Convolutional neural network-based program to predict lymph node metastasis of non-small cell lung cancer using 18F-FDG PET.

PURPOSE: To develop a convolutional neural network (CNN)-based program to analyze maximum intensity projection (MIP) images of 2-deoxy-2-[F-18]fluoro-D-glucose (FDG) positron emission tomography (PET) scans, aimed at predicting lymph node metastasis of non-small cell lung cancer (NSCLC), and to evaluate its effectiveness in providing diagnostic assistance to radiologists. METHODS: We obtained PET images of NSCLC from public datasets, including those of 435 patients with available N-stage information, which were divided into a training set (n = 304) and a test set (n = 131). We generated 36 maximum intensity projection (MIP) images for each patient. A residual network (ResNet-50)-based CNN was trained using the MIP images of the training set to predict lymph node metastasis. Lymph node metastasis in the test set was predicted by the trained CNN as well as by seven radiologists twice: first without and second with CNN assistance. Diagnostic performance metrics, including accuracy and prediction error (the difference between the truth and the predictions), were calculated, and reading times were recorded. RESULTS: In the test set, 67 (51%) patients exhibited lymph node metastases and the CNN yielded 0.748 predictive accuracy. With the assistance of the CNN, the prediction error was significantly reduced for six of the seven radiologists although the accuracy did not change significantly. The prediction time was significantly reduced for five of the seven radiologists with the median reduction ratio 38.0%. CONCLUSION: The CNN-based program could potentially assist radiologists in predicting lymph node metastasis by increasing diagnostic confidence and reducing reading time without affecting diagnostic accuracy, at least in the limited situations using MIP images.

Development of a novel Indium-111 radiolabeled mogamulizumab targeting CCR4 for imaging adult T-cell leukemia/lymphoma in vivo.

OBJECTIVE: Adult T-cell leukemia/lymphoma (ATL), caused by human T-cell lymphotropic virus type I (HTLV-1) infection, is among the most aggressive categories and has the worst prognosis among T-cell lymphomas. Mogamulizumab, an anti-CC chemokine receptor 4 (CCR 4), has been shown to be effective in the treatment of ATL; however, some ATL cases are often resistant, particularly the lymphoma-type ATL. To evaluate drug delivery in vivo and identify the distribution of CCR4-positive cells in the body, we developed a novel mogamulizumab tracer labeled with Indium-111 (111In) via diethylenetriaminepentaacetic acid (DTPA) for single-photon emission computerized tomography (SPECT), named [111In]In-DTPA-mogamulizumab, and evaluated its potential for visualizing CCR4 expression in vivo. METHODS: [111In]In-DTPA-mogamulizumab was added to HCT116/CCR4 or HCT116/empty vector (EV) cells, and their radioactivity was measured 1 h after administration. A blocking study was additionally performed by treating HCT116/CCR4 cells with excess mogamulizumab in addition to [111In]In-DTPA-mogamulizumab. The biodistribution and SPECT imaging of [111In]In-DTPA-mogamulizumab in HCT116/CCR4 and HCT116/EV dual-xenografted BALB/c-nu mice were evaluated for 72 h after intravenous injection. RESULTS: [111In]In-DTPA-mogamulizumab was acquired with a radiochemical purity > 95%. The cellular uptake level of [111In]In-DTPA-mogamulizumab by HCT116/CCR4 cells was significantly higher than that by HCT116/EV cells (HCT116/CCR4: 0.951 ± 0.069, HCT116/EV: 0.006 ± 0.001%dose/mg protein, p < 0.01), and the uptake was significantly suppressed by co-incubation with excess mogamulizumab (0.013 ± 0.003%dose/mg protein, p < 0.01). In the in vivo study, the radioactivity of the HCT116/CCR4 tumor tissue was significantly higher than that of the HCT116/EV tumor tissue at 72 h after the administration of [111In]In-DTPA-mogamulizumab (HCT116/CCR4: 20.5 ± 5.4, HCT116/EV: 5.7 ± 1.0%ID/g), and HCT116/CCR4 tumors were clearly and specifically visualized on SPECT imaging. CONCLUSIONS: We have successfully developed a novel SPECT imaging tracer targeting CCR4, [111In]In-DTPA-mogamulizumab, which showed good specificity and pharmacokinetics, indicating potential in visualizing CCR4 expression in vivo.

Detection efficacy of PET/CT with 18F-FSU-880 in patients with suspected recurrent prostate cancer: a prospective single-center study.

PURPOSE: Our objective was to investigate the efficacy of PET/CT with a novel prostate-specific membrane antigen (PSMA)-targeted PET probe, 18F-FSU-880, for detection and localization of recurrent disease in prostate cancer patients in whom recurrence was suspected based on an increase in plasma prostate-specific antigen (PSA) levels after initial treatment. METHODS: This study was a prospective institutional review board-approved study of 72 patients (age 56-84 years, PSA level 0.22-40.00 ng/ml) with suspected relapse of prostate cancer after primary therapy, including radical prostatectomy (RP) (n = 35) or radiation therapy (RT) (n = 37). Patients underwent PET/CT approximately 1 h and 3 h after injection of 18F-FSU-880 (101.8-380 MBq). The correlation between patient-based detection rate and Gleason score (GS) of the primary tumor and plasma PSA levels at the time of PET/CT was evaluated. Maximum standardized uptake values (SUVmax) of the positive uptakes at 1 h post-injection were compared with those at 3 h post-injection. RESULTS: In total, 51 patients (71%) showed at least one positive PSMA PET result. The PSA-stratified detection rates were 22% (2/9), 36% (4/11), 89% (16/18) and 85% (29/34) for PSA levels of 0.2 to < 0.5, 0.5 to < 1.0, 1.0 to < 2.0 and ≥ 2.0 ng/ml, respectively. The GS-stratified detection rates were 33% (2/6), 67% (16/24), 70% (16/23) and 89% (17/19) for GS 6, 7, 8 and 9, respectively. In lesion-based analysis, 157 positive lesions were detected at 3 h post-injection, 18 in the prostate or prostate bed, 65 in lymph nodes, 71 in the bone and 3 in the lung. Two local recurrences, eight pelvic lymph nodes and one distant lymph node were depicted only at 3 h post-injection. SUV max at 3 h post-injection was significantly higher than SUVmax at 1 h post-injection (p < 0.001). CONCLUSION: Our preliminary data suggest that 18F-FSU-880 might be a promising new PSMA-targeting tracer for detecting recurrence after initial treatment in patients with prostate cancer.

Association between diffuse renal uptake of 18F-FDG and acute kidney injury.

OBJECTIVE: The aim of this study was to investigate the clinical characteristics of patients with diffuse renal uptake (DRU) of 2-deoxy-2-[F-18]fluoro-D-glucose (FDG), with particular focus on renal function. METHODS: We retrospectively analyzed 40 patients who showed DRU on FDG PET/CT and the same number of matched controls. The clinical features, imaging parameters (the mean SUVmax of the kidneys and the kidney size), and laboratory data including renal function parameters (Cr, the serum creatinine level; estimated glomerular filtration ratio (eGFR); Cr-ratio, Cr divided by the baseline; max-Cr-ratio, the maximum serum creatinine level within 30 days divided by the baseline) and C-reactive protein (CRP) were compared between the two groups. In the DRU group, follow-up FDG PET/CT scans were additionally evaluated to determine the presence of DRU. RESULTS: No significant differences were observed in the clinical features except for antibiotic administration, Cr, and eGFR. Significantly more patients underwent antibiotic administration within 30 days in the DRU group (p = 0.002). The mean SUVmax of the kidneys was significantly higher (p < 0.001) and the kidney size was significantly larger in the DRU group (p = 0.003). Cr-ratio and max-Cr-ratio were significantly higher in the DRU group (p = 0.005 and p < 0.001, respectively). CRP was significantly higher in the DRU group (p < 0.001). Eighteen of the 40 patients in the DRU group underwent a second FDG PET/CT scan, and 16 of them did not show DRU. Six of the 18 patients showed acute kidney injury (AKI, i.e., Cr-ratio ≥ 1.5) at the time of the initial scan and recovered before the second scan. None of the six patients showed DRU on the second scan. CONCLUSION: DRU indicates the presence of AKI, could be a reversible finding, and may disappear as renal function improves.