Unfavorable Intermediate- and High-Risk Prostate Cancer Treated With Predominantly Brachytherapy Alone With Long-Term Follow-Up

Abstract

Objective: Assess 10-year outcomes of brachytherapy (BT) with or without supplemental external-beam radiation therapy (S-EBRT) for treatment of unfavorable intermediate-risk prostate cancer (U-IRPC) and high-risk prostate cancer (HRPC).

Materials and Methods: Retrospective analysis using multivariable analysis (MVA) and propensity score matching was performed on 156 patients with U-IRPC and HRPC between 2004 and 2016. Favorable HRPC was defined as T1c-T2c, Gleason group 4, and prostate-specific antigen (PSA) < 10.0. In total, 129 patients underwent BT alone using iodine-125 to 145 Gy, while 27 underwent S-EBRT + BT boost to 110 Gy. S-EBRT dose was 45-46 in 1.8-2.0 Gy fractions to the prostate and seminal vesicles. Freedom from biochemical failure (FFBF) was defined by the Phoenix definition of PSA failure. Complications were assessed using the Radiation Therapy Oncology Group grading scale.

Results: Median follow-up was 8.2 vs 8.3 years for BT vs S-EBRT + BT. FFBF for U-IRPC vs HRPC was 80.7% vs 55.6% (P < .01), and metastases-free survival (MFS) was 94.5% vs 72.6% (P <.01). The S-EBRT + BT group had higher Gleason group (P=.01) and higher percent positive biopsy cores > 50% (P < .01), but also higher use of neoadjuvant androgen deprivation therapy, P < 0.01. On MVA, higher clinical stage (P < .01) and Gleason group (P=.04) independently predicted a lower MFS, whereas higher Charlson score predicted lower overall survival, P=0.01. The adjusted 10-year FFBF and MFS for BT alone vs S-EBRT + BT were 76.8% vs 72.9% (P=.70) and 90.8% vs 87.3% (P=.81). Favorable HRPC had a 10-year FFBF of 91.7% vs unfavorable HRPC of 31.7%, P < 0.01. Prevalence of urinary (P=.04) and rectal (P < .01) complications was higher using S-EBRT, although this was mostly in grades 1 and 2.

Conclusion: Low-dose-rate BT using iodine-125 alone is a reasonable treatment option for U-IRPC and favorable HRPC, which is effective, convenient, and cost-effective.

Keywords: prostate cancer, brachytherapy, high risk, radiation therapy

Introduction

Low-dose-rate (LDR) brachytherapy (BT) using radioactive seeds has shown excellent 10-year results in multiple studies.^1-3 However, most patients in these trials were low risk. The National Comprehensive Cancer Network (NCCN) guidelines recommend local therapy for treating patients with unfavorable intermediate-risk prostate cancer (U-IRPC) and high-risk prostate cancer (HRPC), but do not recommend BT as monotherapy without the use of supplemental external-beam radiation therapy (S-EBRT) or neoadjuvant androgen deprivation therapy (NADT).⁴ These guidelines tend to emphasize the use of pituitary ablation as opposed to prostate ablation. Our prior experience of patients with favorable IRPC and U-IRPC showed impressive results using predominantly BT alone compared with radical prostatectomy (RP) and external-beam radiation therapy (EBRT) with or without NADT.⁵ Our aim was to perform a retrospective analysis of U-IRPC and HRPC, comparing BT with or without S-EBRT, with long-term follow-up.

Materials and Methods

Results

Table 2. Multivariable Analysis of Prognostic Factors

Main Outcomes

None of the survival outcomes were significantly different between BT alone vs S-EBRT + BT boost. The 10-year FFBF for BT vs S-EBRT was not significant in both unadjusted (77.0% vs 71.6%, P = .53) and adjusted models (76.8% vs 72.9%, P = .70). The propensity-adjusted 10-year MFS, prostate cancer-specific survival (PCSS), and overall survival (OS) for BT vs S-EBRT were 90.8% vs 87.3% (P = .81), 98.4% vs 87.3% (P = .36), and 73.4% vs 78.0% (P = .18), respectively (Figure 1, Table 3).

Figure 1.

Adjusted Kaplan-Meier estimates of freedom from biochemical failure, comparing brachytherapy alone vs supplemental external-beam radiation therapy plus brachytherapy.

Table 3. Kaplan-Meier Estimates with 95% Confidence at 10 y by Treatment

Table 4. Patterns of Failure

Table 5. Prevalence of Radiation Therapy Oncology Group Late Complications

Discussion

The NCCN guidelines historically only recommended monotherapy BT for the treatment of low-risk prostate cancer, and only since 2015 recommended its use for favorable IRPC.⁴ The NCCN currently only recommends BT as a boost for U-IRPC and HRPC, along with S-EBRT or whole pelvic radiation, and NADT. Whole pelvic radiation is endorsed by the NCCN, along with 6 months of NADT for U-IRPC and a minimum of 18 months of NADT for HRPC, although conflicting data exist regarding whole pelvic radiation, and none of these trials used BT.^4,17-19 The most recent trial that was positive for pelvic radiation, but did not use BT, showed an improvement in biochemical failure, disease-free survival, and distant metastases-free survival, and also incorporated the use of PSMA-PET to assess distant disease, which is much more sensitive than prior imaging.¹⁹ Increasing the sensitivity of detecting distant disease will influence oncological outcomes, making MFS closer to a surrogate of FFBF or progression-free survival, and less of a predictor of PCSS. For patients not undergoing RP, the standard recommendation has been long-term NADT with EBRT, considered category 1 by the NCCN. One problem with this recommendation is that patients with HRPC are a heterogeneous mix of patients.²⁰ One approach could be to segregate HRPC into a favorable vs unfavorable category, in which favorable could represent those with a reasonable probability of having organ-confined disease or disease into the capsule. With the increasing use of MRI and PSMA-PET, selecting U-IRPC and HRPC that may have localized disease should become more feasible, making these patients amenable to BT alone, which provides more ablative doses than EBRT.⁵ Our preference would be to offer BT alone to those favorable HRPC with Gleason group 4, PSA ≤ 10.0, T1c-T2c disease, including T3a seen on MRI (not T3a on DRE). Also, it seems that our favorable HRPC patients did just as well as those with U-IRPC, suggesting that these could be merged together as a risk group so that favorable HRPC could be categorized as U-IRPC. However, we recommend excluding those with seminal vesicle invasion and patients with a high risk of systemic disease. The majority of our patients received neither S-EBRT nor NADT, and the patients who did undergo NADT were short-term of 3-6 months. Also, none of our patients underwent whole pelvic radiation. A recent randomized study showed no benefit with the addition of S-EBRT to BT for favorable IRPC, but this study did not include U-IRPC or HRPC.²¹ Despite the lack of additional therapies, the BT alone cohort showed excellent FFBF, MFS, and PCSS at 10 years, although for U-HRPC, FFBF and MFS were much lower at 10 years at 31.7% and 43.3%, and no significant difference was found with the addition of S-EBRT, despite 59.3% of S-EBRT + BT boost undergoing NADT vs only 7.0% for the BT-alone group. PCSS remained high for both U-IRPC and HRPC, likely due to the availability of effective salvage therapies that can prolong survival, but are not curative. Although the S-EBRT group did have higher Gleason grouping and higher PPBC > 50%, MVA and propensity score matching did not show any benefit in regards to oncological outcomes with S-EBRT, despite a majority of S-EBRT undergoing NADT. Is it possible that the ablative doses given to the prostate and surrounding capsule are the most important factor in treating U-IRPC and favorable HRPC, while having very high rates of PCSS? The ASCENDE-RT trial treated all patients with NADT of 1 year along with pelvic radiation, comparing high-dose EBRT to the prostate vs BT boost, and this showed improvement of progression-free survival in those who underwent BT, although the BT boost arm did experience more urinary and rectal toxicities.²²,²³ This trial indicates that the ablative doses provided by BT still have an important role in the treatment of U-IRPC and favorable HRPC. A publication of 2 randomized studies by Merrick et al was done on patients with intermediate- and high-risk prostate cancer, which did not show any improvement with the addition of S-EBRT, compared with BT alone.²⁴ It seems that one difference between the ASCENDE-RT trial vs the Merrick study was that the ASCENDE-RT used pelvic radiation, while the Merrick study may have included only the prostate and seminal vesicles when using S-EBRT. Additionally, a Surveillance, Epidemiology, and End Results Medicare analysis of 5835 patients also confirmed the lack of benefit of S-EBRT in addition to BT in regards to PCSS.²⁵

One retrospective study also showed that BT was reasonable for HRPC, which compared 2557 HRPC with a median follow-up of 63.5 months, comparing RP vs EBRT vs BT, with NADT given in 19% vs 93% vs 53%, respectively, P < 0.0001.²⁶ Biochemical relapse-free survival and clinical relapse-free survival (cRFS) were equivalent between BT and EBRT, but both were higher than RP. This may reflect differences on how PSA failures were defined, as the definition for failure is more sensitive for RP compared with EBRT.^12,27 Also, the RP group had more failures in cRFS, as RP patients who underwent adjuvant radiation were counted as failures despite not having failed biochemically.

The main reason for the benefit of BT is that it provides more ablative doses than can be achieved by EBRT with or without NADT by producing a lower PSA nadir of <0.1.²⁸ Although ablative doses may be effective in eradicating prostate cancer, high doses can cause significant morbidity. However, one can properly select patients who can tolerate these ablative doses by using the American Urological Association urinary score to select which patients would have a lower probability of long-term urinary effects.²⁹ In the current study, we had acceptable side effects, and these side effects were more common in those undergoing S-EBRT, although most of these were grades 1 and 2, with severe grade 3 and 4 complications rates being low.

The limitations of this study include that the majority of patients were U-IRPC, with only 20.5% being HRPC. Also, there was an imbalance of GS and PPBC > 50%, with worse patients in the S-EBRT group, showing that the BT-alone patients were subjected to selection bias. We tried to account for these differences by using propensity score matching, although this is a relatively small study with the limitation of being retrospective. Unfortunately, there are few large randomized trials using prostate BT. Low reimbursements, combined with LDR BT requiring more training and skill, give little motivation for physicians to offer LDR BT to their patients.^30-32 This has led to a decline in the use of LDR BT in clinical practice and residency training, with the potential downstream effect of fewer publications on the role of LDR BT in patients with prostate cancer, and potentially leading to the unavailability to many patients of one of the most successful treatment options for prostate cancer. Thus, we publish our 10-year results on the use of LDR BT on patients with more advanced disease, which, if done properly, can yield favorable oncological outcomes with acceptable rates of side effects, suggesting LDR BT to be a reasonable option in the treatment of U-IRPC and selected HRPC.