Aims: To report the results of I(125) prostate brachytherapy from a central, prospectively collected database of three UK institutions.
Materials and methods: All patients treated with I(125) permanent prostate brachytherapy at the Christie Hospital, Manchester (CHM), Cookridge Hospital, Leeds (CKL) and Mount Vernon Hospital, Northwood, London (MVL) since 2003 have been prospectively registered on a detailed central database. Patient, tumour, pre- and post-implant dosimetry data have been recorded. Urinary toxicity as assessed by the International Prostate Symptom Score, catheterisation and urinary stricture rates after implant have been documented and biochemical failure determined, using both the American Society for Therapeutic Radiology and Oncology (ASTRO) consensus and the Phoenix (nadir + 2 ng/ml) definition.
Results: In total, 1535 patients were registered on the database between January 2003 and October 2006, including 432 from CHM, 926 from CKL and 177 from MVL, with a median follow-up of 21 months (range 1-56). Patient and tumour characteristics were similar at all centres. Pre-implant dose indices were comparable between centres, except for the V150, with median values of 51.9, 64.3 and 69.8% at CHM, CKL and MVL, respectively. Median post-implant dose parameters were lower than pre-planned constraints by up to 33.0% at each centre for all values, except at CKL where the V200 was 23.9% higher. The International Prostate Symptom Score increased from a median of 5 at baseline to 18, 6 weeks after implant, but was not significantly different to baseline values by 12 months. Nine per cent of men required catheterisation after implant for a median duration of 53 days, but urinary stricture rates remained low at 1%. Neoadjuvant hormonal manipulation was used in 228 men (15%) for downsizing and 159 (10%) for intermediate/high-risk disease. Collated biochemical failure rates were low at this point of follow-up, with actuarial 2-year ASTRO and Phoenix biochemical failure-free survival rates of 94.4 and 94.5%, respectively, consistent with other large single centre reports. When post-implant dosimetric factors were assessed for a relationship to biochemical failure, no indices consistently predicted for improved ASTRO and Phoenix biochemical failure-free survival rates.
Conclusions: This ongoing collaboration shows that with limited infrastructure (a single industry-sponsored data manager), a large multi-institutional database estimated to represent one-third of implants carried out in the UK during this time can be developed. Patient selection was similar across all centres and adhered to published guidelines. Early biochemical and toxicity outcomes confirm the efficacy and tolerability of I(125) prostate brachytherapy in a large cohort of patients. A further analysis is planned.