The relationship between the bladder volume and optimal treatment planning in definitive radiotherapy for localized prostate cancer

Abstract

Background. There is no current consensus regarding the optimal bladder volumes in definitive radiotherapy for localized prostate cancer. The aim of this study was to clarify the relationship between the bladder volume and optimal treatment planning in radiotherapy for localized prostate cancer. Material and methods. Two hundred and forty-three patients underwent definitive radiotherapy with helical tomotherapy for intermediate- and high-risk localized prostate cancer. The prescribed dose defined as 95% of the planning target volume (PTV) receiving ≧ 100% of the prescription dose was 76 Gy in 38 fractions. The clinical target volume (CTV) was defined as the prostate with a 5-mm margin and 2 cm of the proximal seminal vesicle. The PTV was defined as the CTV with a 5-mm margin. Treatment plans were optimized to satisfy the dose constraints defined by in-house protocols for PTV and organs at risk (rectum wall, bladder wall, sigmoid colon and small intestine). If all dose constraints were satisfied, the plan was defined as an optimal plan (OP). Results. An OP was achieved with 203 patients (84%). Mean bladder volume (± 1 SD) was 266 ml (± 130 ml) among those with an OP and 214 ml (±130 ml) among those without an OP (p = 0.02). Logistic regression analysis also showed that bladder volumes below 150 ml decreased the possibility of achieving an OP. However, the percentage of patients with an OP showed a plateau effect at bladder volumes above 150 ml. Conclusions. Bladder volume is a significant factor affecting OP rates. However, our results suggest that bladder volumes exceeding 150 ml may not help meet planning dose constraints.

The bladder is filled to various volumes during fractionated radiotherapy. Changing bladder volumes affects both bladder dose volumes and the position of adjacent organs (the prostate, seminal vesicles, small intestine and sigmoid colon) [1]. Furthermore, significant variations in bladder volume can affect planned three-dimensional conformal radiotherapy (3D-CRT) and intensity-modulated radiation therapy (IMRT) dose distributions. For all these reasons, bladder volumes must be kept consistent throughout planning and treatment to reduce positional uncertainties related to the prostate and the risk of increased toxicity to the surrounding normal tissue.

There is no current consensus regarding the optimal bladder volumes in definitive radiotherapy for localized prostate cancer. One possible advantage of maintaining a full bladder is that part of the bladder moves away from the target volume, thereby reducing bladder toxicity [2,3]. A full bladder also moves the small intestine and the sigmoid colon out of the irradiation field, reducing toxicity in these organs [1,4–7]. However, if we target larger bladder volumes on planning using computed tomography (CT) and during radiotherapy, such volumes tend to show marked variability [8–10]. On the other hand, excessively small bladder volumes make it difficult to meet planning dose constraints for the bladder and adjacent organs. For these reasons, the optimal bladder volume may be the minimum bladder volume that can satisfy dose constraints. Based on this reasoning, several institutions target a half-full bladder or a comfortably full bladder [8,9]. However, no previous reports have focused on the relationship between the bladder volume and optimal treatment planning.

We evaluated the relationship between the bladder volume on planning CT and the percentage satisfying the dose constraints as a reference what bladder volumes should be targeted.

Material and methods

Between June 2007 and February 2009, 243 patients underwent definitive radiotherapy with helical tomotherapy using the Hi-Art System (Tomotherapy Inc.) for intermediate- and high-risk localized prostate cancer (cT1-4N0M0) according to D'Amico's classification at Edogawa Hospital (Tokyo, Japan) (Table I).

Table I. Patient characteristics.

	no.
cT stage (TNM 6th ed.)
1–2a	101(42%)
2b	32 (13%)
2c	40 (16%)
3a	61 (25%)
3b	8 (33%)
4	1 (0.4%)
Gleason score
2–6	41 (17%)
7	102 (42%)
8–10	100 (41%)
Pretreatment PSA
0–10	104 (43%)
10–20	67 (28%)
> 20	72 (30%)
D'Amico's risk group
Intermediate	71 (29%)
High	172 (71%)
Neoadjuvant hormone therapy
No	81 (33%)
Yes	162 (67%)
Mean age (range)	70 (42–85)
Mean prostate volume (range)	21 ml (6–178)
Mean PTV (range)	112 ml (61–273)
Mean bladder volume (range)	235 ml (45–653)

cT stage, clinical tumor stage; PSA, prostate-specific antigen; PTV, planning target volume.

The patients were irradiated in a supine position, with a knee support. They were instructed to refrain from urinating for 60–90 minutes before the planning computed tomography (CT) scan and before daily irradiation. They were also encouraged to drink an unspecified volume of water to ensure a clear but tolerable urge to urinate before the planning CT scan and before daily irradiation. They were instructed to take laxatives before the planning CT scan, although no specific instructions were issued regarding bowel movements before daily irradiation.

The clinical target volume (CTV) was defined as the prostate that was delineated by the fusion images of CT and magnetic resonance imaging (MRI) with a 5-mm margin and 2 cm of the proximal seminal vesicle. Exceptionally, the whole seminal vesicle was included in the CTV for cases of clinical T3b stage disease. The planning target volume (PTV) was defined as the CTV with a 5-mm margin. The prescribed dose defined as 95% of the PTV receiving ≧ 100% of the prescription dose (D95) was 76 Gy in 38 fractions. The treatment plans were optimized to satisfy the dose constraints defined by in-house protocols for the PTV and organs at risk (OAR) (Table II). No specific protocols were used for the order of prioritization among the constraints. Cases in which all dose constraints were satisfied were defined as an optimal plan (OP).

Table II. Dose constraints.

Target/Organ		Dose constraint
PTV	D95	100% (76 Gy)
	Maximum	< 110% (83.6 Gy)
	Mean	< 105% (79.8 Gy)
Rectum wall*	V40	< 65%
	V60	< 35%
	V70	< 25%
	V78	< 10%
Bladder wall	V40	< 60%
	V70	< 35%
Sigmoid colon	V65	< 0.5 ml
Small bowel	V60	< 0.5 ml

*Rectum wall within 5 mm above and below the PTV, Vx < y% (or ml) means that no more than y% (or ml) of the volume of the organ receive a dose > x Gy.

PTV, planning target volume.

We assessed the relationship between the bladder volumes on planning CT and the percentage of patients achieving an OP. Univariate logistic regression analysis was used to examine the predictive value of covariates including clinical T stage (T1–2a, T2b, T2c, T3a, T3b, and T4), Gleason score (2–6, 7, 8–10), pretreatment PSA (0–10, 10–20, and > 20), D'Amico's risk group (intermediate or high), neoadjuvant hormone therapy (yes or no), age, PTV, and bladder volume. Those showing significant associations in univariate logistic regression analysis were further tested by multivariate logistic regression analysis.

We used GraphPad Prism version 5 (GraphPad Software Inc.) and SPSS version 17 (IBM) for statistical analysis. Differences were deemed significant when two-tailed p-values were less than 0.05.

Results

Of the subjects, 203 patients (84%) met the definitions for an OP. Among these patients, the mean of the mean PTV dose and the maximum dose were 77.4 Gy (range 76.7–79.2 Gy) and 80.7 Gy (range 78.2–83.3 Gy), respectively.

The mean bladder volume (±1 standard deviation; SD) was 266 ml (±130 ml) among those with an OP and 214 ml (±130 ml) among those without an OP (p = 0.02, by unpaired t-test).

Logistic regression analysis also showed that bladder volumes below 150 ml decreased the possibility of achieving an OP (Table III). Figure 1 shows the percentage of patients with an OP according to bladder volumes, indicating that the percentage of patients with an OP showed a plateau effect at bladder volumes above 150 ml. On univariate analysis, higher clinical T stage, younger age, treatment with neoadjuvant hormone therapy, and larger bladder volume were predictors for achieving an OP (Table IV). On multivariate analysis, larger bladder volumes (p = 0.04), younger age (p = 0.01), and higher clinical T stage (p = 0.03) were independent predictors for achieving an OP.

Table III. Logistic regression analysis between bladder volume and the percentage of patients with an optimal plan.

Bladder volume	Number of patients	Patients with an OP	p	Odds ratio (95% CI)
<100 ml	21	15 (71%)	0.069	0.34 (0.11–1.09)
100–149 ml	34	24 (71%)	0.028	0.33 (0.12–0.89)
150–199 ml	43	37 (86%)	0.761	0.85 (0.29–2.50)
200–249 ml	35	30 (86%)	0.739	0.82 (0.26–2.61)
250–299 ml	27	24 (89%)	0.896	1.10 (0.28–4.31)
>300 ml	83	73 (88%)		1

OP, optimal plan.

Figure 1. The percentage of patients with an OP according to bladder volume. Patients were divided into subgroups according to their bladder volume by 50 ml. The percentage of patients with an OP was defined by dividing the number of patients with an OP by the number of patients in each subgroup. The size of each dot represents the number in each subgroup. n, number of patients; OP, optimal plan.

Table IV. Univariate logistic regression analysis of association with achieving an optimal plan.

	Patients with an OP (n, 203)	Patients without an OP (n, 40)	p
cT stage (TNM 6th ed.)			0.03
1–2a	77 (38%)	24 (60%)
2b	26 (13%)	6 (15%)
2c	35 (17%)	5 (13%)
3a	57 (28%)	4 (10%)
3b	7 (3%)	1 (3%)
4	1 (0.5%)	0 (0%)
Gleason score			NS
2–6	39 (19%)	2 (5%)
7	83 (41%)	19 (48%)
8–10	81 (40%)	19 (48%)
Pretreatment PSA			NS
0–10	85 (42%)	19 (48%)
10–20	58 (28%)	9 (23%)
> 20	60 (30%)	12 (30%)
D'Amico's risk group			NS
Intermediate	60 (30%)	11 (28%)
High	143 (70%)	29 (73%)
Neoadjuvant hormone therapy			0.10
No	63 (31%)	18 (45%)
Yes	140 (69%)	22 (55%)
Mean age (range)	70 (42–85)	73 (59–83)	0.01
Mean prostate volume (range)	21 ml (6–178)	22 ml (12–103)	NS
Mean PTV (range)	109 ml (61–225)	115 ml (77–273)	NS
Mean bladder volume (range)	266 ml (45–594)	214 ml (48–653)	0.04

cT stage, clinical tumor stage; OP, optimal plan; PSA, prostate-specific antigen; PTV, planning target volume.

Discussion

We found that bladder volumes among patients with an OP were significantly larger than among patients without an OP. This indicates that bladder volume is a significant factor affecting whether OP is achieved. However, we also found that bladder volumes larger than 150 ml did not contribute to OP rates. We could meet the dose constraints on the bladder even with considerably small bladder volumes. However, small bladders moved the small intestine and the sigmoid colon inside the irradiation field, which made it impossible to meet the dose constraint on those organs. This may explain why we found the plateau effect at bladder volumes above 150 ml.

Our logistic regression analysis did not show a statistically significant difference in the percentage of patients with an OP in the subgroup with the smallest bladder volume. We think the relatively small number of subjects in the subgroup caused the false negative.

Our results suggested that younger age and higher clinical T stage were also independent predictors for achieving an OP. It is difficult to interpret why age affects OP achievement. There may be some anatomic features among younger patients that make it easier to achieve an OP. It is also difficult to interpret why clinical T stage affects OP achievements although we used the same definition of CTV for all clinical T stages except for the few cases of clinical T3b.

The existence of a clear dose effect for genitourinary (GU) toxicity is well-known in cases in which the entire bladder is irradiated [11]. In the case of prostate irradiation, the cranial portion of the bladder is generally spared, whereas the bladder neck and urethra are irradiated at levels close to the prescribed dose. Most of the published results fails to support a correlation between bladder dose volume histograms (DVH) and GU toxicity [12,13], whereas several studies indicate that the absolute volume of the bladder receiving >78 Gy to 80 Gy is most predictive of late GU toxicity [14,15]. Regarding GU toxicity, a half-full bladder and an empty bladder appear to be acceptable bladder volumes [16]. However, an excessively small bladder volume may move the small intestine and sigmoid colon within the high dose irradiated field [1,4–6]. Therefore, we also imposed dose constraints on the small intestine and sigmoid colon.

Several previous studies have reported that the greatest variation in bladder volume is found in patients with large initial bladder volumes [8,9,17]. Significant variations in bladder volume can confound planned dose distributions. A half-full bladder of 150 ml or slightly larger may represent a reasonable target, offering the potential to improve bladder volume consistency without compromising the dose constraints for the adjacent organs.

A limitation of this investigation is the lack of the clinical correlation. We need to investigate the correlation between bladder volumes on planning CT and clinical outcomes in a future study. In most cases, we use a shrinking PTV if we can not satisfy the dose constraints for OARs. Our concern is that the compromise might cause inferior local control and survival rates. However, long-term follow-up is necessary to clarify the clinical impact. We consider achieving an optimal plan a surrogate marker for clinical outcomes; therefore, we report the correlation between bladder volumes and achieving an optimal plan as the first step.

While optimal bladder volumes vary from institution to institution according to the protocol used, we believe that each institution must seek to recognize what bladder volumes are optimal in definitive radiotherapy for localized prostate cancer.

In conclusions, bladder volume is a significant factor affecting the achieving of an optimal plan. However, our results suggest that bladder volumes exceeding 150 ml may not help meet planning dose constraints.

Acknowledgements

A portion of this report was presented at the 52^nd Annual Meeting of the American Society of Therapeutic Radiology and Oncology at San Diego, October 31–November 4, 2010. The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.