A comparison of preplan transrectal ultrasound with preplan-CT in assessing volume and number of seeds needed for real-time ultrasound-based intra-operative planning in prostate 125I seed implantation

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.

Comparison of multiparametric MRI-based and transrectal ultrasound-based preplans with intraoperative ultrasound-based planning for low dose rate interstitial prostate seed implantation

PURPOSE

Transrectal ultrasound images are routinely acquired for low dose rate (LDR) prostate brachytherapy dosimetric preplanning (pTRUS), although diagnostic multiparametric magnetic resonance imaging (mpMRI) may serve this purpose as well. We compared the predictive abilities of TRUS vs MRI relative to intraoperative TRUS (iTRUS) to assess the role of mpMRI in brachytherapy preplanning.

MATERIALS AND METHODS

Retrospective analysis was performed on 32 patients who underwent iTRUS-guided prostate LDR brachytherapy as either mono- or combination therapy. 56.3% had pTRUS-only volume studies and 43.7% had both 3T-mpMRI and pTRUS preplanning. MRI was used for preplanning and its image fusion with iTRUS was also used for intraoperative guidance of seed placement. Differences in gland volume, seed number, and activity and procedure time were examined, as well as the identification of lesions suspicious for tumor foci. Pearson correlation coefficient and Fisher's Z test were used to estimate associations between continuous measures.

RESULTS

There was good correlation of planning volumes between iTRUS and either pTRUS or MRI (r = 0.89, r = 0.77), not impacted by the addition of hormonal therapy (P = 0.65, P = 0.33). Both consistently predicted intraoperative seed number (r = 0.87, r = 0.86). MRI/TRUS fusion did not significantly increase surgical or anesthesia time (P = 0.10, P = 0.46). mpMRI revealed suspicious focal lesions in 11 of 14 cases not visible on pTRUS, that when correlated with histopathology, were incorporated into the plan.

CONCLUSIONS

Relative to pTRUS, MRI yielded reliable preplanning measures, supporting the role of MRI-only LDR treatment planning. mpMRI carries numerous diagnostic, staging and preplanning advantages that facilitate better patient selection and delivery of novel dose escalation and targeted therapy, with no additional surgical or anesthesia time. Prospective studies assessing its impact on treatment planning and delivery can serve to establish mpMRI as the standard of care in LDR prostate brachytherapy planning.

A comparison of preplan MRI and preplan CT-based prostate volume with intraoperative ultrasound-based prostate volume in real-time permanent brachytherapy

Purpose

The present study compared the difference between intraoperative transrectal ultrasound (iTRUS)-based prostate volume and preplan computed tomography (CT), preplan magnetic resonance imaging (MRI)-based prostate volume to estimate the number of seeds needed for appropriate dose coverage in permanent brachytherapy for prostate cancer.

Materials and Methods

Between March 2007 and March 2011, among 112 patients who underwent permanent brachytherapy with ¹²⁵I, 60 image scans of 56 patients who underwent preplan CT (pCT) or preplan MRI (pMRI) within 2 months before brachytherapy were retrospectively reviewed. Twenty-four cases among 30 cases with pCT and 26 cases among 30 cases with pMRI received neoadjuvant hormone therapy (NHT). In 34 cases, NHT started after acquisition of preplan image. The median duration of NHT after preplan image acquisition was 17 and 21 days for cases with pCT and pMRI, respectively. The prostate volume calculated by different modalities was compared. And retrospective planning with iTRUS image was performed to estimate the number of ¹²⁵I seed required to obtain recommended dose distribution according to prostate volume.

Results

The mean difference in prostate volume was 9.05 mL between the pCT and iTRUS and 6.84 mL between the pMRI and iTRUS. The prostate volume was roughly overestimated by 1.36 times with pCT and by 1.33 times with pMRI. For 34 cases which received NHT after image acquisition, the prostate volume was roughly overestimated by 1.45 times with pCT and by 1.37 times with pMRI. A statistically significant difference was found between preplan image-based volume and iTRUS-based volume (p < 0.001). The median number of wasted seeds is approximately 13, when the pCT or pMRI volume was accepted without modification to assess the required number of seeds for brachytherapy.

Conclusion

pCT-based volume and pMRI-based volume tended to overestimate prostate volume in comparison to iTRUS-based volume. To reduce wasted seeds and cost of the brachytherapy, we should take the volume discrepancy into account when we estimate the number of ¹²⁵I seeds for permanent brachytherapy.