A comprehensive review of prostate cancer brachytherapy: defining an optimal technique

A comprehensive review of prostate cancer brachytherapy literature was performed to determine if an optimal method of implantation could be identified, and to compare and contrast techniques currently in use.

A MEDLINE search was conducted to obtain all articles in the English language on prostate cancer brachytherapy from 1985 through 1998. Articles were reviewed and grouped to determine the primary technique of implantation, the method or philosophy of source placement and/or dose specification, the technique to evaluate implant quality, overall treatment results (based upon pretreatment prostate specific antigen, (PSA), and biochemical control) and clinical, pathological or biochemical outcome based upon implant quality.

A total of 178 articles were identified in the MEDLINE database. Of these, 53 studies discussed evaluable techniques of implantation and were used for this analysis. Of these studies, 52% used preoperative ultrasound to determine the target volume to be implanted, 16% used preoperative computerized tomography (CT) scans, and 18% placed seeds with an open surgical technique. An additional 11% of studies placed seeds or needles under ultrasound guidance using interactive real-time dosimetry. The number and distribution of radioactive sources to be implanted or the method used to prescribe dose was determined using nomograms in 27% of studies, a least squares optimization technique in 11%, or not stated in 35%. In the remaining 26%, sources were described as either uniformly, differentially, or peripherally placed in the gland. To evaluate implant quality, 28% of studies calculated some type of dose-volume histogram, 21% calculated the matched peripheral dose, 19% the minimum peripheral dose, 14% used some type of CT-based qualitative review and, in 18% of studies, no implant quality evaluation was mentioned. Six studies correlated outcome with implant dose. One study showed an association of implant dose with the achievement of a PSA nadir < or = 0.5. Two studies showed an improvement in biochemical control with a D90 (dose to 90% of the prostate volume) of 120 to 140 Gy or higher, and 2 additional studies found an association of clinical outcome with implant dose. One study correlated implant quality with biopsy results. Of the articles, 33 discussed evaluable treatment results, but only 16 reported findings based upon pretreatment PSA and biochemical control. Three- to 5-year biochemical control rates ranged from 48% to 100% for pretreatment PSAs < or = 4, 55% to 90% for PSAs between 4 and 10, 30% to 89% for PSAs > 10, < or = 20 and < 10% to 100% for PSAs > 20. Due to substantial differences in patient selection criteria (e.g., median Gleason score, clinical stage, pretreatment PSA), number of patients treated, median follow-up, definitions of biochemical control, and time points for analysis, no single technique consistently produced superior results.

Our comprehensive review of prostate cancer brachytherapy literature failed to identify an optimal treatment approach when studies were analyzed for treatment outcome based upon pretreatment PSA and biochemical control. Although several well-designed studies showed an improvement in outcome with total dose or implant quality, the numerous techniques for implantation and the varied and inconsistent methods to specify dose or evaluate implant quality suggest that standardized protocols should be developed to objectively evaluate this treatment approach. These protocols have recently been suggested and, when implemented, should significantly improve the reporting of treatment data and, ultimately, the efficacy of prostate brachytherapy.