Correlations between whole body volumetric parameters of 68Ga-PSMA PET/CT and biochemical-histopathological parameters in castration-naive and resistant prostate cancer patients

A convolutional neural network-based system for fully automatic segmentation of whole-body [68Ga]Ga-PSMA PET images in prostate cancer

PURPOSE

The aim of this study was development and evaluation of a fully automated tool for the detection and segmentation of mPCa lesions in whole-body [68Ga]Ga-PSMA-11 PET scans by using a nnU-Net framework.

METHODS

In this multicenter study, a cohort of 412 patients from three different center with all indication of PCa who underwent [68Ga]Ga-PSMA-11 PET/CT were enrolled. Two hundred cases of center 1 dataset were used for training the model. A fully 3D convolutional neural network (CNN) is proposed which is based on the self-configuring nnU-Net framework. A subset of center 1 dataset and cases of center 2 and center 3 were used for testing of model. The performance of the segmentation pipeline that was developed was evaluated by comparing the fully automatic segmentation mask with the manual segmentation of the corresponding internal and external test sets in three levels including patient-level scan classification, lesion-level detection, and voxel-level segmentation. In addition, for comparison of PET-derived quantitative biomarkers between automated and manual segmentation, whole-body PSMA tumor volume (PSMA-TV) and total lesions PSMA uptake (TL-PSMA) were calculated.

RESULTS

In terms of patient-level classification, the model achieved an accuracy of 83%, sensitivity of 92%, PPV of 77%, and NPV of 91% for the internal testing set. For lesion-level detection, the model achieved an accuracy of 87-94%, sensitivity of 88-95%, PPV of 98-100%, and F1-score of 93-97% for all testing sets. For voxel-level segmentation, the automated method achieved average values of 65-70% for DSC, 72-79% for PPV, 53-58% for IoU, and 62-73% for sensitivity in all testing sets. In the evaluation of volumetric parameters, there was a strong correlation between the manual and automated measurements of PSMA-TV and TL-PSMA for all centers.

CONCLUSIONS

The deep learning networks presented here offer promising solutions for automatically segmenting malignant lesions in prostate cancer patients using [68Ga]Ga-PSMA PET. These networks achieve a high level of accuracy in whole-body segmentation, as measured by the DSC and PPV at the voxel level. The resulting segmentations can be used for extraction of PET-derived quantitative biomarkers and utilized for treatment response assessment and radiomic studies.

Landmarks in the evolution of prostate biopsy

Approaches and techniques used for diagnostic prostate biopsy have undergone considerable evolution over the past few decades: from the original finger-guided techniques to the latest MRI-directed strategies, from aspiration cytology to tissue core sampling, and from transrectal to transperineal approaches. In particular, increased adoption of transperineal biopsy approaches have led to reduced infectious complications and improved antibiotic stewardship. Furthermore, as image fusion has become integral, these novel techniques could be incorporated into prostate biopsy methods in the future, enabling 3D-ultrasonography fusion reconstruction, molecular targeting based on PET imaging and autonomous robotic-assisted biopsy.

18F-PSMA-1007 PET/CT-derived semi-quantitative parameters for risk stratification of newly diagnosed prostate cancer

Purpose

This study aimed to assess the value of ¹⁸F-PSMA-1007 positron emission tomography/computed tomography (PET/CT)-derived semi-quantitative parameters of primary tumor for risk stratification of newly diagnosed prostate cancer (PCa).

Methods

Sixty patients referred for ¹⁸F-PSMA-1007 PET/CT imaging for primary PCa were retrospectively analyzed and classified into the low-intermediate-risk (LIR) or high-risk (HR) group. The maximum standardized uptake value (SUVmax) of primary tumor, prostate total lesion PSMA (TL-PSMAp), and prostate PSMA-tumor volume (PSMA-TVp) were measured, and group differences were evaluated using the Mann-Whitney U test. Spearman's correlation was performed to assess the correlation between the above parameters with prostate-specific antigen (PSA) levels and Gleason score (GS). Receiver operating characteristic (ROC) curve analysis was used to determine optimal cut-off values for SUVmax, TL-PSMAp, and PSMA-TVp to identify high-risk PCa and compare diagnostic efficacy.

Results

Among 60 patients, 46 were assigned to the HR group and 16 to the LIR group. In all patients, SUVmax, TL-PSMAp, and PSMA-TVp were moderately correlated with pre-treatment PSA values (r = 0.411, p = 0.001; r = 0.663, p < 0.001; and r = 0.549, p < 0.001, respectively). SUVmax and TL-PSMAp were moderately correlated with GS (r = 0.457 and r = 0.448, respectively; p < 0.001), while PSMA-TVp was weakly correlated with GS (r = 0.285, p = 0.027). In the ROC curve analysis, the optimal cut-off values of SUVmax, TL-PSMAp, and PSMA-TVp for identifying high-risk PCa were 9.61, 59.62, and 10.27, respectively, and the areas under the operating curve were 0.828, 0.901, and 0.809, respectively. The sensitivities of SUVmax, TL-PSMAp, and PSMA-TVp were 91.03%, 71.74%, and 63.04%, respectively, and the specificities were 71.43%, 100.00%, and 92.86%, respectively.

Conclusions

TL-PSMAp had a superior ability to identify high-risk PCa. The semi-quantitative parameters of primary tumor on ¹⁸F-PSMA-1007 PET/CT imaging can be an objective imaging reference index to determine PCa risk stratification.

The future of PSMA PET and WB MRI as next-generation imaging tools in prostate cancer

Radiolabelled prostate-specific membrane antigen (PSMA)-based PET-CT has been shown in numerous studies to be superior to conventional imaging in the detection of nodal or distant metastatic lesions. ⁶⁸Ga-PSMA PET-CT is now recommended by many guidelines for the detection of biochemically relapsed disease after radical local therapy. PSMA radioligands can also function as radiotheranostics, and Lu-PSMA has been shown to be a potential new line of treatment for metastatic castration-resistant prostate cancer. Whole-body (WB) MRI has been shown to have a high diagnostic performance in the detection and monitoring of metastatic bone disease. Prospective, randomized, multicentre studies comparing ⁶⁸Ga-PSMA PET-CT and WB MRI for pelvic nodal and metastatic disease detection are yet to be performed. Challenges for interpretation of PSMA include tracer trapping in non-target tissues and also urinary excretion of tracers, which confounds image interpretation at the vesicoureteral junction. Additionally, studies have shown how long-term androgen deprivation therapy (ADT) affects PSMA expression and could, therefore, reduce tracer uptake and visibility of PSMA⁺ lesions. Furthermore, ADT of short duration might increase PSMA expression, leading to the PSMA flare phenomenon, which makes the accurate monitoring of treatment response to ADT with PSMA PET challenging. Scan duration, detection of incidentalomas and presence of metallic implants are some of the major challenges with WB MRI. Emerging data support the wider adoption of PSMA PET and WB MRI for diagnosis, staging, disease burden evaluation and response monitoring, although their relative roles in the standard-of-care management of patients are yet to be fully defined.