Implementation of PSA-based active surveillance in prostate cancer

Mitigating bias in prostate cancer diagnosis using synthetic data for improved AI driven Gleason grading

Prostate cancer (PCa) is a leading cause of cancer-related mortality in men, with Gleason grading critical for prognosis and treatment decisions. Machine learning (ML) models offer potential for automated grading but are limited by dataset biases, staining variability, and data scarcity, reducing their generalizability. This study employs generative adversarial networks (GANs) to generate high-quality synthetic histopathological images to address these challenges. A conditional GAN (dcGAN) was developed and validated using expert pathologist review and Spatial Heterogeneous Recurrence Quantification Analysis (SHRQA), achieving 80% diagnostic quality approval. A convolutional neural network (EfficientNet) was trained on original and synthetic images and validated across TCGA, PANDA Challenge, and MAST trial datasets. Integrating synthetic images improved classification accuracy for Gleason 3 (26%, p = 0.0010), Gleason 4 (15%, p = 0.0274), and Gleason 5 (32%, p < 0.0001), with sensitivity and specificity reaching 81% and 92%, respectively. This study demonstrates that synthetic data significantly enhances ML-based Gleason grading accuracy and improves reproducibility, providing a scalable AI-driven solution for precision oncology.

Enhancing risk stratification models in localized prostate cancer by novel validated tissue biomarkers

BACKGROUND

Localized prostate cancer (PCa) is a largely heterogeneous disease regarding its clinical behavior. Current risk stratification relies on clinicopathological parameters and distinguishing between indolent and aggressive cases remains challenging. To improve risk stratification, we aimed to identify new prognostic markers for PCa.

METHODS

We performed an in silico analysis on publicly available PCa transcriptome datasets. The top 20 prognostic genes were assessed in PCa tissue samples of our institutional cohort (n = 92) using the NanoString nCounter technology. The three most promising candidates were further assessed by immunohistochemistry (IHC) in an institutional (n = 121) and an independent validation cohort from the EMPACT consortium (n = 199). Cancer-specific survival (CSS) and progression-free survival (PFS) were used as endpoints.

RESULTS

Our in silico analysis identified 113 prognostic genes. The prognostic values of seven of the top 20 genes were confirmed in our institutional radical prostatectomy (RPE) cohort. Low CENPO, P2RX5, ABCC5 as well as high ASF1B, NCAPH, UBE2C, and ZWINT gene expressions were associated with shorter CSS. IHC analysis confirmed the significant associations between NCAPH and UBE2C staining and worse CSS. In the external validation cohort, higher NCAPH and ZWINT protein expressions were associated with shorter PFS. The combination of the newly identified tissue protein markers improved standard risk stratification models, such as D'Amico, CAPRA, and Cambridge prognostic groups.

CONCLUSIONS

We identified and validated high tissue levels of NCAPH, UBE2C, and ZWINT as novel prognostic risk factors in clinically localized PCa patients. The use of these markers can improve routinely used risk estimation models.

Characterization of metabolism-associated molecular patterns in prostate cancer

BACKGROUND

Metabolism is a hallmark of cancer and it involves in resistance to antitumor treatment. Therefore, the purposes of this study are to classify metabolism-related molecular pattern and to explore the molecular and tumor microenvironment characteristics for prognosis predicting in prostate cancer.

METHODS

The mRNA expression profiles and the corresponding clinical information for prostate cancer patients from TCGA, cBioPortal, and GEO databases. Samples were classified using unsupervised non-negative matrix factorization (NMF) clustering based on differentially expressed metabolism-related genes (MAGs). The characteristics of disease-free survival (DFS), clinicopathological characteristics, pathways, TME, immune cell infiltration, response to immunotherapy, and sensitivity to chemotherapy between subclusters were explored. A prognostic signature was constructed by LASSO cox regression analysis based on differentially expressed MAGs and followed by the development for prognostic prediction.

RESULTS

A total of 76 MAGs between prostate cancer samples and non-tumorous samples were found, then 489 patients were divided into two metabolism-related subclusters for prostate cancer. The significant differences in clinical characteristics (age, T/N stage, Gleason) and DFS between two subclusters. Cluster 1 was associated with cell cycle and metabolism-related pathways, and epithelial-mesenchymal transition (EMT), etc., involved in cluster 2. Moreover, lower ESTIMATE/immune/stromal scores, lower expression of HLAs and immune checkpoint-related genes, and lower half-maximal inhibitory concentration (IC50) values in cluster 1 compared with cluster 2. The 10 MAG signature was identified and constructed a risk model for DFS predicting. The patients with high-risk scores showed poorer DFS. The area under the curve (AUC) values for 1-, 3-, 5-year DFS were 0.744, 0.731, 0.735 in TCGA-PRAD dataset, and 0.668, 0.712, 0.809 in GSE70768 dataset, 0.763, 0.802, 0.772 in GSE70769 dataset. Besides, risk score and Gleason score were identified as independent factors for DFS predicting, and the AUC values of risk score and Gleason score were respectively 0.743 and 0.738. The nomogram showed a favorable performance in DFS predicting.

CONCLUSION

Our data identified two metabolism-related molecular subclusters for prostate cancer that were distinctly characterized in prostate cancer. Metabolism-related risk profiles were also constructed for prognostic prediction.

miR-192 Is Overexpressed and Promotes Cell Proliferation in Prostate Cancer

OBJECTIVE

Prostate cancer (PCa) is one of the most prevalent types of cancer among men worldwide. The incidence of PCa is increasing in China. Therefore, there is an urgent need to identify novel diagnostic and prognostic markers for PCa to improve the treatment of the disease.

METHODS

The Cancer Genome Atlas (TCGA) and GEO database were used to analyze the expression of miR-192, and the relationship between miR-192 and the clinical features of patients with PCa. Cell cycle and cell proliferation assay were used to detect the functional roles of miR-192 in PCa. Bioinformatic analysis for miR-192-5p was performed using gene ontology and KEGG analysis.

RESULTS

By analyzing the dataset of TCGA, we found that miR-192 was overexpressed in PCa samples compared to normal tissues and was upregulated in high-grade PCa compared to low-grade PCa. We also observed that higher miR-192 expression was associated with a shorter biochemical recurrence-free survival time. Our results also demonstrated that miR-192 promoted PCa cell proliferation and cell cycle progression.

CONCLUSION

These results suggest that miR-192 may be considered for use as a potential diagnostic and therapeutic target of PCa.

Preliminary Case Series of Artemisinin for Prostate Cancer in a Naturopathic Practice

Objective

To determine if oral artemisinin is safe and has a short-term effect on prostate specific antigen (PSA) kinetics in patients with prostate cancer (CaP).

Design

Retrospective case series.

Setting

A private naturopathic urology clinic in Seattle, WA.

Patients

All artemisinin-treated CaP patients were identified retrospectively between 2005 and 2008. A total of 15 patients were identified who had taken artemisinin and included in the study, comprising 5 patients who had previously undergone radical prostatectomy (RP) and were having biochemical recurrences as well as 10 patients with no prior conventional therapy for CaP.

Interventions

High-dose, pulsed oral artemisinin 300-400 mg three times a day every other week for 3-24 months (median 9.5 months, IQR 5-12 months). All patients were treated with an array of other naturopathic therapies.

Outcome measures

The primary outcomes were the PSA doubling time and velocity; secondary outcome measures were signs and symptoms of metastasis and survival.

Results

Of those patients who have previously undergone RP, 2/5 (40%) had improved PSA kinetics after artemisinin therapy. Of those with no prior RP, 5/10 (50%) had improved PSA kinetics. No patient developed signs of metastasis and no patients died. There were no reported adverse effects.

Conclusions

This pilot study provides preliminary evidence to suggest that high-dose, pulsed oral artemisinin therapy may have activity in patients with CaP. A larger controlled trial is warranted to confirm these preliminary beneficial effects.