Dosimetric and radiobiological investigation of permanent implant prostate brachytherapy based on Monte Carlo calculations

Ali-Reza Mehan Haidari; Nelson Miksys; Paul Soubiran; Joanna Ewa Cygler; Oliver Holmes; Gad Perry; Rowan M Thomson

doi:10.1016/j.brachy.2019.06.008

Dosimetric and radiobiological investigation of permanent implant prostate brachytherapy based on Monte Carlo calculations

Brachytherapy. 2019 Nov-Dec;18(6):875-882. doi: 10.1016/j.brachy.2019.06.008. Epub 2019 Aug 7.

Authors

Ali-Reza Mehan Haidari¹, Nelson Miksys², Paul Soubiran³, Joanna Ewa Cygler², Oliver Holmes⁴, Gad Perry⁴, Rowan M Thomson⁵

Affiliations

¹ Department of Physics, Carleton Laboratory for Radiotherapy Physics, Carleton University, Ottawa, Canada.
² Department of Physics, Carleton Laboratory for Radiotherapy Physics, Carleton University, Ottawa, Canada; Department of Medical Physics, The Ottawa Hospital Cancer Centre, The Ottawa Hospital, Ottawa, Canada.
³ Department of Medical Physics, The Ottawa Hospital Cancer Centre, The Ottawa Hospital, Ottawa, Canada.
⁴ Department of Radiation Oncology, The Ottawa Hospital Cancer Centre, The Ottawa Hospital, Ottawa, Canada.
⁵ Department of Physics, Carleton Laboratory for Radiotherapy Physics, Carleton University, Ottawa, Canada. Electronic address: rowan.thomson@carleton.ca.

PMID: 31400953
DOI: 10.1016/j.brachy.2019.06.008

Abstract

Purpose: Permanent implant prostate brachytherapy plays an important role in prostate cancer treatment, but dose evaluations typically follow the water-based TG-43 formalism, ignoring patient anatomy and interseed attenuation. The purpose of this study is to investigate advanced TG-186 model-based dose calculations via retrospective dosimetric and radiobiological analysis for a new patient cohort.

Methods and materials: A cohort of 155 patients treated with permanent implant prostate brachytherapy from The Ottawa Hospital Cancer Centre is considered. Monte Carlo (MC) dose calculations are performed using tissue-based virtual patient models. Dose-volume histogram (DVH) metrics (target, organs at risk) are extracted from 3D dose distributions and compared with those from calculations under TG-43 assumptions (TG43). Equivalent uniform biologically effective dose and tumor control probability are calculated.

Results: For the target, D₉₀ (V₁₀₀) is 136.7 ± 20.6 Gy (85.8% ± 7.8%) for TG43 and 132.8 ± 20.1 Gy (84.1% ± 8.2%) for MC; D₉₀ is 3.0% ± 1.1% lower for MC than TG43. For organs at risk, MC D_1cc = 104.4 ± 27.4 Gy (TG43: 106.3 ± 28.3 Gy) for rectum and 80.8 ± 29.7 Gy (TG43: 78.4 ± 28.4 Gy) for bladder; D_1cc = 185.9 ± 30.2 Gy (TG43: 191.1 ± 32.0 Gy) for urethra. Equivalent uniform biologically effective dose and tumor control probability are generally lower when evaluated using MC doses. The largest dosimetric and radiobiological discrepancies between TG43 and MC are for patients with intraprostatic calcifications, for whom there are low doses (cold spots) in the vicinity of calcifications within the target, identified with MC but not TG43.

Conclusions: DVH metrics and radiobiological indices evaluated with TG43 are systematically inaccurate by upward of several percent compared with MC patient-specific models. Mean cohort DVH metrics and their MC:TG43 variances are sensitive to patient cohort and clinical practice, underlining the importance of further retrospective MC studies toward widespread clinical adoption of advanced model-based dose calculations.

Keywords: Brachytherapy; Dose calculation; Monte Carlo; Permanent seed implant; Prostate; Radiobiological model.

MeSH terms

Aged
Algorithms*
Brachytherapy / methods*
Humans
Male
Middle Aged
Monte Carlo Method
Prostatic Neoplasms / diagnosis
Prostatic Neoplasms / radiotherapy*
Radiometry / methods*
Radiotherapy Planning, Computer-Assisted / methods
Retrospective Studies
Tomography, X-Ray Computed