The Diagnostic Value of 18F-FDG PET/CT Bone Marrow Uptake Pattern in Detecting Bone Marrow Involvement in Pediatric Neuroblastoma Patients

Objectives. To explore the diagnostic value of 18F-FDG PET/CT bone marrow uptake pattern (BMUP) in detecting bone marrow involvement (BMI) in pediatric neuroblastoma (NB) patients. Methods. Ninety-eight NB patients were enrolled in BMI analysis. Four patterns of bone marrow uptake were categorized based on pretreatment cF-FDG PET/CT images. Some crucial inspection indexes and 18F-FDG PET/CT metabolic parameters were analyzed. The BMUP was divided into BMUP1, BMUP2, BMUP3, and BMUP4. Paired-like homeobox 2b (PHOX2B) of bone marrow and blood, bone marrow biopsy (BMB) result, and 18F-FDG PET/CT were compared to detect BMI. All patients were followed up for at least six months. Results. BMUP had excellent consistency among different physicians. Kappa coefficients of two residents and two attending physicians and between the resident and attending physician, were 0.857, 0.891, and 0.845, respectively. The optimal cut-off value of SUVmax-Bone/Liver was 2.08 to diagnose BMI for BMUP3 patients, and the area under curve (AUC) was 0.873. AUC of PHOX2B of bone marrow (PHOX2B of BM), PHOX2B of blood, BMB, and 18F-FDG PET/CT were 0.916, 0.811, 0.806, and 0.904, respectively. There was no significant difference between PHOX2B of BM and PET/CT. Positive predictive value, negative predictive value, sensitivity, and specificity in diagnosis of BMI were 92.9%, 92.9%, 97.0%, and 83.9% for PET/CT and 96.7%, 80.6%, 89.6%, and 93.5% for PHOX2B of BM, respectively. Conclusions. BMUP of pretreatment 18F-FDG PET/CT is a simple and practical method, which has a relatively high diagnostic efficiency in detecting BMI and might decrease unnecessary invasive inspections in some pediatric NB patients.

11C-acetate positron emission tomography is more precise than 18F-fluorodeoxyglucose positron emission tomography in evaluating tumor burden and predicting disease risk of multiple myeloma

The optimal method of tumor burden evaluation in newly diagnosed multiple myeloma (NDMM) is yet to be determined. This study aimed to compare the value of 11C-acetate positron-emission tomography (PET)/computed tomography (CT) (AC-PET and 18F-fluorodeoxyglucose PET/CT (FDG-PET) in the assessment of tumor burden in NDMM. This study evaluated 64 NDMM patients between February 2015 and July 2018. AC-PET and FDG-PET were used to assess myeloma lesions. The clinical data, imaging results, and their correlations were analyzed. Diffuse bone marrow uptake in AC-PET was significantly correlated with biomarkers for tumor burden, including serum hemoglobin (P = 0.020), M protein (P = 0.054), the percentage of bone marrow plasma cells (P < 0.001), and the Durie–Salmon stage of the disease (P = 0.007). The maximum standard uptake value (SUVmax) of focal lesions and high diffuse bone marrow uptake in AC-PET showed stronger correlations with high-risk disease (P = 0.017, P = 0.013) than those in FDG-PET. Moreover, the presence of diffuse bone marrow uptake, more than ten focal lesions, and an SUVmax of focal lesions of > 6.0 in AC-PET, but not in FDG-PET, predicted a higher probability of disease progression and shorter progression-free survival (P < 0.05). AC-PET outperformed FDG-PET in tumor burden evaluation and disease progression prediction in NDMM.

Artificial intelligence could alert for focal skeleton/bone marrow uptake in Hodgkin’s lymphoma patients staged with FDG-PET/CT

To develop an artificial intelligence (AI)-based method for the detection of focal skeleton/bone marrow uptake (BMU) in patients with Hodgkin’s lymphoma (HL) undergoing staging with FDG-PET/CT. The results of the AI in a separate test group were compared to the interpretations of independent physicians. The skeleton and bone marrow were segmented using a convolutional neural network. The training of AI was based on 153 un-treated patients. Bone uptake significantly higher than the mean BMU was marked as abnormal, and an index, based on the total squared abnormal uptake, was computed to identify the focal uptake. Patients with an index above a predefined threshold were interpreted as having focal uptake. As the test group, 48 un-treated patients who had undergone a staging FDG-PET/CT between 2017–2018 with biopsy-proven HL were retrospectively included. Ten physicians classified the 48 cases regarding focal skeleton/BMU. The majority of the physicians agreed with the AI in 39/48 cases (81%) regarding focal skeleton/bone marrow involvement. Inter-observer agreement between the physicians was moderate, Kappa 0.51 (range 0.25–0.80). An AI-based method can be developed to highlight suspicious focal skeleton/BMU in HL patients staged with FDG-PET/CT. Inter-observer agreement regarding focal BMU is moderate among nuclear medicine physicians.

Artificial Intelligence Could Alert for Focal Skeleton/Bone Marrow Uptake in Hodgkin´s Lymphoma Patients Staged With FDG-PET/CT

Purpose: To develop an artificial intelligence (AI)-based method for the detection of focal skeleton/bone marrow uptake (BMU) in patients with Hodgkin´s lymphoma (HL) undergoing staging with FDG-PET/CT. The results of the AI in a separate test group were compared to the interpretations of independent physicians. Methods: The skeleton and bone marrow were segmented using a convolutional neural network. The training of AI was based on 153 un-treated patients. Bone uptake significantly higher than the mean BMU was marked as abnormal, and an index, based on the total squared abnormal uptake, was computed to identify the focal uptake. Patients with an index above a predefined threshold were interpreted as having focal uptake. As the test group, 48 un-treated patients who had undergone a staging FDG-PET/CT between 2017-2018 with biopsy-proven HL were retrospectively included. Ten physicians classified the 48 cases regarding focal skeleton/BMU. Results: The majority of the physicians agreed with the AI in 39/48 cases (81%) regarding focal skeleton/bone marrow involvement. Inter-observer agreement between the physicians was moderate, Kappa 0.51 (range 0.25-0.80). Conclusion: An AI-based method can be developed to highlight suspicious focal skeleton/BMU in HL patients staged with FDG-PET/CT. Inter-observer agreement regarding focal BMU is moderate among nuclear medicine physicians.

[18F]FDG uptake of the normal spinal cord in PET/MR imaging: Comparison with PET/CT imaging

Background The lack of visualization of the spinal cord hinders the evaluation of [18F]Fluoro-deoxy-glucose (FDG) uptake of the spinal cord in PET/CT. By exploiting the capability of MRI to precisely outline the spinal cord, we performed a retrospective study aimed to define normal pattern of spinal cord [18F]FDG uptake in PET/MRI.Methods Forty-one patients with lymphoma without clinical or MRI signs of spinal cord or bone marrow involvement underwent simultaneous PET and MRI acquisition using Siemens Biograph mMR after injection of 3.5 MBq/kg body weight of [18F]FDG for staging purposes. Using a custom-made software, we placed ROIs of 3 and 9 mm in diameter in spinal cord, lumbar CSF and vertebral marrow that were identified on MRI at 5 levels (C2, C5, T6, T12 and L3). The SUVmax, SUVmean and the SUVmax and SUVmean normalized (NSUVmax and NSUVmean) to the liver were measured. For comparison, the same ROIs were placed in PET-CT images obtained immediately before the PET-MRI acquisition following the same tracer injection.Results On PET/MRI using the 3 mm ROI the following average (all level excluding L3) spinal cord median (1st and 3rd quartile) values were measured:SUVmean1.68 (1.39 and 1.83), SUVmax1.92(1.60 and 2.14), NSUVmean1.18(0.93 and 1.36), NSUVmax1.27(1.01 and 1.33). Using the 9 mm ROI the corresponding values were: SUVmean1.41 (1.25-1.55), SUVmax2.41(2.08 and 2.61), NSUVmean0.93 (0.79 and 1.04), NSUVmax1.28(1.02 and 1.39).Using the 3 mm ROI the highest values of PET-MRI SUVmax, SUVmean, NSUVmax and NSUVmean were consistently observed at C5 and the lowest at T6. Using a 9 mm ROI the highest values were consistently observed at C5 and the lowest at T12 or T6. The spinal cord [18F]FDG-uptake values correlated with bone marrow uptake at the same level, especially in case of NSUVmax. Comparison with PET-CT data revealed that the average SUVmax and SUVmean of the spinal cord were similar in PET-MRI and PET-CT. However, the average NSUVmax and NSUVmean of the spinal cord were higher (range 21% - 47%) in PET-MRI than in PET-CT. Conclusions Using a whole-body protocol we defined the maximum and mean [18F]FDG uptake of the normal spinal cord in PET/MRI. While the observed values show the expected longitudinal distribution, they appear to be higher than those measured in PET/CT. Normalization of the SUVmax and SUVmean of the spinal cord to the liver radiotracer uptake could help in multi-institutional comparisons and studies.

FDG uptake of the normal spinal cord in PET/MR imaging

Background: The lack of visualization of the spinal cord hinders the evaluation of18F Fluoro-deoxy-glucose (FDG) uptake of the spinal cord in PET/CT. By exploiting the capability of MRI to precisely outline the spinal cord, we performed a retrospective study aimed to define normal pattern of spinal cord FDG uptake in PET/MRI.Methods: Forty-one patients with lymphoma without clinical or MRI signs of spinal cord or bone marrow involvement underwent simultaneous PET and MRI acquisition using Siemens Biograph mMR after injection of 3.5 MBq/kg body weight of 18FDG for staging purposes. Using a custom-made software, we placed ROIs of 3 and 9 mm in diameter in spinal cord, lumbar CSF and vertebral marrow that were identified on MRI at 5 levels (C2, C5, T6, T12 and L3). The SUVmax, SUVmean and the SUVmax and SUVmean normalized (NSUVmax and NSUVmean) to the liver were measured. For comparison, the same ROIs were placed in PET-CT images obtained immediately before the PET-MRI acquisition following the same tracer injection.Results: On PET/MRI using the 3 mm ROI the following average (all level excluding L3) spinal cord median (1st and 3rd quartile) values were measured: SUVmean1.68 (1.39 and 1.83), SUVmax1.92 (1.60 and 2.14), NSUVmean1.18 (0.93 and 1.36), NSUVmax1.27 (1.01 and 1.33). Using the 9 mm ROI the corresponding values were: SUVmean1.41 (1.25-1.55), SUVmax2.41 (2.08 and2.61), NSUVmean0.93 (0.79 and 1.04), NSUVmax1.28 (1.02 and 1.39). Using the 3 mm ROI the highest values of PET-MRI SUVmax, SUVmean, NSUVmax and NSUVmean were consistently observed at C5 and the lowest at T6. Using a 9 mm ROI the highest values were consistently observed at C5 and the lowest at T12 or T6. The spinal cord FDG-uptake values correlated with bone marrow uptake at the same level, especially in case of NSUVmax. Comparison with PET-CT data revealed that the average SUVmax and SUVmean of the spinal cord were similar in PET-MRI and PET-CT. However, the average NSUVmax and NSUVmean of the spinal cord were higher (range 21% - 47%) in PET-MRI than in PET-CT.Conclusions: Using a whole-body protocol we defined the maximum and mean FDG uptake of the normal spinal cord in PET/MRI. While the observed values show the expected longitudinal distribution, they appear to be higher than those measured in PET/CT. Normalizing the SUVmax and SUVmean of the spinal cord to the liver radiotracer uptake could help in multi-institutional comparisons and studies.