Utility of 64Cu-Sarcophagine-Bombesin PET/CT in Men with Biochemically Recurrent Prostate Cancer and Negative or Equivocal Findings on 68Ga-PSMA-11 PET/CT.

Despite a high detection rate of 68Ga-prostate-specific membrane antigen (PSMA) PET/CT in biochemical recurrence (BCR) of prostate cancer, a significant proportion of men have negative 68Ga-PSMA-11 PET/CT results. Gastrin-releasing peptide receptor, targeted by the copper-chelated bombesin analog 64Cu-sarcophagine-bombesin (SAR-BBN) PET/CT, is also overexpressed in prostate cancer. In this prospective imaging study, we investigate the detection rate of 64Cu-SAR-BBN PET/CT in patients with BCR and negative or equivocal 68Ga-PSMA-11 PET/CT results. Methods: Men with confirmed adenocarcinoma of the prostate, prior definitive therapy, and BCR (defined as a prostate-specific antigen [PSA] level > 0.2 ng/mL) with negative or equivocal 68Ga-PSMA-11 PET/CT results within 3 mo were eligible for enrollment. 64Cu-SAR-BBN PET/CT scans were acquired at 1 and 3 h after administration of 200 MBq of 64Cu-SAR-BBN, with further delayed imaging undertaken optionally at 24 h. PSA (ng/mL) was determined at baseline. All PET (PSMA and bombesin) scans were assessed visually. Images were read with masking of the clinical results by 2 experienced nuclear medicine specialists, with a third reader in cases of discordance. Accuracy was defined using a standard of truth that included biopsy confirmation, confirmatory imaging, or response to targeted treatment. Results: Twenty-five patients were enrolled. Prior definitive therapy was radical prostatectomy (n = 24, 96%) or radiotherapy (n = 1, 4%). The median time since definitive therapy was 7 y (interquartile range [IQR], 4-11 y), and the Gleason score was 7 or less (n = 15, 60%), 8 (n = 3, 12%), or 9 (n = 7, 28%). The median PSA was 0.69 ng/mL (IQR, 0.28-2.45 ng/mL). Baseline PSMA PET scans were negative in 19 patients (76%) and equivocal in 6 (24%). 64Cu-SAR-BBN PET-avid disease was identified in 44% (11/25): 12% (3/25) with local recurrence, 20% (5/25) with pelvic node metastases, and 12% (3/25) with distant metastases. The κ-score between readers was 0.49 (95% CI, 0.16-0.82). Patients were followed up for a median of 10 mo (IQR, 9-12 mo). Bombesin PET/CT results were true-positive in 5 of 25 patients (20%), false-positive in 2 of 25 (8%), false-negative in 7 of 25 (28%), and unverified in 11 of 25 (44%). Conclusion: 64Cu-SAR-BBN PET/CT demonstrated sites of disease recurrence in 44% of BCR cases with negative or equivocal 68Ga-PSMA-11 PET/CT results. Further evaluation to confirm diagnostic benefit is warranted.

Uncertainty estimation using a 3D probabilistic U-Net for segmentation with small radiotherapy clinical trial datasets.

BACKGROUND AND OBJECTIVES: Bio-medical image segmentation models typically attempt to predict one segmentation that resembles a ground-truth structure as closely as possible. However, as medical images are not perfect representations of anatomy, obtaining this ground truth is not possible. A surrogate commonly used is to have multiple expert observers define the same structure for a dataset. When multiple observers define the same structure on the same image there can be significant differences depending on the structure, image quality/modality and the region being defined. It is often desirable to estimate this type of aleatoric uncertainty in a segmentation model to help understand the region in which the true structure is likely to be positioned. Furthermore, obtaining these datasets is resource intensive so training such models using limited data may be required. With a small dataset size, differing patient anatomy is likely not well represented causing epistemic uncertainty which should also be estimated so it can be determined for which cases the model is effective or not. METHODS: We use a 3D probabilistic U-Net to train a model from which several segmentations can be sampled to estimate the range of uncertainty seen between multiple observers. To ensure that regions where observers disagree most are emphasised in model training, we expand the Generalised Evidence Lower Bound (ELBO) with a Constrained Optimisation (GECO) loss function with an additional contour loss term to give attention to this region. Ensemble and Monte-Carlo dropout (MCDO) uncertainty quantification methods are used during inference to estimate model confidence on an unseen case. We apply our methodology to two radiotherapy clinical trial datasets, a gastric cancer trial (TOPGEAR, TROG 08.08) and a post-prostatectomy prostate cancer trial (RAVES, TROG 08.03). Each dataset contains only 10 cases each for model development to segment the clinical target volume (CTV) which was defined by multiple observers on each case. An additional 50 cases are available as a hold-out dataset for each trial which had only one observer define the CTV structure on each case. Up to 50 samples were generated using the probabilistic model for each case in the hold-out dataset. To assess performance, each manually defined structure was matched to the closest matching sampled segmentation based on commonly used metrics. RESULTS: The TOPGEAR CTV model achieved a Dice Similarity Coefficient (DSC) and Surface DSC (sDSC) of 0.7 and 0.43 respectively with the RAVES model achieving 0.75 and 0.71 respectively. Segmentation quality across cases in the hold-out datasets was variable however both the ensemble and MCDO uncertainty estimation approaches were able to accurately estimate model confidence with a p-value < 0.001 for both TOPGEAR and RAVES when comparing the DSC using the Pearson correlation coefficient. CONCLUSIONS: We demonstrated that training auto-segmentation models which can estimate aleatoric and epistemic uncertainty using limited datasets is possible. Having the model estimate prediction confidence is important to understand for which unseen cases a model is likely to be useful.

Prognostic impact of prostate-specific membrane antigen positron emission tomography (PSMA PET) staging for clinically node-positive prostate cancer.

INTRODUCTION: In the current American Joint Committee on Cancer staging system, patients with pelvic nodal metastases are considered stage IV prostate cancer. This study aims to investigate whether men with prostate-specific membrane antigen positron emission tomography (PSMA PET)-detected pelvic node-positive prostate cancer at diagnosis have a better outcome compared to men with node-positive disease identified on conventional imaging. METHODS: This is a retrospective cohort study comparing the outcomes of men with node-positive prostate cancer and disease confined to the pelvis, staged with conventional versus PSMA PET imaging. Men had to be treated definitively with a combination of androgen deprivation therapy and radiation treatment to the prostate and pelvic lymph nodes. Kaplan-Meier and Cox regression analysis was used to compare biochemical failure-free survival (BFFS) and overall survival (OS). RESULTS: Seventy-six men with nodal metastases confined to the pelvis were identified. Fifty-one were detected with PSMA PET while 25 were staged with conventional imaging. PSMA PET staged patients had a lower proportion of Gleason 8-10 disease (78% vs. 96%) as well as a lower median prostate-specific antigen (11 ng/mL vs. 26 ng/mL). BFFS at 4 years was 72% with PSMA PET-detected node-positive disease vs. 38% with conventionally detected node-positive disease. Four-year OS was 93% with PSMA PET staged patients vs. 76% with conventionally staged patients. On multivariate analysis, the PSMA PET staged group was associated with improved BFFS (Adjusted HR = 3.00, 95% CI 1.43, 6.29, P = 0.004) and OS (Adjusted HR = 5.81, 95% CI 1.43, 23.7, P = 0.007). CONCLUSION: Men with PSMA PET-detected node-positive prostate cancer confined to the pelvis have significantly better biochemical control and survival compared to those with node-positive pelvic disease identified through conventional staging.

Evaluating the relationship between contouring variability and modelled treatment outcome for prostate bed radiotherapy.

Objectives.Contouring similarity metrics are often used in studies of inter-observer variation and automatic segmentation but do not provide an assessment of clinical impact. This study focused on post-prostatectomy radiotherapy and aimed to (1) identify if there is a relationship between variations in commonly used contouring similarity metrics and resulting dosimetry and (2) identify the variation in clinical target volume (CTV) contouring that significantly impacts dosimetry.Approach.The study retrospectively analysed CT scans of 10 patients from the TROG 08.03 RAVES trial. The CTV, rectum, and bladder were contoured independently by three experienced observers. Using these contours reference simultaneous truth and performance level estimation (STAPLE) volumes were established. Additional CTVs were generated using an atlas algorithm based on a single benchmark case with 42 manual contours. Volumetric-modulated arc therapy (VMAT) treatment plans were generated for the observer, atlas, and reference volumes. The dosimetry was evaluated using radiobiological metrics. Correlations between contouring similarity and dosimetry metrics were calculated using Spearman coefficient (Γ). To access impact of variations in planning target volume (PTV) margin, the STAPLE PTV was uniformly contracted and expanded, with plans created for each PTV volume. STAPLE dose-volume histograms (DVHs) were exported for plans generated based on the contracted/expanded volumes, and dose-volume metrics assessed.Mainresults. The study found no strong correlations between the considered similarity metrics and modelled outcomes. Moderate correlations (0.5 <Γ< 0.7) were observed for Dice similarity coefficient, Jaccard, and mean distance to agreement metrics and rectum toxicities. The observations of this study indicate a tendency for variations in CTV contraction/expansion below 5 mm to result in minor dosimetric impacts.Significance. Contouring similarity metrics must be used with caution when interpreting them as indicators of treatment plan variation. For post-prostatectomy VMAT patients, this work showed variations in contours with an expansion/contraction of less than 5 mm did not lead to notable dosimetric differences, this should be explored in a larger dataset to assess generalisability.

Robustness of magnetic resonance imaging and positron emission tomography radiomic features in prostate cancer: Impact on recurrence prediction after radiation therapy.

Background and purpose: Radiomic features from MRI and PET are an emerging tool with potential to improve prostate cancer outcomes. However, feature robustness due to image segmentation variations is currently unknown. Therefore, this study aimed to evaluate the robustness of radiomic features with segmentation variations and their impact on predicting biochemical recurrence (BCR). Materials and methods: Multi-scanner, pre-radiation therapy imaging from 142 patients with localised prostate cancer was used. Imaging included T2-weighted (T2), apparent diffusion coefficient (ADC) MRI, and prostate-specific membrane antigen (PSMA)-PET. The prostate gland and intraprostatic tumours were manually and automatically segmented, and differences were quantified using Dice Coefficient (DC). Radiomic features including shape, first-order, and texture features were extracted for each segmentation from original and filtered images. Intraclass Correlation Coefficient (ICC) and Mean Absolute Percentage Difference (MAPD) were used to assess feature robustness. Random forest (RF) models were developed for each segmentation using robust features to predict BCR. Results: Prostate gland segmentations were more consistent (mean DC = 0.78) than tumour segmentations (mean DC = 0.46). 112 (3.6 %) radiomic features demonstrated 'excellent' robustness (ICC > 0.9 and MAPD < 1 %), and 480 features (15.4 %) demonstrated 'good' robustness (ICC > 0.75 and MAPD < 5 %). PET imaging provided more features with excellent robustness than T2 and ADC. RF models showed strong predictive power for BCR with a mean area under the receiver-operator-characteristics curve (AUC) of 0.89 (range 0.85-0.93). Conclusion: When using radiomic features for predictive modelling, segmentation variability should be considered. To develop BCR predictive models, radiomic features from the entire prostate gland are preferable over tumour segmentation-based features.