To evaluate the prognostic values of 18F-FDG PET-derived whole-body imaging features in patients with metastatic lung adenocarcinoma treated with EGFR-targeted therapies. We retrospectively analyzed 249 patients with lung adenocarcinoma who underwent pre-treatment 18F-FDG PET and were treated with EGFR-targeted agents. The patients were divided into analog (n = 150) and digital (n = 99) PET cohorts. Whole-body and primary tumor respiratory-stable imaging features were extracted. The prognostic values of the study variables for progression-free (PFS) and overall survival (OS) were assessed using univariate and multivariate Cox regression analyses across the analog and digital PET cohorts. Total sphericity and primary tumor inverse difference normalized were independent predictors of both PFS and OS. The total metabolic tumor volume was another independent predictor of OS. Combined models integrating these imaging biomarkers with clinical variables outperformed the traditional staging system (c-indices for PFS: 0.649 versus 0.550 for analog and 0.668 versus 0.583 for digital PET cohorts; for OS: 0.694 versus 0.562 for analog and 0.728 versus 0.579 for digital PET cohorts). Our models showed consistent predictive values across subgroups based on sex, EGFR mutation subtype, and clinical stage. Our results indicate that models integrating whole-body 18F-FDG PET features with traditional variables can enhance survival prediction and may support personalized treatment strategies for patients with lung adenocarcinoma treated with EGFR-targeted therapies.

Neuroblastoma is the most common extracranial solid tumor in children and often requires complex, multimodal treatment. A major advancement in its treatment has been the development of metaiodobenzylguanidine (MIBG), a norepinephrine analog, for diagnostic imaging and targeted radiotherapy. This review provides a comprehensive overview of the clinical applications and recent advancements of 131I-MIBG in neuroblastoma treatment. The agent has exhibited substantial efficacy in newly diagnosed and relapsed or refractory disease, through strategies such as monotherapy, tandem dosing, and high-dose administration with stem cell transplantation support. Combination regimens incorporating radiosensitizers, chemotherapeutic agents, and novel targeted approaches-including histone deacetylase inhibitors and immunotherapy-have further broadened its therapeutic scope. Challenges remain, including nonavid tumors, radiation-related risks, and logistical constraints. However, with the recent regulatory approval of 131I-MIBG for neuroblastoma treatment in Japan, its role is expected to expand, paving the way for improved outcomes and more personalized, targeted approaches in pediatric oncology.

This study aimed to develop deep learning (DL) models for CT-free attenuation correction and Monte Carlo-based scatter correction in 99mTc-macroagregated albumin (99mTc-MAA) SPECT imaging, with the goal of enhancing quantitative accuracy for improved treatment planning and pre-therapy dosimetry in 90Y-selctive internal radiation therapy (SIRT). Data from 222 patients who underwent 99mTc-MAA SPECT imaging prior to 90Y-SIRT were included in this study. Uncorrected SPECT images (without attenuation and/or scatter correction) were used as input to a modified 3D shifted-window UNet Transformer (Swin UNETR) architecture. Three separate models were trained to predict attenuation corrected (AC), scatter corrected (SC), and joint attenuation and scatter corrected (ASC) SPECT images. The dataset was split into a training set (~ 80%) and an independent test set (~ 20%). Model training was performed using a five-fold cross-validation framework, with final evaluation conducted on the blind test set. To clinically assess model performance, 3D voxel-wise dosimetry was performed on the test set using the local energy deposition method, assuming 99mTc-MAA as a surrogate for 90Y distribution. Quantitative evaluation included organ- and voxel-level metrics, along with Gamma analysis using three combinations of distance-to-agreement (DTA, mm) and dose-difference (DD, %) criteria. The average (± SD) of the voxel-wise mean error (ME) was ≤ 0.003Gy for all tasks. The Relative Error (RE (%)) for AC, SC, and ASC tasks were 4.64 ± 7.52%, 8.99 ± 26.35%, and 16.45 ± 25.83%, respectively. Voxel-level Gamma evaluations within the whole body using three different criteria sets, including "DTA: 4.79mm, DD: 1%"; "DTA: 10mm, DD: 5%"; and "DTA: 15mm, DD: 10%" yielded pass rates of over 99.60%. The mean absolute error (MAE) for lesions, normal liver and lungs across all tasks were 3.16 ± 3.39, 0.35 ± 0.36, 0.41 ± 0.47Gy for AC, 1.97 ± 2.79, 0.19 ± 0.16, 0.22 ± 0.20Gy, for SC and 5.16 ± 7.10, 0.45 ± 0.51, and 0.34 ± 0.37Gy for ASC, respectively. Multiple models were developed for key SPECT quantification tasks, with potential value in clinical setting lacking reliable CT data or sufficient computational resources for Monte Carlo simulations. The models look promising for potential clinical translation and integration into commercial reconstruction software.

This study aims to systematically evaluate the inter- and intra-observer agreement regarding lesions with uncertain malignancy potential in Ga-68 PSMA PET/CT imaging of prostate cancer patients, utilizing the PSMA-RADS 2.0 classification system, and to emphasize the malignancy evidence associated with these lesions. We retrospectively reviewed Ga-68 PSMA PET/CT images of patients diagnosed with prostate cancer via histopathology between December 2016 and November 2023. Seventy-six patients and 98 target lesions, classified with PSMA-RADS scores 3A-3D by an experienced nuclear medicine specialist, were included. Three observers (two experienced and one less experienced nuclear medicine physician) evaluated the lesions at two different time points, three months apart, in a blinded manner. Intra-observer agreement was assessed using Cohen's kappa, while inter-observer agreement was analyzed using Fleiss' kappa. All observers demonstrated significant intra-observer agreement, with percentages ranging from 67.3 to 80.6% and kappa values between 0.62 and 0.76. The highest intra-observer agreement was noted for bone lesions, while non-regional lymph nodes exhibited the lowest agreement. For inter-observer assessments, moderate agreement was observed across all lesions (Fleiss k = 0.51), with the highest agreement in regional lymph nodes and the lowest in non-regional lymph nodes and bone lesions. Follow-up results indicated malignancy detection rates of 63.4% for 3A, 47% for 3B, 44.5% for 3C, and 41.4% for 3D scored lesions. The PSMA-RADS classification system demonstrated substantial intra-observer agreement and moderate inter-observer agreement for lesions with uncertain malignancy potential, supporting its clinical applicability. However, notable variability and limitations in reporting bone lesions, non-regional lymph nodes, and lung lesions were observed among readers. The ambiguous nature and management of PSMA-RADS 3 lesions underscore the need for further research to enhance risk stratification.

ObjectivesThe identification of metastatic disease in patients with biochemical failure of prostate cancer (PCa) after radical prostatectomy (RP) determines subsequent treatment management. Our objectives were (i) to assess the prevalence of metastatic PCa in patients with biochemical failure following RP and PSA levels ≤ 1 ng/mL, as detected by 18F-PSMA-1007 PET/CT (PSMA-PET), and (ii) to identify predictors of metastatic disease.MethodsFifty-five patients with biochemical recurrence (BCR, n = 47) or persistent disease (n = 8) following RP as their primary and only prior treatment were retrospectively included if presenting with PSA levels ≤ 1 ng/mL. Patients who received any other anticancer treatment were excluded. PSMA-PET findings were categorized as either local recurrence (i.e., prostatic fossa and seminal vesicles) or metastases. Predictors of PSMA-PET positivity were assessed with univariate and multivariate regression analyses both in the whole sample and in two subgroups defined according to PSA levels (Group A = PSA < 0.5 ng/mL; Group B = PSA between 0.5 and 1 ng/mL).ResultsMedian PSA at the time of PET/CT was 0.37 ng/mL (range: 0.13–1.0). PSMA-PET was positive in 22/55 (40%) patients, 14/55 (25%) patients had metastatic disease. Overall, 31 PSMA-positive lesions were identified: 12/31 (39%) and 19/31 (61%) were local recurrences and metastases, respectively. In the whole cohort, ISUP grade > 3 (p = 0.003), pN1 status after surgery (p = 0.011), time to BCR ≤ 26 months (p = 0.03), and persistent disease (p = 0.003) were significantly associated with a higher rate of PSMA-positive metastases. At multivariate analysis, ISUP grade > 3 (p = 0.016) was the only independent predictor of metastases. Metastatic disease was detected in 8/38 (21%) patients in Group A and in 6/17 (35%) patients in Group B, respectively. In Group A, pN1 (p = 0.043) and persistent disease (p = 0.040) were significant predictors of metastases. In Group B, ISUP grade > 3 was the only predictor (p = 0.028).ConclusionsISUP grade > 3, time to BCR ≤ 26 months, pN1 status, and persistent disease after surgery indicated a higher likelihood of PSMA-positive metastatic disease in a homogeneous cohort of patients with biochemical failure and low PSA values. pN1 status and persistent disease were significant predictors of metastatic disease also in patients with PSA levels < 0.5 ng/mL.Supplementary InformationThe online version contains supplementary material available at 10.1007/s12149-026-02153-9.

To determine whether software-based de-filtering can restore the quantitative accuracy of the bone scan index (BSI) and the number of hot spots (HSn) in whole-body scintigraphy images degraded by the Clarity2D noise-reduction filter. In this IRB-approved retrospective study, 101 adults (mean age ± SD: 67 ± 13 years) who underwent 99mTc-HMDP whole-body scintigraphy on a cadmium-zinc-telluride (CZT) SPECT/CT system were analyzed. For each patient, three planar datasets were obtained: (i) unfiltered images, (ii) 40%-blend Clarity2D-filtered images, and (iii) software de-filtered images reconstructed with a deep learning-based inverse filter in VSBONE BSI v3.0. Quantitative indices (BSI and HSn) and lesion masks were automatically extracted. Agreement with the unfiltered reference was evaluated using Pearson correlation, Bland-Altman analysis (bias ± 95% limits), Dice coefficient, and the Hausdorff distance (p < 0.05). Additionally, lesion detection accuracy was quantified using intersection over union (IoU)-based matching to calculate precision, recall, and F1-score. Clarity2D filtering significantly impaired quantitative concordance (BSI r = 0.23, bias = - 1.55 [-6.20 to 3.10]; HSn r = 0.23, bias = - 14.4 lesions). In contrast, de-filtering restored concordance (BSI r = 0.99, bias = - 0.04 [-0.26 to 0.17]; HSn r = 0.98, bias = - 0.04 lesions) and improved spatial overlap (Dice 0.40 to 0.82) while reducing the median Hausdorff distance from 103 pixels (IQR 85-188) to 39 pixels (IQR 1-40) (all p < 0.001). The de-filtered method demonstrated superior lesion detection accuracy compared to Clarity2D (Precision: 0.77 ± 0.37 vs. 0.19 ± 0.29, Recall: 0.81 ± 0.37 vs. 0.43 ± 0.44, F1-score: 0.78 ± 0.36 vs. 0.23 ± 0.30). Furthermore, de-filtering achieved high inter-case stability (median F1-score: 1.0), whereas Clarity2D showed substantial variability (median F1-score: 0.057). The proposed de-filtering algorithm reliably reverses Clarity2D-induced distortions, enabling accurate BSI and HSn measurements and robust lesion detection without additional radiation or acquisition time. This technique has the potential to broaden the clinical adoption of noise-reduction filters while preserving the integrity of downstream quantitative analyses.