(P090) Comparison of Intraoperatively Built Custom-Linked (IBCL) Seeds to Free Seeds for Permanent Prostate Brachytherapy

April 15, 2014

Our prostate brachytherapy technique at the Medical University of South Carolina evolved from implanting free seeds using a Mick Applicator (MA) (Mick Radio-Nuclear Instruments, Inc) to using intraoperatively built custom-linked (IBCL) seeds constructed with the QuickLink device (C.R. Bard, Inc). In this work, we compare dosimetric and early clinical outcomes using free seeds and IBCL seeds.

Simon Brown, BS, Samuel L. Cooper, MD, Michael Ashenafi, MS, Harry Clarke, MD, PhD, David T. Marshall, MD; Medical University of South Carolina

Purpose: Our prostate brachytherapy technique at the Medical University of South Carolina evolved from implanting free seeds using a Mick Applicator (MA) (Mick Radio-Nuclear Instruments, Inc) to using intraoperatively built custom-linked (IBCL) seeds constructed with the QuickLink device (C.R. Bard, Inc). In this work, we compare dosimetric and early clinical outcomes using free seeds and IBCL seeds.

Materials and Methods: From April 2005 to July 2012, a total of 197 patients with clinically localized prostate cancer underwent brachytherapy using real-time ultrasound-guided seed placement and intraoperative dosimetry to optimize target coverage based on the approach of Stock and colleagues at the Mount Sinai School of Medicine in New York. From April 2005 through February 2007, free seeds were placed using an MA. Starting in March 2007, brachytherapy was performed with IBCL seeds constructed using the QuickLink device in the operating room as needed during the implant. Patients were grouped per National Comprehensive Cancer Network (NCCN) risk stratification. All patients underwent postoperative computed tomography (CT)-based dosimetric analysis. Biochemical disease-free survival (bDFS) was calculated using Kaplan-Meier estimates with prostate-specific antigen (PSA) relapse using the Phoenix definition. Postoperative dosimetry for the two methods was compared with the Wilcoxon rank-sum test. Specific endpoint analyses included bDFS, Common Terminology Criteria for Adverse Events version 4.03 (CTCAE v4.03) toxicity criteria grades 3 or greater, urinary retention requiring catheter placement, hematuria, and rectal bleeding.

Results: A total of 197 patients with median follow-up of 3.1 years were identified. Median follow-up for patients implanted with free seeds was 6.3 years and 2.7 years for IBCL seed patients. Median age at diagnosis was 64 years, and 66% of the patients were white. Median PSA at diagnosis was 6.0 ng/mL. NCCN risk group: low 53%, intermediate 40%, and high 7%. In total, 117 patients received I-125 alone, and 79 patients received Pd-103 plus external beam radiotherapy (EBRT). One patient received Pd-103 alone. Median prostate D90 was 172.1 Gy for I-125 and 109.3 Gy for Pd-103. Median prostate V100, urethra D30, and rectal V100 were 93.6%, 132.3%, and 0.55 cc, respectively. IBCL had significantly higher V100 (P = .008) and D90 with I-125 (P = .008) than for free seeds. Five-year bDFS was 90.6% for intermediate-/high-risk patients and 96% for low-risk patients. Five-year bDFS was significantly higher in the low-risk group with IBCL at 100% compared with free seeds at 90.2% (P = .031). The 5-year rates of grade 3 toxicity, urinary retention requiring catheterization, any hematuria, and any rectal bleeding were 3.4%, 4.7%, 10.4%, and 10.2%, respectively, with no significant difference between IBCL or free seeds. The rate of rectal bleeding with radiation therapy (RT) changes on colonoscopy was 1.8% at 5 years. All hematuria resolved on its own, with no RT changes on cystoscopy.

Conclusion: Prostate implants using free seeds or IBCL seeds provide low rates of early biochemical failure and toxicity in patients with clinically localized prostate cancer. IBCL was associated with a higher D90 in patients receiving I-125. Patients with low-risk disease had better bDFS with IBCL seeds compared with patients treated with free seeds, but longer follow-up is needed to confirm these findings.