Segmental dosimetry, toxicity and long-term outcome in patients with prostate cancer treated with permanent seed implants

Authors


Abstract

What's known on the subject? and What does the study add?

  • The development of side effects characteristic for the different treatment methods with impact on the patients' quality of life plays a growing role for individual patients with early stage prostate cancer. Using permanent brachytherapy a high dose to the prostate can be applied with a steep dose gradient to the normal tissue. However, small partial volumes of normal tissue may be exposed to high doses inducing special side effects including lower urinary tract symptoms and/or erectile dysfunction. In the literature there are only few publications so far regarding segmental dosimetry and its influence on side effects and the results are conflicting.
  • We could not identify any dosimetric parameter in segmental dosimetry that may have an influence at certain time intervals on the development of side effects such as lower urinary tract symptoms or erectile dysfunction. However, we could state clearly that the preoperative situation is the most important factor for postoperative outcome.

Objective

  • To report on the side effects of patients with low to low–intermediate risk prostate cancer treated with permanent interstitial brachytherapy with special emphasis on segmental dosimetry.

Patients and Methods

  • A series of 186 consecutive patients treated for early stage prostate cancer receiving definitive I-125 brachytherapy (permanent seed implantation) between November 2001 and April 2005 at our institution were examined for the development of side effects.
  • Morbidity was assessed prospectively using the International Prostate Symptom Score (IPSS) and the International Index of Erectile Function (IIEF-5) in a mean follow-up interval of 30 months.
  • The scores were correlated with segmental dosimetry performed 6 weeks after the implantation.

Results

  • The mean postoperative dose to 90% of the prostate volume (D90) was 180.2 Gy, the mean preoperative IPSS 7.2 and the mean IIEF-5 14.35, with all scores showing a maximum deterioration after 6 weeks with normalization after 24 months.
  • After correlating the segmental dosimetry and the scores at different time intervals, only the baseline scores remained statistically significant in multivariate regression analysis at all time intervals (P < 0.00).

Conclusions

  • We could not demonstrate a correlation of segmental dosimetry with induction of side effects.
  • There is no relationship between dose exposure of partial volumes and the development of radiation-induced toxicities.
  • The preoperative situation regarding lower urinary tract symptoms and erectile function are the most important factors for postoperative outcome.
Abbreviations
IPSS

International Prostate Symptom Score

IIEF-5

International Index of Erectile Function

V100

volume receiving 100% of the prescription dose;

D90

dose to 90% of the volume of the target

Introduction

Prostate cancer is one of the most frequent cancers in men. Due to the screening for PSA the incidence is rising, on the one hand, and, on the other, patients with low risk prostate cancer are seen more frequently [1, 2]. In these patients different therapeutic options exist with high chance of cure [3-6]. In this constellation the development of side effects characteristic for the different treatment methods with impact on the patients' quality of life plays a growing role for the individual patient [7-10]. With the use of permanent brachytherapy a high dose to the prostate can be applied with a steep dose gradient to the normal tissue and excellent cure rates comparable with radical prostatectomy or external beam radiotherapy for low risk cancers [6, 10-15]. A recent publication comparing different therapeutic approaches in patients with prostate cancer in a meta-analysis could even demonstrate the superiority of brachytherapy compared with other treatment modalities in patients with low risk disease [16]. However, small partial volumes of normal tissue may be exposed to high doses inducing special side effects. The aim of our study was to investigate a possible correlation of the dose exposure of partial volumes and the impact on the development of clinically significant side effects by the use of segmental dosimetry.

Patients and Methods

Between November 2001 and April 2005, 305 patients with prostate cancer were treated with permanent brachytherapy at Hannover Medical School. Of these, 186 patients with a minimum follow-up of 24 months and sufficient data obtained by use of questionnaires were taken into account for statistical analysis. Indications for permanent brachytherapy were biopsy-proven adenocarcinoma of the prostate and clinically localized low to intermediate risk prostate cancer (T classification cT1a–cT2a) with PSA serum level < 10 ng/mL and Gleason sum ≤ 6. Exceptions such as T classification cT2b, a PSA serum level 10–20 ng/mL and a Gleason sum of 7 were only made at the discretion of the treating urologist. For reduction of the prostate volume short term (2–4 months pre-implantation) neoadjuvant hormone ablative therapy was given in 12 of the 186 patients (6%). Brachytherapy was administered via transperineal approach using I-125 seeds with a prescription dose of at least 145 Gy, respecting the ESTRO/EAU/EORTC recommendations. Intraoperative dynamic planning and seed placement were performed with biplanar transrectal ultrasound imaging. Post-implant dosimetry with special emphasis on partial volume analysis such as apex and base of the prostate and penile bulb was carried out 6 weeks after implantation by using CT with the Varian VariSeed Software 7.0/7.1 (Varian Medical Systems Inc., Palo Alto, California, USA). Dose to the base of the prostate was defined as the average dose on the first five transverse images taken at 2 mm intervals containing the prostate; dose to the apex of the prostate was defined as the average dose to the last five transverse images containing the prostate. The penile bulb was contoured as the structure that was bounded by the paired crura laterally, the corpora spongiosum anteriorly and the levator ani posteriorly. The partial volumes were delineated by a single person without inter-observer variability.

Outcome of therapy regarding morbidity was measured prospectively with the International Prostate Symptom Score (IPSS, 0–35 points, low score showing good function) and the International Index of Erectile Function (IIEF-5, 1–25 points, high score showing good function). Sixty patients already had moderate to severe impairments of erectile function (IIEF-5 ≤ 11) before implantation but were not excluded from this analysis in order to gain a large number of patients for statistical calculations.

Statistical considerations regarding differences were calculated by the use of univariate and multivariate regression analysis. For multivariate regression analysis all parameters with P > 0.20 in univariate regression analysis were excluded. Additionally, for avoidance of possible collinearities a Pearson's correlational analysis was performed, and these parameters were also excluded. A P value < 0.05 was defined to be statistically significant.

Results

Detailed patients' characteristics are summarized in Table 1. The mean post-implant CT prostate volume was 36.5 mL, the mean post-implant CT volume of the apex of the prostate was 3.6 mL representing 10.0% of the total prostate volume and the mean post-implant CT volume of the base of the prostate was 4.8 mL representing 13.2% of the total prostate volume. Figures 1 and 2 show an example of the dose distribution and the contouring of the partial volumes. The mean preoperative IPSS was 7.2 and the mean IIEF-5 14.35. The detailed segmental dosimetric data are summarized in Table 2. All the scores showed deterioration with a peak 6 weeks after the implantation. At 24 months the baseline scores were nearly reached again with a mean IPSS of 9.99 and a mean IIEF-5 of 12.38 (Figs 3 and 4). Regarding the IPSS the volume of the prostate, the number of seeds, the total activity, the V100 (volume receiving 100% of the prescription dose) of the apex, the D90 (dose to 90% of the volume) of the base and the V100 of the base of the prostate had impact on the IPSS to certain time points in univariate regression analysis. Only the preoperative IPSS was shown to have a statistically significant impact on further course of the IPSS at all time points. Regarding the IIEF-5 the D90 and the V100 of the apex of the prostate, the V100 of the base of the prostate and the pre-existing IIEF-5 had statistically significant impact in univariate regression analysis (Table 3).

Figure 1.

Example of the coronal dose distribution and contouring of the partial volumes (yellow, bladder; red, prostate; light green, base of the prostate; dark green, apex of the prostate; beige, penile bulb; purple, 80 Gy isodose; green, 145 Gy isodose).

Figure 2.

Example of the definition of the penile bulb.

Figure 3.

Course of the different scores during follow-up. The numbers show the number of patients with sufficient questionnaires that could be analysed at the different time points during follow-up. After multivariate regression analysis the baseline IPSS had statistically significant impact on the IPSS level after implantation at all times during the follow-up (P < 0.00) and the baseline IIEF-5 remained with statistically significant impact on the post-implant IIEF-5 at any time (P < 0.00) during the follow-up.

Figure 4.

Course of the IIEF-5 during the follow-up stratified to the pre-existing impairment of erectile function (1–25 points, high score showing good function; no impairment, 25–22 points; mild, 21–17 points; mild to moderate, 16–12 points; moderate, 11–8 points; severe, ≤7 points).

Table 1. Patients' characteristics.
Parameter Range (min–max)
Mean age at diagnosis (years)65.847.8–81.7
T classification  
cT1c1 
cT2a144 
cT2b33 
Unknown8 
PSA level at diagnosis (ng/mL)7.211.38–26
<4 (ng/mL)15 
4–10 (ng/mL)145 
>10 (ng/mL)25 
Unknown1 
Mean Gleason sum5.723–8
554 
6118 
713 
81 
Neoadjuvant hormone therapy12 
Mean pre-implant ultrasound prostate volume (mL)38.815.9–94.3
Mean post-implant CT prostate volume (mL)36.515.1–91.5
Mean post-implant CT volume of the apex of the prostate (mL)3.61.6–7.1
Mean post-implant CT volume of the base of the prostate (mL)4.82.0–9.0
Mean baseline IPSS7.20–32
Mean baseline IIEF-514.351–25
Mean follow-up (months)3012–48
Table 2. Dosimetric data.
Parameter Range (min–max)
Mean number of needles21.213–28
Mean number of seeds66.235–108
Mean post-implant prostate volume (mL)37.0415.14–91.46
Total activity (mCi)31.4216.51–58.36
Mean dose to 90% of the prostate (D90, Gy)180.2166.8–196.2
Mean dose to 90% of apex of the prostate (D90, Gy)162.485–230
Mean dose to 90% of base of the prostate (D90, Gy)148.375–210
Mean dose to 90% of the penile bulb (D90, Gy)88.8115–185
Mean volume of rectum receiving 100% of prescription dose (V100, mL)1.080–6.26
Mean volume of the apex of the prostate receiving 100% of prescription dose (V100, mL)2.80.74–6.0
Mean volume of the base of the prostate receiving 100% of prescription dose (V100, mL)3.50.24–7.97
Mean volume of the penile bulb receiving 100% of prescription dose (V100, mL)0.580.00–2.81
Table 3. Results of the univariate regression analysis.
 6 weeks post-implantation6 months post-implantation12 months post-implantation18 months post-implantation24 months post-implantation
IPSSIIEF-5IPSSIIEF-5IPSSIIEF-5IPSSIIEF-5IPSSIIEF-5
Age0.8880.00040.4630.00030.4960.00030.391<0.00010.7770.0003
Volume of the prostate in ultrasound<0.00010.5340.0200.9970.00050.1720.1230.6900.0690.709
Volume of the prostate in CT<0.00010.5450.0300.7910.00050.2180.0660.7430.0590.978
Number of seeds0.0010.4510.0470.1200.0130.5250.0960.1510.1690.150
Total activity0.00050.8740.0920.4800.0020.3580.1540.6500.1790.807
D90 of the prostate0.5760.5550.1980.7050.4920.2730.3800.4600.3930.640
D90 of the apex of the prostate0.4780.9780.6970.5450.2260.9950.8140.5210.9300.032
V100 of the apex of the prostate0.0040.6900.1030.3080.0140.0030.4050.1960.3200.238
D90 of the base of the prostate0.0930.5910.0410.1680.0630.5160.1350.1380.1840.630
V100 of the base of the prostate0.0030.6020.1050.9400.0030.0200.3230.3070.0490.468
D90 of the penile bulb0.1680.8840.1510.5050.2900.1700.3740.4610.0800.313
D50 of the penile bulb0.0800.9600.1090.7460.8380.3150.7150.6430.4300.091
V100 of the penile bulb0.0810.9520.0690.8620.1690.1690.3260.4790.2270.308
Score before implantation<0.0001<0.0001<0.0001<0.0001<0.0001<0.0001<0.0001<0.0001<0.0001<0.0001

After multivariate regression analysis only including the parameters with P < 0.2 in univariate regression analysis without any collinearities to other parameters the volume of the prostate (P = 0.005) had impact on the IPSS 6 weeks after implantation and the baseline IPSS had a statistically significant impact on the IPSS level after implantation at all times during the follow-up (P < 0.00). The age of the patient had a statistically significant impact on the IIEF-5 at 18 months after the implantation (P = 0.024) whereas the baseline IIEF-5 remained with a statistically significant impact on the post-implant IIEF-5 at any time (P < 0.00) during the follow-up in multivariate regression analysis (Table 4).

Table 4. Results of the multivariate regression analysis after excluding all parameters with P > 0.200 in univariate regression analysis and parameters with collinearities.
 6 weeks post-implantation6 months post-implantation12 months post-implantation18 months post-implantation24 months post-implantation
IPSSIIEF-5IPSSIIEF-5IPSSIIEF-5IPSSIIEF-5IPSSIIEF-5
Age 0.138 0.158 0.405 0.024 0.105
Volume of the prostate in ultrasound          
Volume of the prostate in CT0.005 0.583 0.193 0.683 0.755 
Number of seeds          
Total activity          
D90 of the prostate  0.5520.353   0.7064 0.748
D90 of the apex of the prostate         0.128
V100 of the apex of the prostate0.616 0.830 0.9390.075 0.573  
D90 of the base of the prostate0.177 0.2050.5980.307 0.3620.1420.620 
V100 of the base of the prostate0.964 0.761 0.7570.3010.984   
D90 of the penile bulb0.338 0.161  0.3950.230   
D50 of the penile bulb          
V100 of the penile bulb          
Score before implantation<0.0001<0.0001<0.0001<0.0001<0.0001<0.0001<0.0001<0.0001<0.0001<0.0001

Discussion

The use of permanent brachytherapy has become a very common therapeutic procedure in patients with low risk prostate cancer. Due to the steep dose gradient the surrounding tissue is spared, in contrast with percutaneous irradiation [17]. However, possible side effects include lower urinary tract symptoms and/or erectile dysfunction [14, 15, 18-20]. By the use of permanent brachytherapy in patients with low risk prostate cancer a high cure rate of 93%–97% can be achieved as was also demonstrated in the present analysis [12, 15, 21, 22]. One important parameter for the quality of the implantation concerning cure rate and the occurrence of side effects is the D90 of the prostate [23-27]. In case of a D90 < 140 Gy the cure rate is diminished whereas a D90 > 160 Gy may predict the occurrence of side effects such as erectile dysfunction or lower urinary tract symptoms [28, 29]. The development of side effects can be objectively identified and measured by the use of validated scores before and in regular intervals after implantation such as the IPSS regarding lower urinary tract symptoms and the IIEF-5 regarding erectile function [20, 30, 31]. However, next to this basic quality parameter partial volumes can be defined and identified in the post-implant CT by the use of segmental dosimetry that may also have an influence on the development of certain side effects even though the D90 is between 140 and 160 Gy. In the literature there are only few publications so far regarding segmental dosimetry and its influence on side effects.

Allen et al. [32] examined the possible role of the D90 and its impact on the urinary function, and they could not demonstrate any statistically significant influence in univariate and multivariate analysis on the IPSS. Additionally, Allen et al. could not demonstrate any statistically significant influence of the dosimetry of the urethra and the occurrence of lower urinary tract symptoms measured by the IPSS. Only the baseline IPSS showed a significant correlation [32]. Also Mallick et al. [33] have seen the impact of the pre-existing IPPS on further outcome of the IPSS [33]. Thomas et al. [34] could demonstrate a statistically significant correlation between the maximal rise of the IPSS and the D50 of the base of the prostate, the V100 of the base of the prostate and the volume of the prostate in multivariate analysis. Murakami et al. [35] could show a statistically significant correlation between the postoperative volume of the prostate, the number of seeds, the maximal dose and the D90 and D50 of the basal urethra and its impact on the IPSS. Neill et al. [36] could demonstrate a statistically significant influence of the baseline IPSS and the D30 and V100 of the prostate but not the segmental dosimetry of the prostatic urethra. We could demonstrate a deterioration of the IPSS with a maximum at 6 weeks after the implantation caused by the operative trauma and the irradiation exposure, which was also observed by other authors [37, 38]. Keyes et al. [39] had seen the baseline IPPS, the post-implant maximum IPSS and V150 of the prostate as statistically significant factors with impact on lower urinary tract symptoms [39]. However, the IPSS showed normalization after 24 months in the large majority of the patients. In univariate analysis we could demonstrate a statistically significant correlation between further course of IPSS and the volume of the prostate, the V100 of the apical and basal prostate and the D90 of the basal prostate. However, only the baseline IPSS was found to be statistically significant in univariate and multivariate regression analysis at all times during the follow-up.

Regarding erectile dysfunction after seed application there is a large range in the literature reaching from 39% to 76% for preserving the function, which is also dependent on the follow-up interval [40, 41]. The exact aetiology of developing erectile dysfunction after brachytherapy is not solved clearly; different factors such as neurological or vascular reasons, age, comorbidities, co-medications and implant method were taken into account [42]. Bittner et al. [43] could demonstrate that a low pre-implant IIEF score was associated with a higher incidence of all-cause mortality with pretreatment erectile dysfunction as a surrogate for underlying vascular pathology. Carrier et al. [44] showed in rats that the development of erectile dysfunction seems to be connected with exposure of the proximal penis including the bulb of the penis and the crura and corpora cavernosa to irradiation. We showed that besides the IPSS also erectile dysfunction, objectively measured by the IIEF-5, showed deterioration 6 weeks after the implantation. Merrick et al. [41, 45] demonstrated the effect of the pre-implant IIEF and of the D50 of the penile bulb but not of the D90 of the prostate on the development of erectile dysfunction. However, McDonald et al. [46] showed that age, baseline erectile function and the number of needles but not the dosimetry of the penile bulb have a statistically significant correlation with the occurrence of erectile dysfunction. The authors assume the implant trauma to the penile bulb rather than the exposure of the penile bulb to radiation may have an impact. We also could not demonstrate a statistically significant correlation between the exposure of the penile bulb or number of seeds and the development of erectile dysfunction in univariate or multivariate regression analysis. We demonstrated that age at the time of implantation and the preoperative score in the IIEF-5 had a statistically significant impact on the development of erectile dysfunction in univariate analysis; the baseline IIEF-5 remained statistically significant in multivariate analysis emphasizing the prognostic value of the baseline IIEF-5. Therefore it seems that not the dose as a continuous variable but the underlying IIEF score before implantation as a threshold effect has the most impact on the development of erectile dysfunction, as was also demonstrated by Merrick et al. [41, 45] and Solan et al. [47]. As a caveat of our study it must be taken into account that 60 patients in our cohort still had moderate to severe impairments of erectile function (IIEF-5 ≤ 11) before implantation, and they will naturally not be potent after brachytherapy. Although we compared the difference of the score before and after brachytherapy the score in patients with limited potency only changed a little, in contrast to patients with preserved potency.

However, when comparing with other authors the results of segmental dosimetry for impact on clinical signs, the differences in the definition and contouring of these partial volumes have to be taken into account. We used a rigid definition without respect to the size of the prostate for the definition of the apex and base of the bulbus that may lead to overestimation in small and underestimation in large prostates on the one hand. On the other hand the contouring was performed by a single person to eliminate inter-observer variability. Therefore the volumes and also the dose to the partial volumes may differ substantially between the different studies which makes comparative analysis difficult. Another limitation of our study is the fact that we have not been able to use MRI-based post-implant dosimetry for better visualization of the partial volumes. This would be more accurate than the CT dosimetry due to scatter artifacts. A different limitation of our study and one reason for not finding a difference between the groups is the fact that the trauma from the brachytherapy procedure may alter the volume of the prostate and therefore the volumes and the partial dosimetry in the post-implant CT. These may be the reasons why our study could not demonstrate an influence of the dose to certain partial volumes for the development of side effects.

To conclude, we could not identify any dosimetric parameter in the segmental dosimetry that may have an influence at certain time intervals on the development of side effects such as lower urinary tract symptoms or erectile dysfunction. However, our study has some limitations such as the unavailability of MRI-based post-implant dosimetry, inclusion of patients with pre-existing disturbances in erectile function and the different definitions of partial volumes. Nevertheless we can state clearly that the preoperative situation is the most important factor for postoperative outcome.

Conflict of Interest

None declared.

Ancillary