Retrospective analysis of high-dose-rate brachytherapy for patients with prostate cancer aged ≥ 75 years

Abstract

Background

High-dose-rate brachytherapy (HDR-BT) is a potential treatment for localized prostate cancer. However, high-quality evidence or literature on HDR-BT in older patients is limited. Therefore, we aimed to evaluate HDR-BT treatment outcomes by comparing results between older and younger patients.

Methods

We included 973 patients with primary prostate cancer who were treated with radiation therapy (external beam radiotherapy + HDR-BT) at our institution between November 1997 and December 2021. Propensity score matching was used. The biochemical freedom from failure (bFFF), prostate cancer-specific survival (PCSS), and adverse events were compared between the two groups. The groups comprised 196 matched patients aged ≥ 75 years and 196 patients aged < 75 years. Multivariate analysis was conducted for bFFF.

Results

The median follow-up period after matching was 79.5 and 107.5 months for patients aged ≥ 75 and < 75 years, respectively. Matched 7-year bFFF was 82.9 and 83.1% in those aged ≥ 75 and < 75 years, respectively. However, 7-year PCSS was 100% and 99.3%, respectively, with no significant differences (p = 0.690 and 0.641, respectively). Age was not associated with bFFF in the multivariate analysis (Hazard ratio = 0.976, 95% Confidence interval: 0.674–1.414, p = 0.898). Conversely, low-grade T classification and, Gleason score, low initial prostate-specific antigen levels, hormone therapy, and higher doses (Biologically Effective Dose [BED] ≥ 243.0 Gy) were significantly associated with higher bFFF. The matched 7-year cumulative incidence rate of grade ≥ 2 genitourinary adverse events was 13.5 and 20.4% for those aged ≥ 75 years and < 75 years, respectively. The gastrointestinal toxicity rate was 5.3 and 4.4%, respectively, with no significant differences observed (p = 0.754 and 0.675, respectively).

Conclusions

Compared with the younger group (< 75 years), HDR-BT for the older group (≥ 75 years) was not associated with adverse events. Moreover, high BED irradiation resulted in better bFFF.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12885-025-15533-7.

Background

There were approximately 1.5 million new cases of prostate cancer worldwide, with 397,000 deaths according to 2022 statistics. Prostate cancer is the second most common cancer in men and ranks fifth in cancer-specific mortality [1]. The median age at diagnosis is 66 years; however, the incidence rate is highest among men aged 70–74 years. Therefore, prostate cancer may become a major public health problem, as the proportion of older individuals increases [2–4]. The number of cases of prostate cancer in Japan in 2021 was 95,584, with an age-adjusted incidence rate of 67.9 per 100,000 people, making it the most common cancer among men [5]. A gradual increase in the number of cases is expected, reaching 117,966 in 2040, an increase of 11.1% compared with 2020 [6].

Older patients often have underlying comorbidities compared with younger patients, requiring a balanced consideration between the risk of adverse events and treatment effectiveness in selecting treatment modalities. Surgery and radiation therapy, hormone therapy, and active surveillance are indicated when considering the risk classification for localized prostate cancer.

Radiation therapy is minimally invasive and has been widely used in older patients with prostate cancer [7]. Significant advances have been achieved in treatment technologies such as intensity-modulated radiation therapy (IMRT) and particle beam therapy in external beam radiotherapy (EBRT), demonstrating favorable outcomes [8]. A dose-dependent relationship exists in radiation therapy for prostate cancer, whereby higher doses give better outcomes [9–11]. Higher dose per fraction is biologically advantageous in cases with a low α/β ratio. Combining EBRT with high-dose-rate brachytherapy (HDR-BT) improves biochemical/clinical disease-free survival compared with using only EBRT, and is now an established treatment option for localized prostate cancer [12]. However, younger patients have been the focus of reports on the analysis of treatment outcomes; high-quality evidence or literature on HDR-BT in older patients is limited. Therefore, we aimed to evaluate the treatment outcomes of HDR-BT by comparing the outcomes of older patients with those of younger patients.

Materials and methods

Selection of patients

Our study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments and was approved by the Ethics Committee of Kawasaki Medical School (approval number: 5329-03).

The International Society of Geriatric Oncology recommends that patients aged ≥ 75 years should be considered older when determining the treatment options and indications for prostate cancer [13]. Furthermore, the Japanese Geriatrics Society defines older patients as those aged ≥ 75 years [14]. Radical treatment is generally indicated when the likely life expectancy is ≥ 10 years [15]. This criterion was considered; moreover, the average life expectancy for a 75-year-old man in Japan is 12.13 and 8.98 years for an 80-year-old man [16]. Based on these findings, we defined older in this study as those aged ≥ 75 years.

We included patients with primary prostate cancer who received radiation therapy (HDR-BT + EBRT or HDR-BT alone) at our institution between November 1997 and December 2021. The exclusion criteria were as follows: patients with lymph node or distant metastasis, those who experienced recurrence after hormone therapy, salvage cases, focal therapy cases, follow-up period of < 12 months, palliative irradiation, and patients with no post-treatment follow-up data (Fig. 1). We classified patients into risk categories using the D’Amico classification [17]. The first day of radiation therapy was considered the beginning of the observation period, which continued until the final visit and survival date. Prostate-specific antigen (PSA) recurrence dates were confirmed during the period up to September 30, 2024. Biochemical failure was determined using the Phoenix definition [18].

Fig. 1.

Flow chart of patient selection HDR-BT high-dose-rate brachytherapy, iPSA initial prostate-specific antigen, BED biologically effective dose

Treatment

The HDR-BT technique and radiation treatment plan are described in our previous study [19, 20]. Here, since the introduction of three-dimensional conformal radiotherapy (3D-CRT) into HDR-BT and EBRT planning, EBRT now includes the entire seminal vesicle in the clinical target volume (CTV) for patients with T3b stage, while for other patients, the CTV includes the base of the seminal vesicle. For HDR-BT, irradiation was performed only in the areas where prostate cancer had invaded the seminal vesicles. HDR-BT was planned so that the minimum CTV dose was 95% of the prescribed dose. Up to 120% and 60% of the prescribed dose was allowed in the urethra and rectum, respectively. EBRT was performed using a two-dimensional four-field plan until March 31, 2009. While 3D-CRT was introduced in April 2009, IMRT was introduced in April 2019. In the EBRT plus HDR-BT combination therapy group, EBRT was administered first from November 1997 to June 2011, with HDR-BT administered during treatment. After that period, HDR-BT was administered first, with EBRT initiated within 2 weeks. The treatment plan for HDR-BT was calculated using Planning Tool for Advanced Treatment Oncology (Nucletron, Veenendaal, Netherlands), Oncentra version 3.3.3.86 (Nucletron), and version 4.5.3 (Nucletron). Treatment was conducted using microSelectron v2 (Nucletron).

For patients with prostate volume > 40 ml, 3 months of hormone therapy was initially administered to reduce the prostate volume. Moreover, long-term adjuvant hormone therapy was administered based on risk since 2019.

Initially, we performed pelvic lymph node dissection (PLND) on patients at high risk, as it was believed to influence outcomes. Later, we began omitting PLND when imaging showed no findings suggestive of lymph node metastasis.

Statistical considerations

SPSS version 30 (SPSS Inc., Chicago, IL) and R (R Foundation for Statistical Computing, Vienna, Austria) were used for statistical analysis. Statistical significance was set at p < 0.05. Furthermore, propensity score matching was used to minimize bias due to covariates. The allocation variables included two groups: an older group (≥ 75 years) and a younger group (< 75 years). Covariates were classified as previously described as T classification, Gleason score, initial prostate-specific antigen (iPSA) levels, hormone therapy, and biologically effective dose (BED) (Fig. 1) [21]. Using all covariates in the analysis, the logistic regression model was used to calculate the predicted probability of inclusion in the older group (≥ 75 years). The caliper value was set to 0.2 times the logit-transformed standard deviation of the propensity score. Using the nearest-neighbor matching method, one patient aged ≥ 75 years was matched with another aged < 75 years via non-replacement sampling. The standardized mean difference (SMD) was used to assess the balance of covariates before and after matching. We considered the balance acceptable if the SMD was within 0.15. The SMD for each variable before and after matching is shown in Supplementary Table 1. We compared the biochemical freedom from failure (bFFF), prostate cancer-specific survival (PCSS), and overall survival (OS) between the two groups of 196 matched older patients and 196 patients, both aged < 75 years. Treatment-related deaths were included in PCSS. In a previous multicenter analysis of EBRT combined with HDR-BT for prostate cancer, the BED of the administered dose was calculated and compared using the same method as in this study [22]. The α/β ratio for prostate cancer cells is estimated to be low, around 1.5 Gy [23]. Assuming an α/β ratio of 1.5 Gy, the BED was calculated using the following formula.

In this study, we used BED = 243 Gy—the median prescribed dose—as the cutoff for “high dose”.

Before and after conducting matched data analysis using the following statistical analyses. Percentages were analyzed using the chi-square test and Fisher’s exact test. The Mann–Whitney U test for skewed data was used to compare medians. The Kaplan–Meier method was used to analyze bFFF, PCSS, OS, and accumulated toxicity; comparisons were made using the log-rank test. The crude data were analyzed using multivariate analyses, including Cox proportional hazards models for bFFF. Variables and strata from previous studies were referenced [21, 24].

Toxicity

The Common Terminology Criteria for Adverse Events version 5.0 were used to assess genitourinary (GU) and gastrointestinal (GI) toxicities. GU toxicity includes urinary tract obstruction and hematuria, radiation cystitis, radiation urethritis, and urinary fistula. GI toxicity includes rectal bleeding and radiation colitis. Events that were grade 2 or higher were counted. We conducted long-term follow-up examinations at our hospital, including patients who had received > 10 years of post-treatment. Patients who could no longer attend appointments were contacted by the outpatient physician at our hospital as needed. The physician obtained information by requesting medical records from the healthcare facility where the patient was receiving treatment or via telephone.

Results

Clinical characteristics

Among 1158 patients with prostate cancer who were treated with HDR-BT, 973 were included after excluding 185 who met the exclusion criteria. We compared patient characteristics between the two groups: 196 aged ≥ 75 years and 777 aged < 75 years (Table 1). The median age was 77 years in 196 older patients aged ≥ 75 years (range: 75–84 years), and the median follow-up period was 79.5 months (range: 12–195 months). The median age and follow-up of 777 patients aged < 75 years were 68 years (range: 48–74 years) and 106 months (range: 12–294 months), respectively. The proportions of Gleason scores and D’Amico classifications differed significantly (p = 0.013 and 0.006, respectively). Table 1 shows the details of 196 patients whose backgrounds were matched using propensity score matching. After matching, the median age and follow-up period among 196 patients aged < 75 years were 68 years (range: 54–74 years) and 107.5 months (range: 12–294 months), respectively. Significant differences were not observed except during the follow-up period. Table 2 presents the specific EBRT and HDR-BT protocols, while the complete protocol details are provided in Supplementary Table 2. In both groups, the median BED combining EBRT and HDR-BT before matching was 243.0 Gy (range: 225.0–270.3 Gy) and 243.0 Gy (range: 176.0–270.3 Gy) in those aged < 75 years after matching, with no significant difference between the two groups.

Table 1.

Patient characteristics before and after propensity score matching

		Crude data					Matched data
Variables	Strata	Older group (≥ 75 years)		Younger group (< 75 years)		p-value	Older group (≥ 75 years)		Younger group (< 75 years)		p-value
		n = 196		n = 777			n = 196		n = 196
		No. or Median (range)	%	No. or Median (range)	%		No. or Median (range)	%	No. or Median (range)	%
Age		77 (75–84)		68 (48–74)			77 (75–84)		68 (54–74)
Follow-up	Months	79.5 (12–195)		106 (12–294)		< 0.001*	79.5 (12–195)		107.5 (12–294)		< 0.001*
T category	1	23	11.7	130	16.7	0.059	23	11.7	29	14.8	0.691
	2	141	71.9	480	61.8		141	71.9	136	69.4
	3	31	15.8	155	19.9		31	15.8	31	15.8
	4	1	0.5	12	1.5		1	0.5	0	0.0
iPSA	ng/mL	8.1 (3.9–119)		9.1 (2.2–605)		0.075	8.1 (3.9–119)		8.0 (3.9–107.0)		0.670
Gleason score	−6	49	25.0	248	31.9	0.013 *	49	25.0	47	24.0	0.691
	7	100	51.0	306	39.4		100	51.0	108	55.1
	8-	47	24.0	223	28.7		47	24.0	41	20.9
D’Amico classification	Low	18	9.2	128	16.5	0.006 *	18	9.2	22	11.2	0.284
	Int	73	37.2	219	28.2		73	37.2	85	43.4
	High	105	53.6	430	55.3		105	53.6	89	45.4
Modality	HDR-BT alone	20	10.2	60	7.7	0.184	20	10.2	15	7.7	0.479
	EBRT་HDR-BT	176	89.8	717	92.3		176	89.8	181	92.3
PLND	Yes	49	25.0	238	30.6	0.104	49	25.0	51	26.0	0.908
Hormone therapy	Yes	116	59.2	433	55.7	0.400	116	59.2	118	60.2	0.918

iPSA initial prostate-specific antigen, int intermediate, HDR-BT high-dose-rate brachytherapy, EBRT external beam radiotherapy, PLND pelvic lymph node dissection

* <0.05

Table 2.

Details of the HDR-BT combined with EBRT protocols before and after propensity score matching

				Crude data					Matched data
Variables	Strata			Older group (≥ 75 years)		Younger group (< 75 years)		p-value	Older group (≥ 75 years)		Younger group (< 75 years)		p-value
				n = 196		n = 777			n = 196		n = 196
				No.	%	No.	%		No.	%	No.	%
Irradiation method	EBRT	HDR-BT	BED [Gy] (α/β = 1.5)
Irradiation dose	-	37.5 Gy/5fr	225.0	20	10.2	60	7.7	0.251	20	10.2	15	7.7	0.556
	36.8 Gy/16fr	24 Gy/4fr	213.2	43	21.9	166	21.4		43	21.9	46	23.5
	39.0 Gy/13fr	18 Gy/2fr	243.0	72	36.7	274	35.3		72	36.7	84	42.9
	39.0 Gy/13fr	20 Gy/2fr	270.3	46	23.5	235	30.2		46	23.5	41	20.9
	Others			15	7.7	42	5.4		15	7.7	10	5.1

EBRT external beam radiotherapy, HDR-BT high-dose rate-brachytherapy, BED biologically effective dose, fr fractions

Oncologic endpoints

In the pre-matching dataset, biochemical failure was observed in 34 and 178 cases in the older (≥ 75 years) and younger (< 75 years) groups, respectively. The 7-year bFFF was 82.9% (95% confidence interval [CI]: 76.6–89.2%) and 80.8% (95% CI: 77.9%–83.7%, p = 0.562), respectively, in the older and younger groups, with no significant difference. After matching, it was 82.9% (95% CI: 76.6%–89.2%) and 83.1% (95% CI: 77.6%–88.6%, p = 0.690) in the older and younger groups, with no significant difference (Fig. 2A). For each risk category in the older and younger groups, the 7-year bFFF before matching are as follows: high-risk, 83.2% (95% CI: 75.4–91.0%) and 75.9% (95% CI: 71.6–80.2%, p = 0.252); intermediate-risk, 80.1% (95% CI: 68.1–92.1%) and 87.6% (95% CI: 83.1–92.1%, p = 0.465); and low-risk, 90.0% (95% CI: 71.4–108.6%) and 86.2% (95% CI: 79.7–92.7%, p = 0.737), respectively. After matching, it was 83.2% (95% CI: 75.4–91.0%) and 79.7% (95% CI: 71.1–88.3%, p = 0.991, Fig. 2B) in the high-risk; 80.1% (95% CI: 68.1–92.1%) and 88.5% (95% CI: 81.4–95.6%, p = 0.344, Fig. 2C) in the intermediate-risk; and 90.0% (95% CI: 71.4–100.0%) and 75.9% (95% CI: 57.3%–94.5%, p = 0.342, Fig. 2D) in the low-risk group. No significant differences were observed across groups. Age was not associated with bFFF (HR = 0.976, 95% CI: 0.674–1.414, p = 0.898) in multivariate analysis prior to matching (Table 3). In contrast, low-grade T classification and Gleason score, low iPSA, hormone therapy, and high doses (BED ≥ 243.0 Gy) were significantly associated with high bFFF (Table 3).

Fig. 2.

Matched biochemical freedom from failure rate between the older (≥ 75 years) and younger (< 75 years) groups. A Total population. B High-risk group. C Intermediate risk group. D Low-risk group. The 7-year bFFF did not significantly differ among the groups bFFF biochemical freedom from failure

Table 3.

Univariate and multivariate analyses of factors associated with bFFF before propensity score matching

Variables	Strata	Univariate analysis			Multivariate analysis
		HR	95% CI	p-value	HR	95% CI	p-value
Age	< 75	1	(referent)		1	(referent)
	≥ 75	0.897	0.620–1.297	0.562	0.976	0.674–1.414	0.898
T category	1–2	1	(referent)		1	(referent)
	3–4	2.635	1.993–3.485	< 0.001*	2.131	1.539–2.952	< 0.001 *
Gleason score	≤ 7	1	(referent)		1	(referent)
	8 ≤	1.663	1.259–2.195	< 0.001*	1.429	1.056–1.934	0.021 *
iPSA (ng/mL)	≤ 20	1	(referent)		1	(referent)
	20 <	2.918	2.213–3.848	< 0.001*	2.375	1.736–3.249	< 0.001 *
PLND	Yes	1	(referent)		1	(referent)
	No	0.562	0.428–0.738	< 0.001*	1.017	0.733–1.411	0.919
Hormone therapy	Yes	1	(referent)		1	(referent)
	No	1.272	0.971–1.665	0.091	1.793	1.341–2.397	< 0.001*
BED [Gy]	243.0 ≤	1	(referent)		1	(referent)
	< 243.0	1.967	1.499–2.581	< 0.001*	1.867	1.420–2.454	< 0.001*

bFFF biochemical freedom from failure, iPSA initial prostate-specific antigen, PLND pelvic lymph node dissection, BED biologically effective dose

* <0.05

In the total data, 2 cases of prostate cancer death were observed among the older group and 12 cases among the younger group during the observation period. All of these deaths were in the intermediate- and high-risk groups. Both groups included one grade 5 case. The 7-year PCSS for the older and younger groups before matching were 100.0% (95% CI: 100.0–100.0%) and 99.1% (95% CI: 98.3–99.9%), respectively, with no significant difference (p = 0.930). After matching, the proportions were 100% (95% CI: 100.0–100.0%) and 99.3% (95% CI: 97.9–100.0%), respectively, with no significant difference (p = 0.641; Fig. 3).

Fig. 3.

Matched prostate-cancer-specific survival between the older (≥ 75 years) and younger (< 75 years) groups. The 7-year PCSS after matching showed no significant difference PCSS prostate-cancer-specific survival

The total number of deaths during the follow-up period was 38 and 104 in the older and younger groups, respectively. The 7-year OS before matching was 90.9% (95% CI: 86.4%–95.4%) in the older group and 94.9% (95% CI: 93.3%–96.5%, p < 0.001) in the younger group. The trend showed a significant difference. After matching, it was 90.9% (95% CI: 86.4%–95.4%) and 94.8% (95% CI: 91.5%–98.1%, p < 0.001), respectively, with significantly lower rates in the older group (Fig. 4). The median survival time (MST) was 170 and 249 months in the older and younger groups, respectively.

Fig. 4.

Matched overall survival between the older (≥ 75 years) and younger (< 75 years) groups. The 7-year OS after matching was significantly lower in the older group (≥ 75 years) OS overall survival

Acute and late adverse events

Table 4 shows the GU and GI adverse events. Urinary tract obstruction was the most common GU toxicity, occurring at ≥ grade 2 in 70.3% (26 cases) of older patients (≥ 75 years) and 69.9% (128 cases) of younger patients (< 75 years). It also occurred in 13.3% (26/196) of cases in the older group (≥ 75 years) and in 16.5% (128/777) of the cases in the younger group (< 75 years). Management for urinary tract obstruction included transurethral internal urethrotomy, transurethral urethral dilation with catheter placement, suprapubic catheter insertion, and placement of a urinary catheter alone. We observed one case of grade 5 late GU toxicity in both groups. This was a case of severe anemia and bladder tamponade caused by hemorrhagic cystitis. The 7-year cumulative incidence rate of ≥ grade 2 GU adverse events before matching was 13.5% (95% CI: 8.2%–18.8%) and 16.0% (95% CI: 13.3%–18.7%, p = 0.908) in the older and younger groups, respectively (Fig. 5A). The rates after matching were 13.5% (95% CI: 8.2%–18.8%) and 20.4% (95% CI: 14.5%–26.3%, p = 0.754), respectively, with no significant difference (Fig. 5B). For GI adverse events, the 7-year cumulative incidence rates of ≥ grade 2 before matching were 5.3% (95% CI: 2.0–8.6%) and 3.8% (95% CI: 2.4–5.2%, p = 0.513) in the older and younger groups, respectively (Fig.5C). The rates after matching were 5.3% (95% CI: 2.0–8.6%) and 4.4% (95% CI: 1.5–7.3%, p = 0.675), respectively, with no significant difference (Fig. 5D).

Table 4.

Comparisons of toxicities before and after propensity score matching

		Crude data					Matched data
Toxicities	Grade (CTCAEVer5.0.)	Older group (≥ 75 years)		Younger group (< 75 years)		p-value	Older group (≥ 75 years)		Younger group (< 75 years)		p-value
		n = 196		n = 777			n = 196		n = 196
		No.	%	No.	%		No.	%	No.	%
Genitourinary (GU)	2	32	16.3	170	21.9	0.173	32	16.3	47	24.0	0.233
	3	4	2.0	12	1.5		4	2.0	4	2.0
	4	0	0.0	0	0.0		0	0.0	0	0.0
	5	1	0.5	1	0.1		1	0.5	1	0.5
Gastrointestinal (GI)	2	8	4.1	24	3.1	0.686	8	4.1	6	3.1	0.840
	3	1	0.5	7	0.9		1	0.5	2	1.0
	4	0	0.0	0	0.0		0	0.0	0	0.0
	5	0	0.0	0	0.0		0	0.0	0	0.0

GU genitourinary, GI gastrointestinal

Fig. 5.

Accumulated incidence of grade ≥ 2 toxicity between the older (≥ 75 years) and younger (< 75 years) groups. A Crude data on genitourinary toxicity. B Genitourinary toxicity after matching. C Crude data on gastrointestinal toxicity. D Gastrointestinal toxicity after matching. The 7-year cumulative incidence rate did not significantly differ among the groups GU genitourinary, GI gastrointestinal

Discussion

We showed that the treatment outcomes and adverse events of HDR-BT in the older group (≥ 75 years) did not differ from those in the younger group (< 75 years), and age was not associated with bFFF. There have been limited reports on the long-term treatment outcomes of HDR-BT for prostate cancer in older patients, and our findings provide valuable information for consideration of the treatment modality selection.

Yamazaki et al. compared the treatment outcomes of HDR-BT in older patients aged ≥ 75 years with localized prostate cancer and those aged < 75 years, as well as in those aged ≥ 80 and < 80 years. They reported that equivalent biochemical control rates were achieved in both groups. In addition, we confirmed in our study that bFFF is equivalent in older patients with prostate cancer [21, 24].

Our findings showed that bFFF and PCSS were comparable between groups; however, OS was significantly lower in the older group. There were two cases of death caused by prostate cancer, 36 cases of death due to other causes, and a total of 38 deaths in the older group. The deaths were not caused by reduced treatment efficacy, but rather reflect a high rate of non-cancer-related mortality associated with aging and its accompanying comorbidities.

According to the EAU - EANM - ESTRO - ESUR - ISUP - SIOG Guidelines on Prostate Cancer, radical treatment for localized prostate cancer is primarily performed on intermediate- and high-risk patients, and the prerequisite for achieving therapeutic benefit is a life expectancy of at least 10 years [25]. Furthermore, according to the National Comprehensive Cancer Network Guideline risk classification, radiation therapy is indicated for low- to intermediate-risk patients with an expected survival of ≥ 10 years. For high-risk patients, the expected patient survival is ≥ 5 years if symptoms appear; otherwise, it is ≤ 5 years if the patient is asymptomatic [15]. In our study, the median age of the older group was 77 years (range: 75–84 years). Overall, the patient cohort has a median age of 77 years, with an MST of 170 months, a 7-year PCSS of 100%, and a 7-year OS of 90.9%, indicating favorable outcomes. Our results suggest that our patient selection for treatment was appropriate and consistent with these guidelines.

In our study, we found no association between age and bFFF. Akash et al. reported that age was not a predictive factor for prostate cancer-specific mortality in older men with intermediate-risk prostate cancer who received HDR-BT alone or combined with EBRT [26]. Furthermore, Ashesh B Jani et al. reported that age > 70 years was not associated with the incidence of biochemical failure following definitive radiotherapy. Therefore, age should not be used as an independent factor in determining treatment strategies and counseling for patients with prostate cancer [27]. Furthermore, we suggested that selecting therapeutic strategies based solely on age should be avoided. Furthermore, we showed that low-grade T classification and Gleason score, low iPSA, hormone therapy, and higher doses (BED ≥ 243.0 Gy) were significantly associated with higher bFFF. A dose-dependent relationship exists in radiation therapy for prostate cancer. Nevertheless, to the best of our knowledge, this issue has not been examined in older patients [9–11]. We demonstrated that administering higher BED in HDR-BT + EBRT treatment is associated with a lower incidence of biochemical recurrence. Reports indicate that, despite increasing the dose, the BED ≥ 200 Gy (α/β ratio = 1.5 Gy) did not increase biochemical non-failure rates [28]. Therefore, further study is needed to determine the optimal BED for EBRT + HDR-BT treatment in older patients.

There are various reports on toxicity, and in general, the incidence of GU toxicity is not uncommon, whereas GI toxicity is usually lower [29]. The absorbed dose decreases sharply depending on the distance from the source, owing to the physical characteristics of HDR-BT. This dose distribution theoretically helps protect adjacent normal tissues such as the rectum. Several studies have reported late adverse events following HDR-BT, which are compared with those following stereotactic radiosurgery and low-dose-rate brachytherapy. The incidence of GI toxicity following HDR-BT has been low, consistent with our study findings [30–33]. In contrast, several reports show that grade ≥ 2 adverse events associated with GU toxicity increased cumulatively over the years [34–36]. Urinary tract obstruction is the most common late adverse event in GU toxicity [37]. In our study, it was also the most common grade ≥ 2 GU toxicity, occurring in 26 of 37 (70.3%) adverse events in the older group and in 128 of 183 (69.9%) in the younger group. The onset of urinary tract obstruction may be related to increased dose to the membranous part of the urethra, which can cause fibrotic scarring [38]. Deger et al. reported a 9% incidence of urethral stricture after a median follow-up of 5 years in 442 patients treated with HDR-BT and EBRT combination therapy [39]. Our study showed a higher incidence of urinary tract obstruction ≥ 2 than that reported in previous studies. One possible reason is that the median follow-up period in our study was longer than in previous studies—6.6 and 8.8 years for older and younger groups, respectively—and increased over time. In addition, because a two-dimensional plan was used for treatment, there was a possibility that unexpectedly high doses were delivered to the urethra and bladder. Yamazaki et al. compared older and younger patients treated with HDR-BT and reported that late adverse events were similar in those aged ≥ 75 years [24]. In addition, we found no statistically significant difference in the cumulative incidence of adverse events, consistent with the above findings. Hormone therapy increases the risk of developing hypertension and diabetes, and the risk factor vulnerability increases with age. Lisa et al. reported that having a history of hypertension is a significant predictor of urinary tract obstruction, which increases the incidence of late adverse events, especially when combined with diabetes. Microvascular system changes induced by hypertension may cause poor blood circulation, which may exacerbate urethral damage and strictures [40]. These findings have been reported in previous studies, and urethral stricture—which accounted for many of the GU toxicities in our study—may have contributed to the observed increasing trend over time. However, owing to the limitations associated with retrospective studies, data on underlying conditions such as hypertension or diabetes were not obtained. We did not investigate the effects of these background factors on adverse events; hence, further studies are required. Our study showed that higher BED (≥ 243.0 Gy) was associated with improved bFFF. Nonetheless, a high BED was associated with an increased risk of grade ≥ 2 GU and GI adverse events in a previous study [41]. Compared with younger patients, older counterparts often exhibit a higher prevalence of underlying comorbidities. Therefore, treatment decisions should be made by balancing the risk of adverse events with the potential therapeutic benefits.

One case of grade 5 late GU toxicity was observed in both groups. The case involved severe anemia and bladder tamponade caused by hemorrhagic cystitis. A study of 709 patients with prostate cancer who were treated with radiation therapy revealed that hemorrhagic cystitis was most common in patients who received EBRT (50%), followed by intensity-modulated radiation therapy (13.9%), and brachytherapy (7.6%) [42]. Thereby suggesting that hemorrhagic cystitis was not more common among patients treated with brachytherapy. In the study, grade 5 hemorrhagic cystitis was observed in four cases (0.7%); however, the cause of the onset could not be identified [42]. In our study, one patient (aged ≥ 75 years) had a pacemaker due to complete atrioventricular block and was administered aspirin. This may have contributed to hemorrhagic cystitis. In another patient (aged < 75 years), the underlying cause remained unclear despite considering the patient’s background. A slight change in needle position owing to the physical characteristics of HDR-BT can cause a significant dose change, potentially resulting in inadvertent high-dose radiation. Further studies are needed to identify strategies that reduce serious adverse events involving the bladder and urethra.

The study had some limitations. Patient backgrounds with respect to comorbidities were not considered; the single-center retrospective analysis also contributed to this limitation. Furthermore, unrecognized issues may exist because of the long survey period and the coexistence of various treatment protocols. At treatment initiation, no established protocol for HDR-BT in prostate cancer existed in Japan; accordingly, we explored reducing the number of fractions for HDR-BT and EBRT. Consequently, irradiation protocols varied widely, including cases of HDR-BT alone, which also limits the findings of the study. Regarding adverse events, baseline GU and GI function were not evaluated, and latent, poorly recognized urethral or rectal adverse events may exist, particularly in the older population. Since 2019, long-term hormone combination therapy has been administered based on risk assessment. Given the short follow-up period, this therapy may have significantly influenced bFFF. Nevertheless, our study shows the outcomes of HDR-BT in a large patient cohort, providing important insights for older patients with prostate cancer when selecting treatment options.

Conclusions

We reported the treatment outcomes of HDR-BT in the older group (≥ 75 years) with prostate cancer. In this population, high BED irradiation was associated with improved bFFF. Adverse events did not differ; however, further research is warranted on the impact of optimal dose and patient background factors on adverse events.

Abbreviations

IMRT

Intensity-modulated radiation therapy

EBRT

External beam radiotherapy

HDR-BT

High-dose-rate brachytherapy

3D-CRT

Three-dimensional conformal radiation therapy

CTV

Clinical target volume

PLND

Lymph node dissection

PSA

Prostate-specific antigen

iPSA

Initial prostate-specific antigen

BED

Biologically effective dose

SMD

Standardized mean difference

bFFF

Biochemical freedom from failure

PCSS

Prostate cancer-specific survival

OS

Overall survival

GU

Genitourinary

GI

Gastrointestinal

CI

Confidence interval

MST

Median survival time

Authors’ contributions

KM collected the data, performed the statistical analyses, and drafted the manuscript. TO performed the statistical analyses. NK performed the data curation. WK and RT conceptualized the study. KK reviewed the manuscript and supervised the study. All authors participated in the design of this study and read and approved the final manuscript.

Funding

Not applicable.

Data availability

The datasets of this study are not publicly available because of the institutional review board’s requirements. Still, they are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

Our study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments and was approved by the Ethics Committee of Kawasaki Medical School (approval number: 5329-03). The need for informed consent was exempted owing to the retrospective nature of the study. The choice to opt out was provided through notifications displayed on the institutional’s website.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.