain PET scan classification using CNN, the generalization potential can be seen through external validation. When there is a significant difference between the model classification performance and the external validation, changing the model structure or fine tuning the model can help improve the classification performance, and the optimal model can also be found by collaborating through a web-based open platform.