Journal of Nuclear Medicine最新文献

The 18-kDa translocator protein (TSPO) is gaining recognition as a relevant target in glioblastoma imaging. However, data on the potential prognostic value of TSPO PET imaging in glioblastoma are lacking. Therefore, we investigated the association of TSPO PET imaging results with survival outcome in a homogeneous cohort of glioblastoma patients. Methods: Patients were included who had newly diagnosed, histologically confirmed isocitrate dehydrogenase (IDH)-wild-type glioblastoma with available TSPO PET before either normofractionated radiotherapy combined with temozolomide or hypofractionated radiotherapy. SUV_max on TSPO PET, TSPO binding affinity status, tumor volumes on MRI, and further clinical data, such as O ⁶-alkylguanine DNA methyltransferase (MGMT) and telomerase reverse transcriptase (TERT) gene promoter mutation status, were correlated with patient survival. Results: Forty-five patients (median age, 63.3 y) were included. Median SUV_max was 2.2 (range, 1.0-4.7). A TSPO PET signal was associated with survival: High uptake intensity (SUV_max > 2.2) was related to significantly shorter overall survival (OS; 8.3 vs. 17.8 mo, P = 0.037). Besides SUV_max, prognostic factors for OS were age (P = 0.046), MGMT promoter methylation status (P = 0.032), and T2-weighted MRI volume (P = 0.031). In the multivariate survival analysis, SUV_max in TSPO PET remained an independent prognostic factor for OS (P = 0.023), with a hazard ratio of 2.212 (95% CI, 1.115-4.386) for death in cases with a high TSPO PET signal (SUV_max > 2.2). Conclusion: A high TSPO PET signal before radiotherapy is associated with significantly shorter survival in patients with newly diagnosed IDH-wild-type glioblastoma. TSPO PET seems to add prognostic insights beyond established clinical parameters and might serve as an informative tool as clinicians make survival predictions for patients with glioblastoma.

For patients with advanced-stage metastatic castration-resistant prostate cancer (mCRPC) who do not respond to [¹⁷⁷Lu]Lu-PSMA therapy, there are limited treatment options. Clinical results obtained with [²²⁵Ac]Ac-PSMA are promising. We retrospectively analyzed the outcomes of patients treated with [²²⁵Ac]Ac-PSMA between December 2018 and October 2022. Methods: We evaluated the treatment results of 23 patients (mean age, 70.3 ± 8.8 y) with mCRPC who were refractory to treatment with [¹⁷⁷Lu]Lu-PSMA (2-9 cycles). The safety profile was assessed according to Common Technology Criteria for Adverse Events version 5.0. Treatment efficacy was assessed using prostate-specific membrane antigen PET progression criteria and prostate-specific antigen (PSA) response according to Prostate Cancer Working Group 2 criteria after the first cycle of [²²⁵Ac]Ac-PSMA treatment. Results: All patients received androgen-deprivation therapy, whereas 22 (96%) and 19 (83%) patients received chemotherapy and second-generation antiandrogen therapy, respectively. One patient received 4 cycles, 2 received 3 cycles, 8 received 2 cycles, and 12 received 1 cycle of [²²⁵Ac]Ac-PSMA. The median interval between cycles was 13 wk (range, 8-28 wk). [²²⁵Ac]Ac-PSMA was administered with a mean activity of 7.6 MBq (range, 6.2-10.0 MBq) in each cycle. Patients were at an advanced stage of disease, and tumor burden was very high. Although the best PSA response was observed in 5 patients (26%) after [²²⁵Ac]Ac-PSMA treatment, there was at least some level of decline in PSA observed in 11 patients (58%; n = 19). Treatment response was assessed in patients who underwent [⁶⁸Ga]Ga-PSMA PET/CT imaging. After the first cycle of treatment (n = 18), 50% of patients (n = 9) showed disease progression according to prostate-specific membrane antigen PET progression criteria, and the disease control rate was calculated to be 50%. Median progression-free survival was 3.1 mo, and median overall survival was 7.7 mo. Grade 3 hematologic toxicity occurred in 1 patient, and grade 3 nephrotoxicity was observed in another patient. Parotid SUV_max decreased by 33%, although all patients complained of dry mouth before treatment. Conclusion: We observed that [²²⁵Ac]Ac-PSMA therapy was safe and showed potential even in cases with advanced-stage mCRPC in which all other treatment options were completed.

The deployment of artificial intelligence (AI) has the potential to make nuclear medicine and medical imaging faster, cheaper, and both more effective and more accessible. This is possible, however, only if clinicians and patients feel that these AI medical devices (AIMDs) are trustworthy. Highlighting the need to ensure health justice by fairly distributing benefits and burdens while respecting individual patients' rights, the AI Task Force of the Society of Nuclear Medicine and Molecular Imaging has identified 4 major ethical risks that arise during the deployment of AIMD: autonomy of patients and clinicians, transparency of clinical performance and limitations, fairness toward marginalized populations, and accountability of physicians and developers. We provide preliminary recommendations for governing these ethical risks to realize the promise of AIMD for patients and populations.

This study aimed to develop an analytic approach based on [¹⁸F]FDG PET radiomics using stacking ensemble learning to improve the outcome prediction in diffuse large B-cell lymphoma (DLBCL). Methods: In total, 240 DLBCL patients from 2 medical centers were divided into the training set (n = 141), internal testing set (n = 61), and external testing set (n = 38). Radiomics features were extracted from pretreatment [¹⁸F]FDG PET scans at the patient level using 4 semiautomatic segmentation methods (SUV threshold of 2.5, SUV threshold of 4.0 [SUV4.0], 41% of SUV_max, and SUV threshold of mean liver uptake [PERCIST]). All extracted features were harmonized with the ComBat method. The intraclass correlation coefficient was used to evaluate the reliability of radiomics features extracted by different segmentation methods. Features from the most reliable segmentation method were selected by Pearson correlation coefficient analysis and the LASSO (least absolute shrinkage and selection operator) algorithm. A stacking ensemble learning approach was applied to build radiomics-only and combined clinical-radiomics models for prediction of 2-y progression-free survival and overall survival based on 4 machine learning classifiers (support vector machine, random forests, gradient boosting decision tree, and adaptive boosting). Confusion matrix, receiver-operating-characteristic curve analysis, and survival analysis were used to evaluate the model performance. Results: Among 4 semiautomatic segmentation methods, SUV4.0 segmentation yielded the highest interobserver reliability, with 830 (66.7%) selected radiomics features. The combined model constructed by the stacking method achieved the best discrimination performance. For progression-free survival prediction in the external testing set, the areas under the receiver-operating-characteristic curve and accuracy of the stacking-based combined model were 0.771 and 0.789, respectively. For overall survival prediction, the stacking-based combined model achieved an area under the curve of 0.725 and an accuracy of 0.763 in the external testing set. The combined model also demonstrated a more distinct risk stratification than the International Prognostic Index in all sets (log-rank test, all P < 0.05). Conclusion: The combined model that incorporates [¹⁸F]FDG PET radiomics and clinical characteristics based on stacking ensemble learning could enable improved risk stratification in DLBCL.

Upregulation of prostate-specific membrane antigen (PSMA) in neovasculature has been described in glioblastoma multiforme (GBM), whereas vasculature in nonaffected brain shows hardly any expression of PSMA. It is unclear whether PSMA-targeting tracer uptake on PET is based on PSMA-specific binding to neovasculature or aspecific uptake in tumor. Here, we quantified uptake of various PSMA-targeting tracers in GBM and correlated this with PSMA expression in tumor biopsy samples from the same patients. Methods: Fourteen patients diagnosed with de novo (n = 8) or recurrent (n = 6) GBM underwent a preoperative PET scan after injection of 1.5 MBq/kg [⁶⁸Ga]Ga-PSMA-11 (n = 7), 200 MBq of [¹⁸F]DCFpyl (n = 3), or 200 MBq of [¹⁸F]PSMA-1007 (n = 4). Uptake in tumor and tumor-to-background ratios, with contralateral nonaffected brain as background, were determined. In a subset of patients, PSMA expression levels from different regions in the tumor tissue samples (n = 40), determined using immunohistochemistry (n = 35) or RNA sequencing (n = 13), were correlated with tracer uptake on PET. Results: Moderate to high (SUV_max, 1.3-20.0) heterogeneous uptake was found in all tumors irrespective of the tracer type used. Uptake in nonaffected brain was low, resulting in high tumor-to-background ratios (6.1-359.0) calculated by dividing SUV_max of tumor by SUV_max of background. Immunohistochemistry showed variable PSMA expression on endothelial cells of tumor microvasculature, as well as on dispersed individual cells (of unknown origin), and granular staining of the neuropil. No correlation was found between in vivo uptake and PSMA expression levels (for immunohistochemistry, r = -0.173, P = 0.320; for RNA, r = -0.033, P = 0.915). Conclusion: Our results indicate the potential use of various PSMA-targeting tracers in GBM. However, we found no correlation between PSMA expression levels on immunohistochemistry and uptake intensity on PET. Whether this may be explained by methodologic reasons, such as the inability to measure functionally active PSMA with immunohistochemistry, tracer pharmacokinetics, or the contribution of a disturbed blood-brain barrier to tracer retention, should still be investigated.

CT and bone scintigraphy are not useful for response evaluation of bone metastases to ²²³Ra treatment in metastatic castration-resistant prostate cancer (mCRPC). PET using ⁶⁸Ga prostate-specific membrane antigen 11 (⁶⁸Ga-PSMA) is a promising tool for response evaluation of mCRPC. The aim of this study was to determine the utility of ⁶⁸Ga-PSMA PET/CT for response evaluation of ²²³Ra treatment in patients with mCRPC. Methods: Within this prospective, multicenter, imaging discovery study, 28 patients with mCRPC, eligible for ²²³Ra treatment, were included between 2019 and 2022. Patients received ²²³Ra according to the standard of care. Study procedures included CT, bone scintigraphy, and ⁶⁸Ga-PSMA PET/CT at baseline, after 3 and 6 cycles of ²²³Ra treatment, and on treatment failure. Response to ²²³Ra treatment was visually assessed on all 3 imaging modalities. Total tumor volume within bone (TTV_bone) was determined on ⁶⁸Ga-PSMA PET/CT. Intrapatient heterogeneity in response was studied using a newly developed image-registration tool for sequential images of PET/CT. Results were compared with failure-free survival (good responders vs. poor responders; cutoff, 24 wk) and alkaline phosphatase (ALP) response after 3 cycles. Results: Visual response assessment criteria could not distinguish good responders from poor responders on ⁶⁸Ga-PSMA PET/CT and bone scintigraphy. For ⁶⁸Ga-PSMA PET/CT, TTV_bone at baseline was lower in good responders than in poor responders, whereas TTV_bone increased in both groups during treatment. TTV_bone was higher in patients with new extraosseous metastases during ²²³Ra treatment. Although TTV_bone and ALP correlated at baseline, changes in TTV_bone and ALP on treatment did not. ⁶⁸Ga-PSMA response of TTV_bone showed intrapatient heterogeneity in most patients. Conclusion: mCRPC patients with lower TTV_bone on ⁶⁸Ga-PSMA PET/CT have the best clinical outcome after ²²³Ra treatment. Response is highly heterogeneous in most patients. A decrease in ALP, which occurred in most patients, was not correlated with a decrease in TTV_bone, which might make one question the value of ALP for disease monitoring during ²²³Ra treatment in clinical practice.

Shortwave infrared (900-1,700 nm) fluorescence imaging (SWIRFI) has shown significant advantages over visible (400-650 nm) and near-infrared (700-900 nm) fluorescence imaging (reduced autofluorescence, improved contrast, tissue resolution, and depth sensitivity). However, there is a major lag in the clinical translation of preclinical SWIRFI systems and targeted SWIRFI probes. Methods: We preclinically show that the pH low-insertion peptide conjugated to indocyanine green (pHLIP ICG), currently in clinical trials, is an excellent candidate for cancer-targeted SWIRFI. Results: pHLIP ICG SWIRFI achieved picomolar sensitivity (0.4 nM) with binary and unambiguous tumor screening and resection up to 96 h after injection in an orthotopic breast cancer mouse model. SWIRFI tumor screening and resection had ambient light resistance (possible without gating or filtering) with outstanding signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) values at exposures from 10 to 0.1 ms. These SNR and CNR values were also found for the extended emission of pHLIP ICG in vivo (>1,100 nm, 300 ms). Conclusion: SWIRFI sensitivity and ambient light resistance enabled continued tracer clearance tracking with unparalleled SNR and CNR values at video rates for tumor delineation (achieving a tumor-to-muscle ratio above 20). In total, we provide a direct precedent for the democratic translation of an ambient light resistant SWIRFI and pHLIP ICG ecosystem, which can instantly improve tumor resection.