Clinical Investigation
Predictive Factors for Acute and Late Urinary Toxicity After Permanent Prostate Brachytherapy: Long-Term Outcome in 712 Consecutive Patients

https://doi.org/10.1016/j.ijrobp.2008.05.022Get rights and content

Purpose

To describe the frequency of acute and late Radiation Therapy Oncology Group (RTOG) urinary toxicity, associated predictive factors, and resolution of International Prostate Symptom Score (IPSS) in 712 consecutive prostate brachytherapy patients.

Methods and Materials

Patients underwent implantation between 1998 and 2003 (median follow-up, 57 months). The IPSS and RTOG toxicity data were prospectively collected. The patient, treatment, and implant factors were examined for an association with urinary toxicity. The time to IPSS resolution was examined using Kaplan-Meier curves, and multivariate modeling of IPSS resolution was done using Cox proportional hazards regression analysis. Logistic regression analysis was used to examine the factors associated with urinary toxicity.

Results

The IPSS returned to baseline at a median of 12.6 months. On multivariate analysis, patients with a high baseline IPSS had a quicker resolution of their IPSS. Higher prostate D90 (dose covering 90% of the prostate), maximal postimplant IPSS, and urinary retention slowed the IPSS resolution time. The rate of the actuarial 5-year late urinary (>12 months) RTOG Grade 0, 1, 2, 3, and 4 was 32%, 36%, 24%, 6.2%, and 0.1%, respectively. At 7 years, the prevalence of RTOG Grade 0-1 was 92.5%. Patients with a larger prostate volume, greater number of needles, greater baseline IPSS, and use of hormonal therapy had more acute toxicity. On multivariate analysis, the significant predictors for late greater than or equal to RTOG toxicity 2 were a greater baseline IPSS, maximal postimplant IPSS, presence of acute toxicity, and higher prostate V150 (volume of the prostate covered by 150% of the dose). More recently implanted patients had less acute urinary toxicity and patients given hormonal therapy had less late urinary toxicity (all p < 0.02).

Conclusion

Most urinary symptoms resolved within 12 months after prostate brachytherapy, and significant long-term urinary toxicity was very low. Refined patient selection and greater technical experience in prostate brachytherapy were associated with less urinary toxicity.

Introduction

Permanent low-dose-rate prostate brachytherapy (PB) is a standard option for the treatment of early-stage prostate cancer. PB is often chosen because of its excellent long-term disease control, convenience, and relative absence of severe long-term side effects 1, 2, 3. Additional research into brachytherapy-related toxicity and measures to reduce side effects and preserve patients' quality of life are essential.

Irritative and obstructive urinary symptoms are the main sequelae of PB 4, 5. Acute urinary retention (AUR) is a frequent acute urinary toxicity with published rates of 6–36% (5). Although many publications have described urinary toxicity after PB, few reports have included a large number of patients (>300) 3, 4, 6, 7, 8, described a systematic analysis of treatment and dosimetry predictive factors 4, 7, 8, 9, 10, 11, 12, or have had long-term follow-up 4, 6, 7, 8.

We describe the frequency of acute and late urinary toxicity and associated predictive factors for Radiation Therapy Oncology Group (RTOG) toxicity, International Prostate Symptom Score (IPSS) resolution, and changes in urinary bother score in 712 consecutive PB patients with a median follow-up of 57 months. The incidence and predictive factors for AUR in this cohort have been previously published by our group (5).

Section snippets

Methods and Materials

The Prostate Brachytherapy Program at the British Columbia Cancer Agency was established in 1997. As of March 2008, >2,200 patients had undergone PB. The outcome analysis of the first 1,006 consecutive patients revealed a 5-year Kaplan-Meier freedom from biochemical failure rate of 95.7% 13, 14. The program maintains a large prospective database containing baseline clinical and dosimetric data, follow-up prostate-specific antigen (PSA) level, and testosterone level. Toxicity data were collected

Results

A total of 712 patients (76.4% of all implanted patients) with ≥34 months of follow-up were included in this analysis. The median follow-up was 57.1 months (range, 34–103.9). The demographics are listed in Table 1.

Discussion

To our knowledge, this is the second largest single-institution report of long-term urinary toxicity after PB (7) and the largest study reporting the predictive factors for urinary toxicity. Our group had previously published a detailed analysis of AUR and its predictive factors in the same group of patients; therefore, the AUR analysis was not repeated for the purposes of the present study. In our previous work, we found that AUR rates had dramatically decreased in our program. The AUR rate in

Conclusion

The results of our study have shown that long-term urinary toxicity after PB is very low. In most patients, the IPSS normalized within 12–18 months after implantation. A greater prostate dose, greater maximal postimplant IPSS, and AUR slowed the IPSS resolution time. Patients with a larger prostate volume, a larger number of needles used, a greater baseline IPSS, and given HT had more acute toxicity. Patients with late RTOG Grade 2 or greater toxicity had a greater baseline IPSS, greater

Acknowledgments

BC Cancer Agency Prostate Brachytherapy Program: Drs. Mira Keyes (program head), W. James Morris (quality assurance chair), Michael McKenzie, Alexander Agranovich, Tom Pickles, Eric Berthelet, Howard Pai, Winkle Kwan, Mitchell Liu, Jonn Wu, Ross Halperin; medical physics: Ingrid Spadinger (physics head), Robert Harrison, William Kwa, Vincent Lapointe; data analyst: Barb Baerg; statistician: Veronika Morvan.

References (34)

Cited by (0)

Presented at the ESTRO Leipzig, September 2006 and ABS Chicago, May 2007.

Conflict of interest: none.

View full text