Prostate imaging goals shift as therapeutic options expand

November 2, 2005

Many of the emerging imaging modalities for prostate cancer were first developed for breast cancer patients and then adapted to the prostate. Clearly, almost all aspects of prostate cancer care and diagnosis lag behind those of breast cancer. Breast cancer researchers have successfully carried out a number of groundbreaking randomized controlled trials to define the best treatment approaches, but prostate cancer has yet to see a successful trial comparing even the major forms of prostate cancer treatment: radical prostatectomy and radiation therapy. Adding to the frustration that prostate patients face when no one can tell them the best approach for treating their cancer, two new primary therapies have been sanctioned by the American Urologic Association: cryotherapy and watchful waiting.

Many of the emerging imaging modalities for prostate cancer were first developed for breast cancer patients and then adapted to the prostate. Clearly, almost all aspects of prostate cancer care and diagnosis lag behind those of breast cancer. Breast cancer researchers have successfully carried out a number of groundbreaking randomized controlled trials to define the best treatment approaches, but prostate cancer has yet to see a successful trial comparing even the major forms of prostate cancer treatment: radical prostatectomy and radiation therapy. Adding to the frustration that prostate patients face when no one can tell them the best approach for treating their cancer, two new primary therapies have been sanctioned by the American Urologic Association: cryotherapy and watchful waiting.

As a radiologist who primarily treats prostate cancer patients on a national and international referral basis, I have a unique perspective. The overwhelming majority of patients I see have already been diagnosed and staged by both a urologist and a radiation therapist. I therefore see a representative cross-section of prostate cancer patient evaluations. This experience has shown me that, except for prostate ultrasound, which is used to guide biopsy, imaging has virtually no impact on the care of the primary prostate cancer patient. Prostate MRI both with and without spectroscopy has yet to be used on a routine basis, and ProstaScint scanning has not shown a high enough sensitivity or specificity to affect decision making in patient management.

Still, these disappointments in prostate cancer imaging should not blunt our enthusiasm or optimism for the future. Advances in molecular imaging and radiographic equipment-3T scanners and combinations such as PET/CT-hold great promise for prostate imaging that could have a positive impact on the prostate cancer patient.

Rather than comprehensively review the latest prostate cancer imaging modalities, I will consider here the major clinical trends in care and how these should guide our future efforts in imaging. If the past is any indication, resources for research into prostate cancer imaging will be limited. Our efforts should therefore be optimally directed to answer the needs of both the prostate cancer clinician and the patient.


Prostate cancer imaging has no radiographic equivalent of the mammogram for breast cancer. A surge of enthusiasm that transrectal ultrasound might meet this need was not borne out when TRUS, even coupled with a rectal exam, did not show the sensitivity needed for a screening procedure. Screening for prostate cancer is also a nexus for controversy mainly centered on prostate-specific antigen testing. Recent articles have suggested that elevated PSA reflects the degree of benign prostatic hyperplasia rather than the presence of prostate cancer,1 reducing acceptance of PSA testing as a standard for prostate cancer screening.

It would thus seem that a role for imaging in prostate cancer screening would be worth pursuing. Based on what we know, however, about the costs and capabilities of the likely imaging modality candidates, such as MRI, introducing a valid screening method for prostate cancer may require creative thinking.

Pursuit of a totally novel approach to prostate cancer imaging such as electrical impedance tomography, for example, might be worthwhile. Cancer tissue has an electrical conductivity different from normal or benign tumor tissue. By placing an electrical current into the tissue and then creating a map of the impedance of the tissue, cancer could presumably be detected (Figure 1). Research into this modality is ongoing, and the FDA has already approved a similar device for diagnosing breast cancer.

EIT has some unique characteristics that make it particularly suited as a screening method. Since the EIT device is simply an array of electrodes fixed in a known orientation to the organ, the cost of the hardware of the device itself is negligible, a first for an imaging modality. This makes possible what Professor Boris Rubinsky of the University of California, Berkeley has called distributed network imaging.2 This concept calls for patients being screened to use the imaging device themselves at home. Through an Internet connection, the data are collected and processed at a remote site. The image can then be either read by a radiologist or initially screened by computer, with only the questionable scans being read. Patients would be able to periodically scan themselves and have the data placed into their database to be compared with previous scans. The mental picture of men all over the country using their own rectal imaging probe for prostate cancer screening may be a strange one, but the men I have dealt with would surely prefer this to current alternatives.

EIT might also have an advantage in addressing the thorny problem of cancer detection in the transition zone of the prostate. Benign hypertrophy and cancer have the same basic characteristics on both MRI and TRUS. The electrical characteristics of BPH can differ significantly from those of cancer, however, suggesting that EIT could better characterize cancer and benign central gland abnormalities, a significant potential advantage. This could help solve a major problem in prostate cancer screening, but it remains to be confirmed.


Like real estate, the three most important parameters of prostate cancer imaging are location, location, location. Prostate cancer spreads extracapsularly in a predictable manner, based on inherent weaknesses in the prostatic capsule. It moves into locations where the capsule is traversed by structures such as neurovascular bundles, the confluence of the seminal vesicles leading into the ejaculatory ducts, or areas of inherent capsule thinning at the apex of the gland.

With the advent of nerve-sparing radical prostatectomy to preserve potency, the risk of positive surgical margins increased. One of the failed goals of prostate imaging has been detection to a high degree of certainty of extracapsular extension of prostate cancer, which could aid in patient selection for nerve-sparing RP. This failure is not surprising, as the extracapsular extension is microscopic in most patients, where cancer is adjacent to weak areas of the capsule affected by penetrating vessels and nerves. Even during an operation, it is not possible for the surgeon to tell with accuracy better than chance whether a neurovascular bundle has been infiltrated by tumor.3 It appears doubtful that imaging in the near future will achieve the submillimeter resolution needed to accurately define this problem. For radiologists to continue to pursue this as a major goal is a setup for failure that will falsely devalue prostate imaging.

The best clinical tool to define patients at risk for extracapsular extension is the Partin nomogram. Partin nomograms use the patient's risk factors-PSA level, stage by rectal exam, and Gleason grade-for a risk assessment of extracapsular extension and lymph node and seminal vesicle involvement. What nomograms do not take into account, however, is tumor location. Adding the tumor location into this equation might logically improve the accuracy of these predictions. I consider one of the main goals for prostate imaging to be accurate location of the cancer in relation to the prostate structures at risk in order to improve the predictive value of the Partin nomograms.

Accurate localization of the tumor and its extent will also play a major role in the shift to a more focal treatment of prostate cancer. Recent studies have questioned the efficacy of an aggressive treatment approach, and current management of prostate cancer covers both ends of the treatment spectrum. Patients can elect no treatment at all (i.e., watchful waiting)4 or aggressive whole-gland treatments such as radical prostatectomy, with no middle ground available. If no treatment at all can be advocated for a subset of prostate cancer patients, then the compromise of attempting to destroy just the focus of cancer in the gland should be a viable option.

The use of focal treatment of breast cancer (i.e., lumpectomy) has revolutionized local control of that disease. Lumpectomy showed that the quality of life of the individual patient can successfully be integrated into the equation of cancer treatment without major loss of efficacy.5 Prostate cancer in men raises many of the same issues that breast cancer does in women. Complications of prostate cancer treatment that produce impotence and incontinence affect the male self-image and psyche no less than the loss of a breast affects a woman's. We must ask if a focal prostate-sparing approach can produce good cancer control and significantly lower morbidity.

Cryoablative therapy is now an accepted procedure for the treatment of primary prostate cancer. Since we published the first clinical series of prostate cryoablation in 1993,6 it has been approved by Medicare as a primary treatment for prostate cancer (2000), has been approved specifically for radiation failure patients (2001, the only such treatment), and has gained the sanction of the American Urological Association. With ablation established as a competitive treatment for prostate cancer, we can now explore the flexibility of the modality as a focal treatment.

Prostate cancer has long been considered a multifocal disease and therefore not amenable to focal treatment. Since partial removal of a prostate is not technically feasible, challenging this assumption held little clinical promise. An examination of the pathology literature, however, indicates that approximately 30% of prostate cancers are single in nature, making approximately 60,000 patients a year candidates for a lumpectomy-type approach.7,8 The literature indicates that cancers smaller than 5 mm in diameter are not clinically significant.8 If those patients with a single index tumor and an additional lesion smaller than 5 mm are considered, the number amenable to a lumpectomy approach rises as high as 70% of patients.7,8

Preliminary data regarding whether focally treating prostate cancer provides good cancer control and significantly lower morbidity are very encouraging. At this year's Society of Interventional Radiology meeting, we presented a paper in which a group of 40 patients with one to nine years' follow-up (mean 3.5 years) were treated with just ablation of their primary prostate tumor. The data showed that cancer control was excellent: 95% of patients were biochemically disease-free and only 7% needed additional treatment in another area of the gland. Morbidity was extremely low, 0% of patients experienced incontinence requiring pads (compared with 6% and 16%, respectively, for RP and radiation),9,10 and 80% maintained potency to their satisfaction, compared with 30% for RP sparing one nerve bundle and 50% for radiation (Figure 2).11,12 These data have recently been confirmed by Bahn and Masson, who have reported preliminary results in similar groups of patients (see table).13,14

MRI contrast kinetics holds particular promise in determining the extent and location of prostate cancer. The basis of the technology developed by 3TP is prostate cancer's greater vascularity compared with normal tissue. When the wash-in and washout of contrast are quantified, prostate cancer will have a characteristic kinetic signature. While characteristics of cancer and BPH appear to overlap in the transition zone, the technology seems to have a very high negative predictive value in both the transition zone and the peripheral zone (Figures 3 and 4). Being able to reliably exclude cancer in an area would limit the need for biopsies in those areas.

Based on an informal survey indicating that 70% of physicians performing cryoablation are already offering patients some form of focal ablative therapy, I have little doubt that the determination of the location and extent of prostate cancer in order to optimize focal therapy will be a major goal of imaging in the future.


One of the most difficult clinical situations we face in treating prostate cancer is the post-primary treatment patient with rising PSA. The crux of the problem is determining whether the recurrent disease is confined to the prostate (the patient is still potentially curable) or involves the lymph nodes and/or bones. Our experience with this group is that the bone scan usually does not indicate metastatic disease and the main question to be answered is involvement of the patient's lymph nodes. When ProstaScint scanning first became available, there was great hope that we would finally have an imaging test that could direct us in the management of these patients. Unfortunately, it is the rare urologist still ordering ProstaScint scans. The urologic community has concluded that the sensitivity and specificity of the test have not been high enough to guide clinical decision making.

I am cautiously optimistic that contrast-enhanced MR using iron oxide nanoparticle (Combidex)-based lymph node imaging will assist radiologists in solving this clinical problem. An early study published in The New England Journal of Medicine15 showed that 33 of 33 patients with prostate lymph node metastasis were successfully identified (Figure 5). It is my hope that when Combidex is finally approved by the FDA, clinical experience will confirm these findings and we will finally have a handle on this difficult clinical problem.

With 200,000 new cases of prostate cancer diagnosed in the U.S. each year and virtually no routine imaging for diagnosis or staging being carried out in these patients, radiologists have a great opportunity to positively affect a large patient population. It is critical that research resources be applied to answering the crucial clinical question that emerging new concepts will raise: how best to manage this disease, which every man risks facing.


1. Stamey TA, Caldwell M, McNeal JE, et al. The prostate specific antigen era for prostate cancer is over in the United States. What happened in the last 20 years? J Urol 2004;172:1297-1301.

2. Otten DM, Onik G, Rubinsky B. Distributed network imaging and electrical impedance tomography of minimally invasive surgery. Technol Cancer Res Treat 2004;3(2):125-34.

3. Vaidya A, Hawke C, Tiguert R, et al. Intraoperative T staging in radical retropubic prostatectomy: is it reliable? Urology 2001;57(5):949-954.

4. Messing EM, Thompson I. Follow-up of conservatively managed prostate cancer: watchful waiting and primary hormonal therapy. Urol Clin North Am 2003;30(4):687-702.

5. Santiago RJ, Wu L, Harris E, et al. Fifteen-year results of breast-conserving surgery and definitive irradiation for Stage I and II breast carcinoma: the University of Pennsylvania experience. Int J Radiat Oncol Biol Phys 2004;58(1):233-240.

6. Onik GM, Cohen JK, Reyes GD, et al. Transrectal ultra- sound-guided percutaneous radical cryosurgical ablation of the prostate. Cancer 1993;2:1291-1299.

7. Rukstalis DB, Goldknopf JL, Crowley EM, et al. Prostate cryoablation: a scientific rationale for future modifications. Urology 2002;60:2A 19-25.

8. Noguchi M, Stamey TA, McNeal JE, et al. Prognostic factors for multifocal prostate cancer in radical prostatectomy specimens: Lack of significance of secondary cancers. J Urol 2003;170(2pt1) 459-463.

9. Catalona WJ, Basler JW. Return of erections and urinary continence following nerve sparing radical retropubic prostatectomy. J Urol 1993;150(3):905-907.

10. Talcott JA, Clark JA, Stark PC, Mitchell SP. Long-term treatment related complications of brachytherapy for early prostate cancer: a survey of patients previously treated. J Urol 2001;166(2):494-499.

11. Sanchez-Ortiz RF, Broderick GA, Rovner ES, et al. Erectile function and quality of life after interstitial radiation therapy for prostate cancer. Int J Impot Res 2000;12 Suppl 3:S18-24.

12. Van der Aa F, Joniau S, De Ridder D, Van Poppel. Potency after unilateral nerve sparing surgery: a report on functional and oncological results of unilateral nerve sparing surgery. Prostate Cancer Prostatic Dis 2003;6(1):61-65.

13. Bahn (personal communication, 7/1/05).

14. Masson P, Kristofer L, Prepelica, et al. Focal cryosurgery: encouraging health outcomes for unifocal prostate cancer. AUA 2005 San Antonio Texas.

15. Harisinghani et al. Non invasive detection of occult lymph node metastasis in prostate cancer. NEJM 2003:248 2491-2498.

Dr. Onik is director of surgical imaging at Celebration Health in Celebration, FL. He is a consultant with Endocare.