NovemberDecember 2004Vol11No 6Cancer Control 353 - PDF document

November/December 2004,Vol.11,No. 6Cancer Control 353 No significant relationship exists between the authors and the compa-nies/organizations whose products or services may be referenced inthis article. Abbreviations used in this paper:CT = computed tomography,MRI =magnetic resonance imaging,MRSI = magnetic resonance spectroscopicimaging,PET = positron emission tomography.Assess Tumor in the ProstateSeveral advances in the imaging of prostate cancer have been made in recent years.Diagnostic staging has become increasingly complex and confusing as newer technologies have developed more rapidly thanesearch has been able to confirm or refute the accuracy of these technologies.By the time research has been performed,the technology used for a study has often become outdated and newer and more sophisticated imaginghas become available.viewed the literature on local and nodal staging of prostate cancer,as well as the role of magneticesonance imaging (MRI),magnetic resonance spectroscopic imaging (MRSI),dynamic contrast-enhanced MRI,positron emission tomography (PET),endorectal power Doppler,lymphotropic MRI contrast agents,and future possibilities such as diffusion MRI.This review is not systematic,but rather focused on these imaging modalities.Advances in MRI,ultrasound,and lymphotropic contrast agents have improved our ability to differentiatebetween T2 and T3 prostate tumors.PET imaging has proven less successful at staging prostate cancer.A literatureview suggests patients with moderate risk of extracapsular extension benefit most from endorectal MRI evaluation.Spectroscopy,dynamic imaging,and lymphotropic contrast agents are expected to continue to improve sensitivityand specificity of staging of prostate cancer.Power Doppler evaluation with endorectal ultrasound has proved useful for evaluation during endorectal biopsy for identifying hypervascular tumors for directed biopsy.Diffusion-eighted MRI remains untested clinically and represents a future direction for research.Future studies using these new techniques are needed to demonstrate changes in outcomes in largeRecent advances in imaging techniquesor prostate cancer are reviewed.Michael Mahany.utumn Ptarmigan.Photograph.Denali National Park,Alaska.om the Department of Radiology at the H.Lee Moffitt Cancer Center& Research Institute,Tampa,Florida.Dr.Knapp is now at East PascoMedical Center,Zephyrhills,Florida.Submitted January 5,2004;accepted July 20,2004.Address correspondence to Marla R.Hersh,MD,Department of Radiol-ogy,H.Lee Moffitt Cancer Center & Research Institute,12902 Magno-lia Drive,Tampa,FL 33618.E-mail:hershmr@moffitt.usf.edu November/December 2004,Vol.11,No. 6354Cancer ControlUltrasound and computed tomography (CT) have beenthe ÒstandardÓmeasures to evaluate the prostate inpatients with prostate cancer.Sonographic examinationof the prostate is insufficiently sensitive or specific todetect prostate cancer.ansrectal ultrasound is used toguide sextant biopsy,but its usefulness for staging is limit-ed.CT has traditionally been used to evaluate the extentof local disease.More recently,magnetic resonance imag-ing (MRI) has also been used to evaluate for local disease.Borley et alhave shown that while CT and MRI have ahigh sensitivity for detecting lymph node metastases,theyhave poor sensitivity.Indium-111Ðcapromab pendetide(ProstaScint,Cytogen Corp,Princeton,NJ) has also beenused to evaluate lymph node involvement,relapse afterprostatectomy,and occult malignancy.However,Wilkin-ecently reviewed 42 cases and conclud-ed that ProstaScint did not offer additional benefit inassessing postprostatectomy patients for further therapy.Currently,ProstaScint imaging is not thought to providesufficient additional staging information to warrant its use.Although MRI of the prostate has been available since1984,there is a large variation in reported staging perfor-mance in the literature.These wide variations are pre-sumably due to sample size,variations in technique,studypopulation,and the constant change in state-of-the-arttechnology.This review addresses the new imaging tech-niqu

es for evaluating tumor in the prostate and theirdegree of support in the literature.We also review othertechniques such as ultrasound for their role in staging ofprostate cancer.As with all radiologic modalities,MRI continues to changepidly.Several changes have improved the examinationof the prostate.When endorectal coils became available in1984,they were large and uncomfortable for the patient.Currently,a flexible endorectal coil is positioned in a bal-loon that expands into the rectal vault after placement,holding the coil reliably into position with minimal dis-comfort.Until recently,one of the difficulties withendorectal coil examination was the inability to utilizemore than one coil.If an endorectal coil was employed,other coils could not be utilized,and thus the pelvis couldnot be imaged simultaneously with the prostate.In a com-parison of pelvic phased-array (PPA) coils and integratedendorectal coils,staging accuracy was better with inte-ated endorectal PPA coils than for PPA coils.Newersoftware and technology have allowed simultaneousmulti-channel image acquisition.This advance reduces sig-nal drop-off throughout the pelvis.In a recent meta-analysis of the current literature byEngelbrecht et al,actors improving staging performanceincluded publication year,sample size,histologic gold standard,number of imaging planes,turbo spin echo,endorectal coil,and contrast agents.This meta-analysisdemonstrates the improving staging sensitivity as the tech-nology has improved.Due to time constraints,the numberof imaging planes is usually limited to two.As sequenceshave decreased in acquisition time,the sagittal plane hasbeen added to many evaluations.However,the addition ofthe sagittal plane has not been shown to increase speci-ficity or sensitivity.Turbo spin-echo is a sequence thatallows more rapid acquisition,thereby decreasing motionartifact and increasing patient comfort by shortening thelength of the examination.Interestingly,this study did notshow improved sensitivity with magnetic resonance spec-troscopic imaging (MRSI),dynamic contrast-enhancedimaging,or field strength,possibly because there is notenough improvement with these techniques or becausenot enough studies have been performed with newertechniques such as multivoxel spectroscopy.This meta-analysis suggested that currently,MRI in prostate cancerstaging (T2c vs T3c) has a combined sensitivity and speci-ficity of 71%.At a specificity of 80% on this curve,sensi-tivity was 62%.A sensitivity of 62% is an Òimprovementer the sensitivity of clinical examination,which,by def-inition,has not helped detect periprostatic invasion inpatients with clinically localized disease.ÓGiven these findings,many urologists have elected toproceed with patient treatment of prostate cancer basedonly on Gleason score,prostate-specific antigen (PSA)level,and digital rectal examination.In 1997,only 10% ofthe urologists in the United States used preoperative pro-static MRI,and 31% of urologists reported that MRI wasnot available to them.In 2000,Jager et aldemonstrat-ed that MR staging was cost-effective for patients at inter-mediate risk of extracapsular disease:PSA 10Ð20 ng/mLand Gleason score of 5Ð7.Their study also suggestedcost-effectiveness for patients at low risk.These benefitsderived from the cost savings created whenpatients with clinically T2 disease were converted to T3disease based on imaging and consequently did notundergo surgical prostatectomy.Many urologists will notperform prostatectomy on patients with high risk ofxtracapsular extension despite the lack of evidence ofxtracapsular extension (Fig 1).The surgery-MRI thresh-old has not been evaluated prospectively;however,ret-ospective analysis has suggested that patients with anintermediate to high risk of having T3 or greater diseasebenefit from MRI.Therefore,utilizing MRI for patientsin this category depends on the urologistÕs practice pa-ameters and patient choice.Occasionally,patients withhigh risk of extracapsular extension request prostatecto-despite advice to the contrary.At our institution,weperform endorectal MRI to ev

aluate these patients (Fig2),and we often find definitive evidence of extracapsulartension.This information is shown to the patient toassist in choosing the best treatment. November/December 2004,Vol.11,No. 6Cancer Control 355Other studies confirm this approach and also confirmthe sensitivity and specificity of MRI for staging.A Frenchstudyof 336 patients undergoing radical prostatectomydemonstrated sensitivity of 50% for occult T3 staging and69% for extensive T3 tumor with a specificity of 95%.Thisstudy also showed that MRI was likely to confirm T3 tumorwith three or more sextants involved at biopsy.Theauthors concluded that endorectal MRI was appropriate inpatients with three or more positive sextant biopsies,apalpable tumor,and/or a PSA level greater than 10 ng/mL.The future of MRI imaging includes the advance infield strength.Three Tesla (3T) magnets promise increasedesolution and improved spectroscopy that is likely toenhance specificity and sensitivity in differentiating T2from T3 tumors.Another advance will be the continuedincrease in the number of channels that can be sampledsimultaneously.However,the effect that advanced multi-hannel detection will have on staging of prostate canceris unclear.MRSI utilizes the magnetic field to obtain spectra based onthe in vivo endogenous chemicals present in tissue.Thistechnique is frequently used in neurologic imaging to differentiate tumor from radiationnecrosis.With prostate tissue,MRSImust suppress the water and lipidcontent in order to detect the prosta-tic metabolites citrate,choline,andcreatine.In normal prostatic tissue,citrate is as plentiful as creatine.Choline,a marker of cell membranes,is elevated in prostate cancer,likelythe result of more rapid cell turnover(Fig 3).Until recently,MRSI requireda relatively large volume of tissuefrom which the spectra was analyzedand also required manual placementof the voxel that was to be sampled.This made sampling of the small peripheral zone of theprostate somewhat impractical.Newer technology nowallows multivoxel sampling of the entire gland and alsoprovides spectral arrays from a volume as little as 0.24 cc.The MRSI data can be obtained within the same examina-tion as the endorectal MRI.MRSI evaluation improves estimation of tumor volume (which correlates with extra-capsular extension) and enhances the ability to identifythe location of the tumor within the prostate gland,it may show a relationship between outcome,response totherapy,and the spectral evaluation of the tumor.igneron et aldemonstrated in an abstract that Òa linearcorrelation between the magnitude of the decrease of cit-ate and the elevation of choline with the pathologic Glea-son score.ÓDespite these recent advances in MRSI,esearch has yet to prove its value for patient outcomes ina multicenter randomized trial.ymphotropic Contrast AgentsIn June 2003,Harisinghani et aldescribed utilization ofymphotropic superparamagnetic nanoparticles to identi-fy prostate metastasis to lymph nodes.In this landmarkarticle,the authors reported improved sensitivity of MRIor detection of nodal metastasis from 35.4% to 90.5%.Their study correctly identified all patients with nodal Fig 2. Ñ Coronal T2-weighted STIR (short tau inversion recovery) imagingabsent citrate, indicating carcinoma. (B) is normal with a high citrate peak. PPM 3.02.52.0PPM 3.02.52.0 AB November/December 2004,Vol.11,No. 6356Cancer Controlmetastases.Conventional evaluation of lymph nodes sug-ests malignancy if the short-axis diameter is greater than10 mL or if the short axis diameter is 8 mL and the lymphnode is spherical.However,with the monocrystalline ironxide (Combidex,Advanced Magnetics Inc,Cambridge,Mass),three criteria were used:a decrease in signal inten-sity of less than 30% on T2-weighted fast spin-echo,a het-erogeneous signal giving the node a mottled appearanceor discrete focal defects,and nodes with a central area ofyperintensity with peripheral decrease in signal.findings significantly improve the ability of imaging todetect lymph node metastasis.Future Directions in Magnetic Dynamic contrast-enhan

ced MRI evaluation utilizes aapid T1-weighted MR sequence to sample a single planeof tissue as the tissue enhances from intravenous injec-tion of gadolinium MR contrast (Fig 4).Using differencesin enhancement characteristics,this technique attemptsto differentiate prostate carcinoma from normal periph-eral zone and central gland tissue.The main featuresused for differentiation are the start of enhancement,time to peak,peak enhancement,and washout.The peak-enhancement was the optimal enhancement feature fordifferentiation of carcinoma in the peripheral zone andcentral gland.Diffusion MRI utilizes the ability of MR to detect theBrownian motion of water molecules.The technique hasbecome especially useful in the detection of acuteinfarcts in neurological imaging.Several attempts havebeen made to utilize the technique to evaluate theesponse to radiation therapy in other body sites.A lit-erature search revealed just one journal article describ-ing early evaluation of this technique in prostate cancer.ennings et aldemonstrated that prostate carcinomaxenografts exposed to docetaxel chemotherapy devel-oped an increased apparent diffusion coefficient (ADC).This early study suggests that MR diffusion imaging maybe able to demonstrate the degree of response to radia-tion and chemotherapy.Positron Emission TomographyA recent review by Shvarts et aldiscusses the difficultiesof using positron emission tomography (PET) to stageprostate cancer.The authors note that the use of PET withs note that the use of PET with18F]-fluoro-2-deoxy-the diagnosis of localized prostate cancer has been disap-pointing.Prostate cancer has a low glycolytic rate,whichesults in decreased accumulation of fluorodeoxyglucose.Other research radiopharmaceuticals exist that showpromise for evaluation of locally extensive prostate can-cer.C-choline has been used successfully in identifyingThis radiotracer has ashort half-life,which limits its clinical usefulness.FDG-PEThas been used successfully for the evaluation of metastat-ic disease,tumor burden,and location of disease.Oyamaet aldemonstrated a decrease in PET uptake at metastat-ic sites in patients undergoing androgen ablation.At ourinstitution,PET is not routinely used to stage prostate can-cer or to follow metastatic prostate cancer.A new study utilizing Doppler sonography and powerDoppler techniques has demonstrated that hypervascular-ity correlates with increased Gleason score.This studyprospectively evaluated ninety-four patients with colorDoppler sonography before radical prostatectomy.Thuscolor Doppler sonography evaluation of hypoechoiclesions at time of biopsy is useful in predicting increasedGleason score and the aggressiveness of the tumor.At our institute,we use ultrasound to guide prostate biop-sies (Table).Staging of the pelvis has traditionally beencompleted with CT,but MRI is better in evaluating forymph nodes and the use of MRI is becoming more wide-spread.MRI with the endorectal coil is being used to eval-uate for residual prostate tumor in patients who haveundergone prostatectomy.Bone scan is used to evaluate Scaling: Factor = 1.0 Offset = 0.0erage of mean values: 1: 183.72 2: 164.52erage of areas: 1: 0.26 2: 0.39erage of max values: 1: 211.84 2: 182.98 199.1 1.775 216304458728 Image No.Prostate CancerNormal PeripherySPH 10.5 November/December 2004,Vol.11,No. 6Cancer Control 357Despite the rapid advancement of imaging technolo-large studies demonstrating their usefulness have yet tobe completed.Currently,there is a large multi-institution-al trial underway to evaluate the usefulness of MRSI butthe results are likely several years away.As Langlotz et alnoted in 1995,ÒOne may wonder about the relevance ofesults of large multi-institutional studies charged withassessing technology such as prostatic MRI imaging ÉBecause such projects cannot be designed,implemented,and evaluated in a short time,the results of a scientificinterrogation É are likely to be outdated and of limitedalue by the time they are published.ÓGiven these com-plex issues the clinician is left with utilizing the technolo-av

ailable at their own institution and determining if thetechnology benefits their patients.Current literature sug-ests that patients with moderate to high risk of extracap-sular extension of disease benefit most from endorectalMRI.Spectroscopy remains unproven.Despite these lim-itations,ongoing advances in imaging modalities will helpto allow imaging to equal the gold standard of pathologicevaluation in the staging of prostate cancer.1.Engelbrecht MR,Barentsz JO,Jager GJ,et al.Prostate cancer stag-ing using imaging.2.Borley N,Fabrin K,Sriprasad S,et al.Laparoscopic pelvic lymphnode dissection allows significantly more accurate staging in Òhigh-iskÓprostate cancer compared to MRI or CT.Scand J UrolNephrol3.Purohit RS,Shinohara K,Meng MV,et al.Imaging clinically local-ized prostate cancer.Urol Clin North Am4.Wilkinson S,Chodak G.The role of 111-indium-capromab pende-tide imaging for assessing biochemical failure after radical prosta-tectomy.J Urol5.DÕAmico AV,Whittington R,Malkowicz SB,et al.Critical analysis ofthe ability of the endorectal coil magnetic resonance imaging scanto predict pathologic stage,margin status,and postoperativeprostate-specific antigen failure in patients with clinically organ-confined prostate cancer.6.Engelbrecht MR,Jager GJ,Laheij RJ,et al.Local staging of prostatecancer using magnetic resonance imaging:a meta-analysis.2002;12:2294-2302.Epub 2002 Apr 19.7.Hricak H,White S,Vigneron D,et al.Carcinoma of the prostategland:MR imaging with pelvic phased-array coils versus integrat-ed endorectal pelvic phased-array coils.Radiology8.Langlotz C,Schnall M,Pollack H,et al.Staging of prostate cancer:accuracy of MR imaging.Radiology9.Plawker MW,Fleisher JM,Vapnek EM,et al.Current trends inprostate cancer diagnosis and staging among United States urolo-J Urol10.Jager GJ,Severens JL,Thornbury JR,et al.Prostate cancer staging:should MR imaging be used? A decision analytic approach.ology11.Engelbrecht MR,Jager GJ,Severens JL,et al.Patient selection formagnetic resonance imaging of prostate cancer.Eur Urol.20012.Cornud F,Flam T,Chauveine L,et al.Extraprostatic spread of clini-cally localized prostate cancer:factors predictive of pT3 tumorand of positive endorectal MR imaging examination results.ology13.Kurhanewicz J,Swanson MG,Nelson SJ,et al.Combined magneticesonance imaging and spectroscopic imaging approach to molec-ular imaging of prostate cancer.J Magn Reson Imaging14.Scheidler J,Hricak H,Vigneron DB,et al.Prostate cancer:localiza-tion with three-dimensional proton MR spectroscopic imaging.Clinicopathologic study.Radiology15.Vigneron DB,Males R,Hricak H,et al.Prostate cancer:correlationof 3D MRSI metabolite levels with histologic grade.Proceedings ofthe Radiological Society of North America.Chicago,Ill:1998;181.16.Harisinghani MG,Barentsz J,Hahn PF,et al.Noninvasive detectionof clinically occult lymph-node metastases in prostate cancer.Engl J Med2003:348:2491-2499.Erratum in:17.Engelbrecht MR,Huisman HJ,Laheij RJ,et al.Discrimination ofprostate cancer from normal peripheral zone and central gland tis-sue by using dynamic contrast-enhanced MR imaging.Radiology2003;229:248-254.Epub 2003 Aug 27.18.Jennings D,Hatton BN,Guo J,et al.Early response of prostate car-cinoma xenografts to docetaxel chemotherapy monitored with dif-19.Shvarts O,Han K,Seltzer M,et al.Positron emission tomography inurologic oncology.Cancer Control20.Hara T,Kosaka N,Kishi H,et al.PET imaging of prostate cancerusing carbon-11-choline.J Nucl Med21.Oyama N,Akino H,Suzuki Y,et al.FDG PET for evaluating thehange of glucose metabolism in prostate cancer after androgenablaNucl Med Commun22.Cornud F,Hamida K,Flam T,et al.Endorectal color Doppler sonog-phy and endorectal MR imaging features of nonpalpable prostatecancer:correlation with radical prostatectomy findings.AJR Am JRoentgenol23.Langlotz C,Schnall M,Pollack H,et al.Staging of prostatic cancer:accuracy of MR imaging.Radiology1995;194:645-646;discussion Biopsy direction:UltrasoundLocal staging:Computed tomographyMetastasis:Bone scanResidual prostate:Magnetic resonance imaging with endorectal coi