- Email: [email protected]
Morbidity Associated with Primary High Intensity Focused Ultrasound and Redo High Intensity Focused Ultrasound for Localized Prostate Cancer
Morbidity Associated with Primary High Intensity Focused Ultrasound and Redo High Intensity Focused Ultrasound for Localized Prostate Cancer Viktor Berge,* Louise Dickinson,† Neil McCartan,† Richard G. Hindley, Lien My Diep, Mark Emberton‡ and Hashim Uddin Ahmed† From the Oslo University Hospital (VB, LMD), Oslo, Norway, and Division of Surgery and Interventional Science, University College London and Department of Urology, University College Hospitals National Health Service Foundation Trust (NM), London (LD, ME, HUA) and Department of Urology, North Hampshire and Basingstoke Hospital, Hampshire Hospitals National Health Service Foundation Trust (RGH), Basingstoke, United Kingdom
Abbreviations and Acronyms ADT ¼ androgen deprivation treatment BNI ¼ bladder neck incision GEE ¼ generalized estimating equation HIFU ¼ high intensity focused ultrasound IIEF-15 ¼ International Index of Erectile Function-15 MRI ¼ magnetic resonance imaging PDE-5 ¼ phosphodiesterase-5 PSA ¼ prostate specific antigen RT ¼ radiation therapy TRUS ¼ transrectal ultrasound UTI ¼ urinary tract infection Accepted for publication December 16, 2013. Study received institutional review board approval. * Correspondence: Oslo University Hospital, Oslo, Norway (e-mail: [email protected]). † Financial interest and/or other relationship with SonaCare Medical. ‡ Financial interest and/or other relationship with US HIFU and SonaCare Medical.
Purpose: High intensity focused ultrasound may have a role as an alternative to standard radical therapies for localized prostate cancer. An attribute of high intensity focused ultrasound is that it can be repeated. We determined morbidity after primary and redo high intensity focused ultrasound. Materials and Methods: We performed an academic lead analysis of United Kingdom registry data on high intensity focused ultrasound treatments at 3 centers using patient reported continence and sexual function outcomes. Validated questionnaires were completed before and after each ultrasound treatment. Results: A total of 359 patients received 1 whole gland high intensity focused ultrasound treatment for localized prostate cancer from October 2004 to June 2012, of whom 130 (36.2%) received re-treatment. Median followup was 27 months (range 3 to 81) after re-treatment. When analyzing adverse events, 10.8% of patients experienced urinary tract infection after the first treatment compared to 3.9% after re-treatment (p ¼ 0.009). Urethral dilatation was required in 13.8% and 14.0% of patients after first and redo ultrasound treatments (p ¼ 0.7), and bladder neck incision was required in 9.2% and 11.6%, respectively (p ¼ 0.2). Before and after re-treatment 73.3% and 55.1% of patients had no leak, and 2.7% and 9.0% used daily pads (p <0.001 and p ¼ 0.07, respectively). Analysis of erectile function showed that 56.2% and 56.0% of patients were potent before and after re-treatment, respectively (p ¼ 0.9). Conclusions: Redo high intensity focused ultrasound is associated with an increase in urinary side effects but sexual side effects do not appear to be significantly increased. The number of adverse events seems to be equivalent after first and redo treatments. Meticulous patient selection is of paramount importance when selecting men for redo high intensity focused ultrasound. Key Words: prostate, prostatic neoplasms, high-intensity focused ultrasound ablation, retreatment, adverse effects
THE optimal management strategy for newly diagnosed, clinically localized prostate cancer remains controversial and includes radical prostatectomy, radical radiotherapy
1764
j
www.jurology.com
and active surveillance. As an alternative, minimally invasive therapies such as cryotherapy and HIFU have been proposed as iso-effective cancer control with fewer complications and
0022-5347/14/1916-1764/0 THE JOURNAL OF UROLOGY® © 2014 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH, INC.
http://dx.doi.org/10.1016/j.juro.2013.12.036 Vol. 191, 1764-1769, June 2014 Printed in U.S.A.
MORBIDITY ASSOCIATED WITH HIGH INTENSITY FOCUSED ULTRASOUND
side effects.1,2 Although HIFU still is considered experimental according to European Association of Urology guidelines,3 several studies demonstrated oncologic efficacy comparable to that of radiation therapy with a biochemical failure-free survival rate of between 57% and 76% at 7 to 8-year followup.2,4 The rate of erectile impotence after whole gland HIFU is greater than 50%2,4 and the rate of stress urinary incontinence varies between 7% and 15%.4,5 An advantage of HIFU is that treatment can be repeated (redo HIFU) if the first treatment is not successful. We determine whether redo HIFU results in significant additional morbidity compared to primary HIFU.
PATIENTS AND METHODS Since 2004, an academic led, independent, national United Kingdom HIFU registry has been maintained under the auspices of University College London. In this study we included all patients treated with HIFU using a SonablateÒ 500 for localized prostate cancer at 3 United Kingdom institutions (Princess Grace Hospital, Basingstoke and North Hampshire Hospital, and University College London Hospital National Health Service Foundation Trust) from October 8, 2004 to June 26, 2012. We used standard power settings in the range of 20 to 45 W/cm2 with a duty cycle of 3 seconds on and 3 seconds off or 3 seconds on and 6 seconds off as determined by the operator. Treatment was given on an individual basis with combinations of HIFU blocks derived from the 4 and 3 cm focal lengths to cover the area of recurrence. Men who were unable or unwilling to undergo surgery, radiotherapy or active surveillance were offered HIFU. Contraindications to HIFU were previously reported.6 Men received a single HIFU session or redo HIFU after failed first HIFU based on biochemical, histological or MRI evidence of localized residual disease. To achieve the size restrictions necessary for HIFU treatment cytoreduction with bicalutamide (50 mg once daily) and a 5a-reductase inhibitor (dutasteride or finasteride) for 3 months was done before HIFU. These treatments were stopped on the day of the first HIFU treatment. At the outset all men were fitted with a urethral catheter for 7 to 10 days. After consultation in international user group meetings it was agreed that suprapubic catheters might result in earlier voiding, fewer urinary infections and a lower stricture rate. Since most patients were tertiary referrals and lived some distance away, all men were taught clean intermittent self-catheterization so that they could self-manage decreased flow due to debris passage and short-term prostatic inflammation. Followup after treatment mirrored the regimen used after standard radical therapies, including serum PSA measurement at 6 weeks and then every 3 months for the first year and every 6 months in subsequent followup years. Patients who did not achieve a PSA nadir of less than 0.5 ng/ml and those with PSA less than 0.5 ng/ml with 2 consecutive PSA increases were advised to undergo transrectal prostate biopsy. Of the 359 patients
1765
175 (48.7%) underwent 1 post-HIFU TRUS biopsy and in 108 (30.1%) the TRUS biopsy was positive. Post-HIFU adverse events included need for cystoscopy, BNI, urethral dilatation, UTI and epididymitis. These events were recorded from a review of clinical records on a continual basis by dedicated data management personnel. All patients at the 3 study centers were offered pretreatment and posttreatment validated questionnaires.7 Incontinence data were collected from patient reported outcomes on leakage and pad use using the UCLA-EPIC urinary function question subset.8 From the IIEF-15 data we selected question 2 (“When you had erections with sexual stimulation, how often were your erections hard enough for penetration?”) for analysis, which is accepted as a good indicator of erectile function.9 We defined potency as scoring 2 or above on a scale of 0dno sexual activity, 1dalmost never/never, 2da few times (much less than half the time), 3dsometimes (about half the time), 4dmost times (much more than half the time) and 5dalmost always/always. 1) We compared perioperative outcomes and adverse events between single and redo HIFU. 2) We analyzed the change in urinary continence status from no leak to leak and any pad use. 3) We also evaluated sexual function before and after a single HIFU treatment and subsequently after redo HIFU as well as the differential rates of reported PDE-5 inhibitor use. For completeness so that these outcomes could be placed in context we also report disease control outcomes in the group as a whole. The criteria for HIFU failure was PSA nadir plus 2.0 ng/ml (Phoenix criteria) and/or additional salvage therapy. The difference between single and redo HIFU cases was tested by the 2 independent samples t-test or the Pearson chi-square test, as appropriate. A GEE model with a logit link function was used for a binary dependent variable. Patient identity as a subject variable, time points as a within subject variable and an unstructured correlation structure were selected for the GEE model. The binary dependent variables tested were urinary leakage, pad use, intercourse and PDE-5 inhibitor use. Significance was considered at 0.05 and 2-sided p values are shown. Analysis was done with SPSSÒ, version 17.
RESULTS Patient Characteristics From October 2004 to June 2012 a total of 359 patients were treated with at least 1 whole gland HIFU session for localized prostate cancer. Of the 359 men 96 (26.7%) were pretreated with 3 months of cytoreduction, 130 (36.2%) underwent redo HIFU, 19 (5.3%) underwent 2 redo sessions and 1 (0.3%) underwent 3. The data presented on morbidity and oncologic outcomes were recorded after the first redo HIFU. Preoperative characteristics were similar between men with only 1 HIFU and those with redo HIFU except Gleason score, which was higher in those with redo HIFU (p ¼ 0.004, table 1). Median followup was 45 months (range 3 to 93) in the single
MORBIDITY ASSOCIATED WITH HIGH INTENSITY FOCUSED ULTRASOUND
1766
Table 1. Patient characteristics
No. pts Mean age (range) Mean ml prostate vol (range) Mean SD (ng/ml): Pre-HIFU After first HIFU No. pre-HIFU T stage (%):* T1 T2 T3a No. pre-HIFU Gleason score (%): 5 6 7 (3 þ 4) 7 (4 þ 3) 8 9
1 HIFU
More Than 1 HIFU
p Value
229 65.9 (46.7e87.4) 32.3 (8.4e80.0)
130 64.6 (43.6e80.5) 34.7 (7.0e73.0)
e 0.13 0.08
7.9 6.4 0.8 1.3
8.3 5.0 1.0 1.1
0.54 0.15 0.2
52 130 46
(22.7) (56.8) (20.1)
14 45 23
(17.1) (54.9) (28.0)
3 108 81 20 7 2
(1.4) (49.3) (35.5) (8.8) (3.2) (0.9)
0 45 71 10 3 3
(34.6) (54.6) (7.7) (2.3) (2.3)
0.004
HIFU. There were fewer episodes of UTI after the second HIFU but the rates of post-HIFU cystoscopy, dilatation and BNI were similar for first and redo HIFU (table 2). Patient Questionnaire Response Single HIFU. Of the 359 patients 242 (67.4%) completed UCLA-EPIC questionnaire data on urinary function at baseline and 235 (65.5%) completed the data at last followup after a single HIFU treatment (table 3). Likewise, 124 (34.5%) and 171 patients (47.6%) completed the IIEF-15 questionnaire on sexual function at baseline and last followup, respectively (table 3). Information was available on PDE-5 inhibitor use in 207 (57.7%) and 171 patients (47.6%) at baseline and last followup, respectively (table 3). Redo HIFU. Of 130 patients treated with redo HIFU
* Based on prostate/pelvis multiparametric MRI with locoregional metastasis excluded by staging pelvic MRI in all men and distant metastasis excluded by radioisotope bone scan in all men with Gleason score 7 or greater, PSA 10 ng/ml or greater and/or stage T3a or greater.
HIFU group and 27 months (range 3 to 81) after redo HIFU. Disease Control Outcomes There were no prostate cancer related deaths. In the group with only single HIFU without redo HIFU 60 patients (26.4%) had treatment failure and underwent salvage treatment, including ADT plus salvage RT in 16, ADT plus radical prostatectomy in 2, ADT plus chemotherapy in 1, only salvage RT in 3, only salvage cryotherapy in 1 and only ADT in 13. The remaining patients were on active surveillance. After redo HIFU treatment failed in 56 men (44.8%), of whom 40 underwent TRUS biopsy, which was positive in 22. Salvage therapy included ADT in 20 patients (combined with RT in 2), salvage RT in 17, salvage radical prostatectomy in 1 and second redo HIFU in 19. Perioperative Outcomes and Adverse Events After redo HIFU time, hospital time and catheter time were significantly shorter than after the first
the UCLA-EPIC questionnaire data on urinary function were completed before the first HIFU by 82 patients (66.6%), before redo HIFU by 75 (58.1%) and after redo HIFU by 78 (60.5%) at last followup (table 3). For sexual function IIEF-15 questionnaire data were completed by 40 patients (31.0%) before the first HIFU, by 50 (38.8%) before redo HIFU and by 75 (57.7%) after redo HIFU (table 3). Information was available on PDE-5 inhibitor use for 75 patients (58.1%) at baseline, 74 (57.4%) before redo HIFU and 46 (35.7%) after redo HIFU (table 3). Attrition analysis showed that patients with missing information on urinary continence had higher pre-HIFU PSA (p ¼ 0.02), higher PSA nadir (p ¼ 0.02) and higher Gleason score in the pre-HIFU biopsy (p ¼ 0.03) than who responded (data not shown). Attrition analysis revealed no statistically significant difference in patient characteristics between those who did and did not respond to the IIEF-15 questionnaire and PDE-5 inhibitor use at baseline (data not shown). However, of the 235 patients who did not respond to questions on sexual function 72 (30.6%) received cytoreductive medication before HIFU compared to 24 (19.4%) who responded to the same questions before HIFU (p ¼ 0.02).
Table 2. Treatment characteristics and complications in patients after first and redo HIFU
No. pts* No. cytoreduction (%) No. intraprocedural complications (%) Mean mins treatment time (range) Mean days hospital time (range) Mean days catheter time (range) No. post-HIFU cystoscopy (%) No. UTI (%) No. epididymitis (%) No. BNI with/without resection (%) No. dilatation (%) * No patient had fistula.
HIFU 1
Redo HIFU
p Value
130 34 (26.2) 5 (3.8) 188.6 (90e340) 1.7 (1e7) 17.6 (5e75) 21 (16.2) 14 (10.8) 2 (1.5) 12 (9.2) 18 (13.8)
129 3 (2.3) 7 (5.4) 117.4 (40e285) 1.4 (1e4) 12.3 (1e49) 22 (17.1) 5 (3.9) 2 (1.6) 15 (11.6) 18 (14.0)
<0.001 0.5 <0.001 0.002 0.001 0.9 0.009 0.9 0.2 0.7
MORBIDITY ASSOCIATED WITH HIGH INTENSITY FOCUSED ULTRASOUND
1767
Table 3. Urinary and sexual function, and PDE-5 inhibitor use in 359 patients after first HIFU and in 129 after redo HIFU No. HIFU 1 (%)
Drip free: Yes No Pad free: Yes No Question 2 (did you have erectile function hard enough for penetration over last 4 wks?:* No Yes PDE-5 inhibitors: No Yes
No. Redo HIFU (%)
Pre-HIFU
Post-HIFU
p Value (GEE)
Pre-HIFU 1
Pre-HIFU 2
Post-HIFU 2
p Value (GEE)
208 (86.0) 34 (14.0)
131 (55.7) 104 (44.3)
<0.001
73 (89.2) 9 (10.8)
55 (73.3) 20 (26.7)
43 (55.1) 35 (44.9)
<0.001
238 (98.3) 4 (1.7)
209 (90.1) 23 (9.9)
<0.001
81 (98.8) 1 (1.2)
73 (97.3) 2 (2.7)
71 (91.0) 7 (9.0)
0.07
54 (43.5) 70 (56.5)
86 (50.3) 85 (49.7)
0.2
16 (40.0) 24 (60.0)
22 (44.0) 28 (56.2)
33 (44.0) 42 (56.0)
0.9
173 (83.6) 34 (16.4)
65 (38.0) 106 (62.0)
<0.001
65 (87.8) 9 (12.2)
37 (49.3) 38 (50.7)
23 (50.0) 23 (50.0)
<0.001†
* Less than 1 vs 2 or more. † Pre-HIFU 2 vs post-HIFU 2 p ¼ 0.5.
Function Urinary. There was a statistically significant increase in the rate of leak urinary incontinence and pad use after 1 whole gland HIFU session (table 3). GEE analysis showed that before and after redo HIFU 73.3% and 55.1% of patients, respectively, had no urinary leak (p <0.001). Before and after redo HIFU 2.7% and 9.0% of patients, respectively, used daily pads (p ¼ 0.07, table 3 and part A of figure). Sexual. We found no statistically significant differ-
ence in erectile function sufficient for intercourse before and after the first whole gland HIFU across all patients (table 3). Moreover, redo HIFU did not result in any additional significant deterioration in sexual function (table 3 and part B of figure). There was a statistically significant increase in PDE-5 inhibitor use after first whole gland HIFU (p <0.001). However, after redo HIFU there was no statistically significant difference in PDE-5 inhibitor use (p ¼ 0.5, table 3 and part B of figure). Multivariate logistic regression showed that age (p <0.001) and redo HIFU (p ¼ 0.05) were associated
with decreased erectile function at last followup. There was no statistically significant association between erectile function and covariates such as preHIFU cytoreduction (p ¼ 0.2), pre-HIFU PDE-5 inhibitor use (p ¼ 0.4), post-HIFU PDE-5 inhibitor use (p ¼ 0.2), pad use (p ¼ 0.2), indwelling catheter length (p ¼ 0.8), post-HIFU cystoscopy (p ¼ 0.3), urethral dilatation (p ¼ 0.3) or BNI (p ¼ 0.4). PostHIFU ADT was associated with decreased erectile function at last followup on univariate logistic regression but not on multivariate logistic regression (p ¼ 0.01 and 0.1, respectively).
DISCUSSION To our knowledge this is the first analysis of functional outcomes using validated questionnaires and adverse event reporting after redo HIFU treatment. We noted deterioration in urinary function with an increased risk of any urinary leakage, although we found no statistically significant difference in pad use or sexual function after redo HIFU. Blana et al previously reported similar morbidity associated with redo HIFU in 49 patients with
Patients with redo HIFU. A, urinary function. B, erectile function and PDE-5 inhibitor use.
1768
MORBIDITY ASSOCIATED WITH HIGH INTENSITY FOCUSED ULTRASOUND
13 months of followup using the AblathermÒ HIFU device.5 Stress urinary incontinence increased from 8.2% of cases after the first HIFU to 16.3% after redo HIFU (p ¼ 0.02). In our study higher incontinence after redo HIFU compared to primary HIFU was probably due to the application of more HIFU energy to the prostate apex to eradicate residual tumor. In our study almost half of the patients reported erectile function not sufficient for intercourse before HIFU, which did not change significantly after first or redo HIFU (table 3). Blana et al found 26.5%, 65.3% and 81.6% impotence rates at baseline, and after first and second HIFU, respectively, but without formal assessment using validated questionnaires.5 We previously reported a statistically significant decrease in IIEF-15 score at 3 months compared with baseline (p ¼ 0.01) but no significant difference at other times compared with baseline.6 The need for additional procedures after redo HIFU such as cystoscopy, urethral dilatation and BNI was not increased compared to after single HIFU. This may be surprising since gland perfusion and the specific heat capacity of the tissue are altered after 1 HIFU.10,11 One could speculate that additional HIFU procedures might induce an additional risk of fibrosis and consequent cystoscopic interventions. Attrition analysis showed that patients with missing information on urinary continence probably had more aggressive disease because they had higher pre-HIFU PSA, PSA nadir and Gleason score than patients who responded. This may be interpreted as disappointment and lack of interest in participating in the study because of disease recurrence, although urinary continence may have been satisfactory. Although the main focus of this series was to assess outcomes after redo HIFU, we also report functional outcomes in patients after a single HIFU treatment. Encouragingly, only 9.9% of men used pads at a median followup of 45 months. In a prospective quality of life study using similar validated questionnaires Berge et al found that 25% of patients used pads 3 years after robot-assisted radical prostatectomy.12 The rather high 44.8% treatment failure rate of redo HIFU probably reflects more aggressive disease in the redo group (table 1). This is supported by negative TRUS biopsies in almost half of the patients who underwent biopsy after redo
HIFU, which may imply extraprostatic disease at the time of treatment. Reasons for the slight decrease in hospital time in favor of redo HIFU could be patient familiarity with the procedure and shorter treatment time due to reduced prostate volume after the first treatment. The reason for shorter catheter time was probably less passage of debris after redo HIFU due to less prostate volume and, therefore, less treatment volume. The reasons for fewer UTIs after redo HIFU than after primary HIFU despite the same amount of postoperative dilatation and BNI could be shorter operative time and shorter catheter time in the redo group. Our study has a number of limitations. There is incomplete reporting of outcomes, especially sexual function variables. We cannot be sure that the same result would have arisen had all patients been tested. In the tables and text we tried to document the numerator and denominator so that the sample on which the variable was evaluated would be explicit. An explanation of why so few patients responded to the sexual function questionnaires at baseline may be that 30.6% in the nonresponding group were treated with cytoreductive medication, which deteriorates erectile function and may hamper participation in such a survey. In other words we may have underestimated the degree of sexual dysfunction. Also, while we report disease control outcomes to place functional outcomes in context, there is no widely accepted, validated surrogate indicator of success. We report biochemical and salvage therapies but the series is limited by the median followup duration so that definitive conclusions cannot be made about disease control in our registry series.
CONCLUSIONS Redo HIFU is associated with an increase in urinary leakage but pad use, sexual dysfunction and other adverse events do not appear to increase. These results suggest that repeat HIFU is a viable, acceptable option for residual disease after primary whole gland treatment. However, meticulous patient selection is of paramount importance when selecting men for redo HIFU. Patients should be appropriately counseled on the additional side effect risk of repeat treatments early in the treatment pathway.
REFERENCES 1. Blana A, Walter B, Rogenhofer S et al: Highintensity focused ultrasound for the treatment of localized prostate cancer: 5-year experience. Urology; 2004; 63: 297.
2. Gelet A, Chapelon JY, Bouvier R et al: Transrectal high intensity focused ultrasound for the treatment of localized prostate cancer: factors influencing the outcome. Eur Urol 2001; 40: 124.
3. Heidenreich A, Bastian PJ, Bellmunt J et al: Guidelines on Prostate Cancer. Arnhem, The Netherlands: European Association of Urology 2013.
MORBIDITY ASSOCIATED WITH HIGH INTENSITY FOCUSED ULTRASOUND
4. Crouzet S, Chapelon JY, Rouviere O et al: Whole gland ablation of localized prostate cancer with high-intensity focused ultrasound: oncologic outcomes and morbidity in 1002 patients. Eur Urol 2013; 65: 907. 5. Blana A, Rogenhofer S, Ganzer R et al: Morbidity associated with repeated transrectal highintensity focused ultrasound treatment of localized prostate cancer. World J Urol 2006; 24: 585. 6. Ahmed HU, Zacharakis E, Dudderidge T et al: High-intensity-focused ultrasound in the treatment of primary prostate cancer: the first UK series. Br J Cancer 2009; 101: 19.
1769
7. Rosen RC, Riley A, Wagner G et al: The international index of erectile function (IIEF): a multidimensional scale for assessment of erectile dysfunction. Urology 1997; 49: 822.
10. Chen L, ter Haar HG and Hill CR: Influence of ablated tissue on the formation of high-intensity focused ultrasound lesions. Ultrasound Med Biol 1997; 6: 921.
8. Litwin MS, Hays RD, Fink A et al: The UCLA Prostate Cancer Index: development, reliability, and validity of a health-related quality of life measure. Medical Care 1998; 36: 1002.
11. Curiel L, Chavrier F, Gignoux B et al: Experimental evaluation of lesion prediction modelling in the presence of cavitation bubbles: intended for high-intensity focused ultrasound prostate treatment. Med Biol Eng Comput 2004; 42: 44.
9. Burnett AL, Aus G, Canby-Hagino ED et al: American Urological Association Prostate Cancer Guideline Update Panel Erectile function outcome reporting after clinically localized prostate cancer treatment. J Urol 2007; 178: 597.
12. Berge V, Berg RE, Hoff JR et al: A prospective study of transition from laparoscopic to robotassisted radical prostatectomy: quality of life outcomes after 36-month follow-up. Urology 2013; 81: 781.
Abbreviations and Acronyms ADT ¼ androgen deprivation treatment BNI ¼ bladder neck incision GEE ¼ generalized estimating equation HIFU ¼ high intensity focused ultrasound IIEF-15 ¼ International Index of Erectile Function-15 MRI ¼ magnetic resonance imaging PDE-5 ¼ phosphodiesterase-5 PSA ¼ prostate specific antigen RT ¼ radiation therapy TRUS ¼ transrectal ultrasound UTI ¼ urinary tract infection Accepted for publication December 16, 2013. Study received institutional review board approval. * Correspondence: Oslo University Hospital, Oslo, Norway (e-mail: [email protected]). † Financial interest and/or other relationship with SonaCare Medical. ‡ Financial interest and/or other relationship with US HIFU and SonaCare Medical.
Purpose: High intensity focused ultrasound may have a role as an alternative to standard radical therapies for localized prostate cancer. An attribute of high intensity focused ultrasound is that it can be repeated. We determined morbidity after primary and redo high intensity focused ultrasound. Materials and Methods: We performed an academic lead analysis of United Kingdom registry data on high intensity focused ultrasound treatments at 3 centers using patient reported continence and sexual function outcomes. Validated questionnaires were completed before and after each ultrasound treatment. Results: A total of 359 patients received 1 whole gland high intensity focused ultrasound treatment for localized prostate cancer from October 2004 to June 2012, of whom 130 (36.2%) received re-treatment. Median followup was 27 months (range 3 to 81) after re-treatment. When analyzing adverse events, 10.8% of patients experienced urinary tract infection after the first treatment compared to 3.9% after re-treatment (p ¼ 0.009). Urethral dilatation was required in 13.8% and 14.0% of patients after first and redo ultrasound treatments (p ¼ 0.7), and bladder neck incision was required in 9.2% and 11.6%, respectively (p ¼ 0.2). Before and after re-treatment 73.3% and 55.1% of patients had no leak, and 2.7% and 9.0% used daily pads (p <0.001 and p ¼ 0.07, respectively). Analysis of erectile function showed that 56.2% and 56.0% of patients were potent before and after re-treatment, respectively (p ¼ 0.9). Conclusions: Redo high intensity focused ultrasound is associated with an increase in urinary side effects but sexual side effects do not appear to be significantly increased. The number of adverse events seems to be equivalent after first and redo treatments. Meticulous patient selection is of paramount importance when selecting men for redo high intensity focused ultrasound. Key Words: prostate, prostatic neoplasms, high-intensity focused ultrasound ablation, retreatment, adverse effects
THE optimal management strategy for newly diagnosed, clinically localized prostate cancer remains controversial and includes radical prostatectomy, radical radiotherapy
1764
j
www.jurology.com
and active surveillance. As an alternative, minimally invasive therapies such as cryotherapy and HIFU have been proposed as iso-effective cancer control with fewer complications and
0022-5347/14/1916-1764/0 THE JOURNAL OF UROLOGY® © 2014 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH, INC.
http://dx.doi.org/10.1016/j.juro.2013.12.036 Vol. 191, 1764-1769, June 2014 Printed in U.S.A.
MORBIDITY ASSOCIATED WITH HIGH INTENSITY FOCUSED ULTRASOUND
side effects.1,2 Although HIFU still is considered experimental according to European Association of Urology guidelines,3 several studies demonstrated oncologic efficacy comparable to that of radiation therapy with a biochemical failure-free survival rate of between 57% and 76% at 7 to 8-year followup.2,4 The rate of erectile impotence after whole gland HIFU is greater than 50%2,4 and the rate of stress urinary incontinence varies between 7% and 15%.4,5 An advantage of HIFU is that treatment can be repeated (redo HIFU) if the first treatment is not successful. We determine whether redo HIFU results in significant additional morbidity compared to primary HIFU.
PATIENTS AND METHODS Since 2004, an academic led, independent, national United Kingdom HIFU registry has been maintained under the auspices of University College London. In this study we included all patients treated with HIFU using a SonablateÒ 500 for localized prostate cancer at 3 United Kingdom institutions (Princess Grace Hospital, Basingstoke and North Hampshire Hospital, and University College London Hospital National Health Service Foundation Trust) from October 8, 2004 to June 26, 2012. We used standard power settings in the range of 20 to 45 W/cm2 with a duty cycle of 3 seconds on and 3 seconds off or 3 seconds on and 6 seconds off as determined by the operator. Treatment was given on an individual basis with combinations of HIFU blocks derived from the 4 and 3 cm focal lengths to cover the area of recurrence. Men who were unable or unwilling to undergo surgery, radiotherapy or active surveillance were offered HIFU. Contraindications to HIFU were previously reported.6 Men received a single HIFU session or redo HIFU after failed first HIFU based on biochemical, histological or MRI evidence of localized residual disease. To achieve the size restrictions necessary for HIFU treatment cytoreduction with bicalutamide (50 mg once daily) and a 5a-reductase inhibitor (dutasteride or finasteride) for 3 months was done before HIFU. These treatments were stopped on the day of the first HIFU treatment. At the outset all men were fitted with a urethral catheter for 7 to 10 days. After consultation in international user group meetings it was agreed that suprapubic catheters might result in earlier voiding, fewer urinary infections and a lower stricture rate. Since most patients were tertiary referrals and lived some distance away, all men were taught clean intermittent self-catheterization so that they could self-manage decreased flow due to debris passage and short-term prostatic inflammation. Followup after treatment mirrored the regimen used after standard radical therapies, including serum PSA measurement at 6 weeks and then every 3 months for the first year and every 6 months in subsequent followup years. Patients who did not achieve a PSA nadir of less than 0.5 ng/ml and those with PSA less than 0.5 ng/ml with 2 consecutive PSA increases were advised to undergo transrectal prostate biopsy. Of the 359 patients
1765
175 (48.7%) underwent 1 post-HIFU TRUS biopsy and in 108 (30.1%) the TRUS biopsy was positive. Post-HIFU adverse events included need for cystoscopy, BNI, urethral dilatation, UTI and epididymitis. These events were recorded from a review of clinical records on a continual basis by dedicated data management personnel. All patients at the 3 study centers were offered pretreatment and posttreatment validated questionnaires.7 Incontinence data were collected from patient reported outcomes on leakage and pad use using the UCLA-EPIC urinary function question subset.8 From the IIEF-15 data we selected question 2 (“When you had erections with sexual stimulation, how often were your erections hard enough for penetration?”) for analysis, which is accepted as a good indicator of erectile function.9 We defined potency as scoring 2 or above on a scale of 0dno sexual activity, 1dalmost never/never, 2da few times (much less than half the time), 3dsometimes (about half the time), 4dmost times (much more than half the time) and 5dalmost always/always. 1) We compared perioperative outcomes and adverse events between single and redo HIFU. 2) We analyzed the change in urinary continence status from no leak to leak and any pad use. 3) We also evaluated sexual function before and after a single HIFU treatment and subsequently after redo HIFU as well as the differential rates of reported PDE-5 inhibitor use. For completeness so that these outcomes could be placed in context we also report disease control outcomes in the group as a whole. The criteria for HIFU failure was PSA nadir plus 2.0 ng/ml (Phoenix criteria) and/or additional salvage therapy. The difference between single and redo HIFU cases was tested by the 2 independent samples t-test or the Pearson chi-square test, as appropriate. A GEE model with a logit link function was used for a binary dependent variable. Patient identity as a subject variable, time points as a within subject variable and an unstructured correlation structure were selected for the GEE model. The binary dependent variables tested were urinary leakage, pad use, intercourse and PDE-5 inhibitor use. Significance was considered at 0.05 and 2-sided p values are shown. Analysis was done with SPSSÒ, version 17.
RESULTS Patient Characteristics From October 2004 to June 2012 a total of 359 patients were treated with at least 1 whole gland HIFU session for localized prostate cancer. Of the 359 men 96 (26.7%) were pretreated with 3 months of cytoreduction, 130 (36.2%) underwent redo HIFU, 19 (5.3%) underwent 2 redo sessions and 1 (0.3%) underwent 3. The data presented on morbidity and oncologic outcomes were recorded after the first redo HIFU. Preoperative characteristics were similar between men with only 1 HIFU and those with redo HIFU except Gleason score, which was higher in those with redo HIFU (p ¼ 0.004, table 1). Median followup was 45 months (range 3 to 93) in the single
MORBIDITY ASSOCIATED WITH HIGH INTENSITY FOCUSED ULTRASOUND
1766
Table 1. Patient characteristics
No. pts Mean age (range) Mean ml prostate vol (range) Mean SD (ng/ml): Pre-HIFU After first HIFU No. pre-HIFU T stage (%):* T1 T2 T3a No. pre-HIFU Gleason score (%): 5 6 7 (3 þ 4) 7 (4 þ 3) 8 9
1 HIFU
More Than 1 HIFU
p Value
229 65.9 (46.7e87.4) 32.3 (8.4e80.0)
130 64.6 (43.6e80.5) 34.7 (7.0e73.0)
e 0.13 0.08
7.9 6.4 0.8 1.3
8.3 5.0 1.0 1.1
0.54 0.15 0.2
52 130 46
(22.7) (56.8) (20.1)
14 45 23
(17.1) (54.9) (28.0)
3 108 81 20 7 2
(1.4) (49.3) (35.5) (8.8) (3.2) (0.9)
0 45 71 10 3 3
(34.6) (54.6) (7.7) (2.3) (2.3)
0.004
HIFU. There were fewer episodes of UTI after the second HIFU but the rates of post-HIFU cystoscopy, dilatation and BNI were similar for first and redo HIFU (table 2). Patient Questionnaire Response Single HIFU. Of the 359 patients 242 (67.4%) completed UCLA-EPIC questionnaire data on urinary function at baseline and 235 (65.5%) completed the data at last followup after a single HIFU treatment (table 3). Likewise, 124 (34.5%) and 171 patients (47.6%) completed the IIEF-15 questionnaire on sexual function at baseline and last followup, respectively (table 3). Information was available on PDE-5 inhibitor use in 207 (57.7%) and 171 patients (47.6%) at baseline and last followup, respectively (table 3). Redo HIFU. Of 130 patients treated with redo HIFU
* Based on prostate/pelvis multiparametric MRI with locoregional metastasis excluded by staging pelvic MRI in all men and distant metastasis excluded by radioisotope bone scan in all men with Gleason score 7 or greater, PSA 10 ng/ml or greater and/or stage T3a or greater.
HIFU group and 27 months (range 3 to 81) after redo HIFU. Disease Control Outcomes There were no prostate cancer related deaths. In the group with only single HIFU without redo HIFU 60 patients (26.4%) had treatment failure and underwent salvage treatment, including ADT plus salvage RT in 16, ADT plus radical prostatectomy in 2, ADT plus chemotherapy in 1, only salvage RT in 3, only salvage cryotherapy in 1 and only ADT in 13. The remaining patients were on active surveillance. After redo HIFU treatment failed in 56 men (44.8%), of whom 40 underwent TRUS biopsy, which was positive in 22. Salvage therapy included ADT in 20 patients (combined with RT in 2), salvage RT in 17, salvage radical prostatectomy in 1 and second redo HIFU in 19. Perioperative Outcomes and Adverse Events After redo HIFU time, hospital time and catheter time were significantly shorter than after the first
the UCLA-EPIC questionnaire data on urinary function were completed before the first HIFU by 82 patients (66.6%), before redo HIFU by 75 (58.1%) and after redo HIFU by 78 (60.5%) at last followup (table 3). For sexual function IIEF-15 questionnaire data were completed by 40 patients (31.0%) before the first HIFU, by 50 (38.8%) before redo HIFU and by 75 (57.7%) after redo HIFU (table 3). Information was available on PDE-5 inhibitor use for 75 patients (58.1%) at baseline, 74 (57.4%) before redo HIFU and 46 (35.7%) after redo HIFU (table 3). Attrition analysis showed that patients with missing information on urinary continence had higher pre-HIFU PSA (p ¼ 0.02), higher PSA nadir (p ¼ 0.02) and higher Gleason score in the pre-HIFU biopsy (p ¼ 0.03) than who responded (data not shown). Attrition analysis revealed no statistically significant difference in patient characteristics between those who did and did not respond to the IIEF-15 questionnaire and PDE-5 inhibitor use at baseline (data not shown). However, of the 235 patients who did not respond to questions on sexual function 72 (30.6%) received cytoreductive medication before HIFU compared to 24 (19.4%) who responded to the same questions before HIFU (p ¼ 0.02).
Table 2. Treatment characteristics and complications in patients after first and redo HIFU
No. pts* No. cytoreduction (%) No. intraprocedural complications (%) Mean mins treatment time (range) Mean days hospital time (range) Mean days catheter time (range) No. post-HIFU cystoscopy (%) No. UTI (%) No. epididymitis (%) No. BNI with/without resection (%) No. dilatation (%) * No patient had fistula.
HIFU 1
Redo HIFU
p Value
130 34 (26.2) 5 (3.8) 188.6 (90e340) 1.7 (1e7) 17.6 (5e75) 21 (16.2) 14 (10.8) 2 (1.5) 12 (9.2) 18 (13.8)
129 3 (2.3) 7 (5.4) 117.4 (40e285) 1.4 (1e4) 12.3 (1e49) 22 (17.1) 5 (3.9) 2 (1.6) 15 (11.6) 18 (14.0)
<0.001 0.5 <0.001 0.002 0.001 0.9 0.009 0.9 0.2 0.7
MORBIDITY ASSOCIATED WITH HIGH INTENSITY FOCUSED ULTRASOUND
1767
Table 3. Urinary and sexual function, and PDE-5 inhibitor use in 359 patients after first HIFU and in 129 after redo HIFU No. HIFU 1 (%)
Drip free: Yes No Pad free: Yes No Question 2 (did you have erectile function hard enough for penetration over last 4 wks?:* No Yes PDE-5 inhibitors: No Yes
No. Redo HIFU (%)
Pre-HIFU
Post-HIFU
p Value (GEE)
Pre-HIFU 1
Pre-HIFU 2
Post-HIFU 2
p Value (GEE)
208 (86.0) 34 (14.0)
131 (55.7) 104 (44.3)
<0.001
73 (89.2) 9 (10.8)
55 (73.3) 20 (26.7)
43 (55.1) 35 (44.9)
<0.001
238 (98.3) 4 (1.7)
209 (90.1) 23 (9.9)
<0.001
81 (98.8) 1 (1.2)
73 (97.3) 2 (2.7)
71 (91.0) 7 (9.0)
0.07
54 (43.5) 70 (56.5)
86 (50.3) 85 (49.7)
0.2
16 (40.0) 24 (60.0)
22 (44.0) 28 (56.2)
33 (44.0) 42 (56.0)
0.9
173 (83.6) 34 (16.4)
65 (38.0) 106 (62.0)
<0.001
65 (87.8) 9 (12.2)
37 (49.3) 38 (50.7)
23 (50.0) 23 (50.0)
<0.001†
* Less than 1 vs 2 or more. † Pre-HIFU 2 vs post-HIFU 2 p ¼ 0.5.
Function Urinary. There was a statistically significant increase in the rate of leak urinary incontinence and pad use after 1 whole gland HIFU session (table 3). GEE analysis showed that before and after redo HIFU 73.3% and 55.1% of patients, respectively, had no urinary leak (p <0.001). Before and after redo HIFU 2.7% and 9.0% of patients, respectively, used daily pads (p ¼ 0.07, table 3 and part A of figure). Sexual. We found no statistically significant differ-
ence in erectile function sufficient for intercourse before and after the first whole gland HIFU across all patients (table 3). Moreover, redo HIFU did not result in any additional significant deterioration in sexual function (table 3 and part B of figure). There was a statistically significant increase in PDE-5 inhibitor use after first whole gland HIFU (p <0.001). However, after redo HIFU there was no statistically significant difference in PDE-5 inhibitor use (p ¼ 0.5, table 3 and part B of figure). Multivariate logistic regression showed that age (p <0.001) and redo HIFU (p ¼ 0.05) were associated
with decreased erectile function at last followup. There was no statistically significant association between erectile function and covariates such as preHIFU cytoreduction (p ¼ 0.2), pre-HIFU PDE-5 inhibitor use (p ¼ 0.4), post-HIFU PDE-5 inhibitor use (p ¼ 0.2), pad use (p ¼ 0.2), indwelling catheter length (p ¼ 0.8), post-HIFU cystoscopy (p ¼ 0.3), urethral dilatation (p ¼ 0.3) or BNI (p ¼ 0.4). PostHIFU ADT was associated with decreased erectile function at last followup on univariate logistic regression but not on multivariate logistic regression (p ¼ 0.01 and 0.1, respectively).
DISCUSSION To our knowledge this is the first analysis of functional outcomes using validated questionnaires and adverse event reporting after redo HIFU treatment. We noted deterioration in urinary function with an increased risk of any urinary leakage, although we found no statistically significant difference in pad use or sexual function after redo HIFU. Blana et al previously reported similar morbidity associated with redo HIFU in 49 patients with
Patients with redo HIFU. A, urinary function. B, erectile function and PDE-5 inhibitor use.
1768
MORBIDITY ASSOCIATED WITH HIGH INTENSITY FOCUSED ULTRASOUND
13 months of followup using the AblathermÒ HIFU device.5 Stress urinary incontinence increased from 8.2% of cases after the first HIFU to 16.3% after redo HIFU (p ¼ 0.02). In our study higher incontinence after redo HIFU compared to primary HIFU was probably due to the application of more HIFU energy to the prostate apex to eradicate residual tumor. In our study almost half of the patients reported erectile function not sufficient for intercourse before HIFU, which did not change significantly after first or redo HIFU (table 3). Blana et al found 26.5%, 65.3% and 81.6% impotence rates at baseline, and after first and second HIFU, respectively, but without formal assessment using validated questionnaires.5 We previously reported a statistically significant decrease in IIEF-15 score at 3 months compared with baseline (p ¼ 0.01) but no significant difference at other times compared with baseline.6 The need for additional procedures after redo HIFU such as cystoscopy, urethral dilatation and BNI was not increased compared to after single HIFU. This may be surprising since gland perfusion and the specific heat capacity of the tissue are altered after 1 HIFU.10,11 One could speculate that additional HIFU procedures might induce an additional risk of fibrosis and consequent cystoscopic interventions. Attrition analysis showed that patients with missing information on urinary continence probably had more aggressive disease because they had higher pre-HIFU PSA, PSA nadir and Gleason score than patients who responded. This may be interpreted as disappointment and lack of interest in participating in the study because of disease recurrence, although urinary continence may have been satisfactory. Although the main focus of this series was to assess outcomes after redo HIFU, we also report functional outcomes in patients after a single HIFU treatment. Encouragingly, only 9.9% of men used pads at a median followup of 45 months. In a prospective quality of life study using similar validated questionnaires Berge et al found that 25% of patients used pads 3 years after robot-assisted radical prostatectomy.12 The rather high 44.8% treatment failure rate of redo HIFU probably reflects more aggressive disease in the redo group (table 1). This is supported by negative TRUS biopsies in almost half of the patients who underwent biopsy after redo
HIFU, which may imply extraprostatic disease at the time of treatment. Reasons for the slight decrease in hospital time in favor of redo HIFU could be patient familiarity with the procedure and shorter treatment time due to reduced prostate volume after the first treatment. The reason for shorter catheter time was probably less passage of debris after redo HIFU due to less prostate volume and, therefore, less treatment volume. The reasons for fewer UTIs after redo HIFU than after primary HIFU despite the same amount of postoperative dilatation and BNI could be shorter operative time and shorter catheter time in the redo group. Our study has a number of limitations. There is incomplete reporting of outcomes, especially sexual function variables. We cannot be sure that the same result would have arisen had all patients been tested. In the tables and text we tried to document the numerator and denominator so that the sample on which the variable was evaluated would be explicit. An explanation of why so few patients responded to the sexual function questionnaires at baseline may be that 30.6% in the nonresponding group were treated with cytoreductive medication, which deteriorates erectile function and may hamper participation in such a survey. In other words we may have underestimated the degree of sexual dysfunction. Also, while we report disease control outcomes to place functional outcomes in context, there is no widely accepted, validated surrogate indicator of success. We report biochemical and salvage therapies but the series is limited by the median followup duration so that definitive conclusions cannot be made about disease control in our registry series.
CONCLUSIONS Redo HIFU is associated with an increase in urinary leakage but pad use, sexual dysfunction and other adverse events do not appear to increase. These results suggest that repeat HIFU is a viable, acceptable option for residual disease after primary whole gland treatment. However, meticulous patient selection is of paramount importance when selecting men for redo HIFU. Patients should be appropriately counseled on the additional side effect risk of repeat treatments early in the treatment pathway.
REFERENCES 1. Blana A, Walter B, Rogenhofer S et al: Highintensity focused ultrasound for the treatment of localized prostate cancer: 5-year experience. Urology; 2004; 63: 297.
2. Gelet A, Chapelon JY, Bouvier R et al: Transrectal high intensity focused ultrasound for the treatment of localized prostate cancer: factors influencing the outcome. Eur Urol 2001; 40: 124.
3. Heidenreich A, Bastian PJ, Bellmunt J et al: Guidelines on Prostate Cancer. Arnhem, The Netherlands: European Association of Urology 2013.
MORBIDITY ASSOCIATED WITH HIGH INTENSITY FOCUSED ULTRASOUND
4. Crouzet S, Chapelon JY, Rouviere O et al: Whole gland ablation of localized prostate cancer with high-intensity focused ultrasound: oncologic outcomes and morbidity in 1002 patients. Eur Urol 2013; 65: 907. 5. Blana A, Rogenhofer S, Ganzer R et al: Morbidity associated with repeated transrectal highintensity focused ultrasound treatment of localized prostate cancer. World J Urol 2006; 24: 585. 6. Ahmed HU, Zacharakis E, Dudderidge T et al: High-intensity-focused ultrasound in the treatment of primary prostate cancer: the first UK series. Br J Cancer 2009; 101: 19.
1769
7. Rosen RC, Riley A, Wagner G et al: The international index of erectile function (IIEF): a multidimensional scale for assessment of erectile dysfunction. Urology 1997; 49: 822.
10. Chen L, ter Haar HG and Hill CR: Influence of ablated tissue on the formation of high-intensity focused ultrasound lesions. Ultrasound Med Biol 1997; 6: 921.
8. Litwin MS, Hays RD, Fink A et al: The UCLA Prostate Cancer Index: development, reliability, and validity of a health-related quality of life measure. Medical Care 1998; 36: 1002.
11. Curiel L, Chavrier F, Gignoux B et al: Experimental evaluation of lesion prediction modelling in the presence of cavitation bubbles: intended for high-intensity focused ultrasound prostate treatment. Med Biol Eng Comput 2004; 42: 44.
9. Burnett AL, Aus G, Canby-Hagino ED et al: American Urological Association Prostate Cancer Guideline Update Panel Erectile function outcome reporting after clinically localized prostate cancer treatment. J Urol 2007; 178: 597.
12. Berge V, Berg RE, Hoff JR et al: A prospective study of transition from laparoscopic to robotassisted radical prostatectomy: quality of life outcomes after 36-month follow-up. Urology 2013; 81: 781.