695 
Olivier Rouvière
MD PhD
Professor of Radiology
Department of Genitourinary Radiology
Hôpital E Herriot
Hospices Civils de Lyon
Unit 556
INSERM
Lyon
France
During the past decade, many different minimally invasive techniques have been developed and evaluated as an alternative to open surgery in the treatment of localised malignancies. Among these techniques, high-intensity focused ultrasound (HIFU) ablation is one of the most used.
When a high-intensity ultrasound beam is focused on a specific target point, all of the beam energy is concentrated on that point, which produces a dramatic temperature rise (up to 80ºC in a few seconds), causing coagulation necrosis of the tissues. The normal tissues located between the transducer and the focal point remain unaffected and, thus, this technique has the potential for treating deep tumours by using an external or an endocavitary transducer. The volume of tissue destruction after a single shot is usually small (a few cubic centimetres around the focal point). Since the target that needs to be destroyed is usually larger, it is necessary to repeat the shots and create focal elementary lesions next to one another throughout the target volume.(1,2)
HIFU ablation is now being used to treat solid tumours in different organs, including liver, breast and kidney, and in bone.(2) However, the most advanced clinical application is by far the treatment of organ-confined prostate cancer, with approximately 10,000 patients treated worldwide.(3) In this case, the HIFU beam is delivered through a transrectal transducer. Two different devices are currently used: Ablathermâ„¢ (EDAP, France) and Sonablateâ„¢ (Focus Surgery, USA).
Transrectal HIFU ablation is mainly used in patients with clinically localised prostate cancer that are not suitable candidates for radical prostatectomy. In this group of patients, medium-term results that challenge those of radiation therapy have been published by several independent research teams, with five-year disease-free survival rates of 66-78%.(4-6) A PSA level of ≤10 ng/ml, a Gleason score of ≤6 and a number of positive sextants at biopsy ≤4 are good prognostic factors.(5-7) One of the potential advantages of HIFU ablation over radiation therapy is that the treatment can be repeated, since there is no upper limit of tissue tolerance to repeated ultrasound exposure. Prostates larger than 40cubecm cannot be treated entirely because of the limitations in focal lengths of existing systems. However, because the prostate shrinks in the months following treatment, the part of the prostate that was beyond the focal length for the first session can be treated during a second session. After HIFU ablation, grade III stress incontinence rate is similar to that induced by radiation therapy (0-3.9%). A pretreatment transurethral resection of the prostate significantly decreases postoperative urinary catheterisation time(8) and is now systematically performed at our institution. The most serious HIFU-related complication is urethrorectal fistula, due to necrosis of the rectal wall. Its incidence, initially reported to be 0.7-3.2%,(4,7,9) has been dramatically reduced recently, as exemplified in published studies, owing to implementation of safety features, optimisation of the firing parameters and increased experience of the operator.
Another emerging indication is the treatment of local recurrences after radiation therapy. These recurrences are difficult to treat by conventional means. Further irradiation would lead to risks of serious injury to the urethra, bladder and rectum. Salvage prostatectomy is technically difficult and carries a high rate of morbidity.(10) Therefore, these patients usually end up being treated by hormonal deprivation, which provides effective tumour control but is of limited duration. In this context, HIFU ablation has given promising preliminary results, with 80% of patients with negative post-HIFU biopsy and 44% of absence of disease progression after a mean follow-up of 14.6 months.(11)
These good clinical results would not have been possible without a recent breakthrough in imaging, which allowed accurate targeting of the ablation, assessment of the treated volume and early depiction of recurrences. However, further improvements of the imaging techniques are needed to improve the outcome of prostate cancer HIFU ablation.
Patient selection
It is essential to detect preoperatively those patients with occult metastases, as HIFU ablation would be useless in such cases. This is particularly needed for patients with local recurrence after radiation therapy, because HIFU ablation is more risky in these patients and because recurrent cancer is usually poorly differentiated, with a high risk of occult metastasis. Unfortunately, the imaging techniques currently used (abdominal CT/MRI for node metastasis and bone scan for bone metastasis) lack sensitivity. However, new imaging techniques have yielded very promising results. Bone MRI(12) and 18F-fluoride PET(13) seem more sensitive than bone scan for detecting bone metastases. Pelvic MRI after injection of lymphotropic superparamagnetic nanoparticles has the potential for depicting microscopic invasion of lymph nodes.(14) PET/CT scanning could be an interesting one-stop technique for detecting both lymph nodes and bone metastases.(15) The cost-effectiveness (and therefore their respective role in the assessment of patients who could be treated with HIFU ablation) remain to be determined. However, there is no doubt that some of them will replace the traditional pair, “abdominal CT/bone scan”, in the near future.
Mapping of the intraprostatic distribution of the tumour
Even if the whole prostate has to be treated with HIFU, knowing in advance the position of the tumour nodules in the gland could dramatically increase the outcome of the treatment, by allowing a better targeting of the shots and an increase of the safety margins around the cancerous foci.
However, even in 2007, prostate cancer remains difficult to distinguish from normal surrounding tissues. Transrectal ultrasound, colour/power Doppler, T2-weighted MRI and even MR spectroscopy have all shown limited accuracy in tumour detection. The most important problems include false-positive findings due to some benign conditions (such as postbiopsy bleeding or prostatitis), the natural heterogeneity of benign prostate hyperplasia that makes the depiction of cancer difficult and the lack of natural contrast between some cancers and the normal glands. Interesting results have been published with contrast-enhanced dynamic MRI (Figure 1), but this technique, which has not yet been standardised, needs further assessment.(16,17)
[[HHE07_fig1_R7]]
Research on emerging techniques such as diffusion-weighted MRI, ultrasound (quasistatic) or MR elastography is greatly encouraged.
Assessment of the ablated volume
Contrast-enhanced (nondynamic) MRI is currently the best technique to evaluate the position and extent of the treated volume that appears totally devascularised (Figure 2).(18) However, the homogeneity of the tissue destruction within the treated area cannot be assessed by any current imaging technique. Such a technique would be useful to evaluate the result of the HIFU ablation in the operative room, immediately at the end of the procedure, in order to treat the patient again if necessary. Since coagulation necrosis is stiffer than normal tissue, elastographic techniques might be useful in this indication.
[[HHE07_fig2_R7]]
[[HHE07_fig3_R8]]
Detection of residual/recurrent tumours
If microscopic islets of undestroyed tumour tissue cannot be detected at the end of the HIFU ablation, one should at least be able to depict them, in the post-treatment months, when they are still amenable to a second HIFU application. Colour Doppler has recently been proved to be useful in guiding post-HIFU biopsy towards residual hypervascular foci that are four times more likely to contain viable cancer.(19) Dynamic MRI could also be an alternative; this technique is currently under evaluation.
Preoperative evaluation of perfusion
Theoretical calculations suggest that tissue blood flow plays a crucial role in tissue cooling, even when high temperatures (>55ºC) are obtained. A recent study has confirmed this by showing that the mean preoperative prostate perfusion rate, measured by MRI, was significantly lower in responders to HIFU ablation than in nonresponders (Wiart M, et al. ISMRM meeting presentation, 2006). However, further research is needed to understand how the local perfusion rate influences the treatment outcome and how the firing parameters could be adapted to the prostate perfusion rate of each patient.
MR-guided HIFU ablation
MR-guided treatments could be an interesting one-stop procedure, with a preoperative assessment of the cancer distribution (especially if the good preliminary results of contrast-enhanced dynamic MRI are confirmed), monitoring of the temperature during the procedure (and ideally active feedback control of the firing parameters based on the temperature measurements) and an assessment of the ablated volume (by gadolinium-enhanced nondynamic MRI and, maybe, by MR elastography).
However, MR guidance is much more expensive than the current ultrasound guidance, and its cost-effectiveness will have to be closely evaluated.
References
1. Chapelon JY, et al. Eur Urol 1992;22:147-52.
2. Kennedy JE. Nat Rev Cancer 2005;5:321-7.
3. Chaussy C, et al. Nat Clin Pract Urol 2005;2:191-8.
4. Blana A, et al. Urology 2004;63:297-300.
5. Poissonnier L, et al. Eur Urol 2007;51:381-387.
6. Uchida T, et al. Int J Urol 2006;13:228-33.
7. Gelet A, et al. Eur Urol 2001;40:124-9.
8. Vallancien G, et al. J Urol 2004;171:2265-7.
9. Thuroff S, et al. J Endourol 2003;17:673-7.
10. Schellhammer PF, et al. J Urol 1993;150:1851-5.
11. Gelet A, et al. Urology 2004;63:625-9.
12. Ghanem N, et al. Eur J Radiol 2005;55:41-55.
13. Langsteger W, et al. Semin Nucl Med 2006;36:73-92.
14. Harisinghani MG, et al. N Engl J Med 2003;348:2491-9.
15. Jana S, Blaufox MD. Semin Nucl Med 2006;36:51-72.
16. Futterer JJ, et al. Radiology 2006;241:449-58.
17. Girouin N, et al. Eur Radiol 2007; 17:1498-1509.
18. Rouvière O, et al. Eur Urol 2001;40:265-74.
19. Rouvière O, et al. Eur Urol 2006;50:490-7.
