Comorbidities as predictors of incidental prostate cancer after Holmium laser enucleation of the prostate: diabetes and high-risk cancer

Abstract

Prostate cancer can be diagnosed as an incidental finding during the pathological examination of benign prostatic hyperplasia (BPH) specimens by Holmium laser enucleation of the prostate (HoLEP). BPH and comorbidities such as hypertension, diabetes, and dyslipidemia often coexist in elderly people. We identified which comorbidities can be used to predict the presence of incidental prostate cancer, particularly high-risk cancer, in men who had undergone HoLEP. On the basis of pathological findings of HoLEP specimens, patients with incidental cancer were categorized as low-risk (Gleason ≤6 and T1a) or high-risk (all others). Of the 654 patients who underwent HoLEP, 41 patients (6.3%) were identified as having incidental cancer (25 low-risk and 16 high-risk). There were no significant factors for overall prostate cancers. However, a significantly higher frequency of diabetes was observed in patients with high-risk cancer compared to those with BPH (31% vs. 13%; p = .033). Logistic regression analysis using prostate-specific antigen (PSA) and prostate volume (PV), and smoking showed that diabetes was an independent predictor of high-risk cancer (odds ratio, 3.15; 95% confidence interval, 1.06–9.43). Diabetes may be an important predictor of the presence of high-risk prostate cancer in men with BPH who have undergone HoLEP.

Keywords:

• Benign prostatic hyperplasia
• comorbidities
• diabetes
• Holmium laser
• prostate cancer

Introduction

Prostate cancer is one of the most common malignancies affecting men, and it can be diagnosed as an incidental finding during the pathological examination of benign prostatic hyperplasia (BPH) specimens. Holmium laser enucleation of the prostate (HoLEP) is one of the most widely investigated laser procedures for the treatment of patients with symptomatic BPH. It has the advantage of providing large quantities of prostate tissue for pathological confirmation. In previous studies, incidental prostate cancer was detected in 5–13% of patients who had undergone HoLEP [1–7]. Several studies have reported clinical factors related to prostate, including prostate-specific antigen (PSA) and prostate volume (PV), as predictors of incidental prostate cancer after HoLEP [4–6]. Considering that BPH and comorbidities such as hypertension, diabetes, and dyslipidemia coexist in elderly people, it is useful to investigate what comorbidities can be predictors of incidental prostate cancer in men undergoing HoLEP. However, few studies have investigated the relationship between incidental prostate cancer after HoLEP and comorbidities. Especially, identifying men at increased risk for clinically significant prostate cancer i.e. high-risk cancer, is an important consideration in managing patients with BPH who are considering enucleation surgery.

We sought to determine comorbidities that can be used to predict the presence of incidental prostate cancer in men undergoing HoLEP. We also focused on predictors of high-risk prostate cancer, which often requires active treatment interventions.

Methods

The study population consisted of a consecutive group of 688 patients who were diagnosed with BPH and underwent HoLEP between July 2008 and September 2014. All patients were treated at the St. Luke’s International Hospital. The St. Luke’s International Hospital Internal Review Board approved all study procedures.

Preoperatively, patients underwent PSA screening and abdominal ultrasonography. PV was determined by abdominal ultrasonography using the ellipsoid formula (height × width × length ×0.52). Preoperative prostate biopsy was performed if indicated based on elevated PSA and estimated PV. Body weight and height were measured, and body mass index (BMI) was calculated as weight in kilograms divided by height in meters squared. The presence of comorbidities including hypertension, diabetes, or dyslipidemia was recorded after a medical chart review.

The diagnosis of primary adenocarcinoma of the prostate was confirmed histopathologically by the St. Luke’s International Hospital Department of Pathology. Patients with incidental prostate cancer were categorized as low-risk or high-risk based on the HoLEP specimen pathological findings. A designation of low-risk was limited to those with a Gleason score of 6 or less cancer in less than 5% of the specimen (T1a). All others were categorized as high-risk patients.

Data extracted from medical records included age, BMI, pretreatment PSA levels, estimated PV, smoking habit, medical history (hypertension, diabetes, and dyslipidemia), previous biopsy, previous dutasteride therapy, and pathological findings from HoLEP specimens. We excluded 34 patients for missing BMI (n = 7), PSA (n = 22), and PV (n = 5) data. Consequently, 654 patients were included in the analysis.

Crude associations between predictive variables and incidental prostate cancer were tested using the χ² test or Fisher’s exact test for categorical variables and t-test or Wilcoxon rank sum test for continuous variables. A logistic regression analysis was performed to determine the independent predictive factors for incidental prostate cancer. The multivariate model included factors with p < .20 as potential confounders. We also examined the independent predictors for incidental cancer in the high-risk group. Statistical tests were two-sided and p values < .05 were considered as significant. All statistical analyses were performed using the SAS software (SAS Institute, Cary, NC).

Results

The ages of the patients ranged from 50 to 91 years, with a mean of 70 years. The median (interquartile range) preoperative PSA and PV were 6.15 (3.61–10.55) ng/mL and 66 (50–91) mL, respectively. The mean ± SD BMI was 23.2 ± 2.7 kg/m².

Of the 654 patients who underwent HoLEP, 41 (6.3%) were diagnosed as having prostate cancer. Of the 41 patients with incidental prostate cancer, 25 (3.8%) were classified as low-risk and 16 (2.5%) were classified as high-risk.

Table 1 illustrates the relationship between clinical variables and prostate cancer (combined low-risk and high-risk groups). Patients with a higher PSA and larger PV tended to have an increased risk for incidental prostate cancer (6.71 vs. 6.10 ng/mL; p = .073 and 81 vs. 66 mL; p = .069, respectively). Diabetes was the only significant predictor of high-risk cancer after HoLEP (31% vs. 13%; p = .033).

Table 1. Baseline characteristics in BPH and incidental PCa.

	BPH	PCa		High-risk PCa
	n = 613	n = 41	p	n = 16	p^a
Age, years, mean ± SD	70 ± 7	70 ± 7	.800	71 ± 8	.768
BMI, kg/m^2, mean ± SD	23.1 ± 2.8	23.7 ± 2.6	.171	23.9 ± 2.7	.239
PSA, ng/mL, median (interquartile range)	6.10 (3.59–10.50)	6.71 (5.12–13.00)	.073	6.48 (5.65–11.06)	.152
Preoperative PV, mL, median (interquartile range)	66 (50–90)	81 (54–100)	.069	90 (54–103)	.124
Smoking	240 (39)	20 (48)	.223	9 (56)	.167
Hypertension	236 (39)	19 (46)	.319	6 (38)	.935
Diabetes	79 (13)	8 (20)	.227	5 (31)	.033
Dyslipidemia	96 (16)	10 (24)	.142	4 (25)	.301
Previous biopsy	234 (38)	19 (46)	.299	8 (50)	.337
Previous dutasteride therapy	79 (13)	2 (5)	.216	0 (0)	.243

Values are n (%) unless otherwise stated.

BPH: benign prostate hyperplasia; PCa: prostate cancer.

^a Patients with high-risk PCa compared to those with BPH.

We performed a logistic regression model to determine the independent predictive factors for incidental prostate cancer using factors that had p values < .20. Logistic regression analysis using BMI, PSA, PV, and dyslipidemia revealed that no factors were associated with overall prostate cancers (Table 2).

Table 2. Logistic regression analysis predicting prostate cancer.

	Odds ratio	95% confidence interval	p
BMI, per 1 kg/m^2	1.072	0.956–1.201	.235
PSA, per 1 ng/mL	1.019	0.984–1.055	.291
Preoperative PV, per 1 mL	1.004	0.996–1.012	.306
Dyslipidemia	1.711	0.800–3.659	.166

As shown in Table 3, diabetes remained an independent predictive factor for high-risk cancer after adjustment for PSA, PV, and smoking (odds ratio, 3.15; 95% confidence interval, 1.06–9.43; p = .040).

Table 3. Logistic regression analysis predicting high-risk prostate cancer.

	Odds ratio	95% confidence interval	p
PSA, per 1 ng/mL	1.005	0.948–1.066	.860
Preoperative PV, per 1 mL	1.008	0.996–1.020	.214
Smoking	2.112	0.769–5.802	.147
Diabetes	3.154	1.055–9.433	.040

Discussion

We investigated the predictors of incidental prostate cancer after HoLEP. There were no significant factors for overall prostate cancers. However, we found that diabetes was significantly associated with high-risk prostate cancer, which often requires active treatment interventions.

Several studies have reported that clinical factors related to prostate, including PSA and PV, are predictors for incidental prostate cancer after HoLEP [4–6]. However, little is known about its relationship with comorbidities. Although diabetes has been reported to be associated with an increased risk for many forms of cancer, some studies have shown a significant or non-significant negative association between diabetes and prostate cancer [8–11]. In contrast, our previous study with a longitudinal design suggested a positive association between hemoglobin A_1C and PSA in healthy men [12]. Adiposity measurements including BMI and waist circumference have been reported to be positively associated with aggressive prostate cancer [9,13–16]. One study showed that men with BMI <25 kg/m² with diabetes had an increased risk of aggressive prostate cancer [17]. Most of the subjects in our study (80%) had a BMI <25 kg/m². Further studies may be needed to investigate the association between diabetes and incidental prostate cancer after HoLEP based on BMI and prostate cancer grade.

Approximately one in five men with BPH has hypogonadism, such as low testosterone level [18]. It has also been described that hypogonadism is associated with diabetes [18–20]. Relatively little is known about the relationship between hypogonadism and prostate cancer [18,19,21]. It has been reported that lower serum testosterone level is associated with high-grade prostate cancer [21,22]. Hypogonadism may play a role in developing high-risk prostate cancer in diabetic men.

Smaller PV and higher preoperative PSA have been reported to be predictors of incidental prostate cancer after HoLEP [4,6]. These results contrast with our findings of no significant predictors among them. In our study, patients with prostate cancer tended to have a larger PV. This may be due to differences in the ultrasonic diagnostic methods. While previous studies used transrectal ultrasonography, we used abdominal ultrasonography. However, even when specimen weight was used instead of volume, the results were similar (median 34 g in patients with BPH vs. 46 g in prostate cancer; p = .072). The reason for these differences is not immediately obvious.

This study has several limitations. It was a retrospective observational study with a cross-sectional design. However, comorbidities including diabetes were identified separately before HoLEP and the pathologist did not take comorbidities into account at histopathological diagnosis; thus, it is unlikely that the association between diabetes and high-risk cancer was biased. The indication for biopsy was not clearly specified in advance. However, it is unlikely that the decision-making process related to biopsy distorted the association between comorbidities and incidental prostate cancer. As a high percentage of our patients were referred from outside facilities, biopsies were not standardized and this variable was not completely captured in our database. We investigated a limited population who were treated in one hospital. Only 6.3% of our patients were found to have incidental prostate cancer in their HoLEP specimen, which is similar to the 5–13% of patients reported in previous studies [1–7].

In conclusion, none of the tested comorbidities, including hypertension, diabetes, and dyslipidemia, were significant predictors of incidental prostate cancer after HoLEP. However, we found that diabetes was significantly associated with high-risk cancer. Our findings suggest that diabetes may be an important predictor of the presence of high-risk prostate cancer in men with BPH undergoing HoLEP.

Notes on contributors

Kazuhiro Ohwaki is the Director of the Health Management Center at JCHO Tokyo Shinjuku Medical Center. He is also an Adjunct Research Staff of the Department of Urology at St. Luke’s International Hospital. His research interests include clinical epidemiology and preventive medicine.

Fumiyasu Endo is a Head Physician of the Department of Urology at St. Luke’s International Hospital. His clinical interests include benign prostatic hyperplasia, HoLEP, and robotic surgery.

Masaki Shimbo is an Assistant Head Physician of the Department of Urology at St. Luke’s International Hospital. His clinical interests include general urology, HoLEP, and robotic surgery.

Akiko Fujisaki is an Adjunct Medical Staff of the Department of Urology at St. Luke’s International Hospital. Her clinical interest is in general urology.

Kazunori Hattori is the Director of the Department of Urology at St. Luke’s International Hospital. His clinical interests include robotic surgery, laparoscopic surgery, adrenal gland tumor, and malignant tumor of urinary tract.

Disclosure statement

The authors report no conflicts of interest.