Radical prostatectomy for clinically localized prostate cancer in patients aged 75 years or older: comparison with primary androgen deprivation therapy

Abstract

Objective: To determine whether radical prostatetomy (RP) is suitable for prostate cancer patients with age ≥75 years in comparison to primary androgen deprivation therapy (PADT).

Patients and methods: A cohort study was conducted in clinically localized prostate cancer patients with ≥75 years of age who underwent RP or PADT at six institutions from 2005 to 2013. Patients who had less than 12 months of follow-up, or received neoadjuvant or adjuvant therapy were excluded. We compared clinical characteristics, cancer-specific and overall survivals, and post-treatment complication rates between two groups.

Results: We included 92 and 99 patients in the RP and PADT groups, respectively. In survival analyses, there were no significant differences in cancer-specific and overall survivals (p = .302 and .995, respectively). The incidence of serious adverse events (cardio- or cerebrovascular event, or bone fracture) was higher in the PADT group (p = .001). Multivariable analysis showed that PADT had a worse effect on the serious adverse events (OR 10.12, p = .038).

Conclusions: In selected elderly patients, RP was safe and effective for treatment of localized prostate cancer, as compared to PADT. Surgical treatment options should be considered in elderly patients with respect to life expectancy, rather than chronological age.

Keywords:

• Localized prostate cancer
• 75 years or old
• life expectancy
• radical prostatectomy
• primary androgen deprivation therapy

Introduction

Current recommendation for prostate cancer management is radical prostatectomy (RP) for patients with life expectancy of more than 10 years [1–3]. Life expectancy continues to rise globally and remains critical to informed decision making in many procedures including RP. Life expectancy can be estimated based on the mortality experience of a population during a relatively short period of time [4]. A life table on Social Security Administration is a concise method of estimating life expectancy [4]. According to the life table, life expectancy of male has a tendency of increase; 14.2, 11.0, and 8.2 years were estimated in an age of 70-, 75-, and 80-year, respectively, in 2013. Moreover, life expectancy can be adjusted using the clinician’s assessment of overall health; that of the elderly, best quartile of health, is calculated by increasing 50% [2]. Therefore, the expected lifespan of healthy elderly men aged 70-, 75-, and 80-year is estimated to be 21.4, 16.6, and 12.3 years, respectively.

Nevertheless, most elderly men do not undergo RP. According to the Cancer of the Prostate Strategic Urologic Research Endeavor (CaPSURE) database, 92% of the prostate cancer patients over 75 years of age were classified as clinically localized disease (T1–T2), of which 41% received primary androgen deprivation therapy (PADT) and 3% underwent RP [5]. Androgen deprivation therapy (ADT) comes with many adverse effects, including fatigue, decreased libido, cognitive impairment, weight gain, muscle or bone mass loss, gynecomastia, vasomotor symptoms, anemia, and metabolic derangement [6–13]. In particular, the risk of cardiovascular disease due to ADT is of increasing concern, as it may negate the survival benefit in certain patients [14–16]. Also, most ADTs for prostate cancer have a significant correlation with cerebrovascular accident, and ADT-induced hypogonadism has been associated with decreased bone mineral density, a loss in the structure and strength of bone, and subsequent fractures [17,18].

An interest in RP for elderly patients has been emerging recently. Several authors have shown that RP can be feasible and efficacious in carefully selected elderly patients aged 75 years or older [19–22]. Similarly, we have previously reported that RP in elderly patients aged 75 years or older is comparable to results in patients aged 65–69 years, with minor peri- or postoperative complications and positive oncologic outcomes [19]. This study has been conducted to determine whether RP is suitable for patients with prostate cancer aged 75 years or older through a comparison of the adverse events and survival rates of RP patients with those of PADT patients. As far as we know, this is the first study to compare RP and PADT in the elderly aged 75 years or older.

Patients and methods

Study design and population

This retrospective cohort study was conducted on men 75 years or older who were diagnosed with localized prostate cancer and received RP or PADT from 2005 to 2013 at 6 institutions. This study was approved by the Institutional Review Board (IRB) of Veterans Health Service Medical Center (IRB No. 2017-05-004).

The patients needed to be pathologically diagnosed with prostate cancer through needle biopsy and be at least 75 years of age at the time of diagnosis. All cases in the study were clinically proven to be localized prostate cancer (T1–T2, N0, M0) on magnetic resonance imaging of the prostate and whole-body bone scan with/without abdominopelvic computed tomography. Within 3 months after a diagnosis of clinically localized prostate cancer, initial treatment was applied with either surgery or hormone therapy as monotherapy without an expectant management such as active surveillance or watchful waiting. Patients who had less than 12 months of follow-up, or received neoadjuvant or adjuvant therapy were excluded.

Definitions of treatment

RP as primary treatment included open retropubic, laparoscopic, or robot-assisted laparoscopic surgery. The type of surgery varied according to the operator or institution. Salvage therapy in RP group was performed in cases of biochemical recurrence (serum prostate-specific antigen [PSA] ≥ 0.2 ng/mL), or local recurrence or distant metastasis on radiologic examination. As salvage therapy, external-beam radiation treatment (EBRT), ADT, or combination therapy with EBRT and ADT was selected according to the patient’s condition.

PADT group was treated with gonadotropin-releasing hormone (GnRH) agonist monotherapy or bilateral orchiectomy, antiandrogen monotherapy, or combined androgen blockade. Antiandrogen withdrawal or secondary antiandrogen has been performed in patients with PSA progression or radiologic progression during PADT. There was no salvage prostatectomy, salvage radiotherapy, or cytotoxic chemotherapy in any of the patients.

Data collection and variables

We evaluated the age at the time of diagnosis, PSA before prostate needle biopsy, Gleason score on the biopsy, clinical T stage based on the 2010 American Joint Committee on Cancer (AJCC) system, and Eastern Cooperative Oncology Group (ECOG) performance status and Charlson comorbidity index (CCI) score before primary treatment. Cancer-specific and overall survival was calculated on the basis of the index date, and the date on which treatment was started was set as the index date. Post-treatment complications such as urinary incontinence, new-onset erectile dysfunction, bladder neck contracture or urethral stricture, hernia, hot flush, sweating, mastalgia or gynecomastia, cardio- or cerebrovascular accident, or bone fracture were also reviewed. Urinary incontinence and new-onset erectile dysfunction were investigated at 12 months after primary treatment. Urinary incontinence was defined as either not using diapers or using one protective diaper per day. And new-onset erectile dysfunction was defined as a condition in which the patient complained of a deterioration of the erectile function after the primary treatment or requested an improvement of impaired erectile dysfunction. Among the complications, cardio- or cerebrovascular accident, or bone fracture, which could pose a life-threatening condition, was classified as “serious adverse events”.

Statistical analysis

Mann–Whitney U-test was used for continuous variables and either the chi-square test or Fisher exact test for categorical variables to compare the clinical parameters of RP group and PADT group. Cancer-specific and overall survival rates of the two groups were estimated using the Kaplan–Meier survival analysis and log-rank test. Univariable and multivariable logistic regression analyses were performed to find the predictors of serious adverse events. All statistical analyses were performed using IBM SPSS 18.0 (IBM Co., Armonk, NY), and a p value <.05 was considered statistically significant.

Results

Patient and tumor characteristics

We included 92 patients in the RP group and 99 patients in the PADT group. In the RP group, patients underwent open retropubic (n = 56), laparoscopic (n = 11), or robot-assisted laparoscopic surgery (n = 25). Meanwhile, the PADT group received the hormone therapy using GnRH agonist monotherapy or bilateral orchiectomy (n = 8), antiandrogen monotherapy (n = 19), or combined androgen blockade (n = 72).

Compared with the PADT group, the RP group had significantly lower age (p < .001), lower PSA level (p = .012), and better ECOG performance status grade (p < .001). Biopsy Gleason score, clinical T stage, and CCI were not significantly different between the two groups (Table 1).

Table 1. Baseline characteristics.

	RP group (n = 92)	PADT group (n = 99)	p
Age (years)	76.0 (75.0–77.0)	79.6 (77.0–82.0)	<.001
Prebiopsy PSA (ng/mL)	6.97 (4.84–14.20)	9.69 (6.05–14.97)	.012
Biopsy Gleason score			.353
≤6	35 (38.0%)	34 (34.3%)
7	40 (43.5%)	38 (38.4%)
≥8	17 (18.5%)	27 (27.3%)
Clinical T stage			.909
cT1	19 (20.7%)	23 (23.2%)
cT2a	34 (37.0%)	39 (39.4%)
cT2b	12 (13.0%)	12 (12.1%)
cT2c	27 (29.3%)	25 (25.3%)
ECOG performance status grade			<.001
0	44 (47.8%)	35 (35.4%)
1	47 (51.1%)	44 (44.4%)
≥2	1 (1.1%)	20 (20.2%)
Charlson comorbidity index score			.507
0	57 (62.0%)	59 (59.6%)
1	22 (23.9%)	20 (20.2%)
≥2	13 (14.1%)	20 (20.2%)

Values are presented as median (inter-quartile range), or number (%).

PADT: primary androgen deprivation therapy; ECOG: Eastern Cooperative Oncology Group; PSA: prostate-specific antigen; RP: radical prostatectomy.

Survival analysis

Median follow-up period was 48.5 months in the RP group and 37.0 months in the PADT group. In the Kaplan–Meier analysis, estimates of cancer-specific survival were 9.3 years for the RP group and 9.9 years for the PADT group (p = .302). The overall survival was estimated to be 9.0 years for the RP group and 9.2 years for the PADT group, and there was no significant difference between the two groups (p = .995) (Figure 1).

Figure 1. Survival analysis according to primary treatment modality. (A) Cancer-specific survival. (B) Overall survival. PADT: primary androgen deprivation therapy, RP: radical prostatectomy.

Complications after primary treatment

Urinary incontinence (p < .001), new-onset erectile dysfunction (p = .002), bladder neck contracture or urethral stricture (p = .024), and hernia (p < .001) occurred frequently in the RP group. On the other hand, the incidence of hot flush and sweating was higher in the PADT group (p = .008 and .029, respectively). Particularly, serious adverse events occurred in only 1 case (1.1%) in the RP group, but 13 cases (13.1%) occurred in the PADT group (p = .001) (Table 2).

Table 2. Post-treatment complications.

	RP group (n = 92)	PADT group (n = 99)	p
Urinary incontinence	30 (32.6%)	0 (0.0%)	<.001
New-onset erectile dysfunction	41 (44.6%)	23 (23.2)	.002
Bladder neck contracture or urethral stricture	5 (5.4%)	0 (0.0%)	.024
Hernia	12 (13.0%)	0 (0.0%)	<.001
Hot flush	2 (2.2%)	12 (12.1%)	.008
Sweating	0 (0.0%)	6 (6.1%)	.029
Mastalgia or gynecomastia	4 (4.3%)	11 (11.1%)	.083
Serious adverse events^a	1 (1.1%)	13 (13.1%)	.001
Cardio- or cerebrovascular event	1 (1.1%)	8 (8.1%)	.036
Bone fracture	0 (0.0%)	5 (5.1%)	.060

Values are presented as number (%).

PADT: primary androgen deprivation therapy; RP: radical prostatectomy.

^a Serious adverse events: cardio- or cerebrovascular accident, or bone fracture.

In a univariable regression analysis, advanced age (OR 1.20, p = .007) and choice of PADT as the primary treatment option (OR 13.76, p = .012) had significantly more frequent serious adverse events. However, in the multivariable regression analysis, only PADT as the primary treatment modality had a worse effect on serious adverse events (OR 10.12, p = .038) (Table 3).

Table 3. Univariable and multivariable analysis for prediction of serious adverse events.^a

	Univariable analysis			Multivariable analysis
	OR	(95% CI)	p	OR	(95% CI)	p
Age	1.20	(1.05–1.36)	.007	1.06	(0.91–1.25)	.455
Prebiopsy PSA level^b	1.48	(0.83–2.63)	.180	1.31	(0.68–2.52)	.422
Biopsy Gleason score (≥8 vs. ≤7)	1.97	(0.62–6.21)	.249	1.77	(0.51–6.19)	.373
Clinical T stage (cT2c vs. ≤ cT2b)	2.14	(0.70–6.48)	.180	1.94	(0.58–6.54)	.285
ECOG performance status grade (≥1 vs. 0)	1.29	(0.42–4.02)	.657	1.10	(0.31–3.90)	.882
Charlson comorbidity index score (≥1 vs. 0)	2.19	(0.73–6.58)	.163	2.53	(0.73–8.70)	.142
Primary treatment option (PADT vs. RP)	13.76	(1.76–107.41)	.012	10.12	(1.14–90.10)	.038

ECOG: Eastern Cooperative Oncology Group; PADT: primary androgen deprivation therapy; PSA: prostate-specific antigen; RP: radical prostatectomy.

^a Serious adverse events: cardio- or cerebrovascular accident, or bone fracture.

^b Log-transformed values were used for analysis.

Discussion

This study assessed the suitability of RP as the primary choice for elderly patients aged 75 years or older with clinically localized prostate cancer through two criteria: (1) cancer-specific and overall survival, and (2) post-treatment complications. PSA progression-free survival was excluded from the comparison because there was no objective criterion of PSA progression that could be applied to both RP and PADT groups. In the study, there were no significant differences in survival rates between the two treatments; however, PADT group had significantly more severe adverse events (p = .001). There was no significant report on perioperative morbidity at the time of anesthesia and surgery among RP group patients. We suggest that elderly men with prostate cancer do not have to avoid a surgical treatment if they are healthy and their expected lifespan is long enough.

As mentioned earlier, the life expectancy can be adjusted according to the clinician's assessment of overall health. Because this judgment is dependent on a subjective evaluation of the physician, patient’s functional status is likely to be an important criterion. In the papers supporting the RP as a prostate cancer treatment in patients aged 75 years or older, it was shown that surgical treatment was suitable for “healthy”, “selected”, “well-selected”, or “highly-selected” patients [19–22]. In Table 1, urologists at multiple institutions focused on patients’ performance status as a measure of operability for elderly men (p < .001). On the other hand, some authors reported that the benefits of surgery in patients with CCI ≥2 were relatively low because of high incidence of non-cancer deaths [23] although comorbidities were not analyzed as a significant factor related to decision of treatment policy in our study (p = .507). It is advisable to consider both performance status and comorbidity in making surgical decisions in elderly patients with prostate cancer. The International Society of Geriatric Oncology (SIOG) Prostate Cancer Working Group recommends that the treatment for senior adults should be based on systematic evaluation of health status. The G8 (Geriatric 8) health status screening tool is used for this assessment, which is scored for a total of 17 out of 8 items. “Fit or healthy” older men with a G8 score greater than 14 are eligible for a standard treatment as provided in younger patients, regardless of age [24].

In the CaPSURE database, approximately two-third of men with clinically localized prostate cancer (T1–T3) were maintained ADT at 5 years after PADT, and 4.1% died of prostate cancer [25]. The authors noted that PADT appeared to be effective in controlling clinically localized disease [25]. However, in a registry-based study of 15,170 patients who were diagnosed with clinically localized prostate cancer (T1–T3) and were not treated with curative intent therapy, PADT was found to have no benefit in both all-cause mortality (HR 1.04, p = .33) and cancer-specific mortality (HR 1.03, p = .67) [26]. And PADT was associated with increased risk of overall mortality among the subgroup of men with a low-risk (pretreatment PSA ≤10 ng/mL, Gleason score ≤6, and clinical stage ≤ T2a) of cancer progression (HR 1.41, p = .02) [26]. Meanwhile, Surveillance Epidemiology and End Results (SEER) dataset compared survival rates between RP and PADT in men with clinically localized stage (T1–T2) aged 66–74 years [27]. PADT showed an unfavorable prognosis in both overall survival (HR 2.98, p < .001) and cancer-specific survival (HR 12.47, p < .001) [27]. In our study comparing RP and PADT in clinical T1–T2 cancer patients aged 75 years or older, overall and cancer-specific survivals were not significantly different between the two groups.

We defined “serious adverse events” as cardio- or cerebrovascular accident, or bone fracture in this study. These serious adverse events are generally considered to be fatal and life-threatening than the other complications mentioned in Table 2. Many studies have reported an excess risk for myocardial infarction (MI) and stroke in patients with ADT [15,17,28–32]. MI and stroke are severe complications with 1-year mortality rates of 24% [33] and 21% [34]. In a CaPSURE database-based study of 4892 organ-confined prostate cancer patients, ADT use was found to be an unfavorable factor of cardiac-related death [14]. Meanwhile, ADT was associated with the marked increase in a fracture risk (34–65%) [35,36]. The mortality rate doubled for those experiencing a fracture after their diagnosis of prostate cancer compared with the patients who did not experience a fracture in a study of 80,844 patients from SEER database [36]. In Table 3, a significant predictor of serious adverse events was not the patient age, prostate cancer-related variables (PSA, biopsy Gleason score, clinical stage), or health status measures (performance status, comorbidity), but only the primary treatment option (PADT vs. RP). The PADT was a strongly marked factor to predict the serious adverse events (OR 10.12, p = .038).

There were not many studies evaluating the adequacy of surgery for patients with prostate cancer aged 75 years or older (Table 4). Xylinas et al. [21] reported the oncologic and functional outcomes in 22 patients undergoing laparoscopic RP. No comparison was made with younger patients or with other treatments in their study. Labanaris et al. [22] compared the results of robotic assisted laparoscopic RP with 45 men aged 75 years or older and 2000 overall patient cohort. They found that Gleason score and pathologic T stage were significantly higher in elderly men than in overall cohort, and the proportion of patients with positive surgical margin was also higher in elderly men. In a study published in 2016 [19], we compared the RP results of 89 patients aged 75 years or older and 168 patients aged 65–69 years. It was shown that PSA and Gleason score (both biopsy and RP Gleason score) were higher in the elderly group. And postprostatectomy urinary incontinence was more frequent in the elderly men, but multivariable analysis showed that performance status was more relevant to the incontinence than patient age. We proposed that RP was a suitable option for selected men aged 75 years or older with limited complication rates and good outcomes of biochemical recurrence-free recurrence similar to those for younger patients. Mandel et al. [20] compared the RP results of 265 patients aged 75 years or older and 13,732 patients under 75 years of age. The results of RP Gleason score, organ confinement, lymph node invasion, and positive surgical margin were unfavorable in the older patients, suggesting that the elderly patients had more advanced disease as a whole. And healthy and highly selected men 75 years old or older showed good long-term overall survival, whereupon they concluded that older age in well-selected men should not be contraindicated to RP, especially in patients with high-risk disease.

Table 4. Studies on radical prostatectomy for patients aged 75 years or older.

Study	Year of publication	N	Patients
Xylinas et al. [18]	2010	22	LRP, Age >75 years (single arm)
Labanaris et al. [19]	2012	45	RALRP, Age ≥75 years
		2000	RALRP, Overall cohort
Ryu et al. [16]	2016	89	RP, Age ≥75 years
		168	RP, Age 65–69 years
Mandel et al. [17]	2016	265	RP, Age ≥75 years
		13,732	RP, Age <75 years

LRP: laparoscopic RP; RALRP: robotic assisted laparoscopic RP; RP: radical prostatectomy.

There were several limitations in our study. First, due to a retrospective multi-center nature of the study, we were not able to control for patient’s operation under the same criteria. The selection bias was considered to be relatively small, even though it was a multi-center study, because the researchers almost coincided in opinion on the treatment issue for prostate cancer in the elderly patients. Second, our sample size was limited. Because RP in patients aged 75 years or older was not a commonly accepted treatment and a large number of the elderly men were excluded from consideration of surgery, the included subjects were inevitably a highly selective group. Particularly, the sample size in the PADT arm was small because older patients tended to have more advanced disease at the time of diagnosis than the younger age group, and localized cancer at older ages was more likely to be on expectant management. Third, the median follow-up period was relatively short (48.5 and 37.0 months in RP and PADT group, respectively). It was thought that the study period was rather short to observe survival rates but it was sufficient to look into the treatment-related complications, as the significant results in PADT-related serious adverse events were derived.

In conclusion, RP is an effective and safe modality as a primary treatment for selected elderly patients 75 years or older with localized prostate cancer, as compared to PADT. Elderly patients with prostate cancer should be treated with respect to life expectancy, and surgical treatment should not be ruled out because of patient’s chronological age. Furthermore, it would be prudent to not only study the therapeutic outcomes and adverse effects, but also evaluate cost and psychological stress in that generation.

Disclosure statement

No potential conflict of interest was reported by the authors.