The Capio Prostate Cancer Center Model for Prostate Cancer Diagnostics-Real-world Evidence from 2018 to 2022

Background

The Capio Prostate Cancer Center (Capio PCC) in Stockholm, Sweden, adopts a comprehensive diagnostic approach, utilizing prostate-specific antigen (PSA), Stockholm3, and magnetic resonance imaging (MRI) for prostate cancer risk assessment, followed by targeted and systematic biopsies for high-risk cases.

Objective

This study aims to elucidate the clinical process and real-world outcomes of the Capio PCC model for prostate cancer diagnosis at Capio S:t Göran Hospital.

Design setting and participants

Between 2018 and 2022, a cohort of 12 406 men aged 45-75 yr underwent prostate cancer testing, adhering to Capio PCC's structured diagnostic protocol.

Outcome measurements and statistical analysis

We provide a comprehensive description of the Capio PCC model and present results from its implementation, including assessments of PSA, Stockholm3, MRI scans, and biopsies. A comparative analysis is conducted between the diagnostic outcomes obtained at Capio PCC and those obtained at other regions in Sweden.

Results and limitations

The median participant age was 61 yr (interquartile range [IQR]: 55-67), with PSA levels at 1.6 ng/ml (IQR: 0.8-3.3) and Stockholm3 scores at 4 (IQR: 3-11). Among 1064 men (8.6%) undergoing biopsies, 611 (57% of biopsied) were diagnosed with International Society of Urological Pathology grade ≥ 2 cancer. Notably, employing a Stockholm3 ≥ 15 cutoff for biopsy, in lieu of PSA ≥ 3 ng/ml, reduced biopsy recommendations by 43%. For men with PSA levels between 1.5 and 2.9 ng/ml, 360 (12%) exhibited Stockholm3 scores of ≥ 15, with 72 (56% of biopsied) diagnosed with clinically significant prostate cancer. A comparative analysis with national Swedish prostate cancer detection data indicated that the Capio PCC model (vs Sweden) revealed a distribution of 14% (vs 25%) low-risk, 59% (vs 42%) intermediate-risk, and 26% (vs 30%) high-risk and advanced cancers.

Conclusions

This study underscores the effectiveness of the protocol-driven diagnostic process at Capio PCC, enabling earlier detection of intermediate-risk prostate cancer and reducing the need for MRI assessments compared with standard prostate cancer care in Sweden.

Patient summary

At the Capio Prostate Cancer Center, a novel diagnostic approach incorporating prostate-specific antigen, Stockholm3, magnetic resonance imaging, and targeted biopsies has been implemented to enhance prostate cancer testing and diagnosis in Stockholm, Sweden.

Effect of domain knowledge encoding in CNN model architecture-a prostate cancer study using mpMRI images

Background

Prostate cancer is one of the most common cancers worldwide. Currently, convolution neural networks (CNNs) are achieving remarkable success in various computer vision tasks, and in medical imaging research. Various CNN architectures and methodologies have been applied in the field of prostate cancer diagnosis. In this work, we evaluate the impact of the adaptation of a state-of-the-art CNN architecture on domain knowledge related to problems in the diagnosis of prostate cancer. The architecture of the final CNN model was optimised on the basis of the Prostate Imaging Reporting and Data System (PI-RADS) standard, which is currently the best available indicator in the acquisition, interpretation, and reporting of prostate multi-parametric magnetic resonance imaging (mpMRI) examinations.

Methods

A dataset containing 330 suspicious findings identified using mpMRI was used. Two CNN models were subjected to comparative analysis. Both implement the concept of decision-level fusion for mpMRI data, providing a separate network for each multi-parametric series. The first model implements a simple fusion of multi-parametric features to formulate the final decision. The architecture of the second model reflects the diagnostic pathway of PI-RADS methodology, using information about a lesion's primary anatomic location within the prostate gland. Both networks were experimentally tuned to successfully classify prostate cancer changes.

Results

The optimised knowledge-encoded model achieved slightly better classification results compared with the traditional model architecture (AUC = 0.84 vs. AUC = 0.82). We found the proposed model to achieve convergence significantly faster.

Conclusions

The final knowledge-encoded CNN model provided more stable learning performance and faster convergence to optimal diagnostic accuracy. The results fail to demonstrate that PI-RADS-based modelling of CNN architecture can significantly improve performance of prostate cancer recognition using mpMRI.

Review article: MRI-targeted biopsies for prostate cancer diagnosis and management

PURPOSE

Transrectal ultrasound (TRUS)-guided biopsy has been the traditional biopsy route in the detection of prostate cancer. However, due to concern regarding overdetection of low-risk cancer and missed clinically significant cancers as well as risk of sepsis, alternative approaches have been explored. Transperineal template biopsy-sampling the gland every 5 m to 10 mm-reduces error by sampling the whole prostate but increases risk of detecting clinically insignificant cancers as well as conferring risks of side effects such as urinary retention and bleeding.

METHODS

There are various targeted biopsy techniques, each with different cancer detection rates, costs and learning curves. Current research focuses on refining biopsy methodology to maximize detection of significant cancers, whilst minimising invasiveness and complications. In this article, the up-to-date research data about MRI-targeted prostate biopsy were reviewed to show its utilization in prostate cancer management and diagnosis.

RESULTS AND CONCLUSION

Prostate multiparametric MRI has become an effective tool in the detection of significant cancers and an essential component of the prostate cancer diagnostic pathway incorporating MRI-guided biopsy decisions.

Current routines for antibiotic prophylaxis prior to transrectal prostate biopsy: a national survey to all urology clinics in Sweden

Background: The risk of infection after transrectal ultrasound (TRUS)-guided prostate biopsies is increasing. The aim of the study was to assess the use of antibiotic prophylaxis for prostate biopsy in Sweden. Methods: All public and private urology clinics reporting to the National Prostate Cancer Register of Sweden received a survey on TRUS-biopsy prophylaxis. Results: Of the 84 clinics surveyed, 76 replied (90%). If no risk factors for infection were present, a single dose of ciprofloxacin 750 mg was used by 50 clinics (66%). Multiple doses of ciprofloxacin 500 or 750 mg (n=14; 18%) or a single dose of trimethoprim-sulfamethoxazole 160/800 mg (n=7; 9%) were other common prophylaxes. Most clinics gave the prophylaxes immediately before the biopsy (n=41; 54%). Urine dipstick was used by 30 clinics (39%) and rectal enema by six (8%). In patients with high risk of infection, the survey mirrors a large variety of regiments used. Conclusions: The preference to use a single dose of ciprofloxacin 750 mg is in accordance with the Swedish national guidelines for patients with a low risk of infection. Better compliance to the guideline recommendation to use a urine dipstick would probably increase the number of patients classified as having an increased risk of infection. Being classified as a high-risk patient should lead to an extended duration of antibiotic prophylaxis, however, the variety of regimens used in the high-risk group reflects an inability to treat these patients in a standardized fashion and also highlights a need for more clear-cut guidelines. Pre-biopsy identification of high-risk patients is an important issue to tackle for the urologic clinics in order to reduce the number of infections.