Oncology

Salvage High-intensity FocusedUltrasound for Patients With RecurrentProstate Cancer After Brachytherapy

Vladimir Yutkin, Hashim U. Ahmed, Ian Donaldson, Neil McCartan, Khurram Siddiqui,Mark Emberton, and Joseph L. Chin

OBJECTIVE To report our experience with salvage high-intensity focused ultrasound (HIFU) in patients with

Financial Disclosure: The authoFunding Support: The UCLHgrant, which pays for a data manadata is kept updated in a timely faHIFU (SonaCare Medical, 801centers, the funder had no accsubmission.From the Division of Urolog

Ontario, London, Ontario, CanLondon Hospitals NHS Foundatiof Surgery and Interventional SciAddress correspondence to: Jos

of Surgery, Schulich School of MLondon, ON, Canada N6A 5NSubmitted: March 24, 2014,

ª 2014 Elsevier Inc.All Rights Reserved

local failure after brachytherapy for prostate cancer.

PATIENTS ANDMETHODS

Whole-gland HIFU was administered to prospectively recruited patients with local histologicfailure after brachytherapy at 2 institutions in the United Kingdom and Canada. Functional and

oncologic outcomes of the procedure were analyzed.

RESULTS Nineteen patients underwent the treatment, 12 with Gleason sum 7 and 5 with Gleason sum 8 at

recurrence. Thirteen men had grade-3a or -3b complications by the Clavien system; there were nograde-4 or -5 complications. The most common postoperative complication was dysuria, whichwas self-limited. Three men developed rectourethral fistulae. The overall continence rate was68.4%. At a mean follow-up of 51.6 months, all men were alive. The overall biochemicalrecurrence-free survival rate was 66.7% and 73.3% using the “nadir prostate-specific antigenlevel” þ1.3 ng/mL and þ2 mg/ml criteria, respectively. This study is limited by the small cohortsize, relatively short follow-up period, and heterogeneity of the patient population.

CONCLUSION In this, the largest prospective series to date, we demonstrate that salvage HIFU for locally

recurrent prostate cancer after failed primary brachytherapy has encouraging disease controlresults, albeit with a relatively high complication rate. UROLOGY 84: 1157e1162, 2014.� 2014 Elsevier Inc.

rachytherapy is a well-established primary treat-ment option for clinically localized prostate

Bcancer with the long-term biochemical

recurrence-free survival rate reported to be 77% at12 years.1 However, in case of local failure, these patientsrepresent a challenge to the physician as the complicationrate after the available salvage options is high,2 renderingthe choice of treatment not trivial. Recently, we pub-lished our findings with whole-gland salvage high-intensity focused ultrasound (HIFU) after the failure ofexternal beam radiation therapy (EBRT).3 Due to thepresence of intraprostatic seeds, there has been a doubt

rs declare that they have no relevant financial interests.HIFU registry was funded by an unrestricted researchger to upload patients treated with HIFU and follow-upshion. The UWO HIFU study was supported by USE. Morehead St., Charlotte, NC 28202). In bothess to the data or analyses or publications before

y, Department of Surgery, University of Westernada; the Department of Urology, University Collegeon Trust, London, United Kingdom; and the Divisionence, University College London, United Kingdomeph L. Chin, M.D., Division of Urology, Departmentedicine & Dentistry, University of Western Ontario,3. E-mail: joseph.chin@lhsc.on.caaccepted (with revisions): June 18, 2014

h

about the feasibility and utility of salvage HIFU afterbrachytherapy, because of the concerns regardingcompromised energy delivery to the prostate, reflectivitytoward the rectum, and impaired prostate gland visuali-zation. In this report, we relate our experience withsalvage HIFU after the failure of brachytherapy for clin-ically localized prostate cancer.

PATIENTS AND METHODS

Patients with histologic failure after radiation treatment wereenrolled prospectively and treated with the Sonablate 500 de-vice (SonaCare Medical, Charlotte, NC). Institution ethicsreview board approval was obtained at each location. BetweenApril 2006 and November 2011, we recruited 19 patients forsalvage HIFU after brachytherapy (12 at the University ofWestern Ontario, London, Canada and 7 at University CollegeLondon, London, United Kingdom [UK]).

Patient SelectionMen aged between 40 through 80 years with biochemical failure(nadir prostate-specific antigen [PSA] þ2 ng/mL) after brachy-therapy underwent prostate biopsy for histologic confirmation ofrecurrence. Standard biopsy template was not required in thesebiopsies. Staging was performed using radionuclide bone scanand computerized tomography scan, and men with suspected

ttp://dx.doi.org/10.1016/j.urology.2014.06.0540090-4295/14

1157

metastatic or locally advanced (T3-T4) disease were excluded.Transrectal ultrasonography (TRUS) was performed to ensureprostate volume was under 40 cm3, and men with larger glandswere excluded. All patients were asked to complete the Inter-national Prostate Symptom Score (IPSS) and InternationalIndex of Erectile Function 5 point (IIEF-5) questionnaires bothbefore and at predefined intervals after the HIFU treatment. Atthe University of Western Ontario, patients also were asked toanswer a 36-item health survey questionnaire4 (RAND SF-36).

HIFU TreatmentThe procedure was performed by HIFU surgeons experiencedboth in primary and in salvage procedures.3,5 The Sonablate 500device was used, targeting the whole gland, as reported previ-ously.6 Preoperatively, a suprapubic catheter (SPC) was placedin all patients and retained for 30 days.

Patient CharacteristicsBetween April 2006 and November 2011, twelve patients fromCanada and 7 from UK underwent whole-gland salvage HIFUafter brachytherapy for prostate cancer. At the time of brachy-therapy failure, 18 patients had low- or intermediate-risk diseaseaccording to the D’Amico criteria7 and had a permanentimplant with iodine-125 seeds. The remaining individual wasinitially treated with highedose rate iridium brachytherapycombined with EBRT, after the finding of Gleason sum 8prostate carcinoma.

The study did not mandate a standard biopsy template beforethe procedure. The mean number of cores taken for histologicconfirmation of recurrence was 11.6 (SD ¼ 9.8; median ¼ 10.0;95% confidence interval [CI], 5.5-14.5), with a mean number ofpositive cores of 4.1 (SD ¼ 1.8; median ¼ 4.0; 95% CI, 3.2-4.8).On histologic analysis, 12 patients (63%) had Gleason sum 7and 5 (26%) patients had Gleason 8 disease. In one man fromCanada, the grade score was not assigned because of radiationchanges, and in 1 UK patient it was unavailable.

Mean patient age at the time of HIFU was 65.8 years (SD ¼5.8 years; median ¼ 66.0 years; 95% CI, 63.1-68.9 years). TheAmerican Society of Anesthesiologists (ASA) score was avail-able only for Canadian patients and was ASA 2 in 9 patients(75%) and ASA 3 in 3 patients (25%). Mean preoperative IPSSwas 7.9 (SD ¼ 4.3; median ¼ 8.0; 95% CI, 5.8-10.2) and themean IIEF-5 score before the procedure was 11.8 (SD ¼ 8.4;median ¼ 11.0; 95% CI, 6.7-15.3), with only 4 men having ascore above 20. All patients were continent before the proce-dure and had no significant voiding issues.

Follow-upAll men were followed at 3-month intervals for the first year andevery 6 months thereafter. During the first year, patients wererequested to answer the aforementioned questionnaires at eachfollow-up visit. All men were asked to undergo prostate biopsy6 months after the HIFU procedure. Complications and adverseeffects were graded according to the Clavien-Dindo classifica-tion.8 Briefly, grade I is any deviation from normal postoperativecourse without the need for pharmacologic, surgical, or endo-scopic intervention; some therapeutic interventions are allowed.Grade II is any pharmacologic intervention beyond that allowedin grade I. Grade III requires surgical or endoscopic interven-tion, whereas IIIa is without general anesthesia and IIIb is withgeneral anesthesia. Grade IV is life threatening complication

1158

requiring admission to intensive care unit, and grade V is deathof the patient.PSA response was defined as a decrease in PSA level values

after the procedure. Treatment failure was defined if the 6-month biopsy was positive, if the patient was commenced onandrogen deprivation therapy (ADT) after HIFU, and/or ac-cording to the Stuttgart criterion,9 that is, PSA level exceedingnadir þ 1.2 ng/mL. We also analyzed treatment success or failureaccording to the American Society for Therapeutic Radiologyand Oncology (ASTRO) Phoenix criterion10: PSA levelexceeding nadir þ 2 ng/mL.

StatisticsDescriptive statistics are reported for patient demographic andbaseline characteristics. The association between the total num-ber of complications and pre-HIFU Gleason score of 8 comparedwith 7 and between the presence and absence of pre-HIFU ADTwas made using unpaired t tests assuming unequal variance. Theassociation between total complications and other characteristicswas evaluated using the Pearson correlation coefficient. TheKaplan-Meier technique was used to estimate progression-freeinterval according to the Stuttgart and Phoenix criteria for pro-gression. The Fisher exact test was used for comparing patient pre-HIFU Gleason scores and biochemical failure.All tests were 2-sided and P values <.05 were considered

statistically significant. The software used was SAS, version 9.3(SAS Institute, Cary, NC).

RESULTS

Patients’ CharacteristicsThe mean interval between brachytherapy and salvageHIFU was 6.8 years (SD ¼ 2.4 years; median ¼ 6.0 years;95% CI, 4.2-11.4 years). Mean (SD) PSA level beforesalvage treatment was 5.07 ng/mL (SD ¼ 3.25 ng/mL;median, 4.32; 95% CI, 0.85-15.16), with 4 patients(27%) having commenced ADT before HIFU (whichwas discontinued immediately after HIFU). The meannadir PSA after brachytherapy was 0.70 ng/mL (SD ¼0.51 ng/mL; median ¼ 0.71; 95% CI, 0.18-1.7). Wefound a strong positive correlation between PSAdoubling time and the time elapsed between brachy-therapy and salvage HIFU (r ¼ 0.78).

Perioperative CourseProcedure details were available for 16 patients. Theaverage HIFU procedure duration was 132.0 (31.7) mi-nutes (126.0; 90-212). One patient had intraoperativerectal bleeding that necessitated premature termination ofthe procedure. In all other cases, the procedure wasaccomplished uneventfully. Average hospital stay afterHIFU was 1.1 days (SD ¼ 0.3 day; median ¼ 1.0 day;95% CI, 0.8-1.2).

Complications and Functional OutcomesThe total number of the complications according to theClavien System8 is summarized in Table 1. We examinedthe association between the total number of complica-tions and preoperative variables (Table 2) and identifiedonly 2 with positive association. Administration of ADT

UROLOGY 84 (5), 2014

Table 2. Associations between number of total complications and other characteristics

Patient Characteristic Descriptive Statistics P Value Statistical Test

Pre-HIFU Gleason score of 87—mean � SD (n) 1.9 � 1.2 (12) .108 Unpaired t assuming unequal variances8—mean � SD (n) 2.8 � 0.8 (5)

Pre-HIFU ADTNo—mean � SD (n) 1.8 � 1.3 (15) .037 Unpaired t assuming unequal variancesYes—mean � SD (n) 2.8 � 0.5 (4)

Age at HIFU 0.184 .451 Pearson r; test for correlation >0Pre-HIFU PSA 0.071 .780Pre-HIFU IPSS 0.200 .475Pre-HIFU IIEF �0.195 .487Pre-HIFU SF-36 0.310 .327Procedure time 0.098 .717Interval between brachytherapy and HIFU �0.460 .048

ADT, androgen deprivation treatment; CI, confidence interval; HIFU, high-intensity focused ultrasound; IIEF, International Index of ErectileFunction; IPSS, International Prostate Symptom Score; PSA, prostate-specific antigen; SF-36, 36-item health survey questionnaire.

Bolded P values are statistically significant.

Table 1. Summary of complications by the Clavien system

Clavien Grade1

Clavien Grade2

Clavien Grade3a

Clavien Grade3b

Clavien Grades 4 and5

Number of complications 24 0 7 7 0Number of patients withcomplications (%)

15 (79) 0 (0) 7 (37) 6 (32) 0 (0)

before HIFU and the longer interval between brachy-therapy and salvage HIFU were both associated with ahigher complication rate. There was no correlation be-tween the procedure length and number of complications(r ¼ 0.09).

The most frequent postoperative phenomenon waslower urinary tract symptoms. This symptom complex wasobserved in 9 patients (47%), and in most cases, it wasself-limiting. There was no significant association be-tween either lower urinary tract symptoms or voiding is-sues and various preoperative or intraoperative variables.Thirteen patients (68.4%) did not require pads for urinaryincontinence after the procedure. The second commoncomplication was a need for bladder catheterization—7patients (3 due to rectourethral fistula). One patient wassuccessfully weaned off his catheter without long-termsequela. Of the remaining 6 men, 3 were rendered cath-eter free eventually after an additional surgical procedureand 3 were left with a permanent indwelling catheter.There were no events of osteitis pubis after HIFU.

Rectourethral fistula developed in 3 patients (16%) ata mean of 11.7 months (SD ¼ 15.9; median ¼ 3; range,2-30 months) after the procedure; one of these patientshad highedose rate iridium brachytherapy as a primarytreatment. All 3 were initially managed with urinarydiversion by means of a SPC with eventual successfulfistula repair in 2 patients. The third patient was left withan indwelling SPC without repair attempt. We did notfind an association between fistula formation and anyother variables studied.

Two patients (10%) required implantation of an arti-ficial urinary sphincter because of severe stress urinary

UROLOGY 84 (5), 2014

incontinence; one of them had a recto-urethral fistularepair beforehand.

We did not observe a significant change in IPSS andSF-36 scores at 6 months’ follow-up as compared withpreoperative values. However, of the 4 men with theIIEF-5 scores above 20, only 2 preserved their preproce-dural potency at 6 month’s follow-up.

Oncologic OutcomesMean follow-up duration after the procedure was51.6 months (SD ¼ 24.1 months; median ¼ 59.3 months;range, 12.2-84.3 months) and all men (100%) were alive.All but 1 patient (95%) had PSA response 3 months afterthe HIFU, rendering a PSA response rate of 95%; in theremaining patient, there was no change in PSA values.

According to the Stuttgart definition,9 6 (32%) of the19 patients had biochemical recurrence at a mean follow-up of 4.3 years. Four patients were commenced on ADTimmediately after a confirmed rise in PSA level and 2individuals were followed expectantly, as their PSAvalues were below 5 ng/mL on the last follow-up. PSAvalues were unavailable in 1 patient at the last follow-up.Thus, the overall biochemical recurrence-free survivalrate was calculated to be 66.7%. Median time to recur-rence was 74.8 months (Fig. 1). Using the ASTROPhoenix definition,10 only 5 patients failed biochemi-cally, yielding a biochemical recurrence-free survival rateof 73.6% at a mean of 4.3 years (Fig. 1).

Among the 12 Canadian patients, 9 (75%) had a post-HIFU biopsy (negative in all), and 7 of these menremained in biochemical remission at their last follow-upvisit; however, 2 had had biochemical failure. The

1159

Figure 1. Kaplan-Meier analysis of biochemical progression-free survival rate by Stuttgart and Phoenix criteria.

Table 3. Associations between biochemical failure (last PSA level � nadir >1.2) and other characteristics

No (n ¼ 15) Yes (n ¼ 3) Difference (SD) 95% CI About Difference P Value

Pre-HIFU Gleason Score of 8, n (%) 3 (21.4%) 2 (66.7%) .191Pre-HIFU ADT, n (%) 3 (20.0%) 1 (33.3%) >.999Age at HIFU, y, mean � SD 66.1 (6.1) 64.3 (6.5) 1.8 (6.1) �11.9 to 15.4 .698Pre-HIFU PSA level, ng/mL, mean � SD 5.5 (3.3) 2.9 (2.2) 2.6 (3.2) �1.7 to 6.8 .170Pre-HIFU IPSS, mean � SD 8.3 (3.9) 8.5 (7.8) �0.2 (4.3) �59.7 to 59.3 .981Pre-HIFU IIEF, mean � SD 10.6 (8.9) 16.5 (6.4) �5.9 (8.7) �31.7 to 19.9 .386Pre-HIFU SF-36, mean � SD 104.4 (5.2) 105.0 (8.5) �0.6 (5.6) �58.4 to 57.2 .937Procedure time, min, mean � SD 126.9 (22.0) 154.0 (61.2) �27.1 (30.8) �173.3 to 119.2 .525Interval between brachytherapy andHIFU, y, mean � SD

7.0 (2.5) 5.3 (1.3) 1.8 (2.4) �0.7 to 4.3 .128

PSA doubling time, mo 26.8 (14.7) 15.3 (9.0) 11.5 (13.8) �5.9 to 28.8 .155PSA nadir, mean � SD 0.87 (1.06) 0.41 (0.40) 0.46 (1.00) �0.35 to 1.26 .228

Abbreviations as in Table 1.

reasons for not performing the biopsy included 2 menwith recto-urethral fistula and one in which the proce-dure was abandoned because of rectal bleeding; 2 of themwere started on ADT because of biochemical failure.None of the 7 patients from University College Londonunderwent post-HIFU biopsy at 6 months. We did notfind a significant association between biochemical failureand pre-HIFU Gleason score, PSA nadir, ADT, or othervariables (Table 3). There was no mortality from prostatecancer or any other cause during the follow-up period.

COMMENTPatients with locally recurrent prostate cancer after radi-ation treatment represent a challenging population andeven more so those after brachytherapy. In this study, wereport the largest international series of patients to datewho underwent HIFU with the Sonablate 500 transrectaldevice for radiorecurrent prostate cancer primarily treatedwith brachytherapy.

According to published data, most patients undergoinglocal salvage therapy for radiation recurrent prostatecancer are managed with salvage radical prostatectomy.11-13 As expected, most patients reported were primarilytreated with ERBT. The studies that do give an account

1160

of the functional and surgical complications of prostatec-tomy in post-brachytherapy failure report on the incidenceof rectal injury and urinary fistula between 3% and 33%,depending on the primary treatment method and the sur-gical technique,11,12 with much lower figures after ERBT.In the present series, the fistula formation rate was 16%,which is comparable with post-prostatectomy series reportsof rectal injury after brachytherapy. It might be postulatedthat the presence of brachytherapy seeds can alter heatpaths and tissue (especially rectal) thermoabsorption.Alternatively, the prior brachytherapy might have causedmore perirectal tissue ischemic damage than EBRT, thuspredisposing to fistula formation. Patients treated withprostate cryoablation suffer from 1%-2% of fistula forma-tion,14,15 but only a minority (fewer than 12%) hadbrachytherapy as a primary treatment, which is similar tothat reported in patients treated with HIFU after EBRT.16

Conceivably, patients who had bothmodalities (combinedbrachytherapy and EBRT) as a primary treatment had anextremely high incidence of fistula formation,11 andneither of our groups treat these men with salvage HIFUany longer. Recently, there has been considerable interestin focal salvage therapy to target the recurrent lesion aloneto limit the degree of collateral damage; early outcomesseem to demonstrate a lower morbidity rate.17,18

UROLOGY 84 (5), 2014

We noted spontaneous recovery of complete urinarycontinence in 68.4% of our patients. These numberscompare favorably with published incontinence rate inpost-brachytherapy patients after salvage prostatectomy:90%12 and 61% (need for 1 pad or more).11 In the latterstudy, the authors also reported a 30% incidence ofanastomotic stricture. In our cohort, 2 men (10%) hadpersistent stricture at the bladder neck that was recalci-trant to endoscopic treatment.

In our series, the 5-year biochemical progression-freesurvival (BPFS) was 84% (95% CI, 48%-95.7%) whenthe failure was defined as “PSA nadir” þ 1.2 ng/mL and92.9% (95%CI, 59.1%-99%) with the Phoenix criterion.Chade et al19 reported on their experience with salvageprostatectomy with 89 men who had brachytherapy asprimary treatment and found that the 5-year BPFS was48%. Similarly, Bianco et al16 found 55% BPFS aftersalvage prostatectomy in 100 men, 42 of them afterbrachytherapy. To define failure they used the PSAthresholds 0.1 and 0.2 ng/mL, respectively. Using theSonablate device, Uchida et al20 reported on 22 patients(which included 5 post-brachytherapy patients) with a52% 5-year BPFS for HIFU using the Phoenix defini-tion.10 Previously, we reported a 5-year disease-free sur-vival of 47% using the Phoenix criterion for salvagecryotherapy for radiorecurrent prostate cancer afterEBRT.14 In none of these studies did the authors reportthe results of post-brachytherapy patients separately,probably because they constituted only the minority.Obviously, it is not ideal to compare BPFS using differentcriteria and after dissimilar procedures. Nonetheless, itseems that HIFU is able to provide acceptable outcomes.

Our series has several limitations. First, the size of theprospective study cohort is small. This stems from the factthat brachytherapy is a less common treatment ascompared with EBRT21 and possibly secondary to a ten-dency to refer men with post-brachytherapy failure forexpectant management with delayed ADT.22 Second, thefollow-up period is relatively short; thus, using Stuttgartand ASTRO criteria tends to overestimate success.Hence, our disease control rates have to be viewed withcaution. Third, we have combined patients from 2different centers, thus adding heterogeneity to the data,although we believe the external validity actually im-proves the quality of our data. We recognized that thefollow-up and data acquisition differed in minor details,but it is conceivable that there were unrecognized dif-ferences as well, that is, in inclusion and exclusion criteriaand procedures. An additional possible confounding fac-tor is that 4 patients were given ADT before the salvageHIFU, although it was discontinued immediately after theprocedure with a minimum washout time of 4.7 years(range, 4.7-7 years). We accept that it still mightpotentially mask recurrence or persistence of prostatecancer. The method of administration of the quality-of-life questionnaires might also introduce a bias renderingfalsely improved result; we used a face-to-face interview asopposed to self-completed survey.23

UROLOGY 84 (5), 2014

CONCLUSIONWe describe the largest prospective series to date in menwho underwent salvage HIFU treatment for locallyrecurrent prostate cancer after failed primary brachy-therapy. According to our observations, HIFU hasencouraging disease control outcomes in men withrecurrent prostate cancer after failed brachytherapy andcan be a treatment alternative in experienced hands forpatients not eligible for salvage radical prostatectomy.Patients and physicians should be aware of the potentialfor a relatively high complication rate.

Acknowledgments. The authors thank Mr. Larry Stitt forassistance with statistical analysis.

References

1. Potters L, Morgenstern C, Calugaru E, et al. 12-year outcomesfollowing permanent prostate brachytherapy in patients with clini-cally localized prostate cancer. J Urol. 2008;179:S20-S24.

2. Parekh A, Graham PL, Nguyen PL. Cancer control and complica-tions of salvage local therapy after failure of radiotherapy for prostatecancer: a systematic review. Semin Radiat Oncol. 2013;23:222-234.

3. Uddin Ahmed H, Cathcart P, Chalasani V, et al. Whole-glandsalvage high-intensity focused ultrasound therapy for localizedprostate cancer recurrence after external beam radiation therapy.Cancer. 2012;118:3071-3078.

4. Hays RD, Sherbourne CD, Mazel RM. The RAND 36-item healthsurvey 1.0. Health Econ. 1993;2:217-227.

5. Ahmed HU, Zacharakis E, Dudderidge T, et al. High-intensity-focused ultrasound in the treatment of primary prostate cancer: thefirst UK series. Br J Cancer. 2009;101:19-26.

6. Illing RO, Leslie TA, Kennedy JE, et al. Visually directed high-intensity focused ultrasound for organ-confined prostate cancer: aproposed standard for the conduct of therapy. BJU Int. 2006;98:1187-1192.

7. D’Amico AV, Whittington R, Malkowicz SB, et al. Biochemicaloutcome after radical prostatectomy, external beam radiation ther-apy, or interstitial radiation therapy for clinically localized prostatecancer. JAMA. 1998;280:969-974.

8. Dindo D, Demartines N, Clavien PA. Classification of surgicalcomplications: a new proposal with evaluation in a cohort of 6336patients and results of a survey. Ann Surg. 2004;240:205-213.

9. Blana A, Brown SC, Chaussy C, et al. High-intensity focused ul-trasound for prostate cancer: comparative definitions of biochemicalfailure. BJU Int. 2009;104:1058-1062.

10. Roach M 3rd, Hanks G, Thames H Jr, et al. Defining biochemicalfailure following radiotherapy with or without hormonal therapy inmen with clinically localized prostate cancer: recommendations ofthe RTOG-ASTRO Phoenix Consensus Conference. Int J RadiatOncol Biol Phys. 2006;65:965-974.

11. Stephenson AJ, Scardino PT, Bianco FJ Jr, et al. Morbidity andfunctional outcomes of salvage radical prostatectomy for locallyrecurrent prostate cancer after radiation therapy. J Urol. 2004;172:2239-2243.

12. Heidenreich A, Richter S, Thuer D, et al. Prognostic parameters,complications, and oncologic and functional outcome of salvageradical prostatectomy for locally recurrent prostate cancer after 21st-century radiotherapy. Eur Urol. 2010;57:437-443.

13. Chade DC, Eastham J, Graefen M, et al. Cancer control andfunctional outcomes of salvage radical prostatectomy for radiation-recurrent prostate cancer: a systematic review of the literature. EurUrol. 2012;61:961-971.

14. Williams AK, Martinez CH, Lu C, et al. Disease-free survivalfollowing salvage cryotherapy for biopsy-proven radio-recurrentprostate cancer. Eur Urol. 2011;60:405-410.

1161

15. Pisters LL, Rewcastle JC, Donnelly BJ, et al. Salvage prostate cry-oablation: initial results from the cryo on-line data registry. J Urol.2008;180:559-563; discussion 563-4.

16. Bianco FJ Jr, Scardino PT, Stephenson AJ, et al. Long-termoncologic results of salvage radical prostatectomy for locally recur-rent prostate cancer after radiotherapy. Int J Radiat Oncol Biol Phys.2005;62:448-453.

17. Ahmed HU, Cathcart P, McCartan N, et al. Focal salvage therapyfor localized prostate cancer recurrence after external beam radio-therapy: a pilot study. Cancer. 2012;118:4148-4155.

18. de Castro Abreu AL, Bahn D, Leslie S, et al. Salvage focal andsalvage total cryoablation for locally recurrent prostate cancer afterprimary radiation therapy. BJU Int. 2013;112:298-307.

19. Chade DC, Shariat SF, Cronin AM, et al. Salvage radical prosta-tectomy for radiation-recurrent prostate cancer: a multi-institutionalcollaboration. Eur Urol. 2011;60:205-210.

20. Uchida T, Shoji S, Nakano M, et al. High-intensity focused ul-trasound as salvage therapy for patients with recurrent prostatecancer after external beam radiation, brachytherapy or protontherapy. BJU Int. 2011;107:378-382.

21. Cooperberg MR, Broering JM, Carroll PR. Time trends and localvariation in primary treatment of localized prostate cancer. J ClinOncol. 2010;28:1117-1123.

22. Grossfeld GD, Li YP, Lubeck DP, et al. Predictors of secondarycancer treatment in patients receiving local therapy for prostatecancer: data from cancer of the prostate strategic urologic researchendeavor. J Urol. 2002;168:530-535.

23. Hing CB, Smith TO, Hooper L, et al. A review of how to conduct asurgical survey using a questionnaire. Knee. 2011;18:209-213.

EDITORIAL COMMENT

The treatment of patients that have failed radiation therapytreatment for prostate cancer is a challenging clinical situationthat few urologists choose to tackle with curative therapy. Therelatively low cure rates accompanied by the high side effects oftreatment represent a poor trade-off for most clinicians. Notechnique has convincingly shown superiority over others.Nevertheless, some of these patients can be cured whenappropriately selected and carefully treated.

In this article, the authors examine the specific situation ofthe use of high-intensity focused ultrasound (HIFU) after thefailure of radioactive iodine seed implantation. This does pre-sent some unique challenges for HIFU in that the metallic“seeds” have both the potential to reflect sound energy backtoward the rectum and also to block sound transmission beyondthe “seed” into the prostate. This is clearly an issue worthstudying, and this study is the first report of this kind.

Nineteen patients were treated with whole-gland ablationand complications occurred in a high percentage of patients.Sixteen percent of patients experienced a rectourethral fistula,32% experienced incontinence requiring pads, and 16% of thepatients had to be managed with a long-term indwelling cath-eter to drain their bladder.

Still these complications might be acceptable to some pa-tients if they experienced disease control and an extension intheir quality or quantity of life. This is the major problemwith this manuscript and all others of this sort. Although thistechnique attempted to ablate the entire gland, cryotherapy isusually associated with some level of residual prostate-specificantigen detectability. This significantly complicates our abilityto use prostate-specific antigen levels as an accurate measureof disease eradication. Like cryotherapy, HIFU proponentshave used either the American Society for Therapeutic

1162

Radiology and Oncology Phoenix criteria or the more strin-gent Stuttgart to measure treatment success, and therein liesthe major problem with this report. It is extremely importantto note that the Phoenix criteria was strictly to be applied toradiation therapy and no other technology, and moreimportantly, failure by Phoenix criteria was thought to signifylikely death from prostate cancer not failure of treatment.The 2 are distinctly different. When the Phoenix criteria orthe Stuttgart criteria are used to measure treatment success,they can only be compared with other forms of ablativetherapy, not extirpative surgery like salvage prostatectomy.Many of the patients in their study will have most certainlyfailed to achieve complete tumor eradication by the tech-nique but have not failed by Phoenix or Stuttgart criteria andare unlikely to experience death from prostate cancer in thenear term.These authors should be congratulated for conducting what

must have been a very difficult study to perform and with ac-curate reporting of side effects and relatively long-term follow-up. They have laid the ground work for other studies to improveupon the toxicity of this technology and eventually to provideus with meaningful outcomes that can be compared with thoseof salvage prostatectomy.

Michael O. Koch, M.D., Department of Urology, University ofIndiana School of Medicine, Indianapolis, IN

http://dx.doi.org/10.1016/j.urology.2014.06.055UROLOGY 84: 1162, 2014. � 2014 Elsevier Inc.

REPLY

We completely agree that patients with local disease recurrenceafter brachytherapy have very few curative options. In our series,the patients were meticulously selected, and all procedures wereperformed by surgeons with extensive expertise in local ablativetechniques for prostate cancer. Nonetheless, the fistula rate andneed for long-term indwelling urinary catheter after the proce-dure were significant. Noteworthy is the fact that the cases withfistulas were performed early in our experience. Obviously, itshould be emphasized that vigilance and avoidance of being tooaggressive with the treatment boundaries is of primary impor-tance in all cases.The task of defining treatment failure in local ablative ther-

apy is indeed not a simple one. The reliance on biochemicalrecurrence is often confounded by frequent use of hormonalmanipulation in these patients (not a real issue in our series.)We should point out that in addition to using the Stuttgart andPhoenix definitions, we also performed post-HIFU prostate bi-opsy in all patients at the Canadian site. Unfortunately, in spiteof the negative biopsy results in all cases, only 67% of the menremained biochemically disease free, thus further complicatingthe issue and underscoring the importance of longer clinicalfollow-up.

Vladimir Yutkin, M.D., andJoseph L. Chin, M.D., Division of Urology, Department ofSurgery, University of Western Ontario, London, Ontario,Canada

http://dx.doi.org/10.1016/j.urology.2014.06.056UROLOGY 84: 1162, 2014. � 2014 Elsevier Inc.

UROLOGY 84 (5), 2014