5-year outcomes from a large cohort of HIFU-treated patients in the UK

For several years now, the Ahmed/Emberton-led research team in the United Kingdom (UK) has published the majority of the detailed outcomes data from treatment of men with localized prostate cancer using high-intensity focused ultrasound (HIFU). This group has now given us 5-year outcomes on a cohort of > 550 patients.

Writing in the journal European Urology, Dickinson et al. describe the detailed “medium-term” outcomes from their academic registry study involving whole gland (not focused) treatment of patients with HIFU, using Sonablate 500 technology, at a group of eight UK-based hospitals. It is relevant to note that, in the UK, HIFU is a form of treatment that is only available if the patient is enrolled as a participant in some form of clinical study (which includes this registry program).

Primary, failure-free survival outcomes required patients to not have or need any of the following:

  • Local salvage therapy using either surgery or radiotherapy
  • Systemic therapy for prostate cancer (e.g., with androgen deprivation therapy)
  • Clear evidence of metastatic disease
  • Prostate cancer-specific mortality

The research team also monitored for secondary outcomes included adverse events and genitourinary function.

Here is what Dickinson and his colleagues report:

  • 569 patients with localized prostate cancer were enrolled and treated at one of the eight hospitals.
  • Average (median) patient follow-up was 46 month.
  • Repeat treatments with HIFU were acceptable as part of the treatment intervention over time.
  • The average (mean) age of the patients was 65 years (range, 47 to 87 years).
  • The average (median) PSA level of the patients was 7.0 ng/ml.
  • The number of patients in each National Comprehensive Cancer Network (NCCN) risk category was
    • 161/569 (28 percent) had low-risk disease.
    • 321/569 (56 percent) had intermediate-risk disease.
    • 81/569 (14 percent) had high-risk disease.
  • 163/569 patients (29 percent) required a total of 185 repeat HIFU procedures (implying that about 20 patients had more than one repeat HIFU).
  • Actuarial failure-free survival at 5 years after first HIFU was
    • 87 percent for low-risk patients
    • 63 percent for intermediate-risk patients
    • 58 percent for high-risk patients
    • 70 percent for the entire patient cohort
  • With respect to side effects and complications of treatment,
    • Urinary tract infections were observed after 58/754 HIFU treatments (7.7 percent).
    • Recurrent urinary tract infections were observed after 22/754 HIFU treatments (2.9 percent).
    • Epididymo-orchitis was observed after 22/754 HIFU treatments (2.9 percent).
    • Endoscopic interventions (for any one of several possible reasons) were necessary after 227/754 HIFU treatments (30.1 percent).
    • There was also
  • 183/206 patients known to be pad-free prior to treatment (88 percent) remained pad-free post-treatment
  • 91/236 patients who reported good erectile function prior to treatment (39 percent) reported good erectile function post-treatment.

Dickinson et al. conclude that:

Whole-gland HIFU is a repeatable day-case treatment that confers low rates of urinary incontinence. Disease control at a median of just under 5 years of follow-up demonstrates its potential as a treatment for non-metastatic prostate cancer. Endoscopic interventions and erectile dysfunction rates are similar to other whole-gland treatments.

Now there are differing ways of looking at these data, depending on one’s perspective.

In the first place, this is the only even relatively long-term data from a large cohort of patients treated by initially relatively inexperienced  HIFU providers. In other words, most of the clinicians actually carrying out the HIFU procedures in this series of patients had probably done less than 50 HIFU treatments themselves prior to initiation of the study. To that extent, these data represent “real world” outcomes from HIFU treatment as opposed the the results claimed (although not published) by some of the experienced US-based and European HIFU specialists.

Second, it is easy to look at these data and react by stating that these aren’t very good outcomes — especially for men with intermediate- and high-risk prostate cancer, and that will certainly be a justifiable reaction if the data at 10 years of follow-up were to show a continuing, significant increase in disease recurrence over the ensuing 5 years.

Third, it is clear that there is still some risk for recurrent incontinence in a subset of patients, since 12 percent of the patients who were pad-free prior to treatment were using at least one pad per day post-treatment at an average of 46 months of follow-up.

Fourth, at an average of nearly 4 years of follow-up, well over half the men who claimed good erectile function prior to treatment no longer claimed good erectile function (but at least part of that could be due to patient perception and simple aging).

The other way to look at this is very different. Some men are going to look at these data and say that what they see here are actually pretty good outcomes from less experienced providers for a one-day, in-office procedure (as opposed to multiple treatments using whatever form of external beam radiation therapy or opposed to the known risks of surgery). In particular, if one gets HIFU from a highly experienced HIFU provider — of whom there are currently very few in most countries — would the chances of good oncologic outcomes go up and the chances of side effects and complications go down? Frankly, we don’t have an answer to that question because we have those sorts of long-term data only from one provider in Japan, and “one swallow does not make a summer”.

The “New” Prostate Cancer InfoLink continues to take the position that HIFU is now an available option for the treatment of localized prostate cancer in most countries around the world. Only the individual patient can determine whether it is a “good” option for him. However, if one wants to consider (or have) HIFU as a first-line treatment for localized prostate cancer, we strongly suggest seeking out a specialist with a high degree of experience and skill in the use of this procedure, just as one should do for any other form of interventional treatment for localized prostate cancer.

The importance of this paper is that it has provided us — yet again — with some really detailed data on HIFU outcomes from a relatively large cohort of patients who have been carefully followed over time, and which includes data from patients treated by providers who were still in the “early days” of their experience in actually carrying out this procedure.

13 Responses

  1. This seems to me the worst of both worlds — it has about the same urinary and sexual toxicity as surgery: 12% incontinence and 39% potency preservation, while the cancer control is about 10% points lower in each risk level. With such relatively high levels of recurrence, I think it’s important to consider what the salvage therapies might be and what toxicity attaches to those. Perhaps the full text includes toxicity and cancer control data on the 29% who required re-dos. Perhaps too there is data on those treated early vs those who were treated later when the physicians had more experience. Based on these data, I would never recommend whole-gland HIFU (or probably any other whole gland ablation) as a primary therapy to a patient. If ablation has a role, it is in focal treatment, salvage, or possibly, ablation of metastases.

  2. As I said, Allen, there are two ways to look at these data. Yours is one of them, and I don’t expect people who take that perspective to be able to place any value on the other one.

  3. Still Disappointed in HIFU (except by the Uchida team)

    I’m joining in the negative perspective mentioned by Sitemaster and expressed by Allen. I find these results unimpressive, echoing a string of unimpressive HIFU results worldwide except the recent results of the Uchida team.

    I am also disappointed in the quality of the abstract. I expect some sloppiness from candidates running for president, but how can the authors claim “5 year results” when the median follow-up is 46 months, as Sitemaster noted (and as stated in the abstract). Is it just me, or is 46 months — not quite 4 years, the new equivalent for 5 years by some new method of teaching arithmetic? Moreover, based on the “interquartile” range, a full quarter of these patients having the shortest follow-up had a follow-up of less than 23 months! That means the robustness of the results for patients in the entire group are being diluted by very short follow-up by a large proportion of the group! Ugh!

    And now let’s look at the failure criteria: notice the omission of the need for repeat HIFU as a sign of failure? Now, overall, “success” (“failure-free survival”) was 70% for the entire group, but if you add re-treatment to the failure criteria, and then subtract 29% for patients who had to be retreated, you get an adjusted success percentage of 70% – 29% = 31%. Yuk! And that’s with shakey follow-up of just under 4 years! Double yuk!

    I hope these researcher/physicians will forgive a little sarcasm on my part and will recognize that they could have written the abstract more precisely for a reader now tired of some off-the-mark political rhetoric in the US. As Sitemaster notes, the authors have done a service by tracking and presenting this data. I expect and hope they will be in the forefront of adopting some of the advances pioneered by the Uchida group, advances hailed by Dr. Scionti in America, and will someday in the future achieve real success.

    But for now, I too am thinking that one swallow does not make a summer, and the flock is not yet in sight.

  4. Dear Jim:

    You apparently do not understand that things like 5- and 10-year survival data are commonly based on actuarial projections and can be and are carried out before all of the patients have actually been followed for 5 years of 10 years. This type of actuarial projection is commonplace. Apparently you haven’t noticed this before.

    The other thing is, the results obtained by someone like Uchida, using the very best treatment and doing almost nothing other than HIFU, tend to be much better than the results from a group like this. We see this all the time, for all sorts of treatment. Very few other people have ever been able to replicate the sorts of results that Pat Walsh got for surgery either. Part of this is patient selection; part of it is the skill and experience of the surgeon.

    The point of this paper, from my point of view, is that this is what you could expect if a whole lot of urologists and others started to use HIFU in a widespread way tomorrow. These data do notg reflect what it might be possible for a small number of highly specialized HIFU experts to achieve.

    Since only Uchida and the Ahmed/Emberton group have ever published these types of data, we have very little to go on at the moment.

  5. Point taken. I missed the word “actuarial”, and that negates my reaction to sloppiness in the abstract. I believe I missed it because I was disturbed by the short follow-up for this study — median follow-up of just 46 months, and that is obviously a hard number, not a projection. I’ll stand by my points about the interquartile range and the questionable failure-free criteria.

  6. Jim:

    You may think that the median follow-up is short at 46 months. However, I would point out to you that no one else has published a single paper on a large, detailed, prospective cohort of patients like this, providing all the side effect data, with anything even approaching a 46-month follow-up … not even Uchida.

  7. Please take a look at the Uchida team paper published in 2014.

  8. Jim:

    I have seen this abstract before and I have actually read the entire paper too.

    You need to appreciate that this is a retrospective analysis of Uchida’s data as opposed to a prospective study in which things were being tracked as they actually happened.

    The problem with all retrospective analyses is that you aren’t getting carefully compiled data at the times things happen. You just look into the files and report on what you find there. All retrospective analyses like this have a major inherent problem, which is that the researchers tend to find what they are expecting to find.

    The recent Ahmed/Emberton paper was based on a prospective study plan in which things were being tracked as they actually happened.

  9. The 2014 Uchida paper versus the Dickinson (Emberton) paper

    I see your point that the Dickinson paper is prospective, which I had missed, while the Uchida paper is retrospective, and I appreciate your service in making these important and clarifying distinctions for us. However, to me the Uchida retrospective paper is at the upper end of quality for retrospective studies. Part of the reason I say that is that the Uchida team has been diligent about reporting their results over the years, including unfavorable results with all the warts.

    Here is an important advantage in the Uchida 2014 data over the Dickinson data: the Dickinson data included all patients in the study regardless of how short the follow-up was, including those with up to 2 years of follow-up (from the abstract: “Median follow-up was 46 (interquartile range 23-61) mo.”, which I’m taking in its usual meaning as the figures for the first and third quartiles, in other words, the 25% and 75% points). In sharp contrast, the Uchida results exclude patients followed up for 2 years or a shorter period. That is a key distinction as patients do very well on HIFU in all studies I have seen for the first 2 years; the problem with HIFU generally is that failures start to mount up during the third year, with the upward trend in failure continuing to discouraging levels as time passes, with results that are clearly uncompetitive with established therapies by about the fourth or fifth years (except for the Uchida work with newer technology first published in mid-2014). By including more than one quarter of their group with less than 2 years of follow-up (based on their published figures for the interquartile range of follow-up), the Dickinson team clouds the real failure rate that matters for HIFU, with the inclusion making results look better than they really would be based on longer follow-up. By excluding patients in the first 2 years, the Uchida team is giving us a sharper focus on what their real failure rates are, and those rates are impressive for their group that was treated with the newest technology.

    (It might be worthwhile to run some figures using the Uchida failure rates with older technology for the first 2 years, which they have reported, applied to the number of patients in that first Dickinson quartile, then subtracting the low-failure in that group per Uchida from the overall failure for the entire Dickinson group applied to that quartile, and then adjusting the remaining Dickinson three quartiles to see what the failure rate might look like as adjusted. I’m already not impressed with the Dickinson results so don’t see this exercise as adding insight.)

  10. Dear Jim:

    You are making Mt. Everest out of a molehill. I am in no way suggesting that the Dickinson study data show HIFU in a good light. And I would like to see anyone repeat the Uchida data — because I don’t think they will. Uchida was treating a lot of men who should never have received treatment in the first place. Emberton’s group have been doing the very best they can to make sure that they do not treat men who should be on active surveillance.

    I can make absolutely the reverse argument about the inclusion/exclusion of men followed for less than 2 years. Arguably Dickinson et al.’s experience level was too low and the reason that they had to do so many repeats showed up in the first 2 years of follow-up.

    Can we please drop this? My only point to begin with was that we once again have a unique and highly detailed set of data from the Ahmed/Emberton group. No one else has been doing what they have been doing … which is publishing, prospectively, every bit of data that they can on HIFU. The lesson to everyone else (doctors and patients) is “Caveat emptor“. But I still believe that you can look at these data with two quite different sets of glasses, and nothing you are saying is going to change the minds of those who will look at these data with the glasses you aren’t putting on.

  11. I would like to add to the discussion but will gladly abide by your request to drop it. Thanks for all the work you put into these articles and comments.

  12. How do I find highly experienced HIFU experts for the treatment of localized, early-stage prostate cancer — preferably in the USA; but if not in another country with high treatment standards.

  13. That is a question that is almost impossible to answer. However, if you read through the commentaries referred to in this link, you will find that certain names come up over and over again. Other readers may be willing to make specific recommendations to you.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: