Jelle Barentsz, M.D., Ph.D
Professor of Radiology – Chairman for Research
Department of Radiology
University Medical Center – St. Radboud Nijmegen
Nijmegen, The Netherlands
Edited from PCRI Insights November, 2008 v 11.4
Editor’s Note: This paper discusses ferumoxtran-10 (Combidex). AMAG Pharmaceuticals has discontinued the manufacture of Ferumoxtran-10 in 2010 making Combidex no longer available. They are currently pursuing approval of ferumoxytol as a replacement MRI agent. Watch for available trials.
Confirming the diagnosis of prostate cancer is not always easy.
A patient with complaints and an elevated PSA does not necessarily have prostate cancer. Developments in MR imaging are rapid. This paper describes how the use of MR imaging can be of great value for patients with prostate cancer.
Prostate cancer is one of the most prevalent forms of cancer in men, and this statistic is increasing, partly due to screening and partly due to the aging of the population. Confirming the diagnosis of prostate cancer is not always easy. Complaints of pain or discomfort and/or an elevated serum level of the biomarker prostate specific antigen (PSA) do not necessary mean that the patient has prostate cancer. In many cases (80%), it is “just” a non-malignant disease of the prostate.
Detection and localization
Current medical imaging techniques, such as transrectal ultrasound (TRUS) and conventional MRI, can visualize prostate disease but are barely able to distinguish between cancer and non-malignant disease. Ultrasound-guided biopsy and subsequent histological examination often give a definitive diagnosis of prostate cancer. But since biopsies are taken randomly, there is a risk that the tumor is missed or that the most aggressive part of the tumor is not biopsied. In the Gosselaar study1, prostate cancer was detected in 29% of the men in the first TRUS-guided biopsy. After the second biopsy, it was detected in 19%, and after the third biopsy, again in 19%. Sampling of additional ‘low signal’ areas on TRUS results in a positive tumor in only 3.5 %of men1.
Functional multi-modality MR imaging (See figure 1) includes high resolution MR imaging (1a), dynamic contrast-enhanced (DCE) MRI (1b), MR-spectroscopy (MRS) (1c), and diffusion weighted MR imaging (DWI) (1d). With functional multi-modality MR imaging, it is possible to detect and exactly localize the tumor in the prostate with more than 90% accuracy2. DCE-MRI gives more information regarding the perfusion of the prostate and the tumor. In prostate cancer, there are more blood vessels (and above all more leaky blood vessels) by which the tumor receives more blood and therefore shows more and earlier contrast enhancement. MRS provides quantitative information about choline (elevated in tumor) and citrate concentration (often depressed in tumor) in the prostate. DWI-MRI allows visualization of the amount of free-moving water molecules in tissue; in a tumor, this movement is diminished.
It is therefore not surprising that our first results with multi-modality MR-guided biopsy detected cancer in 59% of the 67 patients who had previously had at least 2 negative TRUS-biopsy sessions. (See figure 2).
Although a large number of men above age 50 will develop prostate cancer, only in a small proportion of patients will their tumor become aggressive. For the majority of men with non-aggressive tumors, one can delay invasive treatment until the tumor becomes aggressive (watchful waiting). It is essential to precisely characterize the tumor. This can be achieved with multi-modality MRI.
With nomograms based on the serum PSA-value, the tumor’s aggression (expressed in Gleason-grade), and the outcome of a digital rectal examination, an attempt is made to predict (1) the aggression and local extension of the tumor, and (2) the presence of lymph node metastasis. Based on these nomograms, the further treatment strategy is determined. Unfortunately, these nomograms are only moderately reliable, so that it is not always possible to make the right treatment decision. By means of functional (multi-modality) MR imaging, information can be acquired regarding the aggression of the tumor. Furthermore, an MR-examination, performed at high field strength (3 Tesla) with use of an endorectal coil (ERC), allows a very accurate determination of minimal (sub-millimeter) extraprostatic spread. The sensitivity and specificity of 3T ERC MR imaging for determination of extraprostatic disease are respectively 87% and 96%3. If it is decided to surgically remove the prostate, it is important to know where the tumor is located and whether it shows extra-prostatic growth (see figure 3). If the tumor is distant from the neurovascular bundles, these bundles can be spared. This decreases the chance of postoperative impotence. If the MRI shows obvious extraprostatic extension, it can be concluded that surgery is less useful so the best choice is hormonal therapy with or without radiotherapy.
Nowadays, prostate cancer is increasingly treated with directed, local radiotherapy. If the precise location of the tumor is known, it is possible to give a local boost to this location in the prostate. This has the advantage that less radiation is given to the surrounding tissue, with fewer side-effects. An MRI can supply this information4.
Lymph Node Metastasis
Hormonal therapy with or without radiation is the best treatment if there are metastases to the lymph nodes. The risk of lymph node metastasis is currently determined with the help of the previously described (inaccurate) nomogram. In patients with an elevated risk for metastasis, additional examinations are required. Currently, the most used imaging techniques for detecting lymph node metastasis are multi-detector CT scan (MDCT) and conventional MRI. The accuracy of MDCT and conventional MRI is not high. This makes supplementary invasive diagnostic examination in the form of surgical pelvic lymph node (PLN) dissection mandatory. There is a new MRI technique using a lymph-node-specific contrast agent (Combidex) (MRL)5,6.
Editor’s note: Combidex has had promising results in Europe, but it is not yet available in the U.S., and it may not get FDA approval soon. For more on Combidex, see:
Combidex consists of iron-oxide containing nanoparticles. When this contrast agent is administered intravenously, it is taken up by macrophages and transported to healthy lymph tissue. The iron causes changes in the magnetic characteristics of the tissue that result in low signal intensity on MR images. Therefore, 24 to 36 hours after Combidex injection, healthy lymph nodes are black on MR images due to the iron within macrophages. Macrophages are absent in lymph nodes with metastasis, and thus these lymph nodes do not have a low signal: the tissue is white (figures 4 and 5).
With MRL, it is possible to examine the entire abdomen instead of only a restricted area surrounding a few pelvic blood vessels, as is the case with PLN dissection. The sensitivity, specificity, negative and positive predictive values are respectively 82%, 93%, 96%, and 69%. In specialized centers, it is even higher: 90%, 94%, 98% and 75%, respectively. The diagnostic accuracy of MRL in the detection of lymph node metastasis is significantly higher than with MDCT6. The high negative predictive value (>96%) of MRL means that after a negative result on MRL, PLN dissection does not have to be performed. Due to the latter, obtaining a diagnosis with MRL is economically cheaper and results in fewer complications than with the current invasive diagnostic technique of MDCT + PLND. In addition, in at least in 30% of patients, thanks to MRL, nodes are detected which are not found by the routine PLN dissection, as they are located in the internal and common iliac, perirectal and para-aortic regions.
The positive and negative nodes can be made visible on a combined CT-angiography-MRL image (see figure 6). In this way, the location of positive nodes relative to the large pelvic vessels can be visualized. This information can help the surgeon to find these nodes or help the radiation oncologist to give selective lymph node radiotherapy. Unfortunately, the iron oxide contrast agent has not yet obtained official registration.
See Sidebar below:
An Alternative to CT Scan and Lymph-Node Dissection in Patients with Prostate Cancer: a Prospective Multicohort Study
If there is a PSA rise in a treated patient, the most important question is: is this the result of a local recurrence, or is it caused by lymph nodes or bone marrow metastases? MRI also has a role in this situation.
• First, bone metastasis can be excluded by means of a ‘whole-body’ MRI.
• If this is negative, then Combidex MRL should be used to exclude node metastasis. In post-treated patients, we found no correlation between PSA value or PSA doubling time and positive nodes. There was, however, a positive correlation between PSA velocity and positive nodes on MRL. Thus, even in patients with a low PSA, but high PSA velocity, a MRL is of use to exclude metastases.
Finally, multi-modality MRI can be performed to determine if there is a local recurrence (see figure 7).
This article describes the potential value of MR imaging for patients with prostate cancer. To summarize, MR imaging affords:
• Accurate detection of tumors, thereby providing increased accuracy and decreasing the number of unnecessary biopsies.
• Better determination of the location and extension of the tumor: this makes targeted radiotherapy or targeted treatments possible, and reduces side-effects of this treatment (damage to the intestine, impotence, incontinence).
• Better prediction of tumor aggression. For patients with a non-aggressive tumor, this means that one has the possibility to defer invasive treatment, and act based on subsequent MR results.
• Non-surgical detection of very small metastases in lymph nodes. This also means there is a possibility for more patients to undergo targeted therapy.
• For ‘PSA-recurrence’ patients, instead of a bone-scan a ‘whole-body’ MRI can be performed, followed by an MRL, and a local “functional” MRI. Based on this MRI, the eventual local treatment can be started.
This paper was made possible due to the work of the Radboud University Medical Center Nijmegen Prostate team:
• Radiology: Prof. Dr. J Barentsz, Prof. Dr. A Heerschap, Dr. Ir. T Scheenen, Dr. J Fütterer, Dr. Ir. H-J Huisman, Dr. Ir. N Karssemeijer, Drs. S Heijmink, Drs. T Hambrock, Drs. D. Yakar, Drs. O Debats, Drs. C Hoeks,
• Urology: Prof. Dr. J Witjes, Drs. I van Oort, Drs. R Somford,
• Radiotherapy: Dr. E van Lin,
• Pathology: Dr. C Hulsbergen, & Drs. C van Niekerk.
1. Gosselaar C, Roobol MJ, Roemeling S, Schröder FH. The Role of the Digital Rectal Examination in Subsequent Screening Visits in the European Randomized Study of Screening for Prostate Cancer (ERSPC). Eur Urol 2008 [Epub ahead of print].
2. Fütterer JJ, Heijmink SW, Scheenen TW, Ve ltman J, Huisman HJ, Vos P, Hulsbergen-Van de Kaa CA, Witjes JA, Krabbe PF, Heerschap A, Barentsz JO. Prostate cancer localization with dynamic contrast-enhanced MR imaging and proton MR spectroscopic imaging. Radiology 2006 241 (2): 449-58.
3. Fütterer JJ, Heijmink SW, Scheenen TW, Jager GJ, Hulsbergen-Van de Kaa CA, Witjes JA, Barentsz JO. Prostate Cancer: Local Staging at 3-T Endorectal MR Imaging-Early Experience. Radiology 2006 238 (1): 184-191.
4. Van Lin EN, Fütterer JJ, Heijmink SW, van der Vight LP, Hoffmann AL, van Kollenburg P, Huisman HJ, Scheenen TW, Witjes JA, Leer JW, Barentsz JO, Visser AG. IMRT boost dose planning on dominant intraprostatic lesions: gold marker-based three-dimensional fusion of CT with dynamic contrast-enhanced and 1H-spectroscopic MRI. International journal of radiation oncology, biology, physics 2006 65 (1): 291-303.
5. Harisinghani MG & Barentsz J (cofirst-authors), Hahn PF, Deserno WM, Tabatabaei S, van de Kaa CH, de la Rosette J, Weissleder R. Noninvasive detection of clinically occult lymph-node metastases in prostate cancer. The New England Journal of Medicine 2003 348(25): 2491-9.
6. Heesakkers RA, Hövels AM, Jager GJ, van den Bosch HC, Witjes JA, Raat HP, Severens JL, Adang EM, van der Kaa CH, Fütterer JJ, Barentsz J. MRI with a lymph-node-specific contrast agent as an alternative to CT scan and lymph-node dissection in patients with prostate cancer: a prospective multi-cohort study. Lancet Oncol. 2008 Aug 15.
An Alternative to CT Scan and Lymph-Node Dissection in Patients with Prostate Cancer: a Prospective Multicohort Study
Editor’s Note: From April 2003 to April 2008, a Radboud University Nijmegen Medical Center team led by Dr. Barentsz tested the hypothesis that a negative MR lymphography result obviates the need for a pelvic lymph node dissection. The following paragraphs, published in a slightly different form in Lancet Oncology, summarize the methodology and conclusions of that study, and amplifies Dr. Barentzs’s article, MRI in Prostate Cancer.
In patients with prostate cancer deemed to be at intermediate or high risk of harbouring nodal metastases, an invasive diagnostic pelvic lymph node (PLN) dissection continues to be the gold standard for the detection of nodal disease. However, recently a new lymph node-specific MR-contrast agent has been shown to be able to detect metastases in small normal-sized nodes (MR lymphography: MRL). In a prospective multi-centre cohort study1, we have compared the diagnostic accuracy of MRL with up-to-date multi-detector CT (MDCT), and have tested the hypothesis that a negative MRL result obviates the need for a PLN dissection.
We included 375 consecutive prostate cancer patients with an intermediate or high risk of harbouring lymph node metastases. The study was conducted in 11 hospitals in the Netherlands between 8 April 2003 and 19 April 2005. All patients were examined by MDCT and MRL and underwent PLN dissection or fine needle aspiration biopsy. Imaging results were correlated with histopathology. (This study is registered with ClinicalTrials.gov, number NCT00185029.)
The prevalence of lymph node metastases was 16% (61/375). For MDCT and MRL respectively, the sensitivity was 34% (95% CI (23-48) and 82% (95% CI 70-90); the specificity was 97% (95% CI 94-98) and 93% (95% CI 89-95); the positive predictive values (PPV) were 66% (95% CI 47-81) and 69% (95% CI 56-79); and the negative predictive values were (NPV) 88% (95% CI 84-91) and 96% (95% CI 93-98). The sensitivity of MRL was significantly better than MDCT (McNemar p< 0.05). Of the 61 patients with lymph node metastases, 50 were detected by MRL, of which 80% (40/50) had metastases in normal-sized lymph nodes (that is, in nodes < 8 mm). The high sensitivity and negative predictive value of an MRL imply that in patients with a negative MRL, the chance of positive lymph nodes is less than 4% (11/302).
The high NPV and sensitivity imply that patients with a negative MRL result have a chance of positive lymph nodes of less than 4%. It has been generally accepted that a PLN dissection should not be performed if the risk of lymph node involvement is less than 5%.2,3,4,5 Based on tables or nomograms with intermediate or high risk for lymph node metastases, this indicates that a PLN dissection can be omitted for patients with a negative MRL. Thus, after a negative MRL, a urologist or radiation oncologist can immediately proceed to local therapy without performing PLN dissection. In this study, it would have precluded a PLN dissection in 84% (302/375) of all included patients. Because a false positive result has serious clinical consequences (namely that a curative therapy might be withheld), a positive MRL finding should always be confirmed by (MR-directed) FNAB or PLND.
Thanks to a more extensive PLN dissection where the MRL showed positive nodes outside the normal surgical area, positive nodes were found in 30% (18/61) of additional patients. These findings are in agreement with Burkhard et al,6 who performed an extended PLN dissection in all of her patients, which yielded positive nodes in 24% additional patients. This is an advantage in terms of costs and morbidity.
The new imaging MRL technique potentially can modify current practice by abolishing the need for a PLN dissection in patients with a negative MRL. No PLN dissection needs to be performed any more. In patients with a positive MRL, the MRL can guide a FNAB or PLN dissection to obtain histopathology or to remove the affected node with minimal surgical effort. This may result in less direct post-surgical morbidity for the patient, less surgery time for the surgeon, lower health care costs, and a more accurate diagnosis.
In summary, we concluded that in patients deemed to be at intermediate or high risk of harboring lymph node metastases and in those centres with experience and dedication in the use and interpretation of MRL, this technique had significantly higher sensitivity and NPV in the detection of lymph node metastases than does MDCT. In such patients after a negative MRL, the post-test probability of having lymph node metastases is sufficiently low (less than 4%) to omit a PLN dissection.
1. Barentsz,J, Heesakkers RA, Hovels AM, Jager GJ et al. MRI with a lymph-node-specific contrast agent as an alternative to CT scan and lymph-node dissection in patients with prostate cancer: a prospective multicohort study. Lancet Oncology, 2008.
2. Bisfoff JT, Reyes A, Thompson IM et al. Pelvic lymphadenectomy can be omitted in slected patients with carcinoma of the prostate: development of a system of patient selection. Urology 1995;45:270-274.
3. Bluestein DL, Bostwick DG, Bergstrahlh EJ, Osterling JE. Eliminating the need for bilateral pelvic lymphadenectomy in select patients with prostate cancer. J Urol 1994; 151: 1315-20.
4. Meng MV, Carroll PR. When is pelvic lymph node dissection necessary before radical prostatectomy? A decision analysis. J Urol 2000; 164: 1235-40.
5. El Galley RE, Keane TE, Petros JA et al. Evaluation of staging lymphadenectomy in prostate cancer. Urology 1998;52:663-67
6. Burkhard FC, Schumacher M, Studer UH. The role of lymphadenectomy in prostate cancer. Nat Clin Pract Urol 2005; 2:336-42.