Prostate-Specific Antigen Evolution After Photoselective Vaporization of the Prostate

Abstract

Background:

Prostate-specific antigen (PSA) variations after photoselective vaporization of the prostate are an unsolved question.

Patients and Methods:

We included prospectively 323 patients who underwent GreenLight® prostate vaporization for prostatic hyperplasia between 2005 and 2013. We excluded patients with prostate cancer and those with concomitant prostate biopsies. Serum PSA levels were measured before and after the procedure at 1, 6, 12, 24, 36, and 48 months.

Results:

The median preoperative PSA level was 4.50 ng/mL; it decreased to 2.41 ng/mL at 1 month (47% reduction), 2.17 ng/mL at 6 months (52% reduction), 2.30 ng/mL at 1 year (49% reduction), 2.40 ng/mL at 2 years (47% reduction), 2.31 ng/mL at 3 years (49% reduction), and 2.54 ng/mL at 4 years (44% reduction) (P values were all <0.0001). The median PSA nadir at 6 months was significantly different from the median PSA level at 1, 2, 3, and 4 years (respectively P=0.0046, P=0.0017, P=0.0006, and P=0.01). Patients who received ≤3000 J/cc had a significant trend to a PSA reascension after 6 months. Patients who received ≥4000 J/cc did not show any significant PSA reascension during the 4 years after procedure. Energy was correlated with the PSA reascension in univariate and multivariate analysis.

Conclusions:

The PSA level significantly decreased by half 1 month after the procedure, reached its nadir at 6 months, and showed a slight progressive reascension during the 4 following years. Applying an energy rate ≥4000 J/cc of prostate induced PSA stability over time whereas energy ≤3000 J/cc induced a reascension of the PSA level after 6 months.

Introduction

The serum prostate-specific antigen (PSA) assay is still the standard for the screening and the management of prostate cancer (PCa). Many urologic events, however, are likely to induce artificial variations of its rate that can lead to an uncertain interpretation of its value. For example: The PSA level might be increased by prostatitis, urinary catheters, digital rectal examination, benign prostatic hyperplasia (BPH), urologic procedures such as prostate biopsies or transurethral resection of the prostate (TURP), without any link with PCa.¹

The emergence of photoselective vaporization of the prostate (PVP) for BPH management is also a potential cause of misinterpretation of the PSA level after the procedure: Thus, it is important to clearly define the natural evolution of the PSA level after PVP to identify its abnormal values.

Unfortunately, the evolution of the PSA level after PVP is not a well-known matter. Only a few studies have assessed the PSA variations and their interpretation through time after PVP: Volkan and associates² first reported that the serum PSA level increased immediately after potassium-titanyl-phosphate (KTP) laser vaporization, then decreased and returned to its preoperative level within 15 days after the procedure. Shim and colleagues³ reported the largest study assessing specifically the PSA evolution after PVP in a series of 278 patients treated by an 80 W KTP laser system. The PSA level increased the first month after the procedure and then decreased significantly over the next 3 months and became stable after 6 months. They also reported that the PSA level had decreased by 37.5% by 12 months after the operation compared with the preoperative level.

On the contrary, the natural evolution of the PSA rates after TURP has been widely assessed through many studies. It has been shown that PSA increases immediately after TURP and decreases afterward to a stable value: Oesterling and coworkers⁴ showed in 1996 that TURP induces an immediate median elevation of 5.9 ng/mL of the PSA rate, and a median time of 18 days was needed to return to a stable value. Aus and colleagues⁵ reported a 70% decrease of the PSA level for 190 patients 3 months after TURP with a value under 4 ng/ml for 90% of the patients. Wolff and associates⁶ also reported that the PSA level decreased from 4.9 ng/mL to 0.6 ng/mL 48 months after TURP, and from 6.8 ng/mL to 2.2 ng/mL 48 months in patients who had a diagnosis of PCa after TURP. In the same way, patients who present a serum PSA level that does not decrease enough or remains high or increases after PVP are supposed to be at risk of presenting with PCa.³

PVP compared with TURP has the disadvantage of not providing any samples for histologic analysis, which might spoil an opportunity to detect PCa. Thus, it is all the more important to have proper knowledge of the PSA evolution after the procedure to not ignore a coexisting prostate malignancy and to define optimal follow-up strategies. The aim of this study was to assess the PSA variations after PVP.

Patients and Methods

This study was conducted prospectively on patients who underwent prostate vaporization with the GreenLight® laser (AMS®, Minnetonka, MN) for HBP at the urology department of our University Hospital between 2005 and 2013. A total of 308 patients were included. We excluded patients with prostatitis or PCa and those who underwent concomitant prostate biopsy. All patients with a preoperative PSA level of ≥4 ng/mL were screened for PCa with 12 transrectal ultrasonography-guided biopsies.

PVP was performed using a GreenLight laser via a 23F continuous-flow dedicated endoscope. An 80 W KTP generator was used from 2005 to 2007; then from 2007 to 2010 a 120 W generator was used; and finally, from 2010 until now a 180 W generator was used. All patients received either general or epidural anesthesia and 1 g of cefuroxime preoperatively. Sodium chloride solution (0.9%) served as the irrigation fluid. At the end of the procedure, a two-way Foley catheter was inserted without irrigation. The catheter was removed at postoperative day 1, after the urine cleared. Follow-up visits were scheduled at 1, 6, and 12 months and then once a year.

Serum PSA levels were measured for every patient before the procedure and after the procedure at 1, 3, 6 months, 2 years, 3 years, and 4 years. Data of patients following any reoperation were excluded. Prostate volume was estimated by transrectal ultrasonography (TRUS) by measuring length (L), width (W), and height (H) and then by calculating volume by use of an ellipsoid formula (L×W×H×π/6).

The PSA evolution was first analyzed for the whole population, then according to the model of generator, and finally according to the energy applied per prostate volume. Cutoffs of ≤3000 J/cc per volume of prostate and ≥4000 J/cc were arbitrarily chosen to define two subgroups that had low and high energy per prostate volume.

Statistical analysis was performed with StatView 5.0 software. The Wilcoxon test was used for median paired nonparametric comparisons. Correlations were performed by the Pearson test. The Kruskall-Wallis test was used for category comparisons. Multivariate analysis was performed by multiple regression analysis. Statistical significance was defined as a P value<0.05.

Institutional ethical approval was obtained before the conduction of the study.

Results

We included 323 patients. PCa was diagnosed in 15 patients during the follow-up, and the patients were excluded from the study. The following results are those of the 308 other patients. Their baseline characteristics are reported in Table 1. The number of patients followed at 1, 6, 12, 24, 36, and 48 months, were respectively: 262, 190, 172, 98, 74, and 42.

Table 1.

Baseline Characteristics

Variables	Medians	Range
Age (years)	74	49–98
Preoperative PSA (ng/mL)	4.5	0.26–32.64
Prostate volume (cc)	66.5	17–330
Energy used (KJ)	266	173–1505
Vaporization time (minutes)	42	9–150
Operative time (minutes)	70	17–220

PSA=prostate-specific antigen.

The PSA rates during the follow-up were compared with preoperative values. A significant decrease was found: Results are reported in Table 2 and Figure 1. Median PSA nadir at 6 months was significantly different from the median PSA value at 1, 2, 3, and 4 years (respectively, P=0.0046, P=0.0017, P=0.0006, and P=0.01). This shows a progressive reascension of the median PSA after 6 months.

FIG. 1.

Follow-up prostate-specific antigen (PSA) values compared with baseline values. *P<0.05 compared with preoperative PSA. °P<0.05 compared with PSA nadir at 6 months. Bars represent standard error.

Table 2.

Median Prostate-Specific Antigen Evolution for All Patients, the Group ≤3000 J/cc, and the Group ≥4000 J/cc

Patients	Time	Median PSA (ng/mL)	Range	Change (%)	P value
All	Preoperative	4.50	0.26–32.64	-	-
	1 month	2.41	0.05–21.8	46	<0.0001
	6 months	2.17	0.17–20	52	<0.0001
	1 year	2.30	0.08–12.7	49	<0.0001
	2 years	2.40	0.1–18.54	47	<0.0001
	3 years	2.31	0.1–15.27	49	<0.0001
	4 years	2.54	0.1–12	44	<0.0001
≤3000 J/cc	Preoperative	5.25	0.4–32.64	-	-
	1 month	2.89	0.16–21.12	−45	<0.0001
	6 months	2.71	0.2–20	−48	<0.0001
	1 year	3.44	0.23–12.7	−34	<0.0001
	2 years	3.47	0.88–18.54	−34	<0.0001
	3 years	3.365	0.49–15.27	−36	<0.0001
	4 years	3.355	1.13–12	−36	0.02
≥4000 J/cc	Preoperative	3.82	0.26–28.86	-	-
	1 month	2.09	0.05–21.8	−45	<0.0001
	6 months	1.78	0.19–8.47	−53	<0.0001
	1 year	1.495	0.12–7.74	−61	<0.0001
	2 years	2.12	0.1–14.4	−44	<0.0001
	3 years	1.61	0.1–5.58	−58	<0.0001
	4 years	2.04	0.1–7.5	−47	0.0005

Prostate-specific antigen (PSA) change is defined as the difference in percentage between the current PSA value and baseline. The P value corresponds to the significance of the difference between the current median PSA value and the baseline.

Two subgroups were assessed: Patients who received ≤3000 J per volume of prostate (cc) (n=94) and those who received ≥4000 J/cc (n=134). In the group that received ≤3000 J/cc, PSA decreased significantly after the procedure when compared with baseline. Results are reported in Table 2. The median PSA nadir occurred at 6 months. There was a significant difference between the median PSA nadir and the median PSA at 2 and 3 years (respectively, P=0.038, P=0.015). This shows a significant trend to a PSA re-ascension 6 months after the procedure for patients who had less than 3000 J/cc of prostate.

In the group that received ≥4000 J/cc, PSA level decreased significantly after the procedure when compared with baseline, with a significant reduction of 45% at 1 month (P<0.0001), a 53% significant reduction at 6 months (P<0.0001), a 61% significant reduction at 1 year (P<0.0001), a 44% significant reduction at 2 years (P<0.0001), a 58% significant reduction at 3 years (P<0.0001), and a 47% significant reduction at 4 years (P=0.0005) (Table 2). We found no significant differences between the median PSA nadir at 1 year and the median PSA at 2, 3, and 4 years. We did not find any significant reascension of the median PSA.

We assessed the PSA evolution according to the model of laser generator used: 80 W, 120 W, 180 W. In each of these groups, there were, respectively, 51, 117, and 140 patients. The decrease was significant during the whole follow-up period for the three groups: 4 years follow-up for the 80 W and 120 W groups, and 2 years for the group 180 W (Table 3). For the 80 W generator, we found a significant difference between the median PSA at 6 months (nadir) and at 1 year (P=0.0194). For the 120 W generator, we found a significant difference between the median PSA at 6 months (nadir) and at 2, 3, and 4 years (respectively, P=0.02, P=0.0039, P=0.003). For the 180 W generator, we did not find any significant difference between the nadir and the following values. There is a nonsignificant trend to a reascension, however. There were no significant differences between the ratio to baseline between the three models at 1 month, 6 months, and 1 year (respectively, P=0.34, P=0.11, P=0.36). There was no significant difference between the 6 months/2 years ratios between the three models (P=0.43).

Table 3.

Prostate-Specific Antigen Evolution According to the Model of Laser Generator

Generator	Time	Median PSA (ng/mL)	Range	PSA change (%)	P value
80 W	Preoperative	2.88	0.4–31.8	-	-
	1 month	1.995	0.2–16.3	−31	0.0015
	6 months	1.14	0.19–7.05	−60	<0.0001
	1 year	1.415	0.13–5.8	−51	0.002
	2 years	1.53	0.1–8.45	−47	0.0016
	3 years	1.67	0.1–9.53	−42	0.0097
	4 years	1.615	0.1–10.17	−44	0.042
120 W	Preoperative	4.325	0.26–31.18	-	-
	1 month	2.375	0.05–13.65	−45.09	<0.0001
	6 months	2.22	0.17–17.36	−48.67	<0.0001
	1 year	2.52	0.08–12.49	−41.73	<0.0001
	2 years	3.09	0.11–14.52	−28.55	<0.0001
	3 years	2.6	0.16–15.27	−39.88	<0.0001
	4 years	2.81	0.16–12	−35.03	<0.0001
180 W	Preoperative	5.06	0.46–32.64	-	-
	1 month	2.675	0.13–21.8	−47.13	<0.0001
	6 months	2.36	0.2–20	−53.36	<0.0001
	1 year	2.72	0.1–12.7	−46.25	<0.0001
	2 years	3.065	0.28–18.54	−39.43	0.0284
	3 years	-	-	-	-
	4 years	-	-	-	-

The median PSA ratio between the 6 month and 2 year values was 0.89 with a first and third quartile respectively at (0.76–1.06), and a range of (0.3–9.0). To find factors correlated with the PSA reascension after 6 months, the Pearson coefficient was calculated for age, vaporization time, energy applied, prostate volume, and energy applied by prostate volume (Table 4). Only the energy applied had a low but significant correlation with the PSA reascension after 6 months (Pearson coefficient=0.3, 95% IC: [0.07–0.47], P=0.009), which shows that PSA either decreases or has a lower reascension when a higher energy is applied. In multivariate analysis, through a multiple regression, the energy applied was still significant with a standardized coefficient at 0.3 and P=0.044 (Table 4). Univariate and multivariate analyses were performed with the 6 months/1 year, 6 months/3 years, and 6 months/4 years ratios without finding any significant results.

Table 4.

Correlation with the Prostate-Specific Antigen Ratio 6 Months/2 Years: Univariate Analysis and Multivariate Analysis by Multiple Regression

	Univariate analysis			Multivariate analysis
	Pearson coefficient	P value	95% CI	Standard error	Standardized coefficient	t value	P value
Age	−0.078	0.484	−0.29;0.14	0.013	−0.024	−0.216	0.829
Vaporization time	0.110	0.328	−0.11;0.32	0.007	−0.092	−0.655	0.515
Energy	0.284	0.009	0.07;0.47	1.20E-06	0.3	2.047	0.044
Prostate volume	−0.172	0.119	−0.37;0.04	0.003	−0.049	−0.406	0.686
Energy/volume	0.198	0.072	−0.02;0.4	6.90E-05	0.123	1.032	0.305

CI=confidence interval.

Discussion

The aim of this study was to assess the PSA variations after PVP. To do so, we prospectively included 308 patients during a period of 8 years. We observed that PSA significantly decreased by half after the procedure. When compared with preoperative values, the PSA level decreased significantly by 47% at 1 month, 52% at 6 months, 49% at 1 year, 47% at 2 years, 49% at 3 years, and 44% at 4 years (all P<0.0001). The median PSA nadir occurred at 6 months and was significantly different from the median PSA level at 1, 2, 3, and 4 years (respectively, P=0.0046, P=0.0017, P=0.0006 and P=0.01), which shows a progressive reascension of the median PSA level after 6 months.

In univariate and multivariate analysis, only the energy seemed to have a low but significant correlation with the PSA reascension between 6 months and 2 years. The standardized coefficient between energy and the PSA-ratio was 0.3; in other words, when the energy increased, the PSA-ratio increased, which means that the PSA level either continued to decrease or at least had a lower reascension.

At the beginning of our experience, we unfortunately delivered in some cases insufficient energy. We observed that these patients had a reascension of the PSA level over time. Consequently, we delivered more energy to remove as much tissue as possible. Eventually, the energy seemed to be an important factor in the PSA evolution; however, we assumed that the prostate volume was an important confusion factor for the energy influence. We therefore decided to consider the energy applied per prostate volume rather than the energy alone. We decided to define two subgroups of patients: Those who received a low amount of energy per prostate volume vs those who received a high amount of energy per volume. Extreme cutoffs of ≤3000 J/cc and ≥4000 J/cc were arbitrary chosen. Actually, these two subgroups did not have any clinical implications; however, they were likely to provide an interesting explicative model for the interpretation of the univariate and multivariate results. We found that applying an energy rate higher than 4000 J/cc of prostate induced PSA stability over time whereas an energy below 3000 J/cc induced a r-ascension of the PSA level 6 months after procedure.

There are some limitations to our study that must be stated. Obviously, it would have been interesting to assess PSA evolution according to the variations of the prostate volume before and after surgery. We lacked data, however, about the volume measurements, and we did not find any correlation between PSA level and volume variations.

Two possible biases were the evolution of the generator technology between 2005 until now and the difference of experience of the team at the beginning and at the end of the inclusion. Aware of these potential biases, we compared the PSA evolution according to the model of the laser generator: We found a significant reascension through time for the 80 W and 120 W groups and trend to a reascension for the 180 W group. In addition, there were no significant differences between the ratio to baseline between the three models at 1 month, 6 months, and 1 year (respectively, P=0.34, P=0.11, P=0.36). There was no significant difference between the 6 months/2 years ratios between the three models (P=0.43) and thus no significant difference in the PSA reascension between the three models.

All this suggests that this phenomenon was not related to the model of generator used. This means that neither technology nor experience did seem to influence the PSA evolution and, moreover, that PSA level was mainly influenced by the amount of energy applied to the prostate. Rieken and coworkers⁷ have reported a significantly higher reduction of PSA with the 180 W generator than with the 80 W and 120 W. Their results, however, have several biases: First, they chose to express the PSA evolution by differences rather than ratios. This is contestable, because their standard deviations were very high for both PSA and prostate volume, which makes the comparisons between differences uninformative. Second, there was a significant difference for the energy applied per prostate volume between the three groups. Finally, it was a retrospective study. Unfortunately, they did not provide any data on the PSA evolution through time.

One of the major points of our study was the long-term prospective outcomes. Several series reported a significant postoperative PSA reduction with a single postoperative measure but did not assess its evolution through time.^7

–10 Mosli and associates¹¹ reported with a 1-year follow-up a 49% reduction of the PSA level 3 months after the procedure that remained stable at 12 months. Only a few studies, however, have assessed the PSA evolution with a long follow-up. Ruszat and colleagues¹² reported a global significant decrease of the PSA by 44% 1 month after PVP. This decrease remained significant and stable during the 48 following months. They observed a trend to a reascension of the PSA level, however, for patients with an initial prostate volume>80 cc. This specific group had the lower rate of energy applied per prostate volume and was the only one that showed a clear reascension of the PSA level during the follow-up: The decrease rate of the PSA level compared with baseline was 43% at 1 month and progressively slipped to 26% at 48 months after the procedure.

Similar results were found by Al-Ansari and coworkers.¹³ They showed significant decrease of the PSA level after PVP: The PSA reduction was 40% the first month after the procedure and 50% at 6 months. Then the PSA level started to increase, and the reduction rate returned to 40% after 3 years. Tasçi and colleagues¹⁴ also reported a global significant decrease of the PSA level by 25% 1 year after PVP. The PSA returned to baseline value 2 years after the procedure. The median energy applied, however, was 2300 J/cc, which was quite low and confirms our results. This means that applying a low energy per prostate volume may lead to a regrowth of the remaining adenoma and a reascension of the PSA value.

Other new laser techniques for BPH treatment such as HoLEP have slightly different outcomes with PSA postoperative evolution⁸: Elmansy and associates¹⁵ reported a significant median PSA drop from 5.44 to 0.91 ng/mL. PVP and holmium laser enucleation of the prostate (HoLEP) were compared in a randomized clinical trial that reported a volume and PSA decrease of 78% and 88% in the HoLEP group and 52% and 60% in the PVP group, respectively.¹⁶ These results are inherent to the techniques itself: HoLEP is an enucleation technique whereas PVP is more like a resection technique. Therefore, it is likely that PVP may leave more remaining prostatic tissue according to the energy per volume applied and is more exposed to a distant postoperative regrowth.

We assume that the natural evolution of the PSA level after PVP for benign prostatic enlargement is a decrease by half the first month with a nadir at 6 months and slight reascension during the 4 following years according to the amount of energy applied. This implies evoking PCa in case of an early increase of the PSA level in the following 4 years after the procedure, especially if there are suspicious arguments in the personal history of the patient and his clinical examination. These results, however, are insufficient to have certainty about the real risk of PCa after PSA reascension post-PVP, but it gives an example of the tendency of the normal PSA evolution after PVP.

Conclusions

PSA significantly decreased by half 1 month after the procedure, reached its nadir at 6 months, and showed a slight progressive reascension during the four following years. Applying an energy rate more than 4000 J/cc of prostate induced PSA stability over time, whereas energy below 3000 J/cc induced a reascension of the PSA level after 1 year. These outcomes give an interesting model of the PSA evolution after PVP and thus might be useful to define optimal follow-up strategies for the prostate.

Footnotes

Disclosure Statement

Dr. Bruyere is a proctor for AMS. For the remaining authors, no competing financial interests exist.

Abbreviations Used

References

Heidenreich

, Bellmunt

, Bolla

, et al. EAU guidelines on prostate cancer. Part 1: Screening, diagnosis, and treatment of clinically localised disease. Eur Urol, 2011; 59:61–71.

Volkan

, Ihsan

, Yilmaz

, et al. Short-term outcomes of high power (80 W) potassium-titanyl-phosphate laser vaporization of the prostate. Eur Urol, 2005; 48:608–613.

Shim

, Kwon

, Kim

, et al. Changes in serum prostate-specific antigen levels after potassium-titanyl-phosphate (KTP) laser vaporization of the prostate. Korean J Urol, 2010; 51:111–114.

Oesterling

, Tekchandani

, Martin

, et al. The periurethral glands do not significantly influence the serum prostate specific antigen concentration. J Urol, 1996; 155:1658–1660.

Aus

, Bergdahl

, Frosing

, et al. Reference range of prostate-specific antigen after transurethral resection of the prostate. Urology, 1996; 47:529–531.

Wolff

, Boekels

, Borchers

, et al. Altered prostate specific antigen reference range after transurethral resection of the prostate. Anticancer Res, 2000; 20:4977–4980.

Rieken

, Bonkat

, Müller

, et al. The effect of increased maximum power output on perioperative and early postoperative outcome in photoselective vaporization of the prostate. Lasers Surg Med, 2013; 45:28–33.

Elterman

, Chughtai

, Lee

, et al. Comparison of techniques for transurethral laser prostatectomy: Standard photoselective vaporization of the prostate versus transurethral laser enucleation of the prostate. J Endourol, 2013; 27:751–755.

Ben-Zvi

, Hueber

, Liberman

, et al. GreenLight XPS 180W vs HPS 120W laser therapy for benign prostate hyperplasia: A prospective comparative analysis after 200 cases in a single-center study. Urology, 2013; 81:853–858. PubMed PMID: 23414692. Epub 2013/02/19. eng.

10.

Hueber

, Ben-Zvi

, Liberman

, et al. Mid term outcomes of initial 250 case experience with GreenLight 120W-HPS photoselective vaporization prostatectomy for benign prostatic hyperplasia: Comparison of prostate volumes <60 cc, 60 cc-100 cc and >100 cc. Can J Urol, 2012; 19:6450–6458. PubMed PMID: 23040627. Epub 2012/10/09. eng.

11.

Mosli

, Abdel-Meguid

, Abdulwahhab

, et al. Photoselective vaporization of the prostate using GreenLight 120-W lithium triborate laser to treat symptomatic benign prostatic hyperplasia: A single-centre prospective study. Can Urol Assoc J, 2012; 19:1–4.

12.

Ruszat

, Seitz

, Wyler

, et al. GreenLight laser vaporization of the prostate: Single-center experience and long-term results after 500 procedures. Eur Urol, 2008; 54:893–901.

13.

Al-Ansari

, Younes

, Sampige

, et al. GreenLight HPS 120-W laser vaporization versus transurethral resection of the prostate for treatment of benign prostatic hyperplasia: A randomized clinical trial with midterm follow-up. Eur Urol, 2010; 58:349–355.

14.

Tasci

, Ilbey

, Luleci

, et al. 120-W GreenLight laser photoselective vaporization of prostate for benign prostatic hyperplasia: Midterm outcomes. Urology, 2011; 78:134–140.

15.

Elmansy

, Elzayat

, Sampalis

, Elhilali

. Prostatic-specific antigen velocity after holmium laser enucleation of the prostate: Possible predictor for the assessment of treatment effect durability for benign prostatic hyperplasia and detection of malignancy. Urology, 2009; 74:1105–1110.

16.

Elmansy

, Baazeem

, Kotb

, et al. Holmium laser enucleation versus photoselective vaporization for prostatic adenoma greater than 60 ml: Preliminary results of a prospective, randomized clinical trial. J Urol, 2012; 188:216–221.