Focal Ablation Targeted to the Index Lesion in Multifocal Localised Prostate Cancer: a Prospective Development Study

2.1. Study design and conduct

Our single-centre study was a prospective development study according to the IDEAL (Idea, Development, Exploration, Assessment, and Long-term) guidelines for evaluating innovation in surgery [13] . The trial was approved by Local Research Ethics Committee A of the University College London Hospitals.

2.2. Patient population

Treatment-naïve men recently diagnosed with low-, intermediate-, or high-risk nonmetastatic prostate cancer (prostate-specific antigen [PSA] ≤20 ng/ml, Gleason ≤ 4 + 3, stage ≤T3aN0M0) were eligible ( Fig. 1 ).

Fig. 1 Trial flow chart. mpMRI = multiparametric magnetic resonance imaging; HIFU = high-intensity focused ultrasound; ADT = androgen deprivation therapy.

2.3. Study interventions

2.3.2. Treatment

Focal ablation of the index lesion was performed using transrectal high-intensity focused ultrasound (HIFU; Sonablate 500, Focus Surgery, Indianapolis, IN, USA) (Supplementary materials). Untreated areas could contain secondary small-volume (CCL ≤5 mm) Gleason 3 + 3 disease [14] , high-grade prostate intraepithelial neoplasia, and/or atypical small acinar proliferation ( Fig. 2 ).

Fig. 2 Schematic diagrams demonstrating the types of focal therapy conducted in this trial. Large red areas represent dominant cancers (so-called index lesions) whilst small green areas represent small, low-grade, secondary lesions. Red transparent boxes represent ablation zones on the high-intensity focused ultrasound device: (A) hemiablation, (B) extended (dogleg) ablation; (C) quadrant ablation; (D) focal ablation.

2.3.3. Follow-up

Contrast-enhanced MRI was carried out at 10–14 d to evaluate the area of ablation, demonstrated by a confluent perfusion deficit ( Fig. 3 ). Clinical review at 1, 3, 6, 9, and 12 mo assessed adverse events, serum PSA, and responses to validated questionnaires. Phosphodiesterase-5 inhibitor (PDE5-I) use was permitted at any time point before treatment and during follow-up to aid erectile function. At 6 mo, mpMRI followed by biopsies targeted to the treated area was scheduled, with a minimum sampling requirement of one core for every 1 ml of residual tissue. Repeat treatment using focal HIFU for treated or untreated areas that showed signs of progression was permitted if biopsies were positive. Further mpMRI was carried out at 12 mo in such cases. As the purpose of the 6-mo biopsies was to determine whether the ablation was successful, our ethics committee did not permit transperineal TPM biopsies of untreated tissue because of the requirement for another general anaesthetic. However, biopsies of untreated tissue were permitted if a new lesion suspicious for cancer was seen on mpMRI or if existing untreated tissue showed signs of progression.

Fig. 3 Contrast-enhanced axial scan of the prostate at mid-gland at 1 wk after high-intensity focused ultrasound ablation of an index lesion in the right prostate lobe. Example of right-sided index lesion treatment with confluent hemiablation necrosis demonstrated on early (1-wk) multiparametric magnetic resonance imaging. The orange line indicates the area of ablation with confluent necrosis, in this case hemiablation. The white line indicates the outline of whole prostate. R = rectum.

2.4. Study endpoints

The primary outcome was the composite rate of genitourinary side effects measured using validated patient questionnaires[15], [16], and [17]. Incontinence was defined using a score of 1–3 for question 1 (any leak) and of 1–3 for question 5 (any pad use) of the University of California, Los Angeles (UCLA) Expanded Prostate Cancer Index Composite (EPIC) continence questionnaire. Erectile dysfunction was defined as a score of 0 or 1 for question 2 of the 15-item International Index of Erectile Function (IIEF-15) questionnaire. Urinary function was evaluated using the International Prostate Symptom Score (IPSS), the IPSS-quality of life questionnaire, and the UCLA-EPIC continence questionnaire. Health-related quality of life was measured using the Functional Assessment of Cancer Therapy-Prostate (FACT-P). Responses to the validated questionnaires were collated using standard methods.

Secondary outcomes were histological parameters of cancer control on 6-mo biopsy, absence of clinically significant disease on mpMRI at 6 and 12 mo, description of PSA kinetics following index lesion ablation, and the proportion of men requiring salvage local radical therapy and systemic therapy for progressive prostate cancer.

2.5. Statistical analysis

The sample size was calculated on a composite endpoint defined as the total proportion of men with erectile dysfunction and/or incontinence. It was estimated that baseline incontinence was about 10% and baseline impotence about 40% in men older than 60 yr[18] and [19]. Overall, we estimated that baseline genitourinary dysfunction would be approximately 50%. Were these men to undergo whole-gland therapy, 50% of those with good baseline function would also suffer some form of genitourinary dysfunction [20] , so total genitourinary dysfunction would be approximately 75% at 12 mo if men were to undergo whole-gland therapy. It was hypothesised that focal therapy would give rise to genitourinary dysfunction in only 5% of those with good baseline function. Therefore, we estimated a total genitourinary dysfunction level of approximately 55% at 12 mo after focal index lesion ablation. Using sample size calculations comparing an estimate proportion (0.55) to a known proportion (0.75), with α = 0.05 and power of 90%, the total number of patients required was 56 [21] .

A Wilcoxon signed rank test (two-tailed) was used to evaluate differences between continuous variables (PSA, questionnaire scores) measured at baseline and at the 12-mo follow-up visit. Statistical significance was set atp≤ 0.05. Changes over time were described using box-and-whisker plots. Subgroup analyses were hypothesis-generating, and with small numbers in each subgroup, it was deemed inappropriate to run statistical tests of significance for such comparisons. Statistical analysis was performed using Stata version 11.2 (Stata Corp, College Station, TX, USA). This analysis was carried out on the available data. However, some functional measures were missing at baseline or at 12-mo follow-up for five men, so we conducted two sensitivity analyses to assess if this was likely to impact our results. We repeated the analysis described above for a “best” outcome scenario and a “worst” outcome scenario by setting missing baseline values to the maximum or minimum possible, respectively.

3.1. Study patients

Fifty-six men were recruited (July 27, 2009–January 26, 2011) with a mean age of 64 yr. According to National Comprehensive Cancer Network criteria [22] , seven men (12.5%) had low-risk, 47 (83.9%) had intermediate-risk, and two (3.6%) had high-risk nonmetastatic prostate cancer ( Fig. 1 , Table 1 ).

3.2. Primary outcome

Overall, 40 patients (71.4%; 95% confidence interval [CI] 59.6–83.2%) were leak-free, pad-free continent and had erections sufficient for penetration at baseline. This composite measure decreased to 33 patients (58.9%; 95% CI 46.0–71.8%) at 12 mo.

3.3. Secondary outcomes

3.3.5. Biochemical, imaging, and histological outcomes

A significant decrease in serum PSA from a median baseline value of 7.4 ng/ml (IQR 5.6–9.5) to 2.4 ng/ml (IQR 1.6–4.1) was observed at 12 mo (p < 0.0001; Fig. 3 ).

At 6 mo, two men refused a post-treatment biopsy, and another two had clinical reasons for omitting biopsies (chemotherapy for newly diagnosed lung cancer in one, and a persistent HIFU cavity connecting with the vas deferens on mpMRI in another). For these four men, mpMRI at 6 mo showed no evidence of measurable residual disease. There was no histological evidence of clinically significant disease (Gleason ≤ 3 + 3, CCL ≤3 mm) in the treated area in 44/52 patients (84.6%; 95% CI 70–92%), or of any cancer in the treated area in 34/52 patients (65.4%; 95% CI 52.5–78.3%). Two had clinically significant disease on biopsies from suspicious areas in the untreated side on mpMRI not present at baseline. Overall, histopathological absence of clinically significant cancer (treated and untreated sides) was observed in 42/52 men (80.8%, 95% CI 67.5–90.4%). The overall rate of absence of measurable disease was 86% (48/56) at the study end following repeat treatment in two patients. However, 43% of men had persistent cancer of any threshold at the study end. Two other men underwent salvage treatment with neoadjuvant hormone ablation therapy and external beam radiotherapy for residual Gleason 4 + 3 and Gleason 4 + 4 disease ( Table 3 ).

Table 3 Histology outcomes at 6 mo after focal index lesion ablation in 52 men who underwent post-treatment prostate biopsy

Parameter	Value
Positive biopsy outcomes on the treated side
Median number of biopsies taken, n (IQR)	5.0 (4.0–6.0)
Positive biopsies (any disease), n (%)	18/52 (34.6)
Positive biopsies (clinically significant disease), n (%)	8/52 (15.4) a
Median CCLmax in positive cores, mm (IQR)	2.0 (1.0–4.0)
Gleason score, n (%)
3 + 3	12/18 (66.7)
3 + 4	5/18 (27.8)
4 + 3	1/18 (5.6)
Positive biopsy outcomes on the untreated side (targeted to suspicious areas on 6-mo MRI)
Positive biopsies (any disease), n (%)	4/52 (7.7)
Positive biopsies (clinically significant disease), n (%)	2/52 (3.8) b
Gleason score (n)
3 + 3	3 c
3 + 4	0
4 + 3	0
4 + 4	1
Absence of any cancer on both sides, n (%; 95% CI)	30/52 (57.7; 43.2–71.3)
Overall absence of clinically significant disease on both sides, n (%; 95% CI)	42/52 (80.8; 67.5–90.4)
Other histological findings, n (%; 95% CI) d
Prostatic acini	2/51 (3.9; 0.5–13.5)
Atrophy	20/51 (39.2; 25.8–53.9)
Fibrosis	39/51 (76.5; 62.5–87.2)
Giant-cell reaction	4/51 (7.8; 2.2–18.9)
Necrosis	11/51 (21.6; 11.3–35.3)

^a These men were counted within the “any disease” category. In other words, eight of the 18 men with positive biopsies on the treated side had clinically significant cancer.

^b These men were counted within the “any disease” category. In other words, two of the four men with positive biopsies on the untreated side had clinically significant cancer.

^c CCL_max in these three cases was <1 mm, 1 mm, and 4 mm.

^d Data only available for 51/52 patients.

IQR = interquartile range; CCL_max = maximum cancer core length; MRI = magnetic resonance imaging; CI = confidence interval.

Of 41 patients with good baseline status, 22 (53.7%, 95% CI 37.4–69.3%) achieved the trifecta outcome (absence of clinically significant disease, pad-free, leak-free continence, and erections sufficient for intercourse) at 12 mo ( Fig. 4 ). The results for sensitivity analyses (data not shown) were very similar and did not affect our overall conclusion.

Fig. 4 Summary of continence, erectile function, and cancer control outcomes and the trifecta rate following high-intensity focused ultrasound ablation of the index lesion. Patient-reported trifecta outcomes were collected using validated questionnaires. (1) Percentage continence data were derived using the questions “Over the past 4 weeks how often did you leak urine?” and “Over the past 4 weeks how many pads or adult diapers per day did you usually use to control leakage?” from the urinary domain of the University of California, Los Angeles Expanded Prostate Cancer Index Composite questionnaire. (2) The percentage of men with erections sufficient for penetration was calculated for men scoring ≥2 for the question “Over the past 4 weeks when you had erections with sexual stimulation, how often were your erections hard enough for penetration?” from the 15-item International Index of Erectile Function. (3) The percentage phosphodiesterase-5 inhibitor (PDE5-I) use (tadalafil, sildenafil, or vardenafil) was calculated using the number of men achieving erections sufficient for penetration from part (2) as the denominator.

4.1. Clinical implications

The outcome from other focal therapy studies in which treatment was applied to all known cancer lesions was recently reviewed [9] . Rates of 95–100% for pad-free, leak-free continence preservation, 54–100% for potency, and 83–100% for overall absence of clinically significant cancer were observed after treatment. Most of these studies applied strict selection criteria for disease risk, biopsy cancer burden, and function at baseline, and therefore have limited external validity. In our current study—in a group of men who were not selected by the investigators on the basis of good baseline function—the functional preservation outcomes were not as high as we have previously reported. Indeed, our trifecta rate is also lower than previously reported, and may not be as encouraging when compared to the rates reported after surgery. With greater numbers in our ongoing larger multicentre trials, we may be to identify men who are unlikely to have good functional outcomes after focal therapy and would be better treated using a radical approach. Our study is unique in that known cancer of low malignant potential was left untreated. We recently showed that if such a strategy were to be adopted, more than 90% of newly diagnosed men with localised prostate cancer might be eligible for focal therapy [27] .

Some have argued that because mpMRI and TPM biopsies are not accurate enough to detect all areas of prostate cancer, the concept of selective therapy lacks legitimacy [28] . However, in most other solid organs, cancer is diagnosed when a clinical manifestation of disease (a lump or an imaging signal) is verified histologically. Inevitably, therapy is directed at the clinical phenotype plus the addition of a margin. This was recently seen in the administration of radiotherapy to the breast after lumpectomy. Historically, whole-breast irradiation was favoured because of the known multifocality of breast cancer. The finding that intraoperative local irradiation of the surgical site was not inferior to whole-breast irradiation challenges the idea that recurrence is due to residual subclinical multifocal lesions away from the index lesion [29] . There is also concern that inclusion of intermediate- and high-risk cases for focal therapy may represent undertreatment for those men who harbour microscopic nodal metastases.

It is yet to be determined whether an RCT of focal therapy compared to radical therapy is deliverable [23] . Previous trials have attempted to compare different interventions in localised prostate cancer; however, many have failed because of a lack of physician and patient equipoise. There are additional concerns regarding appropriate endpoints and the follow-up time frame, especially as a minimum of 10-yr follow-up would be required to evaluate metastases and mortality. The UK National Institute of Health Research has recently funded a pilot RCT to determine whether recruitment might be feasible (ISRCTN99760303). This RCT will evaluate rates of transition to local salvage therapies within a 5-yr period between focal therapy and radical prostatectomy in men with intermediate-risk prostate cancer.

Our findings have implications for prostate cancer management. Some have suggested that low-risk disease should be relabelled as a benign entity because of the rarity of progression and metastases from these lesions[10], [11], and [12]. If this hypothesis proves to be true, we could concentrate on targeting measurable disease based on mpMRI considering the high negative predictive values for clinically significant disease that it confers. However, the notion that Gleason 6 disease cannot metastasise has recently been brought into question following the findings of Haffner et al [30] , who reported on a case in which a small area of Gleason 6 disease within a tumour of very high-grade cancer was responsible for a distant metastasis. There has been considerable debate regarding the validity of these findings [31] .

Our study also represents anin vivohuman model for better understanding of the determinants of progression of low-grade disease in a manner that is not predicated on removing the whole organ. This might allow us to determine if such lesions do progress and, if so, whether they can be identified at the time of diagnosis using detailed molecular typing approaches such as immunohistochemistry and -omics techniques.

2.1. Study design and conduct

Our single-centre study was a prospective development study according to the IDEAL (Idea, Development, Exploration, Assessment, and Long-term) guidelines for evaluating innovation in surgery [13] . The trial was approved by Local Research Ethics Committee A of the University College London Hospitals.

2.2. Patient population

Treatment-naïve men recently diagnosed with low-, intermediate-, or high-risk nonmetastatic prostate cancer (prostate-specific antigen [PSA] ≤20 ng/ml, Gleason ≤ 4 + 3, stage ≤T3aN0M0) were eligible ( Fig. 1 ).

Fig. 1 Trial flow chart. mpMRI = multiparametric magnetic resonance imaging; HIFU = high-intensity focused ultrasound; ADT = androgen deprivation therapy.

2.3. Study interventions

2.3.2. Treatment

Focal ablation of the index lesion was performed using transrectal high-intensity focused ultrasound (HIFU; Sonablate 500, Focus Surgery, Indianapolis, IN, USA) (Supplementary materials). Untreated areas could contain secondary small-volume (CCL ≤5 mm) Gleason 3 + 3 disease [14] , high-grade prostate intraepithelial neoplasia, and/or atypical small acinar proliferation ( Fig. 2 ).

Fig. 2 Schematic diagrams demonstrating the types of focal therapy conducted in this trial. Large red areas represent dominant cancers (so-called index lesions) whilst small green areas represent small, low-grade, secondary lesions. Red transparent boxes represent ablation zones on the high-intensity focused ultrasound device: (A) hemiablation, (B) extended (dogleg) ablation; (C) quadrant ablation; (D) focal ablation.

2.3.3. Follow-up

Contrast-enhanced MRI was carried out at 10–14 d to evaluate the area of ablation, demonstrated by a confluent perfusion deficit ( Fig. 3 ). Clinical review at 1, 3, 6, 9, and 12 mo assessed adverse events, serum PSA, and responses to validated questionnaires. Phosphodiesterase-5 inhibitor (PDE5-I) use was permitted at any time point before treatment and during follow-up to aid erectile function. At 6 mo, mpMRI followed by biopsies targeted to the treated area was scheduled, with a minimum sampling requirement of one core for every 1 ml of residual tissue. Repeat treatment using focal HIFU for treated or untreated areas that showed signs of progression was permitted if biopsies were positive. Further mpMRI was carried out at 12 mo in such cases. As the purpose of the 6-mo biopsies was to determine whether the ablation was successful, our ethics committee did not permit transperineal TPM biopsies of untreated tissue because of the requirement for another general anaesthetic. However, biopsies of untreated tissue were permitted if a new lesion suspicious for cancer was seen on mpMRI or if existing untreated tissue showed signs of progression.

Fig. 3 Contrast-enhanced axial scan of the prostate at mid-gland at 1 wk after high-intensity focused ultrasound ablation of an index lesion in the right prostate lobe. Example of right-sided index lesion treatment with confluent hemiablation necrosis demonstrated on early (1-wk) multiparametric magnetic resonance imaging. The orange line indicates the area of ablation with confluent necrosis, in this case hemiablation. The white line indicates the outline of whole prostate. R = rectum.

2.4. Study endpoints

The primary outcome was the composite rate of genitourinary side effects measured using validated patient questionnaires[15], [16], and [17]. Incontinence was defined using a score of 1–3 for question 1 (any leak) and of 1–3 for question 5 (any pad use) of the University of California, Los Angeles (UCLA) Expanded Prostate Cancer Index Composite (EPIC) continence questionnaire. Erectile dysfunction was defined as a score of 0 or 1 for question 2 of the 15-item International Index of Erectile Function (IIEF-15) questionnaire. Urinary function was evaluated using the International Prostate Symptom Score (IPSS), the IPSS-quality of life questionnaire, and the UCLA-EPIC continence questionnaire. Health-related quality of life was measured using the Functional Assessment of Cancer Therapy-Prostate (FACT-P). Responses to the validated questionnaires were collated using standard methods.

Secondary outcomes were histological parameters of cancer control on 6-mo biopsy, absence of clinically significant disease on mpMRI at 6 and 12 mo, description of PSA kinetics following index lesion ablation, and the proportion of men requiring salvage local radical therapy and systemic therapy for progressive prostate cancer.

2.5. Statistical analysis

The sample size was calculated on a composite endpoint defined as the total proportion of men with erectile dysfunction and/or incontinence. It was estimated that baseline incontinence was about 10% and baseline impotence about 40% in men older than 60 yr[18] and [19]. Overall, we estimated that baseline genitourinary dysfunction would be approximately 50%. Were these men to undergo whole-gland therapy, 50% of those with good baseline function would also suffer some form of genitourinary dysfunction [20] , so total genitourinary dysfunction would be approximately 75% at 12 mo if men were to undergo whole-gland therapy. It was hypothesised that focal therapy would give rise to genitourinary dysfunction in only 5% of those with good baseline function. Therefore, we estimated a total genitourinary dysfunction level of approximately 55% at 12 mo after focal index lesion ablation. Using sample size calculations comparing an estimate proportion (0.55) to a known proportion (0.75), with α = 0.05 and power of 90%, the total number of patients required was 56 [21] .

A Wilcoxon signed rank test (two-tailed) was used to evaluate differences between continuous variables (PSA, questionnaire scores) measured at baseline and at the 12-mo follow-up visit. Statistical significance was set atp≤ 0.05. Changes over time were described using box-and-whisker plots. Subgroup analyses were hypothesis-generating, and with small numbers in each subgroup, it was deemed inappropriate to run statistical tests of significance for such comparisons. Statistical analysis was performed using Stata version 11.2 (Stata Corp, College Station, TX, USA). This analysis was carried out on the available data. However, some functional measures were missing at baseline or at 12-mo follow-up for five men, so we conducted two sensitivity analyses to assess if this was likely to impact our results. We repeated the analysis described above for a “best” outcome scenario and a “worst” outcome scenario by setting missing baseline values to the maximum or minimum possible, respectively.

3.1. Study patients

Fifty-six men were recruited (July 27, 2009–January 26, 2011) with a mean age of 64 yr. According to National Comprehensive Cancer Network criteria [22] , seven men (12.5%) had low-risk, 47 (83.9%) had intermediate-risk, and two (3.6%) had high-risk nonmetastatic prostate cancer ( Fig. 1 , Table 1 ).

3.2. Primary outcome

Overall, 40 patients (71.4%; 95% confidence interval [CI] 59.6–83.2%) were leak-free, pad-free continent and had erections sufficient for penetration at baseline. This composite measure decreased to 33 patients (58.9%; 95% CI 46.0–71.8%) at 12 mo.

3.3. Secondary outcomes

3.3.5. Biochemical, imaging, and histological outcomes

A significant decrease in serum PSA from a median baseline value of 7.4 ng/ml (IQR 5.6–9.5) to 2.4 ng/ml (IQR 1.6–4.1) was observed at 12 mo (p < 0.0001; Fig. 3 ).

At 6 mo, two men refused a post-treatment biopsy, and another two had clinical reasons for omitting biopsies (chemotherapy for newly diagnosed lung cancer in one, and a persistent HIFU cavity connecting with the vas deferens on mpMRI in another). For these four men, mpMRI at 6 mo showed no evidence of measurable residual disease. There was no histological evidence of clinically significant disease (Gleason ≤ 3 + 3, CCL ≤3 mm) in the treated area in 44/52 patients (84.6%; 95% CI 70–92%), or of any cancer in the treated area in 34/52 patients (65.4%; 95% CI 52.5–78.3%). Two had clinically significant disease on biopsies from suspicious areas in the untreated side on mpMRI not present at baseline. Overall, histopathological absence of clinically significant cancer (treated and untreated sides) was observed in 42/52 men (80.8%, 95% CI 67.5–90.4%). The overall rate of absence of measurable disease was 86% (48/56) at the study end following repeat treatment in two patients. However, 43% of men had persistent cancer of any threshold at the study end. Two other men underwent salvage treatment with neoadjuvant hormone ablation therapy and external beam radiotherapy for residual Gleason 4 + 3 and Gleason 4 + 4 disease ( Table 3 ).

Table 3 Histology outcomes at 6 mo after focal index lesion ablation in 52 men who underwent post-treatment prostate biopsy

Parameter	Value
Positive biopsy outcomes on the treated side
Median number of biopsies taken, n (IQR)	5.0 (4.0–6.0)
Positive biopsies (any disease), n (%)	18/52 (34.6)
Positive biopsies (clinically significant disease), n (%)	8/52 (15.4) a
Median CCLmax in positive cores, mm (IQR)	2.0 (1.0–4.0)
Gleason score, n (%)
3 + 3	12/18 (66.7)
3 + 4	5/18 (27.8)
4 + 3	1/18 (5.6)
Positive biopsy outcomes on the untreated side (targeted to suspicious areas on 6-mo MRI)
Positive biopsies (any disease), n (%)	4/52 (7.7)
Positive biopsies (clinically significant disease), n (%)	2/52 (3.8) b
Gleason score (n)
3 + 3	3 c
3 + 4	0
4 + 3	0
4 + 4	1
Absence of any cancer on both sides, n (%; 95% CI)	30/52 (57.7; 43.2–71.3)
Overall absence of clinically significant disease on both sides, n (%; 95% CI)	42/52 (80.8; 67.5–90.4)
Other histological findings, n (%; 95% CI) d
Prostatic acini	2/51 (3.9; 0.5–13.5)
Atrophy	20/51 (39.2; 25.8–53.9)
Fibrosis	39/51 (76.5; 62.5–87.2)
Giant-cell reaction	4/51 (7.8; 2.2–18.9)
Necrosis	11/51 (21.6; 11.3–35.3)

^a These men were counted within the “any disease” category. In other words, eight of the 18 men with positive biopsies on the treated side had clinically significant cancer.

^b These men were counted within the “any disease” category. In other words, two of the four men with positive biopsies on the untreated side had clinically significant cancer.

^c CCL_max in these three cases was <1 mm, 1 mm, and 4 mm.

^d Data only available for 51/52 patients.

IQR = interquartile range; CCL_max = maximum cancer core length; MRI = magnetic resonance imaging; CI = confidence interval.

Of 41 patients with good baseline status, 22 (53.7%, 95% CI 37.4–69.3%) achieved the trifecta outcome (absence of clinically significant disease, pad-free, leak-free continence, and erections sufficient for intercourse) at 12 mo ( Fig. 4 ). The results for sensitivity analyses (data not shown) were very similar and did not affect our overall conclusion.

Fig. 4 Summary of continence, erectile function, and cancer control outcomes and the trifecta rate following high-intensity focused ultrasound ablation of the index lesion. Patient-reported trifecta outcomes were collected using validated questionnaires. (1) Percentage continence data were derived using the questions “Over the past 4 weeks how often did you leak urine?” and “Over the past 4 weeks how many pads or adult diapers per day did you usually use to control leakage?” from the urinary domain of the University of California, Los Angeles Expanded Prostate Cancer Index Composite questionnaire. (2) The percentage of men with erections sufficient for penetration was calculated for men scoring ≥2 for the question “Over the past 4 weeks when you had erections with sexual stimulation, how often were your erections hard enough for penetration?” from the 15-item International Index of Erectile Function. (3) The percentage phosphodiesterase-5 inhibitor (PDE5-I) use (tadalafil, sildenafil, or vardenafil) was calculated using the number of men achieving erections sufficient for penetration from part (2) as the denominator.

4.1. Clinical implications

The outcome from other focal therapy studies in which treatment was applied to all known cancer lesions was recently reviewed [9] . Rates of 95–100% for pad-free, leak-free continence preservation, 54–100% for potency, and 83–100% for overall absence of clinically significant cancer were observed after treatment. Most of these studies applied strict selection criteria for disease risk, biopsy cancer burden, and function at baseline, and therefore have limited external validity. In our current study—in a group of men who were not selected by the investigators on the basis of good baseline function—the functional preservation outcomes were not as high as we have previously reported. Indeed, our trifecta rate is also lower than previously reported, and may not be as encouraging when compared to the rates reported after surgery. With greater numbers in our ongoing larger multicentre trials, we may be to identify men who are unlikely to have good functional outcomes after focal therapy and would be better treated using a radical approach. Our study is unique in that known cancer of low malignant potential was left untreated. We recently showed that if such a strategy were to be adopted, more than 90% of newly diagnosed men with localised prostate cancer might be eligible for focal therapy [27] .

Some have argued that because mpMRI and TPM biopsies are not accurate enough to detect all areas of prostate cancer, the concept of selective therapy lacks legitimacy [28] . However, in most other solid organs, cancer is diagnosed when a clinical manifestation of disease (a lump or an imaging signal) is verified histologically. Inevitably, therapy is directed at the clinical phenotype plus the addition of a margin. This was recently seen in the administration of radiotherapy to the breast after lumpectomy. Historically, whole-breast irradiation was favoured because of the known multifocality of breast cancer. The finding that intraoperative local irradiation of the surgical site was not inferior to whole-breast irradiation challenges the idea that recurrence is due to residual subclinical multifocal lesions away from the index lesion [29] . There is also concern that inclusion of intermediate- and high-risk cases for focal therapy may represent undertreatment for those men who harbour microscopic nodal metastases.

It is yet to be determined whether an RCT of focal therapy compared to radical therapy is deliverable [23] . Previous trials have attempted to compare different interventions in localised prostate cancer; however, many have failed because of a lack of physician and patient equipoise. There are additional concerns regarding appropriate endpoints and the follow-up time frame, especially as a minimum of 10-yr follow-up would be required to evaluate metastases and mortality. The UK National Institute of Health Research has recently funded a pilot RCT to determine whether recruitment might be feasible (ISRCTN99760303). This RCT will evaluate rates of transition to local salvage therapies within a 5-yr period between focal therapy and radical prostatectomy in men with intermediate-risk prostate cancer.

Our findings have implications for prostate cancer management. Some have suggested that low-risk disease should be relabelled as a benign entity because of the rarity of progression and metastases from these lesions[10], [11], and [12]. If this hypothesis proves to be true, we could concentrate on targeting measurable disease based on mpMRI considering the high negative predictive values for clinically significant disease that it confers. However, the notion that Gleason 6 disease cannot metastasise has recently been brought into question following the findings of Haffner et al [30] , who reported on a case in which a small area of Gleason 6 disease within a tumour of very high-grade cancer was responsible for a distant metastasis. There has been considerable debate regarding the validity of these findings [31] .

Our study also represents anin vivohuman model for better understanding of the determinants of progression of low-grade disease in a manner that is not predicated on removing the whole organ. This might allow us to determine if such lesions do progress and, if so, whether they can be identified at the time of diagnosis using detailed molecular typing approaches such as immunohistochemistry and -omics techniques.

2.1. Study design and conduct

Our single-centre study was a prospective development study according to the IDEAL (Idea, Development, Exploration, Assessment, and Long-term) guidelines for evaluating innovation in surgery [13] . The trial was approved by Local Research Ethics Committee A of the University College London Hospitals.

2.2. Patient population

Treatment-naïve men recently diagnosed with low-, intermediate-, or high-risk nonmetastatic prostate cancer (prostate-specific antigen [PSA] ≤20 ng/ml, Gleason ≤ 4 + 3, stage ≤T3aN0M0) were eligible ( Fig. 1 ).

Fig. 1 Trial flow chart. mpMRI = multiparametric magnetic resonance imaging; HIFU = high-intensity focused ultrasound; ADT = androgen deprivation therapy.

2.3. Study interventions

2.3.2. Treatment

Focal ablation of the index lesion was performed using transrectal high-intensity focused ultrasound (HIFU; Sonablate 500, Focus Surgery, Indianapolis, IN, USA) (Supplementary materials). Untreated areas could contain secondary small-volume (CCL ≤5 mm) Gleason 3 + 3 disease [14] , high-grade prostate intraepithelial neoplasia, and/or atypical small acinar proliferation ( Fig. 2 ).

Fig. 2 Schematic diagrams demonstrating the types of focal therapy conducted in this trial. Large red areas represent dominant cancers (so-called index lesions) whilst small green areas represent small, low-grade, secondary lesions. Red transparent boxes represent ablation zones on the high-intensity focused ultrasound device: (A) hemiablation, (B) extended (dogleg) ablation; (C) quadrant ablation; (D) focal ablation.

2.3.3. Follow-up

Contrast-enhanced MRI was carried out at 10–14 d to evaluate the area of ablation, demonstrated by a confluent perfusion deficit ( Fig. 3 ). Clinical review at 1, 3, 6, 9, and 12 mo assessed adverse events, serum PSA, and responses to validated questionnaires. Phosphodiesterase-5 inhibitor (PDE5-I) use was permitted at any time point before treatment and during follow-up to aid erectile function. At 6 mo, mpMRI followed by biopsies targeted to the treated area was scheduled, with a minimum sampling requirement of one core for every 1 ml of residual tissue. Repeat treatment using focal HIFU for treated or untreated areas that showed signs of progression was permitted if biopsies were positive. Further mpMRI was carried out at 12 mo in such cases. As the purpose of the 6-mo biopsies was to determine whether the ablation was successful, our ethics committee did not permit transperineal TPM biopsies of untreated tissue because of the requirement for another general anaesthetic. However, biopsies of untreated tissue were permitted if a new lesion suspicious for cancer was seen on mpMRI or if existing untreated tissue showed signs of progression.

Fig. 3 Contrast-enhanced axial scan of the prostate at mid-gland at 1 wk after high-intensity focused ultrasound ablation of an index lesion in the right prostate lobe. Example of right-sided index lesion treatment with confluent hemiablation necrosis demonstrated on early (1-wk) multiparametric magnetic resonance imaging. The orange line indicates the area of ablation with confluent necrosis, in this case hemiablation. The white line indicates the outline of whole prostate. R = rectum.

2.4. Study endpoints

The primary outcome was the composite rate of genitourinary side effects measured using validated patient questionnaires[15], [16], and [17]. Incontinence was defined using a score of 1–3 for question 1 (any leak) and of 1–3 for question 5 (any pad use) of the University of California, Los Angeles (UCLA) Expanded Prostate Cancer Index Composite (EPIC) continence questionnaire. Erectile dysfunction was defined as a score of 0 or 1 for question 2 of the 15-item International Index of Erectile Function (IIEF-15) questionnaire. Urinary function was evaluated using the International Prostate Symptom Score (IPSS), the IPSS-quality of life questionnaire, and the UCLA-EPIC continence questionnaire. Health-related quality of life was measured using the Functional Assessment of Cancer Therapy-Prostate (FACT-P). Responses to the validated questionnaires were collated using standard methods.

Secondary outcomes were histological parameters of cancer control on 6-mo biopsy, absence of clinically significant disease on mpMRI at 6 and 12 mo, description of PSA kinetics following index lesion ablation, and the proportion of men requiring salvage local radical therapy and systemic therapy for progressive prostate cancer.

2.5. Statistical analysis

The sample size was calculated on a composite endpoint defined as the total proportion of men with erectile dysfunction and/or incontinence. It was estimated that baseline incontinence was about 10% and baseline impotence about 40% in men older than 60 yr[18] and [19]. Overall, we estimated that baseline genitourinary dysfunction would be approximately 50%. Were these men to undergo whole-gland therapy, 50% of those with good baseline function would also suffer some form of genitourinary dysfunction [20] , so total genitourinary dysfunction would be approximately 75% at 12 mo if men were to undergo whole-gland therapy. It was hypothesised that focal therapy would give rise to genitourinary dysfunction in only 5% of those with good baseline function. Therefore, we estimated a total genitourinary dysfunction level of approximately 55% at 12 mo after focal index lesion ablation. Using sample size calculations comparing an estimate proportion (0.55) to a known proportion (0.75), with α = 0.05 and power of 90%, the total number of patients required was 56 [21] .

A Wilcoxon signed rank test (two-tailed) was used to evaluate differences between continuous variables (PSA, questionnaire scores) measured at baseline and at the 12-mo follow-up visit. Statistical significance was set atp≤ 0.05. Changes over time were described using box-and-whisker plots. Subgroup analyses were hypothesis-generating, and with small numbers in each subgroup, it was deemed inappropriate to run statistical tests of significance for such comparisons. Statistical analysis was performed using Stata version 11.2 (Stata Corp, College Station, TX, USA). This analysis was carried out on the available data. However, some functional measures were missing at baseline or at 12-mo follow-up for five men, so we conducted two sensitivity analyses to assess if this was likely to impact our results. We repeated the analysis described above for a “best” outcome scenario and a “worst” outcome scenario by setting missing baseline values to the maximum or minimum possible, respectively.

3.1. Study patients

Fifty-six men were recruited (July 27, 2009–January 26, 2011) with a mean age of 64 yr. According to National Comprehensive Cancer Network criteria [22] , seven men (12.5%) had low-risk, 47 (83.9%) had intermediate-risk, and two (3.6%) had high-risk nonmetastatic prostate cancer ( Fig. 1 , Table 1 ).

3.2. Primary outcome

Overall, 40 patients (71.4%; 95% confidence interval [CI] 59.6–83.2%) were leak-free, pad-free continent and had erections sufficient for penetration at baseline. This composite measure decreased to 33 patients (58.9%; 95% CI 46.0–71.8%) at 12 mo.

3.3. Secondary outcomes

3.3.5. Biochemical, imaging, and histological outcomes

A significant decrease in serum PSA from a median baseline value of 7.4 ng/ml (IQR 5.6–9.5) to 2.4 ng/ml (IQR 1.6–4.1) was observed at 12 mo (p < 0.0001; Fig. 3 ).

At 6 mo, two men refused a post-treatment biopsy, and another two had clinical reasons for omitting biopsies (chemotherapy for newly diagnosed lung cancer in one, and a persistent HIFU cavity connecting with the vas deferens on mpMRI in another). For these four men, mpMRI at 6 mo showed no evidence of measurable residual disease. There was no histological evidence of clinically significant disease (Gleason ≤ 3 + 3, CCL ≤3 mm) in the treated area in 44/52 patients (84.6%; 95% CI 70–92%), or of any cancer in the treated area in 34/52 patients (65.4%; 95% CI 52.5–78.3%). Two had clinically significant disease on biopsies from suspicious areas in the untreated side on mpMRI not present at baseline. Overall, histopathological absence of clinically significant cancer (treated and untreated sides) was observed in 42/52 men (80.8%, 95% CI 67.5–90.4%). The overall rate of absence of measurable disease was 86% (48/56) at the study end following repeat treatment in two patients. However, 43% of men had persistent cancer of any threshold at the study end. Two other men underwent salvage treatment with neoadjuvant hormone ablation therapy and external beam radiotherapy for residual Gleason 4 + 3 and Gleason 4 + 4 disease ( Table 3 ).

Table 3 Histology outcomes at 6 mo after focal index lesion ablation in 52 men who underwent post-treatment prostate biopsy

Parameter	Value
Positive biopsy outcomes on the treated side
Median number of biopsies taken, n (IQR)	5.0 (4.0–6.0)
Positive biopsies (any disease), n (%)	18/52 (34.6)
Positive biopsies (clinically significant disease), n (%)	8/52 (15.4) a
Median CCLmax in positive cores, mm (IQR)	2.0 (1.0–4.0)
Gleason score, n (%)
3 + 3	12/18 (66.7)
3 + 4	5/18 (27.8)
4 + 3	1/18 (5.6)
Positive biopsy outcomes on the untreated side (targeted to suspicious areas on 6-mo MRI)
Positive biopsies (any disease), n (%)	4/52 (7.7)
Positive biopsies (clinically significant disease), n (%)	2/52 (3.8) b
Gleason score (n)
3 + 3	3 c
3 + 4	0
4 + 3	0
4 + 4	1
Absence of any cancer on both sides, n (%; 95% CI)	30/52 (57.7; 43.2–71.3)
Overall absence of clinically significant disease on both sides, n (%; 95% CI)	42/52 (80.8; 67.5–90.4)
Other histological findings, n (%; 95% CI) d
Prostatic acini	2/51 (3.9; 0.5–13.5)
Atrophy	20/51 (39.2; 25.8–53.9)
Fibrosis	39/51 (76.5; 62.5–87.2)
Giant-cell reaction	4/51 (7.8; 2.2–18.9)
Necrosis	11/51 (21.6; 11.3–35.3)

^a These men were counted within the “any disease” category. In other words, eight of the 18 men with positive biopsies on the treated side had clinically significant cancer.

^b These men were counted within the “any disease” category. In other words, two of the four men with positive biopsies on the untreated side had clinically significant cancer.

^c CCL_max in these three cases was <1 mm, 1 mm, and 4 mm.

^d Data only available for 51/52 patients.

IQR = interquartile range; CCL_max = maximum cancer core length; MRI = magnetic resonance imaging; CI = confidence interval.

Of 41 patients with good baseline status, 22 (53.7%, 95% CI 37.4–69.3%) achieved the trifecta outcome (absence of clinically significant disease, pad-free, leak-free continence, and erections sufficient for intercourse) at 12 mo ( Fig. 4 ). The results for sensitivity analyses (data not shown) were very similar and did not affect our overall conclusion.

Fig. 4 Summary of continence, erectile function, and cancer control outcomes and the trifecta rate following high-intensity focused ultrasound ablation of the index lesion. Patient-reported trifecta outcomes were collected using validated questionnaires. (1) Percentage continence data were derived using the questions “Over the past 4 weeks how often did you leak urine?” and “Over the past 4 weeks how many pads or adult diapers per day did you usually use to control leakage?” from the urinary domain of the University of California, Los Angeles Expanded Prostate Cancer Index Composite questionnaire. (2) The percentage of men with erections sufficient for penetration was calculated for men scoring ≥2 for the question “Over the past 4 weeks when you had erections with sexual stimulation, how often were your erections hard enough for penetration?” from the 15-item International Index of Erectile Function. (3) The percentage phosphodiesterase-5 inhibitor (PDE5-I) use (tadalafil, sildenafil, or vardenafil) was calculated using the number of men achieving erections sufficient for penetration from part (2) as the denominator.

4.1. Clinical implications

The outcome from other focal therapy studies in which treatment was applied to all known cancer lesions was recently reviewed [9] . Rates of 95–100% for pad-free, leak-free continence preservation, 54–100% for potency, and 83–100% for overall absence of clinically significant cancer were observed after treatment. Most of these studies applied strict selection criteria for disease risk, biopsy cancer burden, and function at baseline, and therefore have limited external validity. In our current study—in a group of men who were not selected by the investigators on the basis of good baseline function—the functional preservation outcomes were not as high as we have previously reported. Indeed, our trifecta rate is also lower than previously reported, and may not be as encouraging when compared to the rates reported after surgery. With greater numbers in our ongoing larger multicentre trials, we may be to identify men who are unlikely to have good functional outcomes after focal therapy and would be better treated using a radical approach. Our study is unique in that known cancer of low malignant potential was left untreated. We recently showed that if such a strategy were to be adopted, more than 90% of newly diagnosed men with localised prostate cancer might be eligible for focal therapy [27] .

Some have argued that because mpMRI and TPM biopsies are not accurate enough to detect all areas of prostate cancer, the concept of selective therapy lacks legitimacy [28] . However, in most other solid organs, cancer is diagnosed when a clinical manifestation of disease (a lump or an imaging signal) is verified histologically. Inevitably, therapy is directed at the clinical phenotype plus the addition of a margin. This was recently seen in the administration of radiotherapy to the breast after lumpectomy. Historically, whole-breast irradiation was favoured because of the known multifocality of breast cancer. The finding that intraoperative local irradiation of the surgical site was not inferior to whole-breast irradiation challenges the idea that recurrence is due to residual subclinical multifocal lesions away from the index lesion [29] . There is also concern that inclusion of intermediate- and high-risk cases for focal therapy may represent undertreatment for those men who harbour microscopic nodal metastases.

It is yet to be determined whether an RCT of focal therapy compared to radical therapy is deliverable [23] . Previous trials have attempted to compare different interventions in localised prostate cancer; however, many have failed because of a lack of physician and patient equipoise. There are additional concerns regarding appropriate endpoints and the follow-up time frame, especially as a minimum of 10-yr follow-up would be required to evaluate metastases and mortality. The UK National Institute of Health Research has recently funded a pilot RCT to determine whether recruitment might be feasible (ISRCTN99760303). This RCT will evaluate rates of transition to local salvage therapies within a 5-yr period between focal therapy and radical prostatectomy in men with intermediate-risk prostate cancer.

Our findings have implications for prostate cancer management. Some have suggested that low-risk disease should be relabelled as a benign entity because of the rarity of progression and metastases from these lesions[10], [11], and [12]. If this hypothesis proves to be true, we could concentrate on targeting measurable disease based on mpMRI considering the high negative predictive values for clinically significant disease that it confers. However, the notion that Gleason 6 disease cannot metastasise has recently been brought into question following the findings of Haffner et al [30] , who reported on a case in which a small area of Gleason 6 disease within a tumour of very high-grade cancer was responsible for a distant metastasis. There has been considerable debate regarding the validity of these findings [31] .

Our study also represents anin vivohuman model for better understanding of the determinants of progression of low-grade disease in a manner that is not predicated on removing the whole organ. This might allow us to determine if such lesions do progress and, if so, whether they can be identified at the time of diagnosis using detailed molecular typing approaches such as immunohistochemistry and -omics techniques.

2.1. Study design and conduct

Our single-centre study was a prospective development study according to the IDEAL (Idea, Development, Exploration, Assessment, and Long-term) guidelines for evaluating innovation in surgery [13] . The trial was approved by Local Research Ethics Committee A of the University College London Hospitals.

2.2. Patient population

Treatment-naïve men recently diagnosed with low-, intermediate-, or high-risk nonmetastatic prostate cancer (prostate-specific antigen [PSA] ≤20 ng/ml, Gleason ≤ 4 + 3, stage ≤T3aN0M0) were eligible ( Fig. 1 ).

Fig. 1 Trial flow chart. mpMRI = multiparametric magnetic resonance imaging; HIFU = high-intensity focused ultrasound; ADT = androgen deprivation therapy.

2.3. Study interventions

2.3.2. Treatment

Focal ablation of the index lesion was performed using transrectal high-intensity focused ultrasound (HIFU; Sonablate 500, Focus Surgery, Indianapolis, IN, USA) (Supplementary materials). Untreated areas could contain secondary small-volume (CCL ≤5 mm) Gleason 3 + 3 disease [14] , high-grade prostate intraepithelial neoplasia, and/or atypical small acinar proliferation ( Fig. 2 ).

Fig. 2 Schematic diagrams demonstrating the types of focal therapy conducted in this trial. Large red areas represent dominant cancers (so-called index lesions) whilst small green areas represent small, low-grade, secondary lesions. Red transparent boxes represent ablation zones on the high-intensity focused ultrasound device: (A) hemiablation, (B) extended (dogleg) ablation; (C) quadrant ablation; (D) focal ablation.

2.3.3. Follow-up

Contrast-enhanced MRI was carried out at 10–14 d to evaluate the area of ablation, demonstrated by a confluent perfusion deficit ( Fig. 3 ). Clinical review at 1, 3, 6, 9, and 12 mo assessed adverse events, serum PSA, and responses to validated questionnaires. Phosphodiesterase-5 inhibitor (PDE5-I) use was permitted at any time point before treatment and during follow-up to aid erectile function. At 6 mo, mpMRI followed by biopsies targeted to the treated area was scheduled, with a minimum sampling requirement of one core for every 1 ml of residual tissue. Repeat treatment using focal HIFU for treated or untreated areas that showed signs of progression was permitted if biopsies were positive. Further mpMRI was carried out at 12 mo in such cases. As the purpose of the 6-mo biopsies was to determine whether the ablation was successful, our ethics committee did not permit transperineal TPM biopsies of untreated tissue because of the requirement for another general anaesthetic. However, biopsies of untreated tissue were permitted if a new lesion suspicious for cancer was seen on mpMRI or if existing untreated tissue showed signs of progression.

Fig. 3 Contrast-enhanced axial scan of the prostate at mid-gland at 1 wk after high-intensity focused ultrasound ablation of an index lesion in the right prostate lobe. Example of right-sided index lesion treatment with confluent hemiablation necrosis demonstrated on early (1-wk) multiparametric magnetic resonance imaging. The orange line indicates the area of ablation with confluent necrosis, in this case hemiablation. The white line indicates the outline of whole prostate. R = rectum.

2.4. Study endpoints

The primary outcome was the composite rate of genitourinary side effects measured using validated patient questionnaires[15], [16], and [17]. Incontinence was defined using a score of 1–3 for question 1 (any leak) and of 1–3 for question 5 (any pad use) of the University of California, Los Angeles (UCLA) Expanded Prostate Cancer Index Composite (EPIC) continence questionnaire. Erectile dysfunction was defined as a score of 0 or 1 for question 2 of the 15-item International Index of Erectile Function (IIEF-15) questionnaire. Urinary function was evaluated using the International Prostate Symptom Score (IPSS), the IPSS-quality of life questionnaire, and the UCLA-EPIC continence questionnaire. Health-related quality of life was measured using the Functional Assessment of Cancer Therapy-Prostate (FACT-P). Responses to the validated questionnaires were collated using standard methods.

Secondary outcomes were histological parameters of cancer control on 6-mo biopsy, absence of clinically significant disease on mpMRI at 6 and 12 mo, description of PSA kinetics following index lesion ablation, and the proportion of men requiring salvage local radical therapy and systemic therapy for progressive prostate cancer.

2.5. Statistical analysis

The sample size was calculated on a composite endpoint defined as the total proportion of men with erectile dysfunction and/or incontinence. It was estimated that baseline incontinence was about 10% and baseline impotence about 40% in men older than 60 yr[18] and [19]. Overall, we estimated that baseline genitourinary dysfunction would be approximately 50%. Were these men to undergo whole-gland therapy, 50% of those with good baseline function would also suffer some form of genitourinary dysfunction [20] , so total genitourinary dysfunction would be approximately 75% at 12 mo if men were to undergo whole-gland therapy. It was hypothesised that focal therapy would give rise to genitourinary dysfunction in only 5% of those with good baseline function. Therefore, we estimated a total genitourinary dysfunction level of approximately 55% at 12 mo after focal index lesion ablation. Using sample size calculations comparing an estimate proportion (0.55) to a known proportion (0.75), with α = 0.05 and power of 90%, the total number of patients required was 56 [21] .

A Wilcoxon signed rank test (two-tailed) was used to evaluate differences between continuous variables (PSA, questionnaire scores) measured at baseline and at the 12-mo follow-up visit. Statistical significance was set atp≤ 0.05. Changes over time were described using box-and-whisker plots. Subgroup analyses were hypothesis-generating, and with small numbers in each subgroup, it was deemed inappropriate to run statistical tests of significance for such comparisons. Statistical analysis was performed using Stata version 11.2 (Stata Corp, College Station, TX, USA). This analysis was carried out on the available data. However, some functional measures were missing at baseline or at 12-mo follow-up for five men, so we conducted two sensitivity analyses to assess if this was likely to impact our results. We repeated the analysis described above for a “best” outcome scenario and a “worst” outcome scenario by setting missing baseline values to the maximum or minimum possible, respectively.

3.1. Study patients

Fifty-six men were recruited (July 27, 2009–January 26, 2011) with a mean age of 64 yr. According to National Comprehensive Cancer Network criteria [22] , seven men (12.5%) had low-risk, 47 (83.9%) had intermediate-risk, and two (3.6%) had high-risk nonmetastatic prostate cancer ( Fig. 1 , Table 1 ).

3.2. Primary outcome

Overall, 40 patients (71.4%; 95% confidence interval [CI] 59.6–83.2%) were leak-free, pad-free continent and had erections sufficient for penetration at baseline. This composite measure decreased to 33 patients (58.9%; 95% CI 46.0–71.8%) at 12 mo.

3.3. Secondary outcomes

3.3.5. Biochemical, imaging, and histological outcomes

A significant decrease in serum PSA from a median baseline value of 7.4 ng/ml (IQR 5.6–9.5) to 2.4 ng/ml (IQR 1.6–4.1) was observed at 12 mo (p < 0.0001; Fig. 3 ).

At 6 mo, two men refused a post-treatment biopsy, and another two had clinical reasons for omitting biopsies (chemotherapy for newly diagnosed lung cancer in one, and a persistent HIFU cavity connecting with the vas deferens on mpMRI in another). For these four men, mpMRI at 6 mo showed no evidence of measurable residual disease. There was no histological evidence of clinically significant disease (Gleason ≤ 3 + 3, CCL ≤3 mm) in the treated area in 44/52 patients (84.6%; 95% CI 70–92%), or of any cancer in the treated area in 34/52 patients (65.4%; 95% CI 52.5–78.3%). Two had clinically significant disease on biopsies from suspicious areas in the untreated side on mpMRI not present at baseline. Overall, histopathological absence of clinically significant cancer (treated and untreated sides) was observed in 42/52 men (80.8%, 95% CI 67.5–90.4%). The overall rate of absence of measurable disease was 86% (48/56) at the study end following repeat treatment in two patients. However, 43% of men had persistent cancer of any threshold at the study end. Two other men underwent salvage treatment with neoadjuvant hormone ablation therapy and external beam radiotherapy for residual Gleason 4 + 3 and Gleason 4 + 4 disease ( Table 3 ).

Table 3 Histology outcomes at 6 mo after focal index lesion ablation in 52 men who underwent post-treatment prostate biopsy

Parameter	Value
Positive biopsy outcomes on the treated side
Median number of biopsies taken, n (IQR)	5.0 (4.0–6.0)
Positive biopsies (any disease), n (%)	18/52 (34.6)
Positive biopsies (clinically significant disease), n (%)	8/52 (15.4) a
Median CCLmax in positive cores, mm (IQR)	2.0 (1.0–4.0)
Gleason score, n (%)
3 + 3	12/18 (66.7)
3 + 4	5/18 (27.8)
4 + 3	1/18 (5.6)
Positive biopsy outcomes on the untreated side (targeted to suspicious areas on 6-mo MRI)
Positive biopsies (any disease), n (%)	4/52 (7.7)
Positive biopsies (clinically significant disease), n (%)	2/52 (3.8) b
Gleason score (n)
3 + 3	3 c
3 + 4	0
4 + 3	0
4 + 4	1
Absence of any cancer on both sides, n (%; 95% CI)	30/52 (57.7; 43.2–71.3)
Overall absence of clinically significant disease on both sides, n (%; 95% CI)	42/52 (80.8; 67.5–90.4)
Other histological findings, n (%; 95% CI) d
Prostatic acini	2/51 (3.9; 0.5–13.5)
Atrophy	20/51 (39.2; 25.8–53.9)
Fibrosis	39/51 (76.5; 62.5–87.2)
Giant-cell reaction	4/51 (7.8; 2.2–18.9)
Necrosis	11/51 (21.6; 11.3–35.3)

^a These men were counted within the “any disease” category. In other words, eight of the 18 men with positive biopsies on the treated side had clinically significant cancer.

^b These men were counted within the “any disease” category. In other words, two of the four men with positive biopsies on the untreated side had clinically significant cancer.

^c CCL_max in these three cases was <1 mm, 1 mm, and 4 mm.

^d Data only available for 51/52 patients.

IQR = interquartile range; CCL_max = maximum cancer core length; MRI = magnetic resonance imaging; CI = confidence interval.

Of 41 patients with good baseline status, 22 (53.7%, 95% CI 37.4–69.3%) achieved the trifecta outcome (absence of clinically significant disease, pad-free, leak-free continence, and erections sufficient for intercourse) at 12 mo ( Fig. 4 ). The results for sensitivity analyses (data not shown) were very similar and did not affect our overall conclusion.

Fig. 4 Summary of continence, erectile function, and cancer control outcomes and the trifecta rate following high-intensity focused ultrasound ablation of the index lesion. Patient-reported trifecta outcomes were collected using validated questionnaires. (1) Percentage continence data were derived using the questions “Over the past 4 weeks how often did you leak urine?” and “Over the past 4 weeks how many pads or adult diapers per day did you usually use to control leakage?” from the urinary domain of the University of California, Los Angeles Expanded Prostate Cancer Index Composite questionnaire. (2) The percentage of men with erections sufficient for penetration was calculated for men scoring ≥2 for the question “Over the past 4 weeks when you had erections with sexual stimulation, how often were your erections hard enough for penetration?” from the 15-item International Index of Erectile Function. (3) The percentage phosphodiesterase-5 inhibitor (PDE5-I) use (tadalafil, sildenafil, or vardenafil) was calculated using the number of men achieving erections sufficient for penetration from part (2) as the denominator.

4.1. Clinical implications

The outcome from other focal therapy studies in which treatment was applied to all known cancer lesions was recently reviewed [9] . Rates of 95–100% for pad-free, leak-free continence preservation, 54–100% for potency, and 83–100% for overall absence of clinically significant cancer were observed after treatment. Most of these studies applied strict selection criteria for disease risk, biopsy cancer burden, and function at baseline, and therefore have limited external validity. In our current study—in a group of men who were not selected by the investigators on the basis of good baseline function—the functional preservation outcomes were not as high as we have previously reported. Indeed, our trifecta rate is also lower than previously reported, and may not be as encouraging when compared to the rates reported after surgery. With greater numbers in our ongoing larger multicentre trials, we may be to identify men who are unlikely to have good functional outcomes after focal therapy and would be better treated using a radical approach. Our study is unique in that known cancer of low malignant potential was left untreated. We recently showed that if such a strategy were to be adopted, more than 90% of newly diagnosed men with localised prostate cancer might be eligible for focal therapy [27] .

Some have argued that because mpMRI and TPM biopsies are not accurate enough to detect all areas of prostate cancer, the concept of selective therapy lacks legitimacy [28] . However, in most other solid organs, cancer is diagnosed when a clinical manifestation of disease (a lump or an imaging signal) is verified histologically. Inevitably, therapy is directed at the clinical phenotype plus the addition of a margin. This was recently seen in the administration of radiotherapy to the breast after lumpectomy. Historically, whole-breast irradiation was favoured because of the known multifocality of breast cancer. The finding that intraoperative local irradiation of the surgical site was not inferior to whole-breast irradiation challenges the idea that recurrence is due to residual subclinical multifocal lesions away from the index lesion [29] . There is also concern that inclusion of intermediate- and high-risk cases for focal therapy may represent undertreatment for those men who harbour microscopic nodal metastases.

It is yet to be determined whether an RCT of focal therapy compared to radical therapy is deliverable [23] . Previous trials have attempted to compare different interventions in localised prostate cancer; however, many have failed because of a lack of physician and patient equipoise. There are additional concerns regarding appropriate endpoints and the follow-up time frame, especially as a minimum of 10-yr follow-up would be required to evaluate metastases and mortality. The UK National Institute of Health Research has recently funded a pilot RCT to determine whether recruitment might be feasible (ISRCTN99760303). This RCT will evaluate rates of transition to local salvage therapies within a 5-yr period between focal therapy and radical prostatectomy in men with intermediate-risk prostate cancer.

Our findings have implications for prostate cancer management. Some have suggested that low-risk disease should be relabelled as a benign entity because of the rarity of progression and metastases from these lesions[10], [11], and [12]. If this hypothesis proves to be true, we could concentrate on targeting measurable disease based on mpMRI considering the high negative predictive values for clinically significant disease that it confers. However, the notion that Gleason 6 disease cannot metastasise has recently been brought into question following the findings of Haffner et al [30] , who reported on a case in which a small area of Gleason 6 disease within a tumour of very high-grade cancer was responsible for a distant metastasis. There has been considerable debate regarding the validity of these findings [31] .

Our study also represents anin vivohuman model for better understanding of the determinants of progression of low-grade disease in a manner that is not predicated on removing the whole organ. This might allow us to determine if such lesions do progress and, if so, whether they can be identified at the time of diagnosis using detailed molecular typing approaches such as immunohistochemistry and -omics techniques.

2.1. Study design and conduct

Our single-centre study was a prospective development study according to the IDEAL (Idea, Development, Exploration, Assessment, and Long-term) guidelines for evaluating innovation in surgery [13] . The trial was approved by Local Research Ethics Committee A of the University College London Hospitals.

2.2. Patient population

Treatment-naïve men recently diagnosed with low-, intermediate-, or high-risk nonmetastatic prostate cancer (prostate-specific antigen [PSA] ≤20 ng/ml, Gleason ≤ 4 + 3, stage ≤T3aN0M0) were eligible ( Fig. 1 ).

Fig. 1 Trial flow chart. mpMRI = multiparametric magnetic resonance imaging; HIFU = high-intensity focused ultrasound; ADT = androgen deprivation therapy.

2.3. Study interventions

2.3.2. Treatment

Focal ablation of the index lesion was performed using transrectal high-intensity focused ultrasound (HIFU; Sonablate 500, Focus Surgery, Indianapolis, IN, USA) (Supplementary materials). Untreated areas could contain secondary small-volume (CCL ≤5 mm) Gleason 3 + 3 disease [14] , high-grade prostate intraepithelial neoplasia, and/or atypical small acinar proliferation ( Fig. 2 ).

Fig. 2 Schematic diagrams demonstrating the types of focal therapy conducted in this trial. Large red areas represent dominant cancers (so-called index lesions) whilst small green areas represent small, low-grade, secondary lesions. Red transparent boxes represent ablation zones on the high-intensity focused ultrasound device: (A) hemiablation, (B) extended (dogleg) ablation; (C) quadrant ablation; (D) focal ablation.

2.3.3. Follow-up

Contrast-enhanced MRI was carried out at 10–14 d to evaluate the area of ablation, demonstrated by a confluent perfusion deficit ( Fig. 3 ). Clinical review at 1, 3, 6, 9, and 12 mo assessed adverse events, serum PSA, and responses to validated questionnaires. Phosphodiesterase-5 inhibitor (PDE5-I) use was permitted at any time point before treatment and during follow-up to aid erectile function. At 6 mo, mpMRI followed by biopsies targeted to the treated area was scheduled, with a minimum sampling requirement of one core for every 1 ml of residual tissue. Repeat treatment using focal HIFU for treated or untreated areas that showed signs of progression was permitted if biopsies were positive. Further mpMRI was carried out at 12 mo in such cases. As the purpose of the 6-mo biopsies was to determine whether the ablation was successful, our ethics committee did not permit transperineal TPM biopsies of untreated tissue because of the requirement for another general anaesthetic. However, biopsies of untreated tissue were permitted if a new lesion suspicious for cancer was seen on mpMRI or if existing untreated tissue showed signs of progression.

Fig. 3 Contrast-enhanced axial scan of the prostate at mid-gland at 1 wk after high-intensity focused ultrasound ablation of an index lesion in the right prostate lobe. Example of right-sided index lesion treatment with confluent hemiablation necrosis demonstrated on early (1-wk) multiparametric magnetic resonance imaging. The orange line indicates the area of ablation with confluent necrosis, in this case hemiablation. The white line indicates the outline of whole prostate. R = rectum.

2.4. Study endpoints

The primary outcome was the composite rate of genitourinary side effects measured using validated patient questionnaires[15], [16], and [17]. Incontinence was defined using a score of 1–3 for question 1 (any leak) and of 1–3 for question 5 (any pad use) of the University of California, Los Angeles (UCLA) Expanded Prostate Cancer Index Composite (EPIC) continence questionnaire. Erectile dysfunction was defined as a score of 0 or 1 for question 2 of the 15-item International Index of Erectile Function (IIEF-15) questionnaire. Urinary function was evaluated using the International Prostate Symptom Score (IPSS), the IPSS-quality of life questionnaire, and the UCLA-EPIC continence questionnaire. Health-related quality of life was measured using the Functional Assessment of Cancer Therapy-Prostate (FACT-P). Responses to the validated questionnaires were collated using standard methods.

Secondary outcomes were histological parameters of cancer control on 6-mo biopsy, absence of clinically significant disease on mpMRI at 6 and 12 mo, description of PSA kinetics following index lesion ablation, and the proportion of men requiring salvage local radical therapy and systemic therapy for progressive prostate cancer.

2.5. Statistical analysis

The sample size was calculated on a composite endpoint defined as the total proportion of men with erectile dysfunction and/or incontinence. It was estimated that baseline incontinence was about 10% and baseline impotence about 40% in men older than 60 yr[18] and [19]. Overall, we estimated that baseline genitourinary dysfunction would be approximately 50%. Were these men to undergo whole-gland therapy, 50% of those with good baseline function would also suffer some form of genitourinary dysfunction [20] , so total genitourinary dysfunction would be approximately 75% at 12 mo if men were to undergo whole-gland therapy. It was hypothesised that focal therapy would give rise to genitourinary dysfunction in only 5% of those with good baseline function. Therefore, we estimated a total genitourinary dysfunction level of approximately 55% at 12 mo after focal index lesion ablation. Using sample size calculations comparing an estimate proportion (0.55) to a known proportion (0.75), with α = 0.05 and power of 90%, the total number of patients required was 56 [21] .

A Wilcoxon signed rank test (two-tailed) was used to evaluate differences between continuous variables (PSA, questionnaire scores) measured at baseline and at the 12-mo follow-up visit. Statistical significance was set atp≤ 0.05. Changes over time were described using box-and-whisker plots. Subgroup analyses were hypothesis-generating, and with small numbers in each subgroup, it was deemed inappropriate to run statistical tests of significance for such comparisons. Statistical analysis was performed using Stata version 11.2 (Stata Corp, College Station, TX, USA). This analysis was carried out on the available data. However, some functional measures were missing at baseline or at 12-mo follow-up for five men, so we conducted two sensitivity analyses to assess if this was likely to impact our results. We repeated the analysis described above for a “best” outcome scenario and a “worst” outcome scenario by setting missing baseline values to the maximum or minimum possible, respectively.

3.1. Study patients

Fifty-six men were recruited (July 27, 2009–January 26, 2011) with a mean age of 64 yr. According to National Comprehensive Cancer Network criteria [22] , seven men (12.5%) had low-risk, 47 (83.9%) had intermediate-risk, and two (3.6%) had high-risk nonmetastatic prostate cancer ( Fig. 1 , Table 1 ).

3.2. Primary outcome

Overall, 40 patients (71.4%; 95% confidence interval [CI] 59.6–83.2%) were leak-free, pad-free continent and had erections sufficient for penetration at baseline. This composite measure decreased to 33 patients (58.9%; 95% CI 46.0–71.8%) at 12 mo.

3.3. Secondary outcomes

3.3.5. Biochemical, imaging, and histological outcomes

A significant decrease in serum PSA from a median baseline value of 7.4 ng/ml (IQR 5.6–9.5) to 2.4 ng/ml (IQR 1.6–4.1) was observed at 12 mo (p < 0.0001; Fig. 3 ).

At 6 mo, two men refused a post-treatment biopsy, and another two had clinical reasons for omitting biopsies (chemotherapy for newly diagnosed lung cancer in one, and a persistent HIFU cavity connecting with the vas deferens on mpMRI in another). For these four men, mpMRI at 6 mo showed no evidence of measurable residual disease. There was no histological evidence of clinically significant disease (Gleason ≤ 3 + 3, CCL ≤3 mm) in the treated area in 44/52 patients (84.6%; 95% CI 70–92%), or of any cancer in the treated area in 34/52 patients (65.4%; 95% CI 52.5–78.3%). Two had clinically significant disease on biopsies from suspicious areas in the untreated side on mpMRI not present at baseline. Overall, histopathological absence of clinically significant cancer (treated and untreated sides) was observed in 42/52 men (80.8%, 95% CI 67.5–90.4%). The overall rate of absence of measurable disease was 86% (48/56) at the study end following repeat treatment in two patients. However, 43% of men had persistent cancer of any threshold at the study end. Two other men underwent salvage treatment with neoadjuvant hormone ablation therapy and external beam radiotherapy for residual Gleason 4 + 3 and Gleason 4 + 4 disease ( Table 3 ).

Table 3 Histology outcomes at 6 mo after focal index lesion ablation in 52 men who underwent post-treatment prostate biopsy

Parameter	Value
Positive biopsy outcomes on the treated side
Median number of biopsies taken, n (IQR)	5.0 (4.0–6.0)
Positive biopsies (any disease), n (%)	18/52 (34.6)
Positive biopsies (clinically significant disease), n (%)	8/52 (15.4) a
Median CCLmax in positive cores, mm (IQR)	2.0 (1.0–4.0)
Gleason score, n (%)
3 + 3	12/18 (66.7)
3 + 4	5/18 (27.8)
4 + 3	1/18 (5.6)
Positive biopsy outcomes on the untreated side (targeted to suspicious areas on 6-mo MRI)
Positive biopsies (any disease), n (%)	4/52 (7.7)
Positive biopsies (clinically significant disease), n (%)	2/52 (3.8) b
Gleason score (n)
3 + 3	3 c
3 + 4	0
4 + 3	0
4 + 4	1
Absence of any cancer on both sides, n (%; 95% CI)	30/52 (57.7; 43.2–71.3)
Overall absence of clinically significant disease on both sides, n (%; 95% CI)	42/52 (80.8; 67.5–90.4)
Other histological findings, n (%; 95% CI) d
Prostatic acini	2/51 (3.9; 0.5–13.5)
Atrophy	20/51 (39.2; 25.8–53.9)
Fibrosis	39/51 (76.5; 62.5–87.2)
Giant-cell reaction	4/51 (7.8; 2.2–18.9)
Necrosis	11/51 (21.6; 11.3–35.3)

^a These men were counted within the “any disease” category. In other words, eight of the 18 men with positive biopsies on the treated side had clinically significant cancer.

^b These men were counted within the “any disease” category. In other words, two of the four men with positive biopsies on the untreated side had clinically significant cancer.

^c CCL_max in these three cases was <1 mm, 1 mm, and 4 mm.

^d Data only available for 51/52 patients.

IQR = interquartile range; CCL_max = maximum cancer core length; MRI = magnetic resonance imaging; CI = confidence interval.

Of 41 patients with good baseline status, 22 (53.7%, 95% CI 37.4–69.3%) achieved the trifecta outcome (absence of clinically significant disease, pad-free, leak-free continence, and erections sufficient for intercourse) at 12 mo ( Fig. 4 ). The results for sensitivity analyses (data not shown) were very similar and did not affect our overall conclusion.

Fig. 4 Summary of continence, erectile function, and cancer control outcomes and the trifecta rate following high-intensity focused ultrasound ablation of the index lesion. Patient-reported trifecta outcomes were collected using validated questionnaires. (1) Percentage continence data were derived using the questions “Over the past 4 weeks how often did you leak urine?” and “Over the past 4 weeks how many pads or adult diapers per day did you usually use to control leakage?” from the urinary domain of the University of California, Los Angeles Expanded Prostate Cancer Index Composite questionnaire. (2) The percentage of men with erections sufficient for penetration was calculated for men scoring ≥2 for the question “Over the past 4 weeks when you had erections with sexual stimulation, how often were your erections hard enough for penetration?” from the 15-item International Index of Erectile Function. (3) The percentage phosphodiesterase-5 inhibitor (PDE5-I) use (tadalafil, sildenafil, or vardenafil) was calculated using the number of men achieving erections sufficient for penetration from part (2) as the denominator.

4.1. Clinical implications

The outcome from other focal therapy studies in which treatment was applied to all known cancer lesions was recently reviewed [9] . Rates of 95–100% for pad-free, leak-free continence preservation, 54–100% for potency, and 83–100% for overall absence of clinically significant cancer were observed after treatment. Most of these studies applied strict selection criteria for disease risk, biopsy cancer burden, and function at baseline, and therefore have limited external validity. In our current study—in a group of men who were not selected by the investigators on the basis of good baseline function—the functional preservation outcomes were not as high as we have previously reported. Indeed, our trifecta rate is also lower than previously reported, and may not be as encouraging when compared to the rates reported after surgery. With greater numbers in our ongoing larger multicentre trials, we may be to identify men who are unlikely to have good functional outcomes after focal therapy and would be better treated using a radical approach. Our study is unique in that known cancer of low malignant potential was left untreated. We recently showed that if such a strategy were to be adopted, more than 90% of newly diagnosed men with localised prostate cancer might be eligible for focal therapy [27] .

Some have argued that because mpMRI and TPM biopsies are not accurate enough to detect all areas of prostate cancer, the concept of selective therapy lacks legitimacy [28] . However, in most other solid organs, cancer is diagnosed when a clinical manifestation of disease (a lump or an imaging signal) is verified histologically. Inevitably, therapy is directed at the clinical phenotype plus the addition of a margin. This was recently seen in the administration of radiotherapy to the breast after lumpectomy. Historically, whole-breast irradiation was favoured because of the known multifocality of breast cancer. The finding that intraoperative local irradiation of the surgical site was not inferior to whole-breast irradiation challenges the idea that recurrence is due to residual subclinical multifocal lesions away from the index lesion [29] . There is also concern that inclusion of intermediate- and high-risk cases for focal therapy may represent undertreatment for those men who harbour microscopic nodal metastases.

It is yet to be determined whether an RCT of focal therapy compared to radical therapy is deliverable [23] . Previous trials have attempted to compare different interventions in localised prostate cancer; however, many have failed because of a lack of physician and patient equipoise. There are additional concerns regarding appropriate endpoints and the follow-up time frame, especially as a minimum of 10-yr follow-up would be required to evaluate metastases and mortality. The UK National Institute of Health Research has recently funded a pilot RCT to determine whether recruitment might be feasible (ISRCTN99760303). This RCT will evaluate rates of transition to local salvage therapies within a 5-yr period between focal therapy and radical prostatectomy in men with intermediate-risk prostate cancer.

Our findings have implications for prostate cancer management. Some have suggested that low-risk disease should be relabelled as a benign entity because of the rarity of progression and metastases from these lesions[10], [11], and [12]. If this hypothesis proves to be true, we could concentrate on targeting measurable disease based on mpMRI considering the high negative predictive values for clinically significant disease that it confers. However, the notion that Gleason 6 disease cannot metastasise has recently been brought into question following the findings of Haffner et al [30] , who reported on a case in which a small area of Gleason 6 disease within a tumour of very high-grade cancer was responsible for a distant metastasis. There has been considerable debate regarding the validity of these findings [31] .

Our study also represents anin vivohuman model for better understanding of the determinants of progression of low-grade disease in a manner that is not predicated on removing the whole organ. This might allow us to determine if such lesions do progress and, if so, whether they can be identified at the time of diagnosis using detailed molecular typing approaches such as immunohistochemistry and -omics techniques.

2.1. Study design and conduct

Our single-centre study was a prospective development study according to the IDEAL (Idea, Development, Exploration, Assessment, and Long-term) guidelines for evaluating innovation in surgery [13] . The trial was approved by Local Research Ethics Committee A of the University College London Hospitals.

2.2. Patient population

Treatment-naïve men recently diagnosed with low-, intermediate-, or high-risk nonmetastatic prostate cancer (prostate-specific antigen [PSA] ≤20 ng/ml, Gleason ≤ 4 + 3, stage ≤T3aN0M0) were eligible ( Fig. 1 ).

Fig. 1 Trial flow chart. mpMRI = multiparametric magnetic resonance imaging; HIFU = high-intensity focused ultrasound; ADT = androgen deprivation therapy.

2.3. Study interventions

2.3.2. Treatment

Focal ablation of the index lesion was performed using transrectal high-intensity focused ultrasound (HIFU; Sonablate 500, Focus Surgery, Indianapolis, IN, USA) (Supplementary materials). Untreated areas could contain secondary small-volume (CCL ≤5 mm) Gleason 3 + 3 disease [14] , high-grade prostate intraepithelial neoplasia, and/or atypical small acinar proliferation ( Fig. 2 ).

Fig. 2 Schematic diagrams demonstrating the types of focal therapy conducted in this trial. Large red areas represent dominant cancers (so-called index lesions) whilst small green areas represent small, low-grade, secondary lesions. Red transparent boxes represent ablation zones on the high-intensity focused ultrasound device: (A) hemiablation, (B) extended (dogleg) ablation; (C) quadrant ablation; (D) focal ablation.

2.3.3. Follow-up

Contrast-enhanced MRI was carried out at 10–14 d to evaluate the area of ablation, demonstrated by a confluent perfusion deficit ( Fig. 3 ). Clinical review at 1, 3, 6, 9, and 12 mo assessed adverse events, serum PSA, and responses to validated questionnaires. Phosphodiesterase-5 inhibitor (PDE5-I) use was permitted at any time point before treatment and during follow-up to aid erectile function. At 6 mo, mpMRI followed by biopsies targeted to the treated area was scheduled, with a minimum sampling requirement of one core for every 1 ml of residual tissue. Repeat treatment using focal HIFU for treated or untreated areas that showed signs of progression was permitted if biopsies were positive. Further mpMRI was carried out at 12 mo in such cases. As the purpose of the 6-mo biopsies was to determine whether the ablation was successful, our ethics committee did not permit transperineal TPM biopsies of untreated tissue because of the requirement for another general anaesthetic. However, biopsies of untreated tissue were permitted if a new lesion suspicious for cancer was seen on mpMRI or if existing untreated tissue showed signs of progression.

Fig. 3 Contrast-enhanced axial scan of the prostate at mid-gland at 1 wk after high-intensity focused ultrasound ablation of an index lesion in the right prostate lobe. Example of right-sided index lesion treatment with confluent hemiablation necrosis demonstrated on early (1-wk) multiparametric magnetic resonance imaging. The orange line indicates the area of ablation with confluent necrosis, in this case hemiablation. The white line indicates the outline of whole prostate. R = rectum.

2.4. Study endpoints

The primary outcome was the composite rate of genitourinary side effects measured using validated patient questionnaires[15], [16], and [17]. Incontinence was defined using a score of 1–3 for question 1 (any leak) and of 1–3 for question 5 (any pad use) of the University of California, Los Angeles (UCLA) Expanded Prostate Cancer Index Composite (EPIC) continence questionnaire. Erectile dysfunction was defined as a score of 0 or 1 for question 2 of the 15-item International Index of Erectile Function (IIEF-15) questionnaire. Urinary function was evaluated using the International Prostate Symptom Score (IPSS), the IPSS-quality of life questionnaire, and the UCLA-EPIC continence questionnaire. Health-related quality of life was measured using the Functional Assessment of Cancer Therapy-Prostate (FACT-P). Responses to the validated questionnaires were collated using standard methods.

Secondary outcomes were histological parameters of cancer control on 6-mo biopsy, absence of clinically significant disease on mpMRI at 6 and 12 mo, description of PSA kinetics following index lesion ablation, and the proportion of men requiring salvage local radical therapy and systemic therapy for progressive prostate cancer.

2.5. Statistical analysis

The sample size was calculated on a composite endpoint defined as the total proportion of men with erectile dysfunction and/or incontinence. It was estimated that baseline incontinence was about 10% and baseline impotence about 40% in men older than 60 yr[18] and [19]. Overall, we estimated that baseline genitourinary dysfunction would be approximately 50%. Were these men to undergo whole-gland therapy, 50% of those with good baseline function would also suffer some form of genitourinary dysfunction [20] , so total genitourinary dysfunction would be approximately 75% at 12 mo if men were to undergo whole-gland therapy. It was hypothesised that focal therapy would give rise to genitourinary dysfunction in only 5% of those with good baseline function. Therefore, we estimated a total genitourinary dysfunction level of approximately 55% at 12 mo after focal index lesion ablation. Using sample size calculations comparing an estimate proportion (0.55) to a known proportion (0.75), with α = 0.05 and power of 90%, the total number of patients required was 56 [21] .

A Wilcoxon signed rank test (two-tailed) was used to evaluate differences between continuous variables (PSA, questionnaire scores) measured at baseline and at the 12-mo follow-up visit. Statistical significance was set atp≤ 0.05. Changes over time were described using box-and-whisker plots. Subgroup analyses were hypothesis-generating, and with small numbers in each subgroup, it was deemed inappropriate to run statistical tests of significance for such comparisons. Statistical analysis was performed using Stata version 11.2 (Stata Corp, College Station, TX, USA). This analysis was carried out on the available data. However, some functional measures were missing at baseline or at 12-mo follow-up for five men, so we conducted two sensitivity analyses to assess if this was likely to impact our results. We repeated the analysis described above for a “best” outcome scenario and a “worst” outcome scenario by setting missing baseline values to the maximum or minimum possible, respectively.

3.1. Study patients

Fifty-six men were recruited (July 27, 2009–January 26, 2011) with a mean age of 64 yr. According to National Comprehensive Cancer Network criteria [22] , seven men (12.5%) had low-risk, 47 (83.9%) had intermediate-risk, and two (3.6%) had high-risk nonmetastatic prostate cancer ( Fig. 1 , Table 1 ).

3.2. Primary outcome

Overall, 40 patients (71.4%; 95% confidence interval [CI] 59.6–83.2%) were leak-free, pad-free continent and had erections sufficient for penetration at baseline. This composite measure decreased to 33 patients (58.9%; 95% CI 46.0–71.8%) at 12 mo.

3.3. Secondary outcomes

3.3.5. Biochemical, imaging, and histological outcomes

A significant decrease in serum PSA from a median baseline value of 7.4 ng/ml (IQR 5.6–9.5) to 2.4 ng/ml (IQR 1.6–4.1) was observed at 12 mo (p < 0.0001; Fig. 3 ).

At 6 mo, two men refused a post-treatment biopsy, and another two had clinical reasons for omitting biopsies (chemotherapy for newly diagnosed lung cancer in one, and a persistent HIFU cavity connecting with the vas deferens on mpMRI in another). For these four men, mpMRI at 6 mo showed no evidence of measurable residual disease. There was no histological evidence of clinically significant disease (Gleason ≤ 3 + 3, CCL ≤3 mm) in the treated area in 44/52 patients (84.6%; 95% CI 70–92%), or of any cancer in the treated area in 34/52 patients (65.4%; 95% CI 52.5–78.3%). Two had clinically significant disease on biopsies from suspicious areas in the untreated side on mpMRI not present at baseline. Overall, histopathological absence of clinically significant cancer (treated and untreated sides) was observed in 42/52 men (80.8%, 95% CI 67.5–90.4%). The overall rate of absence of measurable disease was 86% (48/56) at the study end following repeat treatment in two patients. However, 43% of men had persistent cancer of any threshold at the study end. Two other men underwent salvage treatment with neoadjuvant hormone ablation therapy and external beam radiotherapy for residual Gleason 4 + 3 and Gleason 4 + 4 disease ( Table 3 ).

Table 3 Histology outcomes at 6 mo after focal index lesion ablation in 52 men who underwent post-treatment prostate biopsy

Parameter	Value
Positive biopsy outcomes on the treated side
Median number of biopsies taken, n (IQR)	5.0 (4.0–6.0)
Positive biopsies (any disease), n (%)	18/52 (34.6)
Positive biopsies (clinically significant disease), n (%)	8/52 (15.4) a
Median CCLmax in positive cores, mm (IQR)	2.0 (1.0–4.0)
Gleason score, n (%)
3 + 3	12/18 (66.7)
3 + 4	5/18 (27.8)
4 + 3	1/18 (5.6)
Positive biopsy outcomes on the untreated side (targeted to suspicious areas on 6-mo MRI)
Positive biopsies (any disease), n (%)	4/52 (7.7)
Positive biopsies (clinically significant disease), n (%)	2/52 (3.8) b
Gleason score (n)
3 + 3	3 c
3 + 4	0
4 + 3	0
4 + 4	1
Absence of any cancer on both sides, n (%; 95% CI)	30/52 (57.7; 43.2–71.3)
Overall absence of clinically significant disease on both sides, n (%; 95% CI)	42/52 (80.8; 67.5–90.4)
Other histological findings, n (%; 95% CI) d
Prostatic acini	2/51 (3.9; 0.5–13.5)
Atrophy	20/51 (39.2; 25.8–53.9)
Fibrosis	39/51 (76.5; 62.5–87.2)
Giant-cell reaction	4/51 (7.8; 2.2–18.9)
Necrosis	11/51 (21.6; 11.3–35.3)

^a These men were counted within the “any disease” category. In other words, eight of the 18 men with positive biopsies on the treated side had clinically significant cancer.

^b These men were counted within the “any disease” category. In other words, two of the four men with positive biopsies on the untreated side had clinically significant cancer.

^c CCL_max in these three cases was <1 mm, 1 mm, and 4 mm.

^d Data only available for 51/52 patients.

IQR = interquartile range; CCL_max = maximum cancer core length; MRI = magnetic resonance imaging; CI = confidence interval.

Of 41 patients with good baseline status, 22 (53.7%, 95% CI 37.4–69.3%) achieved the trifecta outcome (absence of clinically significant disease, pad-free, leak-free continence, and erections sufficient for intercourse) at 12 mo ( Fig. 4 ). The results for sensitivity analyses (data not shown) were very similar and did not affect our overall conclusion.

Fig. 4 Summary of continence, erectile function, and cancer control outcomes and the trifecta rate following high-intensity focused ultrasound ablation of the index lesion. Patient-reported trifecta outcomes were collected using validated questionnaires. (1) Percentage continence data were derived using the questions “Over the past 4 weeks how often did you leak urine?” and “Over the past 4 weeks how many pads or adult diapers per day did you usually use to control leakage?” from the urinary domain of the University of California, Los Angeles Expanded Prostate Cancer Index Composite questionnaire. (2) The percentage of men with erections sufficient for penetration was calculated for men scoring ≥2 for the question “Over the past 4 weeks when you had erections with sexual stimulation, how often were your erections hard enough for penetration?” from the 15-item International Index of Erectile Function. (3) The percentage phosphodiesterase-5 inhibitor (PDE5-I) use (tadalafil, sildenafil, or vardenafil) was calculated using the number of men achieving erections sufficient for penetration from part (2) as the denominator.

4.1. Clinical implications

The outcome from other focal therapy studies in which treatment was applied to all known cancer lesions was recently reviewed [9] . Rates of 95–100% for pad-free, leak-free continence preservation, 54–100% for potency, and 83–100% for overall absence of clinically significant cancer were observed after treatment. Most of these studies applied strict selection criteria for disease risk, biopsy cancer burden, and function at baseline, and therefore have limited external validity. In our current study—in a group of men who were not selected by the investigators on the basis of good baseline function—the functional preservation outcomes were not as high as we have previously reported. Indeed, our trifecta rate is also lower than previously reported, and may not be as encouraging when compared to the rates reported after surgery. With greater numbers in our ongoing larger multicentre trials, we may be to identify men who are unlikely to have good functional outcomes after focal therapy and would be better treated using a radical approach. Our study is unique in that known cancer of low malignant potential was left untreated. We recently showed that if such a strategy were to be adopted, more than 90% of newly diagnosed men with localised prostate cancer might be eligible for focal therapy [27] .

Some have argued that because mpMRI and TPM biopsies are not accurate enough to detect all areas of prostate cancer, the concept of selective therapy lacks legitimacy [28] . However, in most other solid organs, cancer is diagnosed when a clinical manifestation of disease (a lump or an imaging signal) is verified histologically. Inevitably, therapy is directed at the clinical phenotype plus the addition of a margin. This was recently seen in the administration of radiotherapy to the breast after lumpectomy. Historically, whole-breast irradiation was favoured because of the known multifocality of breast cancer. The finding that intraoperative local irradiation of the surgical site was not inferior to whole-breast irradiation challenges the idea that recurrence is due to residual subclinical multifocal lesions away from the index lesion [29] . There is also concern that inclusion of intermediate- and high-risk cases for focal therapy may represent undertreatment for those men who harbour microscopic nodal metastases.

It is yet to be determined whether an RCT of focal therapy compared to radical therapy is deliverable [23] . Previous trials have attempted to compare different interventions in localised prostate cancer; however, many have failed because of a lack of physician and patient equipoise. There are additional concerns regarding appropriate endpoints and the follow-up time frame, especially as a minimum of 10-yr follow-up would be required to evaluate metastases and mortality. The UK National Institute of Health Research has recently funded a pilot RCT to determine whether recruitment might be feasible (ISRCTN99760303). This RCT will evaluate rates of transition to local salvage therapies within a 5-yr period between focal therapy and radical prostatectomy in men with intermediate-risk prostate cancer.

Our findings have implications for prostate cancer management. Some have suggested that low-risk disease should be relabelled as a benign entity because of the rarity of progression and metastases from these lesions[10], [11], and [12]. If this hypothesis proves to be true, we could concentrate on targeting measurable disease based on mpMRI considering the high negative predictive values for clinically significant disease that it confers. However, the notion that Gleason 6 disease cannot metastasise has recently been brought into question following the findings of Haffner et al [30] , who reported on a case in which a small area of Gleason 6 disease within a tumour of very high-grade cancer was responsible for a distant metastasis. There has been considerable debate regarding the validity of these findings [31] .

Our study also represents anin vivohuman model for better understanding of the determinants of progression of low-grade disease in a manner that is not predicated on removing the whole organ. This might allow us to determine if such lesions do progress and, if so, whether they can be identified at the time of diagnosis using detailed molecular typing approaches such as immunohistochemistry and -omics techniques.

2.1. Study design and conduct

Our single-centre study was a prospective development study according to the IDEAL (Idea, Development, Exploration, Assessment, and Long-term) guidelines for evaluating innovation in surgery [13] . The trial was approved by Local Research Ethics Committee A of the University College London Hospitals.

2.2. Patient population

Treatment-naïve men recently diagnosed with low-, intermediate-, or high-risk nonmetastatic prostate cancer (prostate-specific antigen [PSA] ≤20 ng/ml, Gleason ≤ 4 + 3, stage ≤T3aN0M0) were eligible ( Fig. 1 ).

Fig. 1 Trial flow chart. mpMRI = multiparametric magnetic resonance imaging; HIFU = high-intensity focused ultrasound; ADT = androgen deprivation therapy.

2.3. Study interventions

2.3.2. Treatment

Focal ablation of the index lesion was performed using transrectal high-intensity focused ultrasound (HIFU; Sonablate 500, Focus Surgery, Indianapolis, IN, USA) (Supplementary materials). Untreated areas could contain secondary small-volume (CCL ≤5 mm) Gleason 3 + 3 disease [14] , high-grade prostate intraepithelial neoplasia, and/or atypical small acinar proliferation ( Fig. 2 ).

Fig. 2 Schematic diagrams demonstrating the types of focal therapy conducted in this trial. Large red areas represent dominant cancers (so-called index lesions) whilst small green areas represent small, low-grade, secondary lesions. Red transparent boxes represent ablation zones on the high-intensity focused ultrasound device: (A) hemiablation, (B) extended (dogleg) ablation; (C) quadrant ablation; (D) focal ablation.

2.3.3. Follow-up

Contrast-enhanced MRI was carried out at 10–14 d to evaluate the area of ablation, demonstrated by a confluent perfusion deficit ( Fig. 3 ). Clinical review at 1, 3, 6, 9, and 12 mo assessed adverse events, serum PSA, and responses to validated questionnaires. Phosphodiesterase-5 inhibitor (PDE5-I) use was permitted at any time point before treatment and during follow-up to aid erectile function. At 6 mo, mpMRI followed by biopsies targeted to the treated area was scheduled, with a minimum sampling requirement of one core for every 1 ml of residual tissue. Repeat treatment using focal HIFU for treated or untreated areas that showed signs of progression was permitted if biopsies were positive. Further mpMRI was carried out at 12 mo in such cases. As the purpose of the 6-mo biopsies was to determine whether the ablation was successful, our ethics committee did not permit transperineal TPM biopsies of untreated tissue because of the requirement for another general anaesthetic. However, biopsies of untreated tissue were permitted if a new lesion suspicious for cancer was seen on mpMRI or if existing untreated tissue showed signs of progression.

Fig. 3 Contrast-enhanced axial scan of the prostate at mid-gland at 1 wk after high-intensity focused ultrasound ablation of an index lesion in the right prostate lobe. Example of right-sided index lesion treatment with confluent hemiablation necrosis demonstrated on early (1-wk) multiparametric magnetic resonance imaging. The orange line indicates the area of ablation with confluent necrosis, in this case hemiablation. The white line indicates the outline of whole prostate. R = rectum.

2.4. Study endpoints

The primary outcome was the composite rate of genitourinary side effects measured using validated patient questionnaires[15], [16], and [17]. Incontinence was defined using a score of 1–3 for question 1 (any leak) and of 1–3 for question 5 (any pad use) of the University of California, Los Angeles (UCLA) Expanded Prostate Cancer Index Composite (EPIC) continence questionnaire. Erectile dysfunction was defined as a score of 0 or 1 for question 2 of the 15-item International Index of Erectile Function (IIEF-15) questionnaire. Urinary function was evaluated using the International Prostate Symptom Score (IPSS), the IPSS-quality of life questionnaire, and the UCLA-EPIC continence questionnaire. Health-related quality of life was measured using the Functional Assessment of Cancer Therapy-Prostate (FACT-P). Responses to the validated questionnaires were collated using standard methods.

Secondary outcomes were histological parameters of cancer control on 6-mo biopsy, absence of clinically significant disease on mpMRI at 6 and 12 mo, description of PSA kinetics following index lesion ablation, and the proportion of men requiring salvage local radical therapy and systemic therapy for progressive prostate cancer.

2.5. Statistical analysis

The sample size was calculated on a composite endpoint defined as the total proportion of men with erectile dysfunction and/or incontinence. It was estimated that baseline incontinence was about 10% and baseline impotence about 40% in men older than 60 yr[18] and [19]. Overall, we estimated that baseline genitourinary dysfunction would be approximately 50%. Were these men to undergo whole-gland therapy, 50% of those with good baseline function would also suffer some form of genitourinary dysfunction [20] , so total genitourinary dysfunction would be approximately 75% at 12 mo if men were to undergo whole-gland therapy. It was hypothesised that focal therapy would give rise to genitourinary dysfunction in only 5% of those with good baseline function. Therefore, we estimated a total genitourinary dysfunction level of approximately 55% at 12 mo after focal index lesion ablation. Using sample size calculations comparing an estimate proportion (0.55) to a known proportion (0.75), with α = 0.05 and power of 90%, the total number of patients required was 56 [21] .

A Wilcoxon signed rank test (two-tailed) was used to evaluate differences between continuous variables (PSA, questionnaire scores) measured at baseline and at the 12-mo follow-up visit. Statistical significance was set atp≤ 0.05. Changes over time were described using box-and-whisker plots. Subgroup analyses were hypothesis-generating, and with small numbers in each subgroup, it was deemed inappropriate to run statistical tests of significance for such comparisons. Statistical analysis was performed using Stata version 11.2 (Stata Corp, College Station, TX, USA). This analysis was carried out on the available data. However, some functional measures were missing at baseline or at 12-mo follow-up for five men, so we conducted two sensitivity analyses to assess if this was likely to impact our results. We repeated the analysis described above for a “best” outcome scenario and a “worst” outcome scenario by setting missing baseline values to the maximum or minimum possible, respectively.

3.1. Study patients

Fifty-six men were recruited (July 27, 2009–January 26, 2011) with a mean age of 64 yr. According to National Comprehensive Cancer Network criteria [22] , seven men (12.5%) had low-risk, 47 (83.9%) had intermediate-risk, and two (3.6%) had high-risk nonmetastatic prostate cancer ( Fig. 1 , Table 1 ).

3.2. Primary outcome

Overall, 40 patients (71.4%; 95% confidence interval [CI] 59.6–83.2%) were leak-free, pad-free continent and had erections sufficient for penetration at baseline. This composite measure decreased to 33 patients (58.9%; 95% CI 46.0–71.8%) at 12 mo.

3.3. Secondary outcomes

3.3.5. Biochemical, imaging, and histological outcomes

A significant decrease in serum PSA from a median baseline value of 7.4 ng/ml (IQR 5.6–9.5) to 2.4 ng/ml (IQR 1.6–4.1) was observed at 12 mo (p < 0.0001; Fig. 3 ).

At 6 mo, two men refused a post-treatment biopsy, and another two had clinical reasons for omitting biopsies (chemotherapy for newly diagnosed lung cancer in one, and a persistent HIFU cavity connecting with the vas deferens on mpMRI in another). For these four men, mpMRI at 6 mo showed no evidence of measurable residual disease. There was no histological evidence of clinically significant disease (Gleason ≤ 3 + 3, CCL ≤3 mm) in the treated area in 44/52 patients (84.6%; 95% CI 70–92%), or of any cancer in the treated area in 34/52 patients (65.4%; 95% CI 52.5–78.3%). Two had clinically significant disease on biopsies from suspicious areas in the untreated side on mpMRI not present at baseline. Overall, histopathological absence of clinically significant cancer (treated and untreated sides) was observed in 42/52 men (80.8%, 95% CI 67.5–90.4%). The overall rate of absence of measurable disease was 86% (48/56) at the study end following repeat treatment in two patients. However, 43% of men had persistent cancer of any threshold at the study end. Two other men underwent salvage treatment with neoadjuvant hormone ablation therapy and external beam radiotherapy for residual Gleason 4 + 3 and Gleason 4 + 4 disease ( Table 3 ).

Table 3 Histology outcomes at 6 mo after focal index lesion ablation in 52 men who underwent post-treatment prostate biopsy

Parameter	Value
Positive biopsy outcomes on the treated side
Median number of biopsies taken, n (IQR)	5.0 (4.0–6.0)
Positive biopsies (any disease), n (%)	18/52 (34.6)
Positive biopsies (clinically significant disease), n (%)	8/52 (15.4) a
Median CCLmax in positive cores, mm (IQR)	2.0 (1.0–4.0)
Gleason score, n (%)
3 + 3	12/18 (66.7)
3 + 4	5/18 (27.8)
4 + 3	1/18 (5.6)
Positive biopsy outcomes on the untreated side (targeted to suspicious areas on 6-mo MRI)
Positive biopsies (any disease), n (%)	4/52 (7.7)
Positive biopsies (clinically significant disease), n (%)	2/52 (3.8) b
Gleason score (n)
3 + 3	3 c
3 + 4	0
4 + 3	0
4 + 4	1
Absence of any cancer on both sides, n (%; 95% CI)	30/52 (57.7; 43.2–71.3)
Overall absence of clinically significant disease on both sides, n (%; 95% CI)	42/52 (80.8; 67.5–90.4)
Other histological findings, n (%; 95% CI) d
Prostatic acini	2/51 (3.9; 0.5–13.5)
Atrophy	20/51 (39.2; 25.8–53.9)
Fibrosis	39/51 (76.5; 62.5–87.2)
Giant-cell reaction	4/51 (7.8; 2.2–18.9)
Necrosis	11/51 (21.6; 11.3–35.3)

^a These men were counted within the “any disease” category. In other words, eight of the 18 men with positive biopsies on the treated side had clinically significant cancer.

^b These men were counted within the “any disease” category. In other words, two of the four men with positive biopsies on the untreated side had clinically significant cancer.

^c CCL_max in these three cases was <1 mm, 1 mm, and 4 mm.

^d Data only available for 51/52 patients.

IQR = interquartile range; CCL_max = maximum cancer core length; MRI = magnetic resonance imaging; CI = confidence interval.

Of 41 patients with good baseline status, 22 (53.7%, 95% CI 37.4–69.3%) achieved the trifecta outcome (absence of clinically significant disease, pad-free, leak-free continence, and erections sufficient for intercourse) at 12 mo ( Fig. 4 ). The results for sensitivity analyses (data not shown) were very similar and did not affect our overall conclusion.

Fig. 4 Summary of continence, erectile function, and cancer control outcomes and the trifecta rate following high-intensity focused ultrasound ablation of the index lesion. Patient-reported trifecta outcomes were collected using validated questionnaires. (1) Percentage continence data were derived using the questions “Over the past 4 weeks how often did you leak urine?” and “Over the past 4 weeks how many pads or adult diapers per day did you usually use to control leakage?” from the urinary domain of the University of California, Los Angeles Expanded Prostate Cancer Index Composite questionnaire. (2) The percentage of men with erections sufficient for penetration was calculated for men scoring ≥2 for the question “Over the past 4 weeks when you had erections with sexual stimulation, how often were your erections hard enough for penetration?” from the 15-item International Index of Erectile Function. (3) The percentage phosphodiesterase-5 inhibitor (PDE5-I) use (tadalafil, sildenafil, or vardenafil) was calculated using the number of men achieving erections sufficient for penetration from part (2) as the denominator.

4.1. Clinical implications

The outcome from other focal therapy studies in which treatment was applied to all known cancer lesions was recently reviewed [9] . Rates of 95–100% for pad-free, leak-free continence preservation, 54–100% for potency, and 83–100% for overall absence of clinically significant cancer were observed after treatment. Most of these studies applied strict selection criteria for disease risk, biopsy cancer burden, and function at baseline, and therefore have limited external validity. In our current study—in a group of men who were not selected by the investigators on the basis of good baseline function—the functional preservation outcomes were not as high as we have previously reported. Indeed, our trifecta rate is also lower than previously reported, and may not be as encouraging when compared to the rates reported after surgery. With greater numbers in our ongoing larger multicentre trials, we may be to identify men who are unlikely to have good functional outcomes after focal therapy and would be better treated using a radical approach. Our study is unique in that known cancer of low malignant potential was left untreated. We recently showed that if such a strategy were to be adopted, more than 90% of newly diagnosed men with localised prostate cancer might be eligible for focal therapy [27] .

Some have argued that because mpMRI and TPM biopsies are not accurate enough to detect all areas of prostate cancer, the concept of selective therapy lacks legitimacy [28] . However, in most other solid organs, cancer is diagnosed when a clinical manifestation of disease (a lump or an imaging signal) is verified histologically. Inevitably, therapy is directed at the clinical phenotype plus the addition of a margin. This was recently seen in the administration of radiotherapy to the breast after lumpectomy. Historically, whole-breast irradiation was favoured because of the known multifocality of breast cancer. The finding that intraoperative local irradiation of the surgical site was not inferior to whole-breast irradiation challenges the idea that recurrence is due to residual subclinical multifocal lesions away from the index lesion [29] . There is also concern that inclusion of intermediate- and high-risk cases for focal therapy may represent undertreatment for those men who harbour microscopic nodal metastases.

It is yet to be determined whether an RCT of focal therapy compared to radical therapy is deliverable [23] . Previous trials have attempted to compare different interventions in localised prostate cancer; however, many have failed because of a lack of physician and patient equipoise. There are additional concerns regarding appropriate endpoints and the follow-up time frame, especially as a minimum of 10-yr follow-up would be required to evaluate metastases and mortality. The UK National Institute of Health Research has recently funded a pilot RCT to determine whether recruitment might be feasible (ISRCTN99760303). This RCT will evaluate rates of transition to local salvage therapies within a 5-yr period between focal therapy and radical prostatectomy in men with intermediate-risk prostate cancer.

Our findings have implications for prostate cancer management. Some have suggested that low-risk disease should be relabelled as a benign entity because of the rarity of progression and metastases from these lesions[10], [11], and [12]. If this hypothesis proves to be true, we could concentrate on targeting measurable disease based on mpMRI considering the high negative predictive values for clinically significant disease that it confers. However, the notion that Gleason 6 disease cannot metastasise has recently been brought into question following the findings of Haffner et al [30] , who reported on a case in which a small area of Gleason 6 disease within a tumour of very high-grade cancer was responsible for a distant metastasis. There has been considerable debate regarding the validity of these findings [31] .

Our study also represents anin vivohuman model for better understanding of the determinants of progression of low-grade disease in a manner that is not predicated on removing the whole organ. This might allow us to determine if such lesions do progress and, if so, whether they can be identified at the time of diagnosis using detailed molecular typing approaches such as immunohistochemistry and -omics techniques.

2.1. Study design and conduct

Our single-centre study was a prospective development study according to the IDEAL (Idea, Development, Exploration, Assessment, and Long-term) guidelines for evaluating innovation in surgery [13] . The trial was approved by Local Research Ethics Committee A of the University College London Hospitals.

2.2. Patient population

Treatment-naïve men recently diagnosed with low-, intermediate-, or high-risk nonmetastatic prostate cancer (prostate-specific antigen [PSA] ≤20 ng/ml, Gleason ≤ 4 + 3, stage ≤T3aN0M0) were eligible ( Fig. 1 ).

Fig. 1 Trial flow chart. mpMRI = multiparametric magnetic resonance imaging; HIFU = high-intensity focused ultrasound; ADT = androgen deprivation therapy.

2.3. Study interventions

2.3.2. Treatment

Focal ablation of the index lesion was performed using transrectal high-intensity focused ultrasound (HIFU; Sonablate 500, Focus Surgery, Indianapolis, IN, USA) (Supplementary materials). Untreated areas could contain secondary small-volume (CCL ≤5 mm) Gleason 3 + 3 disease [14] , high-grade prostate intraepithelial neoplasia, and/or atypical small acinar proliferation ( Fig. 2 ).

Fig. 2 Schematic diagrams demonstrating the types of focal therapy conducted in this trial. Large red areas represent dominant cancers (so-called index lesions) whilst small green areas represent small, low-grade, secondary lesions. Red transparent boxes represent ablation zones on the high-intensity focused ultrasound device: (A) hemiablation, (B) extended (dogleg) ablation; (C) quadrant ablation; (D) focal ablation.

2.3.3. Follow-up

Contrast-enhanced MRI was carried out at 10–14 d to evaluate the area of ablation, demonstrated by a confluent perfusion deficit ( Fig. 3 ). Clinical review at 1, 3, 6, 9, and 12 mo assessed adverse events, serum PSA, and responses to validated questionnaires. Phosphodiesterase-5 inhibitor (PDE5-I) use was permitted at any time point before treatment and during follow-up to aid erectile function. At 6 mo, mpMRI followed by biopsies targeted to the treated area was scheduled, with a minimum sampling requirement of one core for every 1 ml of residual tissue. Repeat treatment using focal HIFU for treated or untreated areas that showed signs of progression was permitted if biopsies were positive. Further mpMRI was carried out at 12 mo in such cases. As the purpose of the 6-mo biopsies was to determine whether the ablation was successful, our ethics committee did not permit transperineal TPM biopsies of untreated tissue because of the requirement for another general anaesthetic. However, biopsies of untreated tissue were permitted if a new lesion suspicious for cancer was seen on mpMRI or if existing untreated tissue showed signs of progression.

Fig. 3 Contrast-enhanced axial scan of the prostate at mid-gland at 1 wk after high-intensity focused ultrasound ablation of an index lesion in the right prostate lobe. Example of right-sided index lesion treatment with confluent hemiablation necrosis demonstrated on early (1-wk) multiparametric magnetic resonance imaging. The orange line indicates the area of ablation with confluent necrosis, in this case hemiablation. The white line indicates the outline of whole prostate. R = rectum.

2.4. Study endpoints

The primary outcome was the composite rate of genitourinary side effects measured using validated patient questionnaires[15], [16], and [17]. Incontinence was defined using a score of 1–3 for question 1 (any leak) and of 1–3 for question 5 (any pad use) of the University of California, Los Angeles (UCLA) Expanded Prostate Cancer Index Composite (EPIC) continence questionnaire. Erectile dysfunction was defined as a score of 0 or 1 for question 2 of the 15-item International Index of Erectile Function (IIEF-15) questionnaire. Urinary function was evaluated using the International Prostate Symptom Score (IPSS), the IPSS-quality of life questionnaire, and the UCLA-EPIC continence questionnaire. Health-related quality of life was measured using the Functional Assessment of Cancer Therapy-Prostate (FACT-P). Responses to the validated questionnaires were collated using standard methods.

Secondary outcomes were histological parameters of cancer control on 6-mo biopsy, absence of clinically significant disease on mpMRI at 6 and 12 mo, description of PSA kinetics following index lesion ablation, and the proportion of men requiring salvage local radical therapy and systemic therapy for progressive prostate cancer.

2.5. Statistical analysis

The sample size was calculated on a composite endpoint defined as the total proportion of men with erectile dysfunction and/or incontinence. It was estimated that baseline incontinence was about 10% and baseline impotence about 40% in men older than 60 yr[18] and [19]. Overall, we estimated that baseline genitourinary dysfunction would be approximately 50%. Were these men to undergo whole-gland therapy, 50% of those with good baseline function would also suffer some form of genitourinary dysfunction [20] , so total genitourinary dysfunction would be approximately 75% at 12 mo if men were to undergo whole-gland therapy. It was hypothesised that focal therapy would give rise to genitourinary dysfunction in only 5% of those with good baseline function. Therefore, we estimated a total genitourinary dysfunction level of approximately 55% at 12 mo after focal index lesion ablation. Using sample size calculations comparing an estimate proportion (0.55) to a known proportion (0.75), with α = 0.05 and power of 90%, the total number of patients required was 56 [21] .

A Wilcoxon signed rank test (two-tailed) was used to evaluate differences between continuous variables (PSA, questionnaire scores) measured at baseline and at the 12-mo follow-up visit. Statistical significance was set atp≤ 0.05. Changes over time were described using box-and-whisker plots. Subgroup analyses were hypothesis-generating, and with small numbers in each subgroup, it was deemed inappropriate to run statistical tests of significance for such comparisons. Statistical analysis was performed using Stata version 11.2 (Stata Corp, College Station, TX, USA). This analysis was carried out on the available data. However, some functional measures were missing at baseline or at 12-mo follow-up for five men, so we conducted two sensitivity analyses to assess if this was likely to impact our results. We repeated the analysis described above for a “best” outcome scenario and a “worst” outcome scenario by setting missing baseline values to the maximum or minimum possible, respectively.

3.1. Study patients

Fifty-six men were recruited (July 27, 2009–January 26, 2011) with a mean age of 64 yr. According to National Comprehensive Cancer Network criteria [22] , seven men (12.5%) had low-risk, 47 (83.9%) had intermediate-risk, and two (3.6%) had high-risk nonmetastatic prostate cancer ( Fig. 1 , Table 1 ).

3.2. Primary outcome

Overall, 40 patients (71.4%; 95% confidence interval [CI] 59.6–83.2%) were leak-free, pad-free continent and had erections sufficient for penetration at baseline. This composite measure decreased to 33 patients (58.9%; 95% CI 46.0–71.8%) at 12 mo.

3.3. Secondary outcomes

3.3.5. Biochemical, imaging, and histological outcomes

A significant decrease in serum PSA from a median baseline value of 7.4 ng/ml (IQR 5.6–9.5) to 2.4 ng/ml (IQR 1.6–4.1) was observed at 12 mo (p < 0.0001; Fig. 3 ).

At 6 mo, two men refused a post-treatment biopsy, and another two had clinical reasons for omitting biopsies (chemotherapy for newly diagnosed lung cancer in one, and a persistent HIFU cavity connecting with the vas deferens on mpMRI in another). For these four men, mpMRI at 6 mo showed no evidence of measurable residual disease. There was no histological evidence of clinically significant disease (Gleason ≤ 3 + 3, CCL ≤3 mm) in the treated area in 44/52 patients (84.6%; 95% CI 70–92%), or of any cancer in the treated area in 34/52 patients (65.4%; 95% CI 52.5–78.3%). Two had clinically significant disease on biopsies from suspicious areas in the untreated side on mpMRI not present at baseline. Overall, histopathological absence of clinically significant cancer (treated and untreated sides) was observed in 42/52 men (80.8%, 95% CI 67.5–90.4%). The overall rate of absence of measurable disease was 86% (48/56) at the study end following repeat treatment in two patients. However, 43% of men had persistent cancer of any threshold at the study end. Two other men underwent salvage treatment with neoadjuvant hormone ablation therapy and external beam radiotherapy for residual Gleason 4 + 3 and Gleason 4 + 4 disease ( Table 3 ).

Table 3 Histology outcomes at 6 mo after focal index lesion ablation in 52 men who underwent post-treatment prostate biopsy

Parameter	Value
Positive biopsy outcomes on the treated side
Median number of biopsies taken, n (IQR)	5.0 (4.0–6.0)
Positive biopsies (any disease), n (%)	18/52 (34.6)
Positive biopsies (clinically significant disease), n (%)	8/52 (15.4) a
Median CCLmax in positive cores, mm (IQR)	2.0 (1.0–4.0)
Gleason score, n (%)
3 + 3	12/18 (66.7)
3 + 4	5/18 (27.8)
4 + 3	1/18 (5.6)
Positive biopsy outcomes on the untreated side (targeted to suspicious areas on 6-mo MRI)
Positive biopsies (any disease), n (%)	4/52 (7.7)
Positive biopsies (clinically significant disease), n (%)	2/52 (3.8) b
Gleason score (n)
3 + 3	3 c
3 + 4	0
4 + 3	0
4 + 4	1
Absence of any cancer on both sides, n (%; 95% CI)	30/52 (57.7; 43.2–71.3)
Overall absence of clinically significant disease on both sides, n (%; 95% CI)	42/52 (80.8; 67.5–90.4)
Other histological findings, n (%; 95% CI) d
Prostatic acini	2/51 (3.9; 0.5–13.5)
Atrophy	20/51 (39.2; 25.8–53.9)
Fibrosis	39/51 (76.5; 62.5–87.2)
Giant-cell reaction	4/51 (7.8; 2.2–18.9)
Necrosis	11/51 (21.6; 11.3–35.3)

^a These men were counted within the “any disease” category. In other words, eight of the 18 men with positive biopsies on the treated side had clinically significant cancer.

^b These men were counted within the “any disease” category. In other words, two of the four men with positive biopsies on the untreated side had clinically significant cancer.

^c CCL_max in these three cases was <1 mm, 1 mm, and 4 mm.

^d Data only available for 51/52 patients.

IQR = interquartile range; CCL_max = maximum cancer core length; MRI = magnetic resonance imaging; CI = confidence interval.

Of 41 patients with good baseline status, 22 (53.7%, 95% CI 37.4–69.3%) achieved the trifecta outcome (absence of clinically significant disease, pad-free, leak-free continence, and erections sufficient for intercourse) at 12 mo ( Fig. 4 ). The results for sensitivity analyses (data not shown) were very similar and did not affect our overall conclusion.

Fig. 4 Summary of continence, erectile function, and cancer control outcomes and the trifecta rate following high-intensity focused ultrasound ablation of the index lesion. Patient-reported trifecta outcomes were collected using validated questionnaires. (1) Percentage continence data were derived using the questions “Over the past 4 weeks how often did you leak urine?” and “Over the past 4 weeks how many pads or adult diapers per day did you usually use to control leakage?” from the urinary domain of the University of California, Los Angeles Expanded Prostate Cancer Index Composite questionnaire. (2) The percentage of men with erections sufficient for penetration was calculated for men scoring ≥2 for the question “Over the past 4 weeks when you had erections with sexual stimulation, how often were your erections hard enough for penetration?” from the 15-item International Index of Erectile Function. (3) The percentage phosphodiesterase-5 inhibitor (PDE5-I) use (tadalafil, sildenafil, or vardenafil) was calculated using the number of men achieving erections sufficient for penetration from part (2) as the denominator.

4.1. Clinical implications

The outcome from other focal therapy studies in which treatment was applied to all known cancer lesions was recently reviewed [9] . Rates of 95–100% for pad-free, leak-free continence preservation, 54–100% for potency, and 83–100% for overall absence of clinically significant cancer were observed after treatment. Most of these studies applied strict selection criteria for disease risk, biopsy cancer burden, and function at baseline, and therefore have limited external validity. In our current study—in a group of men who were not selected by the investigators on the basis of good baseline function—the functional preservation outcomes were not as high as we have previously reported. Indeed, our trifecta rate is also lower than previously reported, and may not be as encouraging when compared to the rates reported after surgery. With greater numbers in our ongoing larger multicentre trials, we may be to identify men who are unlikely to have good functional outcomes after focal therapy and would be better treated using a radical approach. Our study is unique in that known cancer of low malignant potential was left untreated. We recently showed that if such a strategy were to be adopted, more than 90% of newly diagnosed men with localised prostate cancer might be eligible for focal therapy [27] .

Some have argued that because mpMRI and TPM biopsies are not accurate enough to detect all areas of prostate cancer, the concept of selective therapy lacks legitimacy [28] . However, in most other solid organs, cancer is diagnosed when a clinical manifestation of disease (a lump or an imaging signal) is verified histologically. Inevitably, therapy is directed at the clinical phenotype plus the addition of a margin. This was recently seen in the administration of radiotherapy to the breast after lumpectomy. Historically, whole-breast irradiation was favoured because of the known multifocality of breast cancer. The finding that intraoperative local irradiation of the surgical site was not inferior to whole-breast irradiation challenges the idea that recurrence is due to residual subclinical multifocal lesions away from the index lesion [29] . There is also concern that inclusion of intermediate- and high-risk cases for focal therapy may represent undertreatment for those men who harbour microscopic nodal metastases.

It is yet to be determined whether an RCT of focal therapy compared to radical therapy is deliverable [23] . Previous trials have attempted to compare different interventions in localised prostate cancer; however, many have failed because of a lack of physician and patient equipoise. There are additional concerns regarding appropriate endpoints and the follow-up time frame, especially as a minimum of 10-yr follow-up would be required to evaluate metastases and mortality. The UK National Institute of Health Research has recently funded a pilot RCT to determine whether recruitment might be feasible (ISRCTN99760303). This RCT will evaluate rates of transition to local salvage therapies within a 5-yr period between focal therapy and radical prostatectomy in men with intermediate-risk prostate cancer.

Our findings have implications for prostate cancer management. Some have suggested that low-risk disease should be relabelled as a benign entity because of the rarity of progression and metastases from these lesions[10], [11], and [12]. If this hypothesis proves to be true, we could concentrate on targeting measurable disease based on mpMRI considering the high negative predictive values for clinically significant disease that it confers. However, the notion that Gleason 6 disease cannot metastasise has recently been brought into question following the findings of Haffner et al [30] , who reported on a case in which a small area of Gleason 6 disease within a tumour of very high-grade cancer was responsible for a distant metastasis. There has been considerable debate regarding the validity of these findings [31] .

Our study also represents anin vivohuman model for better understanding of the determinants of progression of low-grade disease in a manner that is not predicated on removing the whole organ. This might allow us to determine if such lesions do progress and, if so, whether they can be identified at the time of diagnosis using detailed molecular typing approaches such as immunohistochemistry and -omics techniques.

2.1. Study design and conduct

Our single-centre study was a prospective development study according to the IDEAL (Idea, Development, Exploration, Assessment, and Long-term) guidelines for evaluating innovation in surgery [13] . The trial was approved by Local Research Ethics Committee A of the University College London Hospitals.

2.2. Patient population

Treatment-naïve men recently diagnosed with low-, intermediate-, or high-risk nonmetastatic prostate cancer (prostate-specific antigen [PSA] ≤20 ng/ml, Gleason ≤ 4 + 3, stage ≤T3aN0M0) were eligible ( Fig. 1 ).

Fig. 1 Trial flow chart. mpMRI = multiparametric magnetic resonance imaging; HIFU = high-intensity focused ultrasound; ADT = androgen deprivation therapy.

2.3. Study interventions

2.3.2. Treatment

Focal ablation of the index lesion was performed using transrectal high-intensity focused ultrasound (HIFU; Sonablate 500, Focus Surgery, Indianapolis, IN, USA) (Supplementary materials). Untreated areas could contain secondary small-volume (CCL ≤5 mm) Gleason 3 + 3 disease [14] , high-grade prostate intraepithelial neoplasia, and/or atypical small acinar proliferation ( Fig. 2 ).

Fig. 2 Schematic diagrams demonstrating the types of focal therapy conducted in this trial. Large red areas represent dominant cancers (so-called index lesions) whilst small green areas represent small, low-grade, secondary lesions. Red transparent boxes represent ablation zones on the high-intensity focused ultrasound device: (A) hemiablation, (B) extended (dogleg) ablation; (C) quadrant ablation; (D) focal ablation.

2.3.3. Follow-up

Contrast-enhanced MRI was carried out at 10–14 d to evaluate the area of ablation, demonstrated by a confluent perfusion deficit ( Fig. 3 ). Clinical review at 1, 3, 6, 9, and 12 mo assessed adverse events, serum PSA, and responses to validated questionnaires. Phosphodiesterase-5 inhibitor (PDE5-I) use was permitted at any time point before treatment and during follow-up to aid erectile function. At 6 mo, mpMRI followed by biopsies targeted to the treated area was scheduled, with a minimum sampling requirement of one core for every 1 ml of residual tissue. Repeat treatment using focal HIFU for treated or untreated areas that showed signs of progression was permitted if biopsies were positive. Further mpMRI was carried out at 12 mo in such cases. As the purpose of the 6-mo biopsies was to determine whether the ablation was successful, our ethics committee did not permit transperineal TPM biopsies of untreated tissue because of the requirement for another general anaesthetic. However, biopsies of untreated tissue were permitted if a new lesion suspicious for cancer was seen on mpMRI or if existing untreated tissue showed signs of progression.

Fig. 3 Contrast-enhanced axial scan of the prostate at mid-gland at 1 wk after high-intensity focused ultrasound ablation of an index lesion in the right prostate lobe. Example of right-sided index lesion treatment with confluent hemiablation necrosis demonstrated on early (1-wk) multiparametric magnetic resonance imaging. The orange line indicates the area of ablation with confluent necrosis, in this case hemiablation. The white line indicates the outline of whole prostate. R = rectum.

2.4. Study endpoints

The primary outcome was the composite rate of genitourinary side effects measured using validated patient questionnaires[15], [16], and [17]. Incontinence was defined using a score of 1–3 for question 1 (any leak) and of 1–3 for question 5 (any pad use) of the University of California, Los Angeles (UCLA) Expanded Prostate Cancer Index Composite (EPIC) continence questionnaire. Erectile dysfunction was defined as a score of 0 or 1 for question 2 of the 15-item International Index of Erectile Function (IIEF-15) questionnaire. Urinary function was evaluated using the International Prostate Symptom Score (IPSS), the IPSS-quality of life questionnaire, and the UCLA-EPIC continence questionnaire. Health-related quality of life was measured using the Functional Assessment of Cancer Therapy-Prostate (FACT-P). Responses to the validated questionnaires were collated using standard methods.

Secondary outcomes were histological parameters of cancer control on 6-mo biopsy, absence of clinically significant disease on mpMRI at 6 and 12 mo, description of PSA kinetics following index lesion ablation, and the proportion of men requiring salvage local radical therapy and systemic therapy for progressive prostate cancer.

2.5. Statistical analysis

The sample size was calculated on a composite endpoint defined as the total proportion of men with erectile dysfunction and/or incontinence. It was estimated that baseline incontinence was about 10% and baseline impotence about 40% in men older than 60 yr[18] and [19]. Overall, we estimated that baseline genitourinary dysfunction would be approximately 50%. Were these men to undergo whole-gland therapy, 50% of those with good baseline function would also suffer some form of genitourinary dysfunction [20] , so total genitourinary dysfunction would be approximately 75% at 12 mo if men were to undergo whole-gland therapy. It was hypothesised that focal therapy would give rise to genitourinary dysfunction in only 5% of those with good baseline function. Therefore, we estimated a total genitourinary dysfunction level of approximately 55% at 12 mo after focal index lesion ablation. Using sample size calculations comparing an estimate proportion (0.55) to a known proportion (0.75), with α = 0.05 and power of 90%, the total number of patients required was 56 [21] .

A Wilcoxon signed rank test (two-tailed) was used to evaluate differences between continuous variables (PSA, questionnaire scores) measured at baseline and at the 12-mo follow-up visit. Statistical significance was set atp≤ 0.05. Changes over time were described using box-and-whisker plots. Subgroup analyses were hypothesis-generating, and with small numbers in each subgroup, it was deemed inappropriate to run statistical tests of significance for such comparisons. Statistical analysis was performed using Stata version 11.2 (Stata Corp, College Station, TX, USA). This analysis was carried out on the available data. However, some functional measures were missing at baseline or at 12-mo follow-up for five men, so we conducted two sensitivity analyses to assess if this was likely to impact our results. We repeated the analysis described above for a “best” outcome scenario and a “worst” outcome scenario by setting missing baseline values to the maximum or minimum possible, respectively.

3.1. Study patients

Fifty-six men were recruited (July 27, 2009–January 26, 2011) with a mean age of 64 yr. According to National Comprehensive Cancer Network criteria [22] , seven men (12.5%) had low-risk, 47 (83.9%) had intermediate-risk, and two (3.6%) had high-risk nonmetastatic prostate cancer ( Fig. 1 , Table 1 ).

3.2. Primary outcome

Overall, 40 patients (71.4%; 95% confidence interval [CI] 59.6–83.2%) were leak-free, pad-free continent and had erections sufficient for penetration at baseline. This composite measure decreased to 33 patients (58.9%; 95% CI 46.0–71.8%) at 12 mo.

3.3. Secondary outcomes

3.3.5. Biochemical, imaging, and histological outcomes

A significant decrease in serum PSA from a median baseline value of 7.4 ng/ml (IQR 5.6–9.5) to 2.4 ng/ml (IQR 1.6–4.1) was observed at 12 mo (p < 0.0001; Fig. 3 ).

At 6 mo, two men refused a post-treatment biopsy, and another two had clinical reasons for omitting biopsies (chemotherapy for newly diagnosed lung cancer in one, and a persistent HIFU cavity connecting with the vas deferens on mpMRI in another). For these four men, mpMRI at 6 mo showed no evidence of measurable residual disease. There was no histological evidence of clinically significant disease (Gleason ≤ 3 + 3, CCL ≤3 mm) in the treated area in 44/52 patients (84.6%; 95% CI 70–92%), or of any cancer in the treated area in 34/52 patients (65.4%; 95% CI 52.5–78.3%). Two had clinically significant disease on biopsies from suspicious areas in the untreated side on mpMRI not present at baseline. Overall, histopathological absence of clinically significant cancer (treated and untreated sides) was observed in 42/52 men (80.8%, 95% CI 67.5–90.4%). The overall rate of absence of measurable disease was 86% (48/56) at the study end following repeat treatment in two patients. However, 43% of men had persistent cancer of any threshold at the study end. Two other men underwent salvage treatment with neoadjuvant hormone ablation therapy and external beam radiotherapy for residual Gleason 4 + 3 and Gleason 4 + 4 disease ( Table 3 ).

Table 3 Histology outcomes at 6 mo after focal index lesion ablation in 52 men who underwent post-treatment prostate biopsy

Parameter	Value
Positive biopsy outcomes on the treated side
Median number of biopsies taken, n (IQR)	5.0 (4.0–6.0)
Positive biopsies (any disease), n (%)	18/52 (34.6)
Positive biopsies (clinically significant disease), n (%)	8/52 (15.4) a
Median CCLmax in positive cores, mm (IQR)	2.0 (1.0–4.0)
Gleason score, n (%)
3 + 3	12/18 (66.7)
3 + 4	5/18 (27.8)
4 + 3	1/18 (5.6)
Positive biopsy outcomes on the untreated side (targeted to suspicious areas on 6-mo MRI)
Positive biopsies (any disease), n (%)	4/52 (7.7)
Positive biopsies (clinically significant disease), n (%)	2/52 (3.8) b
Gleason score (n)
3 + 3	3 c
3 + 4	0
4 + 3	0
4 + 4	1
Absence of any cancer on both sides, n (%; 95% CI)	30/52 (57.7; 43.2–71.3)
Overall absence of clinically significant disease on both sides, n (%; 95% CI)	42/52 (80.8; 67.5–90.4)
Other histological findings, n (%; 95% CI) d
Prostatic acini	2/51 (3.9; 0.5–13.5)
Atrophy	20/51 (39.2; 25.8–53.9)
Fibrosis	39/51 (76.5; 62.5–87.2)
Giant-cell reaction	4/51 (7.8; 2.2–18.9)
Necrosis	11/51 (21.6; 11.3–35.3)

^a These men were counted within the “any disease” category. In other words, eight of the 18 men with positive biopsies on the treated side had clinically significant cancer.

^b These men were counted within the “any disease” category. In other words, two of the four men with positive biopsies on the untreated side had clinically significant cancer.

^c CCL_max in these three cases was <1 mm, 1 mm, and 4 mm.

^d Data only available for 51/52 patients.

IQR = interquartile range; CCL_max = maximum cancer core length; MRI = magnetic resonance imaging; CI = confidence interval.

Of 41 patients with good baseline status, 22 (53.7%, 95% CI 37.4–69.3%) achieved the trifecta outcome (absence of clinically significant disease, pad-free, leak-free continence, and erections sufficient for intercourse) at 12 mo ( Fig. 4 ). The results for sensitivity analyses (data not shown) were very similar and did not affect our overall conclusion.

Fig. 4 Summary of continence, erectile function, and cancer control outcomes and the trifecta rate following high-intensity focused ultrasound ablation of the index lesion. Patient-reported trifecta outcomes were collected using validated questionnaires. (1) Percentage continence data were derived using the questions “Over the past 4 weeks how often did you leak urine?” and “Over the past 4 weeks how many pads or adult diapers per day did you usually use to control leakage?” from the urinary domain of the University of California, Los Angeles Expanded Prostate Cancer Index Composite questionnaire. (2) The percentage of men with erections sufficient for penetration was calculated for men scoring ≥2 for the question “Over the past 4 weeks when you had erections with sexual stimulation, how often were your erections hard enough for penetration?” from the 15-item International Index of Erectile Function. (3) The percentage phosphodiesterase-5 inhibitor (PDE5-I) use (tadalafil, sildenafil, or vardenafil) was calculated using the number of men achieving erections sufficient for penetration from part (2) as the denominator.