High-intensity focused ultrasound for localized prostate cancer: initial experience with a 2-year follow-up

Authors


Benjamin Challacombe, The Urology Centre, Guy’s Hospital, Great Maze Pond, London SE1 9RT, UK.
e-mail: benchallacombe@doctors.org.uk

Abstract

OBJECTIVE

To report on the short-term functional and oncological results, from one institution, of high-intensity focused ultrasound (HIFU) for treating localized prostate cancer.

PATIENTS AND METHODS

Over a 3-year period, 43 patients with localized prostate cancer were scheduled for HIFU in the primary (31) and salvage (12) settings using a second-generation AblathermTM device (EDAP, Lyon, France). Oncological failure was defined by several criteria, including biochemical failure (assessed using both the Phoenix definition of the nadir + 2 ng/mL) and the current Food and Drug Administration (FDA) trial endpoint of a prostate-specific antigen (PSA) level of ≥0.5 ng/mL, or starting salvage therapy, or the presence of cancer on biopsy after treatment.

RESULTS

Three patients had their procedures abandoned due to technical limitations/rectal wall thickness. The mean PSA levels in the primary and salvage groups were 9.2 and 5.1 ng/mL, respectively. The mean HIFU treatment time in the primary and salvage groups was 71.1 and 63.3 min, respectively. Using the Phoenix definition of biochemical failure, HIFU treatment failed in 13 patients in the primary group (46%) and five in the salvage group. Using the FDA trial endpoint, HIFU failed in 21 patients in the primary group (75%) and eight in the salvage group. One man died from metastatic prostate cancer 18 months after salvage HIFU. There were two urethral strictures in the primary (7%) and one in the salvage treatment group. There were two prostato-rectal fistulae in the salvage HIFU group.

CONCLUSIONS

HIFU is proposed to be a minimally invasive low-morbidity ablative treatment for localized prostate cancer, and with good efficacy. The present limited series is unable to support these claims. There were significant rates of complications and oncological failure in both the primary and salvage setting. As a result we have suspended our programme pending further evidence of its safety and efficacy.

Abbreviations
HIFU

high-intensity focused ultrasound

EBRT

external beam radiotherapy

RP

radical prostatectomy

SHIM

Sexual Health Inventory for Men

ADT

androgen deprivation therapy

FDA

USA Food and Drug Administration

ASTRO

American Society for Therapeutic Radiology and Oncology

NICE

National Centre for Clinical Excellence.

INTRODUCTION

The proportion of men diagnosed with early prostate cancer is increasing due to several factors, including PSA screening and increased patient awareness. As a result, clinicians and patients are constantly seeking improvements in treatment. The key goal in managing localized prostate cancer is an oncologically effective intervention with minimal morbidity. Existing techniques are being refined to reduce their associated morbidities whilst maintaining oncological efficacy.

Transrectal high-intensity focused ultrasound (HIFU) is under evaluation as a treatment option for localized prostate cancer in both the primary and salvage setting. The treatment uses the principle of ultrasound energy to image and subsequently ablate the prostate by generating sequential targeted lesions [1], by heat-induced coagulative necrosis of the prostate, and cavitation [2]. The energy is delivered using a transrectal probe under general or regional anaesthesia and is often preceded by a limited TURP. It is attractive due its minimally invasive approach, short hospital stay, reduced convalescence and the potential for re-treatment. Initial reports suggest that HIFU treatment is associated with low morbidity, including preservation of continence and erectile function [3]. However, there is little information in the long-term oncological efficacy; thus we report our experience of HIFU for localized prostate cancer in the primary and salvage setting.

PATIENTS AND METHODS

After obtaining local institutional approval and clinical governance clearance, a HIFU programme was introduced in our department with the intention of treating selected patients with localized prostate cancer. Initial training was received at an approved AblathermTM (EDAP, Lyon, France) HIFU training centre. This high-volume centre also provided mentoring for our initial cases. Selection was broadly divided into primary treatment, and salvage after external beam radiotherapy (EBRT). All patients were given counselling about the investigational nature of this treatment and the relative lack of long-term oncological outcome data. For primary treatment, the selection was based on the following variables: cancer localized to the prostate, PSA level, clinical stage, comorbidity and fitness for general anaesthesia. Patients were unsuitable or unwilling to accept established treatments such as radical prostatectomy (RP) and RT, including EBRT and brachytherapy. Patients for salvage HIFU after EBRT were selected after PSA relapse and biopsy-confirmed recurrence of prostate cancer in the absence of identifiable metastatic disease.

A second generation Ablatherm device was used in all patients. After inducing general anaesthesia, the patient was placed in the lithotomy position and had cysto-urethroscopy before a limited standard TURP. This reduces the postoperative catheterization time, might decrease prostatic calcification, and debulks larger glands for a technically better treatment [4]. The patient was then transferred to the Ablatherm platform and secured in the right lateral position with appropriate protection of all pressure points. A dual-head ultrasound probe is positioned in the rectum. The probe incorporates a firing transducer (3 MHz) with a focal distance of 45 mm, and an imaging transducer (7.5 MHz), and is surrounded by a coupling solution (AblasonicTM; EDAP, Lyon, France) in a balloon. Captured transverse images are computed into longitudinal images. The patient lies in the right lateral position during treatment, which allows gas bubbles to remain out of the firing line before being removed in the circulating Ablasonic fluid. This fluid is cooled in a 5 °C bath, producing a peri-probe temperature of 16–18 °C to cool the rectal wall as protection against inadvertent heating. A further safety feature is an infrared motion detector, placed on the left iliac crest, that stops the treatment if triggered during the procedure. The lesion size can be altered by the operator from 19 to 24 mm long with a constant diameter of 1.7 mm. Treatment is delivered in four to six treatment blocks, depending on the size of the gland [2]. A 16 F Foley two-way catheter was used for bladder drainage and removed a few days later in the clinic. The care plan was to discharge patients on the first day after surgery.

Patients were followed up in the clinic, with PSA measurements at 3 months and every 6 months thereafter. The functional outcome was assessed using the IPSS and Sexual Health Inventory for Men (SHIM) scores.

Oncological failure was defined by several criteria, including biochemical failure, starting salvage therapy (RT or androgen deprivation therapy, ADT), or the presence of cancer on biopsy after treatment; the biopsies were not taken routinely but were triggered by an increasing PSA level. Biochemical failure was assessed using both the Phoenix definition (PSA nadir + 2 ng/mL) [5] and the current USA Food and Drug Administration (FDA) trial endpoint of a PSA level of ≥0.5 ng/mL.

RESULTS

Between 2005 and 2007, 43 patients were scheduled for transrectal HIFU at Guy’s Hospital. Although 31 patients were scheduled for primary treatment and 12 for salvage treatment, three were found to be unsuitable during intraoperative planning, due to excessive rectal wall thickness or rigidity, and treatment did not proceed. The perioperative and demographic characteristics are summarized in Table 1.

Table 1.  The patient characteristics, perioperative data, complications rate and oncological outcomes
VariablePrimary HIFUSalvage HIFU
No. of patients2812
Mean (sd, range):
 Age, years (range)70.1 (5.1, 60–79)70.6 (6.6, 56–79)
 PSA level, ng/mL9.2 (6.7, 0.1–27.1)5.1 (4.3, 0.1–15.5)
 Prostate volume, mL30.0 (7.6, 15–48)21.5 (5.7, 14–34)
D’Amico risk category, n (%)
 Low6 (21) 
 Intermediate16 (57) 
 High6 (21) 
Neoadjuvant ADT10 (36)5
Peri-operative data
Mean (sd, range):
 Total HIFU treatment time71.1 (33.3, 31–154)63.1 (33.2, 30–126)
 Duration of catheterization, days4.1 (5.4, 1–21)5.25 (1.2, (3–7)
 Hospital stay, days2.1 (0.4, 2–4)2.3, (0.65, 2–4)
Synchronous TURP, yesall11/12
Complication rate, % or n115/12
Complications and functional outcome  
Urinary retention11
Urethral stricture21
Prostatitis01
Urorectal fistula02
At 12 months
Continence: pad-freeall10/12
Erectile function: SHIM score ≥2114 (50)none
Oncological outcome
Mean (sd):
Follow-up, months24.9 (4.9)22 (5.9)
PSA nadir, ng/mL1.3 (2.7)2.6 (3.9)
Biopsy after HIFU, n10 (80% positive)0
Phoenix definition of failure13/28 (46%)5/12
FDA definition of failure21/28 (75%)8/12

Of the 28 patients who had primary HIFU treatment the proportion in the low-, intermediate- and high-risk categories of D’Amico et al.[6] were 21%, 57% and 21%, respectively, with a mean (sd) PSA level of 9.2 (6.7) ng/mL. Ten patients (36%) had received neoadjuvant ADT for 3 months and this was discontinued immediately after HIFU. The mean (sd) prostate volume at the time of treatment was 30.0 (7.6) mL.

Of the salvage group (12 patients), where all patients were high-risk by definition, the mean (sd) PSA level was 5.1 (4.3) ng/mL. Five patients had received neoadjuvant ADT. The mean (sd) prostate volume was 21.5 (5.7) mL.

The mean (sd) total treatment time (excluding TURP) was 71.1 (33.3) min for the primary treatment group and 63.1 (33.3) min for the salvage group. All patients in the primary group had a synchronous TURP, as did 11 of the 12 in the salvage group (Table 1).

There was one episode of acute urinary retention in each group, managed by a further period of catheterization. There were two urethral strictures in the primary (7%) and one in the salvage treatment group, managed by urethral dilatation or optical urethrotomy. One patient in the salvage group developed significant prostatitis lasting 6 months, and required a prolonged period of antibiotics before resolution.

There were two urinary fistulae in the salvage HIFU group. One patient developed urinary leakage via the rectum 8 weeks after HIFU and was initially managed by urethral catheterization, as he declined urinary or bowel diversion. He subsequently had a salvage RP and reconstruction with temporary colostomy. A second patient initially developed a painful anterior fistula to the lower limbs from the prostatic cavity via the symphysis pubis at 4 weeks after HIFU, requiring urinary diversion. Subsequently he developed pyocystis requiring bladder washout and anterior sphincterotomy, an adductor abscess requiring incision and drainage, and eventually a prostatorectal fistula palliated with fecal diversion.

At 12 months after HIFU all 28 patients in the primary and 10 of 12 in the salvage group were pad-free. Two patients in the primary group had transient stress urinary incontinence that resolved with no intervention within 3 months. In the salvage group two patients were still requiring two to three pads per day at 12 months of follow-up, one of whom has had an artificial urinary sphincter placed. The mean change in IPSS was −8.7 in the primary and −4.1 in the salvage group.

Potency was defined as a SHIM score of ≥21. Ten of 28 men in the primary group were potent before HIFU, of whom five were potent at 12 months afterward. None of the men in the salvage group regained potency after treatment. The complications and functional outcomes are also summarized in Table 1.

For oncological outcomes (Table 1), the 28 patients treated with primary HIFU had a mean (sd) follow-up of 24.9 (4.9) months. Using the Phoenix definition of failure, HIFU failed in 13 of 28 (46%) patients. Applying the current FDA trial primary endpoint of biochemical failure, the HIFU failed in 21 of 28 patients (75%).

In the 12 patients treated with salvage HIFU the mean follow-up was 22 (5.9) months. Using the Phoenix definition of failure, HIFU failed in five, and using the FDA primary endpoint of biochemical failure, HIFU failed in eight. Ten patients had a prostate biopsy after HIFU, due to an increasing PSA level, eight of whom had residual cancer. Of the two patients with no cancer on their biopsy, one had evidence of metastatic disease on a bone scan and one developed biochemical failure requiring ADT.

One man has died from metastatic prostate cancer 18 months after salvage HIFU treatment. He had a biochemical recurrence at 6 months, with a PSA level of 51 ng/mL, and had been treated for complex urocutaneous and urorectal fistulae with urinary and fecal diversion (see above). There were no HIFU re-treatments in this series.

DISCUSSION

The management of localized prostate cancer offers several challenges, particularly in achieving the balance between oncological cure whilst minimizing functional morbidity. The so-called ‘trifecta’ of outcomes reflects the key goals of oncological cure, with maintenance of urinary continence and erectile function [7]. Established interventions for managing localized prostate cancer, e.g. RP and EBRT, have undergone technical refinements in recent years to improve these outcomes.

HIFU is a minimally invasive ablative technology that has emerged as a treatment option for managing localized prostate cancer in both the primary and salvage setting. It is attractive as a single-session procedure, with the potential for re-treatment if required. Initial reports suggested low morbidity and good functional results, whilst achieving good oncological outcomes [3,8]. The European multicentre study reported on 402 patients who were treated with HIFU using the Ablatherm device for T1-T2 localized prostate cancer [3]. Of the 288 for whom biopsy data were assessable, 87% had negative sextant biopsies. Of these, 212 reached the 6-month follow up with a median PSA level of 0.6 ng/mL. Recto-urethral fistula was reported in five patients (1.2%) and severe incontinence, requiring an artificial urinary sphincter, in four (1%).

Uchida et al.[9] reported outcome data for 63 patients undergoing HIFU for localized prostate cancer, using the SonablateTM device. The negative biopsy rate in that study was also 87%, although 25% developed biochemical failure at a mean follow-up of <2 years.

The first report of a long-term outcome for patients undergoing HIFU for localized prostate cancer was published in 2008. Blana et al.[10] reported on 140 patients treated with the Ablatherm device with a mean follow-up of 6.4 years. The biochemical recurrence rate was 33.6% using the Phoenix definition (nadir + 2 ng/mL). However, with a mean follow-up of 3.9 years, Misrai et al.[11] reported on 119 patients treated with the Ablatherm device, who had a biochemical failure rate of 43.7% using the same definition of biochemical failure. Table 2 shows the HIFU outcomes published previously [3,9–12].

Table 2.  HIFU outcomes published previously
RefNo. of patientsMean age, yearsMean PSA level, ng/mLRisk category (low/intermediate/high), %Mean follow-up, yearsOncological outcomeComment
  • *

    ASTRO definition of biochemical failure (BF), three successive PSA increases above the nadir;

  • Phoenix definition of BF, nadir + 2.0 ng/mL;

  • PSA ≥0.5 ng/mL at 12 months; BCR, biochemical recurrence.

[3]4026910.9NA1.287.2%–ve biopsyMean number of treatments, 1.4
[9] 6371 8.535/41/241.925% BF* 
[10]14070 7.051/49/06.433.6% BF20% of those with −ve biopsies developed BCR
[11]11968 8.255/42/33.943.7% BF*Poor results in intermediate and high-risk groups
[12]17265NA28/37/35121.7% BF36% urethral stricture rate

The results above illustrate a key problem with the evaluation of HIFU as a treatment option for localized prostate cancer, i.e. the definition of oncological success. With RP there is some agreement that a PSA level of ≥0.2 ng/mL constitutes biochemical failure [13], whilst after EBRT the American Society for Therapeutic Radiology and Oncology (ASTRO) have published a consensus statement (the ‘Phoenix definition’) defining biochemical failure as the PSA nadir + 2 ng/mL [5]. However, there is no agreement as to what constitutes biochemical failure after ablative treatments such as HIFU or cryotherapy. The largest report to date of the long-term outcome after HIFU for localized prostate cancer uses the Phoenix definition to define oncological success [10]. One of the editorials following this paper questioned the appropriateness of the Phoenix definition for HIFU. In the consensus statement, ASTRO specify that the ‘nadir + 2 ng/mL’ definition ‘is not recommended for other modalities such as cryotherapy’. It can be inferred that its use for other ablative methods such as HIFU is questionable. The current FDA trial of HIFU in the USA uses a PSA level of ≥0.5 ng/mL to define biochemical failure, and that is the primary endpoint of that trial.

In the present study we chose to express biochemical failure using both the Phoenix definition and the FDA trial threshold of a PSA level of ≥0.5 ng/mL. For the primary and salvage HIFU groups this translates into a biochemical failure rate of 46–75% and 42–67%, respectively. This highlights the discrepancy between criteria used to define oncological success after HIFU treatment. Even using biopsy after treatment as a measure of success is fraught with difficulty, as 20% of patients in the series of Blana et al.[10] with a negative biopsy subsequently failed using the Phoenix definition. There is clearly a need for consensus on the definition used for oncological success in series of patients treated with HIFU.

There were no perioperative deaths in the present series. The procedure was well tolerated in general and rates of incontinence and urethral stricture formation were comparable with other series. Indeed, a large series from London presented at the BAUS Annual Meeting in 2008 [12] reported overall urethral stricture rates as high as 35% using the Sonablate device.

There were no urorectal injuries in the primary HIFU group, but there were two fistulae in the salvage group, despite using the post-RT algorithm on the Ablatherm device. The urorectal fistula rate in patients undergoing salvage HIFU has been reported as 0–6%[14–16]. The potential for inadvertent rectal injury is highlighted by Zacharakis et al.[16] who attributed one of their rectal injuries to inadequate anaesthesia when using the Sonablate device for salvage HIFU. Andrich et al.[17] recently reported their experience of eight patients with a urorectal fistula after salvage HIFU who had complex reconstructive surgery, with high morbidity. All patients were incontinent of urine after reconstruction.

The National Centre for Clinical Excellence (NICE) in the UK initially endorsed HIFU as a treatment for localized prostate cancer in 2005, stating that its use could be supported, provided that normal arrangements were in place for clinical audit. However, the NICE prostate cancer guidelines released in February 2008 no longer supported the use of HIFU outside of controlled clinical trials. The current European Association of Urology guidelines list HIFU as ‘investigational or experimental’[18].

In conclusion, HIFU is proposed to be a minimally invasive low-morbidity ablative treatment for localized prostate cancer, and with good efficacy. The present limited series is unable to support these claims. We found significant rates of complications and oncological failure in both the primary and salvage setting. As a result we have suspended our programme. The safety and efficacy of HIFU must be established within randomized controlled trials against existing established interventions before its role in the management of localized prostate cancer can be truly defined.

CONFLICT OF INTEREST

None declared.

Ancillary