Manual vs. automated implantation of seeds in prostate brachytherapy: Oncologic results from a single-center study

A machine-learning approach based on 409 treatments to predict optimal number of iodine-125 seeds in low-dose-rate prostate brachytherapy

Purpose

Low-dose-rate brachytherapy is a key treatment for low-risk or favorable intermediate-risk prostate cancer. The number of radioactive seeds inserted during the procedure depends on prostate volume, and is not easy to predict without pre-planning. Consequently, a large number of unused seeds may be left after treatment. The objective of the present study was to predict the exact number of seeds for future patients using machine learning and a database of 409 treatments.

Material and methods

Database consisted of 18 dosimetric and efficiency parameters for each of 409 cases. Nine predictive algorithms based on machine-learning were compared in this database, which was divided into training group (80%) and test group (20%). Ten-fold cross-validation was applied to obtain robust statistics. The best algorithm was then used to build an abacus able to predict number of implanted seeds from expected prostate volume only. As an evaluation, the abacus was also applied on an independent series of 38 consecutive patients.

Results

The best coefficients of determination R² were given by support vector regression, with values attaining 0.928, 0.948, and 0.968 for training set, test set, and whole set, respectively. In terms of predicted seeds in test group, mean square error, median absolute error, mean absolute error, and maximum error were 2.55, 0.92, 1.21, and 7.29, respectively. The use of obtained abacus in 38 additional patients resulted in saving of 493 seeds (393 vs. 886 remaining seeds).

Conclusions

Machine-learning-based abacus proposed in this study aims at estimating the necessary number of seeds for future patients according to past experience. This new abacus, based on 409 treatments and successfully tested in 38 new patients, is a good alternative to non-specific recommendations.

Diagnostic performance of imaging techniques for detecting of local recurrence after prostate brachytherapy

PURPOSE

The purpose of this study was to evaluate MRI and fluorocholine PET/CT diagnostic performances for the detection of local recurrence following prostate brachytherapy for localised prostate cancer.

MATERIAL AND METHODS

In this single-centre study, we retrospectively reviewed data from 21 patients treated by brachytherapy for localised prostate cancer and diagnosed with biochemical recurrence according to Phoenix Criteria, who underwent MRI and fluorocholine PET/CT. We included patients with local relapse suspicion according to imaging exams, with biopsy for the final assessment of local recurrence. Patient analysis data were supplemented by segment analysis using an 8-segment model.

RESULTS

The fluorocholine PET/CT was positive for 81% and negative for 19% of patients. The sensitivity and specificity were 92% and 33% with diagnosis accuracy of 67%. The MRI was positive for 57% and negative for 43% of patients. The sensitivity and specificity were 67% and 56% with diagnosis accuracy of 62%. There was no statistically significant difference between fluorocholine PET/CT and MRI accuracy (P=0.63). On a segment-based analysis, the sensitivity and specificity were 44% and 82% for fluorocholine PET/CT with diagnosis accuracy of 78%. For MRI, specificity was 91% diagnosis accuracy was 82%.

CONCLUSION

Both MRI and fluorocholine PET/CT permit to highlight local recurrence sites after prostate brachytherapy. Confirmation biopsies are, however, necessary since this accuracy is insufficient.

Biochemical outcomes and predictive factors by risk group after permanent iodine-125 seed implantation: Prospective cohort study in 2,316 patients

PURPOSE

To evaluate the biochemical freedom from failure (bFFF) by risk group and treatment modality and the predictive factors of bFFF by risk group in patients with prostate cancer undergoing permanent seed implantation (PI) with or without external beam radiation therapy (EBRT) in a nationwide prospective cohort study (Japanese Prostate Cancer Outcome Study of Permanent Iodine-125 [I-125] Seed Implantation) in Japan during the first 2 years.

METHODS AND MATERIALS

The analyses included 2,316 participants in 42 institutions; bFFF was evaluated using the Phoenix definition and calculated using the Kaplan-Meier method, and the Cox proportional hazards model was used to identify the factors associated with bFFF.

RESULTS

Median followup period was 60.0 months. The 5-year bFFF rates in all patients, 1,028 low-risk patients, 1,114 intermediate-risk patients, and 133 high-risk patients were 93.6%, 94.9%, 92.7%, and 91.1%, respectively. The 5-year bFFF rates in the PI group and EBRT combination therapy group were 93.7% and 93.3%, respectively. In a multivariate analysis, younger age, higher Gleason score (GS), higher percent positive biopsies (%PB), and lower prostate V₁₀₀ (p = 0.0012, 0.0030, 0.0026, and 0.0368) in all patients; younger age, higher pretreatment prostate-specific antigen, and lower prostate V₁₀₀ (p = 0.0002, 0.0048, and 0.0012) in low-risk patients; higher GS, higher %PB, and no hormonal treatment (p = 0.0005, 0.0120, and 0.0022) in intermediate-risk patients; and higher GS and higher %PB (p = 0.0329 and 0.0120) in high-risk patients were significantly associated with bFFF.

CONCLUSIONS

PI with or without EBRT resulted in excellent short-term biochemical outcomes in all risk groups, especially in high-risk patients. Age, pretreatment prostate-specific antigen, and prostate V₁₀₀ in low-risk patients; GS, %PB, and hormonal treatment in intermediate-risk patients; and GS and %PB in high-risk patients were independently affected bFFF.