Cribriform pattern disease over-represented in pelvic lymph node metastases identified on 68GA PSMA-PET/CT

Automated segmentation of lesions and organs at risk on [68Ga]Ga-PSMA-11 PET/CT images using self-supervised learning with Swin UNETR

BACKGROUND

Prostate-specific membrane antigen (PSMA) PET/CT imaging is widely used for quantitative image analysis, especially in radioligand therapy (RLT) for metastatic castration-resistant prostate cancer (mCRPC). Unknown features influencing PSMA biodistribution can be explored by analyzing segmented organs at risk (OAR) and lesions. Manual segmentation is time-consuming and labor-intensive, so automated segmentation methods are desirable. Training deep-learning segmentation models is challenging due to the scarcity of high-quality annotated images. Addressing this, we developed shifted windows UNEt TRansformers (Swin UNETR) for fully automated segmentation. Within a self-supervised framework, the model's encoder was pre-trained on unlabeled data. The entire model was fine-tuned, including its decoder, using labeled data.

METHODS

In this work, 752 whole-body [68Ga]Ga-PSMA-11 PET/CT images were collected from two centers. For self-supervised model pre-training, 652 unlabeled images were employed. The remaining 100 images were manually labeled for supervised training. In the supervised training phase, 5-fold cross-validation was used with 64 images for model training and 16 for validation, from one center. For testing, 20 hold-out images, evenly distributed between two centers, were used. Image segmentation and quantification metrics were evaluated on the test set compared to the ground-truth segmentation conducted by a nuclear medicine physician.

RESULTS

The model generates high-quality OARs and lesion segmentation in lesion-positive cases, including mCRPC. The results show that self-supervised pre-training significantly improved the average dice similarity coefficient (DSC) for all classes by about 3%. Compared to nnU-Net, a well-established model in medical image segmentation, our approach outperformed with a 5% higher DSC. This improvement was attributed to our model's combined use of self-supervised pre-training and supervised fine-tuning, specifically when applied to PET/CT input. Our best model had the lowest DSC for lesions at 0.68 and the highest for liver at 0.95.

CONCLUSIONS

We developed a state-of-the-art neural network using self-supervised pre-training on whole-body [68Ga]Ga-PSMA-11 PET/CT images, followed by fine-tuning on a limited set of annotated images. The model generates high-quality OARs and lesion segmentation for PSMA image analysis. The generalizable model holds potential for various clinical applications, including enhanced RLT and patient-specific internal dosimetry.

Lymphotropic Pattern of Prostate-specific Membrane Antigen-detected Metastases Among Biochemically Recurrent Radical Prostatectomy Patients with Cribriform Disease

BACKGROUND

Cribriform morphology portends worse oncologic outcomes, and has unique cellular intrinsic pathway alterations and tumor microenvironments that may impact metastatic spread patterns.

OBJECTIVE

To determine whether the presence of cribriform morphology in prostatectomy specimens of patients with biochemical recurrence after radical prostatectomy (RP) is associated with the presence of metastasis on prostate-specific membrane antigen (PSMA) positron emission tomography/computed tomography (PET/CT) and a distinct pattern of spread.

DESIGN, SETTING, AND PARTICIPANTS

A cross-sectional analysis was conducted of all prostate cancer patients with biochemical recurrence after RP undergoing 18F-DCFPyL-PET/CT between December 2018 and February 2021 at the Princess Margaret Cancer Centre.

OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS

Outcomes were presence of any metastasis in the overall cohort and lymphatic versus bone/visceral metastases among patients with metastatic disease. The associations between the presence of intraductal (IDC) and/or invasive cribriform (ICC) carcinoma on the RP specimen and study outcomes were evaluated using logistic regression analyses.

RESULTS AND LIMITATIONS

The cohort included 176 patients. IDC and ICC were observed in 77 (43.8%) and 80 (45.5%) RP specimens, respectively. The median time from RP to PSMA-PET/CT was 5.0 yr. The median serum prostate-specific antigen level at PSMA-PET/CT was 1.12 ng/ml. Overall, metastasis was observed in 77 patients, of whom 58 were had lymphatic-only metastasis. On a multivariable analysis, presence of IDC on RP was associated with increased odds of overall metastasis (odds ratio [OR]: 2.17; 95% confidence interval [CI]: 1.07-4.45; p = 0.033). Presence of ICC on RP was associated with significantly increased odds of lymphatic versus bone/visceral metastases (OR: 3.13; 95% CI: 1.09-21.7; p = 0.004).

CONCLUSIONS

Presence of cribriform morphology on RP specimens of patients with biochemical failure after RP is associated with increased odds of PSMA-PET/CT-detected metastases with a lymphatic predominant pattern of spread. These findings have implications for the design and evaluation of post-RP salvage therapies.

PATIENT SUMMARY

We found that microscopic cribriform appearance correlates with disease spread on imaging in prostate cancer patients with recurrence and has a predilection for spread to lymph nodes, as opposed to bone or visceral organs.

Distant Nodes Seen on PSMA PET-CT Staging Predicts Post-Treatment Progression in Men with Newly Diagnosed Prostate Cancer-A Prospective Cohort Study

PSMA PET-CT scans are now recommended in international urological guidelines for primary staging and re-staging of prostate cancer. However, there is little published literature on the clinical outcomes for patients after treatment decisions made using PSMA PET-CT results. This is a multisite, prospective cohort study investigating the clinical outcomes of men who received treatment plans based on PSMA PET-CT results for primary staging. Men with biopsy proven prostate cancer received a PSMA PET-CT scan for primary staging. Treatment plans were recommended by multidisciplinary teams (MDT). After treatment, these men were followed with 6 monthly PSA tests and imaging or biopsies if recommended by MDT. The primary outcome was treatment progression defined as the addition or change of any treatment modalities such as androgen deprivation therapy, radiation therapy or chemotherapy. In total, 80% of men did not have any treatment progression after enactment of treatment based on PSMA PET-CT primary staging results at 29 months of follow up. Men who had distant nodes seen on PSMA PET-CT had a 5 times increased risk of treatment progression. Larger studies with longer follow up are needed to validate our results and optimise the way clinicians use PSMA PET-CT results to guide management.