This study aimed to evaluate the prognostic value of dynamic [18F]FDG PET/CT metabolic parameters and the neutrophil-to-lymphocyte ratio (NLR), individually and in combination, in patients with non-small cell lung cancer (NSCLC) receiving neoadjuvant immunotherapy plus chemotherapy. We retrospectively analyzed 168 NSCLC patients treated with neoadjuvant immunochemotherapy followed by surgery. Pre- and post-treatment PET metabolic parameters assessed by iPERCIST and peripheral blood NLR were collected. Receiver operating characteristic (ROC) analysis and Kaplan–Meier survival analysis with Cox regression were performed to identify predictors of disease-free survival (DFS). During a median follow-up of 25.2 months (95
To evaluate the performance of the deep progressive reconstruction method (DPR) for low-dose 18F-FDG PET imaging in pediatric patients. Forty-one pediatric patients who underwent clinical 18F-FDG PET/CT were retrospectively enrolled. List-mode PET data acquired at 2 min per bed position were reconstructed using OSEM and DPR, generating the standard-dose images labelled OSEM_full and DPR_full, respectively. To simulate low-dose conditions, the acquisition time of the original list-mode data was reduced to 1/n (n = 2, 3, 4) and reconstructed with DPR, yielding DPR_1/2, DPR_1/3, and DPR_1/4. For all reconstruction groups, quantitative metrics including the standard uptake value (SUV) and coefficient of variation (COV) in the liver, blood pool, and gluteus maximus, as well as lesion tumour-to-background ratio (TBR) and contrast-to-noise ratio (CNR). Subjective image quality was evaluated based on overall image quality, lesion conspicuity, and image noise. Age-stratified subgroup analysis was performed, with OSEM_full as the reference standard. DPR reconstructions showed good agreement in quantitative accuracy with reference imaging. Compared with OSEM_full, DPR_full and DPR_1/2 significantly lowered background COV, DPR_1/3 showed comparable background noise, while DPR_1/4 presented increased noise. All DPR groups yielded significantly higher lesion TBR, and DPR_full, DPR_1/2, DPR_1/3 improved lesion CNR, whereas DPR_1/4 showed reduced CNR. Subjectively, DPR_full and DPR_1/2 achieved significantly higher image quality scores, with over 60
High-energy photons emitted by 123I may penetrate the septa of a collimator, thereby degrading quantitative accuracy. Collimator choice affects the evaluation of the heart-to-mediastinum (H/M) ratio in cardiac sympathetic imaging with 123I-metaiodobenzylguanidine (123I-MIBG). We investigated the validity of previously reported convenient methods for correcting collimator-dependent differences in the H/M ratio. Twenty patients who underwent cardiac 123I-MIBG imaging for evaluation of neurodegenerative diseases on a gamma camera system provided by GE Healthcare were studied. Anterior chest images were acquired successively with three collimators: the low-energy high-resolution and sensitivity (LEHRS), extended low-energy general-purpose (ELEGP), and medium-energy general-purpose (MEGP) collimators. The H/M ratios were estimated using various correction methods: 123I-dual-window (IDW) correction using the high-energy subwindow, empiric correction based on 123I point-source imaging, combined correction using the high-energy subwindow and point-source imaging, and phantom-based method using imaging of a dedicated phantom. Empiric correction and combined correction had been established using patient data mainly obtained with a system provided by Siemens Healthineers. The uncorrected H/M ratios obtained with the LEHRS collimator correlated closely with those obtained with the MEGP collimator but were substantially smaller, indicating severe underestimation with the LEHRS collimator, attributable to septal penetration. The underestimation remained but decreased after any correction. The H/M ratios after correction were largest with combined correction, followed by empiric correction, phantom-based correction, and IDW correction. The absolute errors after correction were smallest for combined correction, followed by empiric correction, phantom-based correction, and IDW correction. The underestimation was substantially smaller with the ELEGP collimator than with the LEHRS collimator, and any type of correction reduced the underestimation. Convenient correction methods using 123I point-source imaging were valid for reducing collimator-dependent differences in the H/M ratio in cardiac 123I-MIBG imaging, demonstrating their applicability to a GE system.
Objective Whole-body PET metrics quantify the magnitude of metastatic burden but ignore its anatomical distribution. Whether resolving metabolic burden according to anatomical compartment provides incremental prognostic information beyond conventional whole-body PET metrics remains unknown. We therefore tested whether the anatomical distribution of metabolic burden carries prognostic information beyond aggregate whole-body load in de novo metastatic HER2-positive breast cancer.Methods Fifty-seven consecutive treatment-naive patients who underwent baseline & sup1;F-8-FDG PET/CT before first-line anti-HER2 therapy were retrospectively analyzed. Compartment-specific metabolic parameters and clinical variables were evaluated using multivariable Cox regression. Model discrimination was internally validated over 1000 bootstrap resamples, and the incremental value of compartment-specific burden over whole-body metrics was evaluated with likelihood-ratio tests.Results Twenty patients (35.1%) died during a median follow-up of 50.4 months. LN_MTV, Bone_TLG, and HR status independently predicted OS, yielding a model with a modest optimism-corrected C-index of 0.685 when the entire variable-selection sequence was repeated within each bootstrap resample. Adding compartment-specific metabolic burden improved models based on whole-body MTV or TLG (p = 0.003 and p = 0.001, respectively), whereas whole-body MTV provided no additional prognostic information once compartment-specific burden was considered (p = 0.90). Bone_TLG, but not binary bone involvement, independently predicted OS, and its prognostic contribution increased over time.Conclusion In this exploratory analysis, the anatomical distribution of metabolic tumor burden appeared to carry prognostic information beyond conventional whole-body PET metrics. These findings are hypothesis-generating, and prospective multicenter validation is warranted before compartment-specific metabolic phenotyping could inform clinical risk stratification in HER2-positive metastatic breast cancer.
Objective This Phase I/IIa study evaluated the safety, pharmacokinetics, internal radiation dosimetry, and image quality of cyclotron-produced Ga-6(8)-PSMA-11 synthesized using an automated module in a Japanese population. Methods In this single-center trial, nine Japanese men (3 healthy volunteers, 6 prostate cancer patients) received a single intravenous injection of Ga-6(8)-PSMA-11 (4.0 MBq/kg). Safety, pharmacokinetics (blood/urine), and internal radiation dosimetry (serial PET/CT) were assessed. Diagnostic performance was evaluated by lesion detection rates and image quality by quantitative metrics, including noise-equivalent count (NEC), to optimize the protocol. Results No adverse events were observed. The automated synthesis process consistently yielded high-quality Ga-6(8)-PSMA-11, with a decay-corrected radiochemical yield of 2841 +/- 584 MBq and a radiochemical purity > 99%. The kidneys received the highest absorbed dose, with an effective dose comparable to that reported in published data. The agent demonstrated rapid clearance, primarily via renal excretion (58.99% +/- 11.34% in 6 h), with a biological half-life of approximately 3.8 h in whole blood. All histopathology-confirmed primary lesions were detected. Quantitative analysis indicated that high image quality was maintained even with substantially reduced acquisition times. Conclusions Cyclotron-produced Ga-6(8)-PSMA-11 is safe and exhibits favorable pharmacokinetics and dosimetry in a Japanese population. This exploratory Phase I/IIa trial suggests that the production method appears feasible.
