Use of [ C]Choline PET-CT as a Noninvasive Method …clincancerres.aacrjournals.org/content/clincanres/17/24/7673.full.pdf · Imaging, Diagnosis, Prognosis Use of [11C]Choline PET-CT

Imaging, Diagnosis, Prognosis

Use of [11C]Choline PET-CT as a Noninvasive Method forDetecting Pelvic Lymph Node Status from Prostate Cancerand Relationship with Choline Kinase Expression

Kaiyumars Contractor1, Amarnath Challapalli1, Tara Barwick2, Mathias Winkler1, Giles Hellawell1,Steve Hazell3, Giampaolo Tomasi1, Adil Al-Nahhas2, Paola Mapelli1, Laura M. Kenny1, Paul Tadrous4,R. Charles Coombes1, Eric O. Aboagye1, and Stephen Mangar1

AbstractPurpose: To evaluate the accuracy and biological basis for [11C]choline-PET-CT in the nodal staging of

high risk localized prostate cancer patients.

Experimental Design: Twenty-eight patients underwent dynamic [11C]choline-PET-CT of the pelvis and

lower abdomen prior to extended laparoscopic pelvic lymph node dissection (eLPL). The sensitivity and

specificity of [11C]choline PET, [11C]choline PET-CT, andMRI for nodal detection were calculated. Average

and maximal standardized uptake values (SUVave, SUVmax) were compared with choline kinase alpha

(CHKa) and Ki67 immunohistochemistry scores.

Results: Four hundred and six lymph nodes (LN), in 26 patients, were assessable. Twenty-seven (6.7%)

involved pelvic nodes at eLPL were detected in 9 patients. Seventeen of the 27 involved nodes were

subcentimeter. The sensitivity and specificity on apernodal basiswere 18.5%and98.7%, 40.7%and98.4%,

and 51.9% and 98.4% for MRI, [11C]choline PET, and [11C]choline PET-CT, respectively. Sensitivity was

higher for [11C]choline PET-CT compared with MRI (P ¼ 0.007). A higher nodal detection rate, including

subcentimeter nodes, was seen with [11C]choline PET-CT than MRI. Malignant lesions showed CHKaexpression in both cytoplasm and nucleus. SUVave and SUVmax strongly correlated with CHKa staining

intensity (r ¼ 0.68, P < 0.0001 and r ¼ 0.63, P ¼ 0.0004, respectively). In contrast, Ki67 expression was

generally low in all tumors.

Conclusion: This study establishes the relationship between [11C]choline PET-CT uptake with choline

kinase expression in prostate cancer and allows it to be used as a noninvasive means of staging pelvic LNs,

being highly specific and more sensitive than MRI, including the detection of subcentimeter disease. Clin

Cancer Res; 17(24); 7673–83. �2011 AACR.

Introduction

The evaluation of lymph nodes (LN) has important ther-apeutic and prognostic significance in patients diagnosedwith prostate cancer. Although a curative approach can beadopted for thosewithorgan-confinednode-negativedisease

withmodalities such as surgery, external beam radiotherapy,or brachytherapy, those with node-positive disease ultimate-ly relapsewithmetastatic disease (relapse rate 30%–50%at 5years, 90% at 10 years; refs. 1, 2). As such, the presence of LNinvolvement reduces the 5-year disease-free survival from85% to approximately 50%,with a shift in focus of treatmentto long-term androgen deprivation with the addition ofpelvic radiotherapy to reduce loco-regional recurrence (3,4). Pelvic LN dissection is currently the gold standard forevaluating the presence of nodal involvement (5, 6). Thisprocedure can either be open or laparoscopic and is usuallylimited to the external iliac and obturator nodes, though amore extended procedure to include the internal iliac nodesis usually advocated for those with a higher risk of nodaldisease (7). Either way, both these methods are invasive,associated with morbidity (8) and, importantly, may not beable to sample all potential LN areas.

It is thus important to have a sensitive and reliablenoninvasive means of detecting nodal involvement. Thecriteria for nodal characterization using cross-sectionalimaging, such as computerized tomography (CT) or

Authors' Affiliations: Department of 1Surgery and Cancer, 2Radiology/Nuclear medicine, 3Cellular Pathology, Imperial College London and Impe-rial College Healthcare NHS Trust, Hammersmith and Charing CrossHospitals; and 4Cellular Pathology, Northwest London Hospitals NHSTrust, London, United Kingdom

Note: Supplementary data for this article are available at Clinical CancerResearch Online (http://clincancerres.aacrjournals.org/).

K. Contractor and A. Challapalli have contributed equally to this work andare joint first authors.

Corresponding Author: Eric O. Aboagye, Department of Surgery andCancer, Imperial College London, Hammersmith Hospital, Du Cane Road,LondonW12 0NN, UnitedKingdom. Phone: 44-20-8383-3759; Fax: 44-20-8383-1783; E-mail: [email protected]

doi: 10.1158/1078-0432.CCR-11-2048

�2011 American Association for Cancer Research.

ClinicalCancer

Research

www.aacrjournals.org 7673

on July 28, 2018. © 2011 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst October 28, 2011; DOI: 10.1158/1078-0432.CCR-11-2048

http://clincancerres.aacrjournals.org/

Magnetic Resonance Imaging (MRI), relies primarily onmorphologic assessment on the basis of size and shape,with a nodal short axis diameter of 1 cm generallyaccepted as an upper limit of normal. A threshold of 1cm in the short axis diameter for oval nodes and 0.8 cmfor round nodes has been recommended as criteria fordiagnosis of prostate cancer nodal metastases (9). Arecent meta-analysis on the diagnostic accuracy ofcross-sectional imaging in the staging of pelvic LNs inprostate cancer reported a high pooled specificity for MRIof 0.82 with a low and heterogenous pooled sensitivity of0.39 (10). The lack of sensitivity belies the fact that nodalinvolvement is not always correlated with enlargementand enlarged nodes may also be due to a benign etiology.Neither MRI nor lymphangiography has shown highersensitivity than CT scanning in the detection of nodalmetastases (10, 11). The use of an MR contrast agentcontaining ultrasmall particles of iron oxide (ferumox-tran10-Sinerem, USPIO) has been shown to yield sensi-tivity and specificity above 90% in the detection ofprostate cancer LN metastases (12). However, this is notwidely available and its intravenous infusion is notwithout side effects (13). While further studies usingdiffusion weighted MR undoubtedly have improvedintraprostatic tumor detection and localization, thismethod has been less satisfactory for assessing pelvicnodal disease (14).

Positron emission tomography (PET) offers functionalinformation about tissue activity, thereby having the poten-tial to provide superior staging information as well as theability to monitor the response to treatment. The clinicalexperience with [18F]fluorodeoxyglucose (FDG) PET inprostate cancer is limited due to variable uptake of [18F]FDGin prostate cancer and the rapid excretion of FDG in urine,causing an accumulation of activity in the bladder (15–17).

[11C]choline is a relatively new radiopharmaceutical forPET imaging, and its utility in visualizing and stagingprostate cancer has been published (18, 19). Malignanttransformation is postulated to be associated with changesin pathways of choline transport, utilization, and increased

choline kinase alpha (CHKa) expression that will lead to anincreased uptake of choline (20, 21). As illustrated by anumber of MR Spectroscopy studies (21–23), CHKa con-verts choline to phosphocholine in cells that is elevatedduring transformation and progression. The tumor PETsignal from [11C]choline, however, comprises of free[11C]choline and [11C]phosphocholine, as well as the oxi-dation product, [11C]betaine (24). The PET signal (tumor[11C]choline uptake), therefore, largely reflects transportand phosphorylation of [11C]choline and, to a lesser extent(given that liver and kidneys producemost of the circulating[11C]betaine), [11C]choline oxidation. Unlike [18F]FDG, ithas low renal elimination and therefore, visualization of theprostate and surrounding nodes may be enhanced by thelow accumulation of tracer within the bladder (16). Pre-liminary studies of [11C]choline-PET in pelvic nodal stagingin prostate cancer patients have shown early promise (25–27). However, no study to date has established a directrelationship between CHKa expression and [11C]cholineuptake in prostate tumors.

This prospective study compares the use of [11C]cholinePET-CT with MRI in determining pelvic nodal status inpatients with high risk localized prostate cancer undergoingsurgical staging with extended laparoscopic pelvic lymphnode dissection (eLPL; reference standard). We also soughtto document the early kinetics of [11C]choline fromdynam-ic imaging up to 60 minutes postradiotracer injection. Inaddition, the association between [11C]choline uptake[standardized uptake values (SUV)] and immunohis-tochemistry scores for CHKa and Ki67 expression in pros-tate tumors and involved nodes were compared.

Materials and Methods

PatientsPatients with histologically confirmed prostate cancer

staged as either high risk localized [either prostate-specificantigen (PSA) >20 ng/mL or Gleason score 8 to 10 or TNMstage� T2]/locally advanced (nodal disease on stagingMRIof the pelvis) were eligible for the study. Patients withvisceral or bonemetastases were ineligible. Ethical approvalfor the study was granted by the Hospital Research EthicsCommittee. All patients gave written informed consent toparticipate in the study, which was carried out according tothe Declaration of Helsinki guidelines. The administrationof radioactivity for the PET scans was approved by theAdministration of Radioactive Substances Advisory Com-mittee, United Kingdom.

Imaging protocol[11C]choline was synthesized at Hammersmith Imanet

according to the method described by Pascali and collea-gues (28). To minimize post-biopsy effects, all imagingstudies were done at least 6 weeks after the transrectalbiopsy. Subjects were asked to fast for 6 hours prior to theprocedure (as bowel choline uptake interferes with inter-pretationof [11C]choline images). All patientswere scannedon a PET-CT (GE-Discovery RX) scanner after being

Translational Relevance

This study establishes the feasibility of using[11C]choline-PET CT as a noninvasive means of stagingpelvic lymph nodes in high-risk prostate cancer, beinghighly specific and more sensitive than PET alone orMRI, including the detection of subcentimeter disease.The high specificity could potentially be helpful clini-cally in terms of selecting out patients who may notrequire pelvic radiotherapy.We also showed, for the firsttime in prostate cancer biopsies, that tumor radiolabeledcholine uptake is related to choline kinase alpha expres-sion in prostate cancer. This relationship could beexploited to develop new drugs for prostate cancer.

Contractor et al.

Clin Cancer Res; 17(24) December 15, 2011 Clinical Cancer Research7674




positioned such that the field of view (FOV) included thewhole pelvis and the lower abdomen. This was followed bya diagnostic quality CT scan (settings were 300 mA, 120kVp, 0.8 sec/rotation i.e., 65 mA.s, 8 � 2.5-mm slices andpitch 1.35), which was used for attenuation correction andcoregistration with the PET images. [11C]choline wasadministered by a bolus intravenous injection over 10 to30 seconds. PET scanning (3-dimensional acquisition) wascommenced over 2 bed positions (3 minutes per bedposition) starting from the distal margin of the pelvic floor,covering the pelvis, and lower abdomen (axial FOV per bedposition, 15.7 cm; transaxial, 70 cm) for 65 minutes. RawPET data were corrected for scatter and attenuation andreconstructed with an iterative OSEM (ordered subsetexpectation maximum) algorithm comprising 8 iterationsand 21 subsets. Decay corrected images were then viewedusing Analyze software (Analyze Version 7; BiomedicalImaging Resource). From summed images, regions of inter-est (ROI) were drawn manually around visible tumors inthe prostate, and any visible pelvic nodes. The [11C]cholineradioactivity concentration within the ROIs was then deter-mined and normalized for injected radioactivity and bodyweight to obtain SUVave and SUVmax.

MRI acquisitionAll patients underwent standard noncontrast staging

MRI of the pelvis from aortic bifurcation to pubic sym-physis comprising of T1-weighted axial images; axial,sagittal, and coronal T2-weighted images and small FOVaxial T2-weighted images through the prostate. Theimaging was done on a 1.5-Tesla Philips scanner in 5

patients and a 1.5-Tesla Siemens-Magnetom scanner in 21patients.

Extended laparoscopic extraperitoneal pelviclymphadenectomy

This was done in a standard predefined protocol bythe Urologists within an average of 22 days (2–49 days) ofthe [11C]choline PET-CT. Nodal status was discussed withthe surgeon before lymphadenectomy using informationfrom both MRI and the [11C]choline PET-CT images. TheeLPL included nodes along the external and internal iliacvessels to the ureter proximally, obturator nerve medially,and the genitofemoral nerve laterally. All nodes removedwere carefully labeled for size and anatomical location.Nodes were fixed, paraffin embedded, stained with hema-toxylin and eosin, and reported as negative or positive formetastasis by a histopathologist with a specialist interest inurologic malignancy. The samples were also subjected toadditional immunohistochemistry with Ki67 and CHKa(vide infra).

Image interpretationThe images of the [11C]choline PET-CT were interpreted

prospectively to outline the ROIs, carry out SUV analysis,and discuss outcomewith surgeons preoperatively. Further-more, all the imaging data (MRI, [11C]choline PET and[11C]choline PET-CT) were pooled and evaluated by a dualaccredited nuclear medicine radiologist, blinded to theresults of the histopathology, on separate occasions to avoidreporting bias. The criteria used for assessing nodal involve-ment are given in Table 1.

Table 1. Criteria for nodal involvement and ROC analysis

Imaging modality Criteria for nodal involvement

MRI Size ratio criteriaa (9, 12)Benign: nodes less than 8 mm short axisMalignant: nodes >10 mm short axis and round nodes >8 mm (ratio of the short to long axis >0.8)

[11C]Choline PET Focal uptake outside the normal physiologic distribution of tracer in locations corresponding to nodalchains

[11C]Choline PET-CT Nodeswith increased tracer uptake above the background, even when <10mm in short-axis diameter

5 point scale for ROC analysis

Scale MRI [11C]Choline PET/PET-CTb

1 Nodes <4 mm or not seen Definitely normal2 Nodes ¼ 4–5.9 mm Probably normal (more likely to be physiologic)3 Nodes ¼ 6–7.9 mm Indeterminate (equally physiologic/pathologic)4 Nodes �8 mm but <10 mm Probably abnormal (more likely to be pathologic)5 Nodes �10 mm Definitely malignant

aShort-axis and long-axis diameters of the identifiable LNs were measured using electronic calipers on the scanner console.bDefinitely normal, probably normal, and indeterminate were considered benign and probably abnormal and definitely abnormal wereconsidered malignant.

[11C]Choline PET-CT for Nodal Evaluation in Prostate Cancer

www.aacrjournals.org Clin Cancer Res; 17(24) December 15, 2011 7675




ImmunohistochemistryCHK immunohistochemistry was carried out using a

primary polyclonal, human anti-CHKa antibody (cata-logue no. HPA024153; Sigma-Aldrich), as per manufac-turer’s instructions. The positive control used was bron-chial tissue as per manufacturer’s specifications. Slideswere then scored independently by 3 pathologists usingthe intensity of cytoplasmic and nuclear staining in pros-tate tumor cells from 1 to 3 (1þ, low intensity; 2þ,moderate intensity; 3þ, high intensity including nuclearstaining). Ki67 staining was done using anti-Ki67 anti-body (NCL-Ki67-MM1; Novocastra Laboratories). Tonsiltissue was used as a positive control. Ki67 score wasdetermined by dividing the total number of Ki67-positivetumor cells with the total number of tumor cells countedin 8 high powered fields (200� magnification) using anOlympus microscope. The final score was expressed as apercentage.

Statistical considerationsThemean, SD, medians, range, and frequencies were used

as descriptive statistics. The sensitivity, specificity, and num-ber of correctly recognized cases with MRI, [11C]choline PETand [11C]choline PET-CT in nodal detection were calculatedfor a per patient and per node analysis. The comparison ofeach imaging method was done using the McNemar testimplemented in its uncorrected exact form, on the basis ofbinomial distribution (29). Receiver operating characteristic(ROC) analysis and the area under the curve (AUC) wasdetermined by recalculating sensitivity and specificity forMRI, PET, and PET-CT along the 5-point grading scale for aper patient and a per nodal analysis usingMedCalc statistical

software (version 11.6.1). SUV60, ave and SUV60, max werecompared with CHKa and Ki67 scores using Spearman’scorrelation test and a P value of �0.05 was consideredsignificant.

Results

PatientsTwenty-eight patients underwent [11C]choline PET-CT

after fulfilling the inclusion criteria. Two patients could notundergo surgery after [11C]choline PET-CT as one becameunwell and the other changed his mind about undergoingsurgery. Thus 26 patients underwent [11C]choline PET-CTfollowed by eLPL/sampling (1 had LN sampling rather thandissection due to fibrotic and calcified LNs). All patientssubsequently had neoadjuvant androgen deprivation fol-lowed by radical radiotherapy to the prostate and the pelvis.The median (mean; range) age of subjects was 67 years(67.7; 51 to 83 years), Gleason score of primary prostatebiopsies was 7 (7.6; 6–9) and the pretreatment PSA levelswere 26.25 (44.25; 8.1–209).

The interval between the [11C]choline PET-CT and eLPLwas an average of 22 days (2–49 days). From the 26patients, a total of 406 pelvic LNs sampled were availablefor pathology, with a median of 16 (range: 3–36) nodesharvested per patient. Twenty-seven (6.7%) involved pel-vic nodes at eLPL were detected in 9 patients (Table 2). Ofthe involved nodes 17 of the 27 LN were less than 10 mmin size. The average nodal size of the histologically pos-itive nodes was 9.8 mm, with an average tumor focus of5.7 mm.

The [11C]choline PET-CT was well tolerated with noimmediate or delayed complications observed.

Table 2. Characteristics of patients with histologically positive nodes (9/26)

Pt no. Age (y) GS iPSA cT pN No ofþ LN

Site of þ LN MRI Size(mm)

PET PET-CT Size(mm)

1 73 7 8.54 T3a N1 1 1-R Obt R Obt 11 TP 1-R Obt TP R Obt 11 TP10 82 8 13.5 T3 N1 5 3-R Obt, 1-R II,

1-R GF2-R EI 21,8 TP 3-R EI TP 3-R EI 20,11,7 TP

13 56 7 50 T1c N1 1 1-L Obt FN FN FN15 65 9 209 T2b N1 7 4-L EI, 3-R II R II 19 TP 1-L EI,

3-R IITP 1-L EI,

3-R II9,18,11,5 TP

17 76 7 169 T4 N1 1 1-R Obt L Obt 9 FP R Obt TP R Obt 9 TP20 76 7 21 T2b N1 1 1-R Obt FN R Obt TP R Obt 4 TP24 61 9 45 T3 N1 8 1-R II, 2-R Obt,

3-L EI, 2-L ObtFN 1-R Obt TP 1-R Obt,

2-R II6,6,4 TP

25 76 9 24.5 T2b N1 2 2-R EI FN FN FN27 51 7 44.8 T3b N1 1 1-L II L II 10 TP FN L II 10 TPMean 68.4 7.8 65 13.8 9.4Median 73 7 44.8 11 9

GS-Gleasonscore; iPSA, initial prostate-specificantigen; cT, clinical tumor stage; pN, pathologic nodal stage; LN, lymphnode;R, right;L, left; Obt, obturator; II, internal iliac; GF, genitofemoral; þ, positive.

Contractor et al.





Time points for SUV measurementThe average andmaximumSUVat 60minutes (SUV60, ave,

and SUV60, max) were determined. Due to the rapid systemicmetabolism of [11C]choline (30), SUV has also beendetermined at an earlier time point. The time versus radio-activity curves (TAC) achieve a steady state after approxi-mately 15minutes (Supplementary Figs. S1 and S2).Hence,SUV15, ave, and SUV15, max were reported (SupplementaryFig. S3).

[11C]choline uptake within themalignant prostate andpelvic nodesIn addition to visualization of nodal uptake, primary

prostate tumors in all 26 patients were well visualized withgood tumor-to-background ratios (Fig. 1 and Supplemen-tary Fig. S4). Themedian (mean� SD; range) SUV60, ave andSUV60, max were 4.85 (4.92 � 1.75; 2.19–9.28) and 9.97(11.05�3.72; 4.73–20.54), respectively (median SUV15, ave

and SUV15, max were 4.82 and 8.80, respectively). Dynamic

TACs for [11C]choline in primary prostate tumors and thenodal metastases showed good retention of radio activityafter plateauing at approximately 15 minutes until 60minutes with SUVave (Supplementary Fig. S1 and S2).However, with SUVmax there was a suggestion of increasingactivity at 60minutes, whichmaybedue to the contributionof [11C]betaine.

Diagnostic performance of MRI, [11C]choline PET, and[11C]choline PET-CT in detection of nodal disease

On a per patient basis, the sensitivity and specificity were50% and 72.2%; 66.7% and 76.4%, and 77.8% and 82.4%,respectively, for MRI, [11C]choline PET, and [11C]cholinePET-CT. On a per nodal basis, the sensitivity and specificitywere 18.5% and 98.7%; 40.7% and 98.4%; and 51.9% and98.4%, respectively, for MRI, [11C]choline PET, and[11C]choline PET-CT (Supplementary Table S1). No statis-tical difference between any 2modalitieswas detected in thepatient analysis, mainly owing to the relatively low number

Aiviiiiii

Biviiiiii

C

ii

iviiii ii

B

B

B

Figure 1. A, T1 weighted MRI (i), axial [11C]choline PET (ii), CT (iii), and PET-CT fused (iv) shows focal uptake in a 4-mm right obturator node (arrowed)clearly separate to the ureter on the PET only (ii); in retrospect visible on MRI (i) but not called as well below size criteria. B, coronal T2 weighted MRI (i),[11C]choline PET maximum intensity projection (MIP) (ii), CT (iii), and PET-CT fused (iv) shows focal uptake in cluster of right external iliac nodes (arrowed).Note uptake in prostate extending to seminal vesicle (green arrow on coronal MIP). C, T1 weightedMRI (i) axial [11C]choline PET (ii), CT (iii), and PET-CT fused(iv) shows focal uptake in 10-mm left obturator node (arrowed) which was false positive. B; Bladder activity.






of subjects. In the per nodal analysis, the sensitivity wassignificantly improved with the use of [11C]choline PET-CT(P ¼ 0.007) and [11C]choline PET (P ¼ 0.07) comparedwith MRI imaging, without a decrease in the specificity(P ¼ 1, 1, and 0.48 for [11C]choline PET vs. MRI,[11C]choline PET-CT vs. MRI and [11C]choline PET-CT vs.[11C]choline PET comparisons, respectively).

ROC analysis (Fig. 2) showed the overall diagnosticperformance improved in the following order: MRI <[11C]choline PET < [11C]choline PET-CT.

Table 3 shows the detection rate of MRI, [11C]cholinePET, and [11C]choline PET-CT for nodal metastases accord-ing to the diameter of the involved LNs. A higher LNdetection rate, including the detection of subcentimeter

nodes, was seen with [11C]choline PET-CT than MRI. Themean diameter of the positive LNs on histopathology was9.8 mm and that of the true positive (TP) LNs was 13.8 and9.4 mm, respectively, on MRI and [11C]choline PET-CT(using CT component for size definition).

Sites of nodal involvementThe majority of the nodes were detected within the

standard surgical template. However, 4 of 26 patients(15.4%) had focal increased uptake above the region ofeLPL (common iliac (CI) region and lower para aorticregion–median SUV60 ave and SUV60 max of 1.12 and3.91; median SUV15, ave and SUV15, max of 2.61 and 6.51,respectively) as detected on imaging and therefore were not

Per

pat

ien

t b

asis

Per

no

dal

bas

is

AUC-0.820AUC-0.625

AUC-0.677 AUC-0.745 AUC-0.766

AUC-0.830

Figure 2. ROC curve analysis showed the AUC to be 0.625, 0.820, and 0.830, respectively, for MRI, [11C]choline PET and [11C]choline PET-CT on a perpatient basis and 0.677, 0.745, and 0.766, respectively, on a per nodal analysis. The overall diagnostic performance improved in the following orderMRI < [11C]choline PET < [11C]choline PET-CT.

Table 3. Detection rate of the three imaging modalities by the size of the node

Size of infiltrated nodes (mm) No. of LN MRI þ (%) [11C]choline PET þ (%) [11C]choline PET-CTþ (%)

0.1–1.9 1 0 (0) 0 (0) 0 (0)2–4.9 4 0 (0) 0 (0) 1 (25)5–9.9 12 0 (0) 4 (33) 4 (33)�10 10 5 (50) 7 (70) 9 (90)

þ, positive.

Contractor et al.





sampled. Eight of 26 (31%) patients had nodes detectedbelow the surgical template, ofwhich3patients haddiscreteunilateral uptake in the inguinal LNs (median SUV60 ave

and SUV60 max of 1.21 and 2.50; median SUV15, ave andSUV15, max of 1.54 and 2.63, respectively); significantlylower as compared with TP pelvic nodes (P values of0.002, 0.0002, 0.004, and 0.0002, respectively, for SUV15

ave, SUV15 max, SUV60 ave and SUV60 max in the 2-sided t test)which was interpreted as probably reactive uptake andtherefore were considered nonmetastatic (SupplementaryFig. S5). One patient had a 5-mm tumor focus in a geni-tofemoral node, which was outside the FOV.

Nodal analysis on MRIIn 4 of 9 patients, MRI was positive for 5 malignant true

positive (TP) nodes with a median maximum diameter of11 mm (range: 9–21 mm; mean: 13.8 mm).In 22 malignant nodes using size criteria, MRI was false

negative (FN). Eighteen of 22 (82%) nodes were subcenti-meter and were reported as normal. Four nodes more than10 mm were missed. This was due to a cluster of 3 nodesreported as one (Fig. 1), lateral extension of tumor obscur-ing the obturator node and 2 nodes measuring 12 and 15mm on histology, which measured 4 and 5 mm on MRI,highlighting the pitfall of gross nodal measurements whichmay include surrounding perinodal fat and soft tissue.In 4 patients, MRI was false positive (FP; a total of 5

nodes). This was due to a probable sampling error in 2patients (Fig. 1), reactive external iliac (EI) nodes in 1patient and a positive round reactive obturator node whichwas negative on PET-CT.

Nodal analysis on [11C]choline PETIn 6 of the 9 patients, [11C]choline PET alone was TP for

11 of 27 malignant LNs.In the 16 FN malignant nodes, 13 were due to micro-

metastases, 2weremistaken for focal ureteric activity, whichwas resolved with PET-CT, and 1 node was in the saturationband (i.e., where there was an overlap when the 2 bedpositions were fused). This saturation band, obscuringsome parts of the imaged area, is not a general feature ofPET-CT but probably related to specific equipment settingsor reconstruction.In 4 of 17 patients [11C]choline PET was FP (total of 6

nodes). There are varying reasons for this: one FP node wasdue to focal uptake in a calcified vessel mistaken for a nodewhich was resolved with PET-CT; 2 nodes were reactive EInodes; one node was situated in the saturation band and forthe remaining nodes in 2 patients, there was a probablesampling error given that 15 and 28 nodes were removed intotal fromthosepatients, respectively. ThemedianSUV15, ave:SUV60, ave, andSUV15,max: SUV60,maxof theFPLNswere2.54:2.70, and 5.07: 6.51, respectively. SUVs of the TP LNs tendedto be higher (median SUV15, ave : SUV60, ave – 2.99: 2.64and SUV15, max: SUV60, max – 7.04: 7.77, respectively) thanSUVsof theFPLNs,althoughstatistical significancewasneverreached (P values of 0.48, 0.28, 0.56, and 0.22 for SUV15 ave,SUV15 max, SUV60 ave and SUV60 max in the 2-sided t test).

Nodal analysis on [11C]choline PET-CTIn 7 of 9 patients, [11C]choline PET-CT was TP for 14

malignant LNs (Fig. 1). The median maximum diameter ofthe malignant LNs detected was 9 mm (range: 4–20 mm;mean: 9.4 mm).

In 13 malignant nodes, [11C]choline PET-CT was FN asexplained in the preceding paragraph. In 3 patients,[11C]choline PET-CT was FP in 6 nodes. In one patient, afurther FP node close to the saturation band was called onPET-CT but not PET only. The other 5 nodes in 2 patientswere FP on both PET only and combined PET-CT asexplained above.

Ki67 and CHKa expression in prostate tumors andnodal metastases

Biopsy samples from 20 prostate cores and 7 metastaticnodes were available for immunostaining (SupplementaryTable S2). There was cytoplasmic CHKa staining in allprostate tumor cells that varied in intensity from 1 to 3(Fig. 3A) compared with a positive control (SupplementaryFig. S6). In one section, some benign glandular areas werealso weakly stained (Fig. 3B). In one section, an increasednuclear staining for CHKa with increasing Gleason scoreswas also observed, especially between Gleason 3 and 5 (Fig.3C) visually differentiating the 2 grades. There was norelationship between cytoplasmic intensity and nuclearstaining of CHKa. In fact, in one tumor, an area of prostaticintraepithelial neoplasia (PIN) showed cytoplasmic as wellas nuclear staining (Fig. 3D). In pelvic nodes, benign nodesshowed no CHKa staining (Fig. 3E), whereas malignantnodes showed moderate cytoplasmic staining (Fig. 3F).Ki67 staining revealed (Fig. 3G and H) that most primaryand nodal prostate tumors had a low proliferation index(median 3%, range: 1%–17%).

Spearman’s correlation test was used to test the associa-tion between [11C]choline SUV and tumor immunohis-tochemistry, PSA andGleason scores (Supplementary TableS3). There was a positive correlation between SUV60, ave

or SUV60 max (Supplementary Fig. S7), with cytoplasmicCHKa intensity in prostate tumors (r¼ 0.68,P<0.0001 andr¼ 0.63, P¼ 0.0004, respectively). This positive correlationwas seen even at early time points (SUV15 ave, SUV15 max:r ¼ 0.55, P ¼ 0.003, and r ¼ 0.46, P ¼ 0.02, respectively).[11C]choline SUV also weakly correlated with serum PSAlevels at diagnosis. There was no correlation of [11C]cholineSUV with either Ki67 or Gleason scores. The associationbetween immunohistochemistry scores for CHKa and Ki67with Gleason’s scores or PSA was assessed. Only Gleasonscores and Ki67 indices showed a positive correlation(r ¼ 0.55, P ¼ 0.01).

Discussion

This study supports the feasibility of using [11C]cholinePET-CT in determining pelvic LN status in patients withhigh-risk prostate cancer. Thismethod is specific and showsearly promise in yielding a greater diagnostic accuracy thaneitherMRIorPETonly scanning. This is especially evident in






the detection of subcentimeter disease, although the sensi-tivity is not sufficient to exclude lymphadenectomy, asmetastases less than 6 mm in particular may be missed.However, it has the potential to highlight nodal uptake

outside the surgical template for LN dissection, especiallyin the CI and para-aortic area as shown in this study, whichcan have significant consequences in terms of patientmanagement.

A B

C D

E F

G H

Figure 3. CHK immunostainingshowing (A) cytoplasmic staining(B)CHKa expression in benign (B) andmalignant (M) acini (C) differentialCHKa in Gleason stage 3 and 5, (D)CHKa in PIN, (E) no staining in abenign node, (F) malignant focus innode, (G) low Ki67 staining inprostate tumor and, (H) low Ki67staining in metastatic node.Magnifications of 200�.

Contractor et al.





The somewhat disappointing performance of [18F]FDGPET in the setting of prostate cancer has prompted interestin newer PET tracers, such as [18F] and [11C]choline, for thedetection of primary tumor within the prostate andthe staging of pelvic nodal disease. For the detection of theprimary tumor, some authors have reported 100% sensi-tivity (19, 31, 32), whereas others report lower detectionrates ranging from 19% to 58%, depending on whetherresults were reported on a per patient or per lesion basis(33–36). Supplementary Table S4 summarizes the pub-lished studies assessing LN stage and shows varied andconflicting results (25–27, 37–39). Likewise for staging ofpelvic nodal disease, the reported sensitivity and specificityranged from 50% to 80% and 90% to 96%, respectively, instudies that employed PET alone based on a per patientanalysis (25, 26). The variation in sensitivity may be in partdue to patient selection.In this study, we have assessed the value of MRI,

[11C]choline PET, and [11C]choline PET-CT imaging in thepreoperative staging of high-risk prostate cancer patients.We have shown an overall sensitivity and specificity of[11C]choline PET-CT on a per patient basis, of 77.7% and82.4%, respectively, in the detection of nodal metastases.These results were superior than both MRI (50% and72.2%) and [11C]choline PET (66.6% and 76.4%),although were not significantly different probably due torelatively low patient numbers. For MRI, the sensitivity andspecificity achieved in our study are in keeping with previ-ously reported data (12, 40). Dynamic contrast MRI mayhelp with tumor localization within the prostate, but thereare no specific reports on the additional benefit in nodalstaging. More importantly, on a per nodal basis (27/406),the sensitivity was significantly higher for [11C]choline PET-CT (51.9%) compared with MRI (18.5%; P¼ 0.007) with agreater confidence for identifying subcentimeter involvedLNs, which occurred in 30% of the cases. However, in thisstudy apart from one 4-mm node, we were unable to detectlow volume metastases of less than 5 mm in diameter,probably reflecting the limited spatial resolution of thecurrent generation of scanners.Inoneof thefirst published series,De Jong and colleagues

obtained promising results with [11C]choline PET in thepreoperative nodal staging of 67 patients, with a sensitivityof 80% in a per patient-based analysis. Metastatic LNsranging from 0.5 to 3 cm in size with a mean SUV of 4.7(2.9–9.1) were reported. FP activity in 2 patients was attrib-uted to inflammatory change and focal bowel activity.However, in their study, about 50% of the node positivepatients had a PSA of >50 ng/mL (range: 3–500), comparedwith our mean PSA value of 44.25 ng/mL (range: 8.1–209),which may have contributed to a selection bias and mayunderrepresent the cohort of high risk localized prostatecancer patients for which radiotherapy to the pelvis wouldbe indicated (25). Conversely, Hacker and colleagues (38)reported a very low sensitivity of 10% in a study of 20patients assessed with F-18 fluorocholine. The mean diam-eter of metastatic LNs in their study was 3.8 mm, which iswell below the resolution of PET.

In more recent studies utilizing [11C]choline PET-CT,Schiavina and colleagues (27) evaluated 57 intermediateor high-risk prostate cancer patients prior to surgical treat-ment. They reported a sensitivity of 60% and a specificity of98% for the detection of nodal metastases. Husarik andcolleagues (39) evaluated 111 patients with prostate cancerin a [18F]choline PET-CT study, 43 of whom had staging forassessment of primary disease. The PET-CT findings werecorrelated to the histopathologic findings of 115 sampledLNs in 25 patients, with sensitivity and specificity on a perpatient basis of 33% and 100%, respectively.

Beheshti and colleagues (37) evaluated 130 patientswith intermediate or high-risk prostate cancer with[18F]fluorocholine (FCH) PET-CT prior to extended pelvicnode dissection with sensitivity and specificity in the detec-tion of malignant nodes of 45% and 96%, respectively.Furthermore, they reported a change in management in15%of cases. The authors also noted discrete FCHuptake ininguinal LNs which was interpreted as probable reactiveuptake and therefore excluded from data analysis. This wassimilarly observed in our study cohort (8 of 26 patients),although the visible inguinal nodes had significantly lowerSUVs than both the metastatic LNs and the malignantprostate. As nodal dissection does not routinely removeinguinal nodes as part of standard practice, it may bedifficult to ascertain whether these were involved. Theunderlying assumption is that inguinal nodes were allwithin physiologic limits of less than10 mm in dimensionbased on the fact that prostate tumors normally do notspread to inguinal nodes (41).

Twostudieshave reportedonaper-nodal analysis.Husarikand colleagues in their study including 25 patients stagedwith [18F]fluorocholine reported a low sensitivity of 20%(1 of 5 involved nodes) and a specificity of 100%. All the FNnodeshad tumor fociof less than5mm.ThemeanSUVmax ofthe detected LNs was 5.04; range: 4.9–5.2). Notably, onlyobturator nodes were removed rather than a more extensivelymphadenectomy, and the authors did not comment onFCH positive nodes outside the obturator region. Schiavinaand colleagues, in the study mentioned earlier, reported asensitivity of 41.4% and a specificity of 99.8%on a per nodalanalysis. The mean diameter (in mm) of the metastaticdeposit of TP nodes was significantly higher than that of FNnodes (9.2 vs. 4.2; P ¼ 0.001). Our per-nodal results ofsensitivity and specificity with [11C]choline were similar at51.9%and98.4%.A limitationofour studywas the technicaldifficulties encountered with the interpretation of findingson the PET scans in the region of the saturation band (wherethere was an overlap when the 2 bed positions were fused),which accounted for someof the FP results on the PET alone.In the 2 patients in whom MRI and PET-CT were FP for a26-mm and 10-mm node, there is the possibility, despitecareful use of surgical templates, that these nodes were notsampled. The median SUV60, max of FP LNs was 6.51,compared with 7.77 for the TP nodes.

This study is one of the first to evaluate the time-depen-dent uptake of [11C]choline in prostate tumors up to 60minutes. Dynamic TACs for [11C]choline in primary






prostate tumors and the nodal metastases showed a goodsustained retention of activity after plateauing at approxi-mately 15 minutes until 60 minutes with SUVave. However,with SUVmax, there is a hint of increasing activity at 60minutes which may be due to the contribution of[11C]betaine.

We showed for the first time in prostate tumor samplesthat tumor radiolabeled choline uptake is closely relatedto CHKa expression in prostate cancer. Both semiquan-titative parameters of choline uptake in tumors correlatedwell with CHKa scores (best with SUV60, ave r ¼ 0.68, P <0.0001, Spearman’s test). It was observed that benignprostatic tissue as well as PIN in the malignant coresshowed cytoplasmic and nuclear staining. This may rep-resent the range of CHKa expression in normal andpremalignant tissues. In certain tissue sections, nuclearstaining was observed, particularly in PIN and in certainhigh Gleason grade tumors and although we do notfully understand this phenomena, a possible reason isthat, as with other cellular proteins such as ERK1/2,phosophorylated CHKa may translocate to the nucleus.This hypothesis needs further evaluation. This study alsoshowed that proliferation in prostate tumors was low, asreflected by the low Ki67 index in most tumors. This wascontrary to the high CHKa expression. For this precisereason, there was no correlation between [11C]cholineSUV and Ki67 scores in tumors. A possible explanationfor this is that for prostate malignancies, CHKa expres-sion is a proliferation-independent marker of the prostatetumor phenotype. This is contrary to the evidence in othercell/tumor types linking CHKa or choline metabolitesand proliferation (42–44). Of note, one study hasreported an association between choline uptake and Ki67scores in prostate tumors (45). Piert and colleaguesshowed that tumor-to-benign prostate background ratiowas significantly high in tumors with a Ki67 score of morethan 5% (P < 0.01). In our study, Ki67 indices were inthe range of 1% to 17%. Seven cores had a Ki67 index ofmore than 5% with a mean SUV60, ave of 4.7 whichis higher than that reported by Piert and colleagues. Ki67did, however, correlate with Gleason score (r ¼ 0.55,P ¼ 0.01, Spearman’s test).

Themain drawback to [11C]choline is the relatively shorthalf-life (20.9minutes) and thus the compound needs to beused close to where it is manufactured. Newer more stableand specific choline compounds are in development (24).

To conclude, this detailed study establishes the feasibilityof [11C]choline PET-CT as a noninvasive means of stagingpelvic LNs in prostate cancer, being highly specific (98.4%)and more sensitive than PET alone or MRI. The highspecificity is potentially helpful clinically in terms of select-ing out those patients with high-risk prostate cancer whomay not need pelvic radiotherapy. Although it cannotcurrently replace MRI as a staging tool, its ability to detectsubcentimeter nodes and a differential SUV value betweeninvolved and physiologic LNs allows for this functionalimaging methodology to assess the radiation response toinvolved nodes. The relationship between CHKa expres-sion and [11C]choline uptake, together with the avid intra-tumoral uptake of choline shown in this study, meritsfurther investigation in a larger patient population and inpatients with other risk profiles.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Acknowledgments

The authors thank Ms Laura Maher for helping with patient recruitment,Dr.DavidPinato andMr.David Peston for advice on immunohistochemistryand Ms. Kasia Kozlowski for editorial assistance. The authors also thank theradiographers, radio-chemists, blood laboratory staff at HammersmithIMANET and finally, all the patients who have taken part in this study.

Grant Support

The work received support from the United Kingdom Medical ResearchCouncil grant (U1200.02.005.00001.01), Cancer Research UK grant (C37/A5610; C2536/A10337), Experimental CancerMedicineCentres grant (C37/A7283), and Biomedical Research Centre (BRC) award to the ImperialCollege Academic Health Sciences Centre (AHSC).

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received August 8, 2011; revised September 29, 2011; accepted October13, 2011; published OnlineFirst October 28, 2011.

References1. Danella JF, deKernion JB, Smith RB, Steckel J. The contemporary

incidence of lymph node metastases in prostate cancer: implicationsfor laparoscopic lymph node dissection. J Urol 1993;149:1488–91.

2. Leibel SA, Fuks Z, Zelefsky MJ, Whitmore WF Jr. The effects of localand regional treatment on the metastatic outcome in prostatic carci-nomawith pelvic lymph node involvement. Int J Radiat Oncol Biol Phys1994;28:7–16.

3. Kumar S, Shelley M, Harrison C, Coles B, Wilt TJ, Mason MD. Neo-adjuvant and adjuvant hormone therapy for localised and locallyadvanced prostate cancer. Cochrane Database Syst Rev 2006:CD006019.

4. Robnett TJ, Whittington R, Malkowicz SB, Brereton HD, Van ArsdalenK, Drach G, et al. Long-term use of combined radiation therapy andhormonal therapy in themanagement of stageD1prostate cancer. Int JRadiat Oncol Biol Phys 2002;53:1146–51.

5. Stone NN, Stock RG. Laparoscopic pelvic lymph node dissection inthe staging of prostate cancer. Mt Sinai J Med 1999;66:26–30.

6. Allaf ME, Partin AW, Carter HB. The importance of pelvic lymph nodedissection in men with clinically localized prostate cancer. Rev Urol2006;8:112–9.

7. Wyler SF, Sulser T, Seifert HH, Ruszat R, Forster TH, Gasser TC, et al.Laparoscopic extended pelvic lymph node dissection for high-riskprostate cancer. Urology 2006;68:883–7.

8. Parkin J, Keeley FX Jr, Timoney AG. Laparoscopic lymph node sam-pling in locally advanced prostate cancer. BJU Int 2002;89:14–7;discussion 7-8.

9. Jager GJ, Barentsz JO, Oosterhof GO, Witjes JA, Ruijs SJ. Pelvicadenopathy in prostatic and urinary bladder carcinoma: MR imagingwith a three-dimensional TI-weighted magnetization-prepared-rapidgradient-echo sequence. AJR Am J Roentgenol 1996;167:1503–7.

Contractor et al.





10. Hovels AM, Heesakkers RA, Adang EM, Jager GJ, Strum S, Hoogev-een YL, et al. The diagnostic accuracy of CT and MRI in the staging ofpelvic lymph nodes in patients with prostate cancer: a meta-analysis.Clin Radiol 2008;63:387–95.

11. Rorvik J, Halvorsen OJ, Albrektsen G, Haukaas S. Lymphangiographycombined with biopsy and computer tomography to detect lymphnode metastases in localized prostate cancer. Scand J Urol Nephrol1998;32:116–9.

12. Harisinghani MG, Barentsz J, Hahn PF, Deserno WM, Tabatabaei S,van de Kaa CH, et al. Noninvasive detection of clinically occultlymph-node metastases in prostate cancer. N Engl J Med 2003;348:2491–9.

13. BerndH, De Kerviler E, Gaillard S, Bonnemain B. Safety and tolerabilityof ultrasmall superparamagnetic iron oxide contrast agent: compre-hensive analysis of a clinical development program. Invest Radiol2009;44:336–42.

14. Roy C, Bierry G, Matau A, Bazille G, Pasquali R. Value of diffusion-weighted imaging to detect small malignant pelvic lymph nodes at 3 T.Eur Radiol 2010;20:1803–11.

15. Effert PJ, Bares R, Handt S, Wolff JM, Bull U, Jakse G. Metabolicimaging of untreated prostate cancer by positron emission tomogra-phy with 18fluorine-labeled deoxyglucose. J Urol 1996;155:994–8.

16. Hain SF. Positron emission tomography in uro-oncology. CancerImaging 2005;5:1–7.

17. Hoh CK, Seltzer MA, Franklin J, deKernion JB, Phelps ME, BelldegrunA. Positron emission tomography in urological oncology. J Urol1998;159:347–56.

18. Hara T, Kosaka N, Kishi H. PET imaging of prostate cancer usingcarbon-11-choline. J Nucl Med 1998;39:990–5.

19. Reske SN, Blumstein NM, Neumaier B, Gottfried HW, Finsterbusch F,Kocot D, et al. Imaging prostate cancer with 11C-choline PET/CT.J Nucl Med 2006;47:1249–54.

20. Ramirez de Molina A, Rodriguez-Gonzalez A, Gutierrez R, Martinez-Pineiro L, Sanchez J, Bonilla F, et al. Overexpression of choline kinaseis a frequent feature in human tumor-derived cell lines and in lung,prostate, and colorectal human cancers. Biochem Biophys Res Com-mun 2002;296:580–3.

21. Ackerstaff E, Pflug BR, Nelson JB, Bhujwalla ZM. Detection ofincreased choline compounds with proton nuclear magnetic reso-nance spectroscopy subsequent to malignant transformation ofhuman prostatic epithelial cells. Cancer Res 2001;61:3599–603.

22. Aboagye EO, Bhujwalla ZM. Malignant transformation alters mem-brane choline phospholipid metabolism of humanmammary epithelialcells. Cancer Res 1999;59:80–4.

23. Glunde K, Jie C, Bhujwalla ZM. Molecular causes of the aberrantcholine phospholipid metabolism in breast cancer. Cancer Res 2004;64:4270–6.

24. Leyton J, Smith G, Zhao Y, Perumal M, Nguyen QD, Robins E, et al.[18F]fluoromethyl-[1,2-2H4]-choline: a novel radiotracer for imagingcholine metabolism in tumors by positron emission tomography.Cancer Res 2009;69:7721–8.

25. de Jong IJ, Pruim J, Elsinga PH, VaalburgW,MensinkHJ. Preoperativestaging of pelvic lymph nodes in prostate cancer by 11C-choline PET.J Nucl Med 2003;44:331–5.

26. Kotzerke J, Prang J, Neumaier B, Volkmer B, Guhlmann A, Kleinsch-midt K, et al. Experience with carbon-11 choline positron emissiontomography in prostate carcinoma. Eur J Nucl Med 2000;27:1415–9.

27. Schiavina R, Scattoni V, Castellucci P, Picchio M, Corti B, Briganti A,et al. 11C-choline positron emission tomography/computerizedtomography for preoperative lymph-node staging in intermediate-riskand high-risk prostate cancer: comparison with clinical staging nomo-grams. Eur Urol 2008;54:392–401.

28. Pascali C, Bogni A, Itawa R, Cambi�e M, Bombardieri E.[11C]Methylation on a C18 Sep-Pak cartridge: a convenient way to

produce [N-methyl-11C]choline. J Labelled Comp Rad 2000;43:195–203.

29. Hawass NE. Comparing the sensitivities and specificities of twodiagnostic procedures performed on the same group of patients. BrJ Radiol 1997;70:360–6.

30. Kenny LM,Contractor KB, Hinz R, Stebbing J, Palmieri C, Jiang J, et al.Reproducibility of [11C]choline-positron emission tomography andeffect of trastuzumab. Clin Cancer Res 2010;16:4236–45.

31. Kwee SA, Coel MN, Lim J, Ko JP. Prostate cancer localization with18fluorine fluorocholine positron emission tomography. J Urol 2005;173:252–5.

32. Yamaguchi T, Lee J, Uemura H, Sasaki T, Takahashi N, Oka T, et al.Prostate cancer: a comparative study of 11C-choline PET and MRimaging combined with proton MR spectroscopy. Eur J Nucl Med MolImaging 2005;32:742–8.

33. Farsad M, Schiavina R, Castellucci P, Nanni C, Corti B, Martorana G,et al. Detection and localizationof prostate cancer: correlationof (11)C-choline PET/CT with histopathologic step-section analysis. J NuclMed 2005;46:1642–9.

34. Giovacchini G, Picchio M, Coradeschi E, Scattoni V, Bettinardi V,Cozzarini C, et al. [(11)C]choline uptake with PET/CT for the initialdiagnosis of prostate cancer: relation to PSA levels, tumor stageand anti-androgenic therapy. Eur J Nucl Med Mol Imaging 2008;35:1065–73.

35. Martorana G, Schiavina R, Corti B, FarsadM, Salizzoni E, Brunocilla E,et al. 11C-choline positron emission tomography/computerizedtomography for tumor localization of primary prostate cancer in com-parison with 12-core biopsy. J Urol 2006;176:954–60; discussion 60.

36. ScherB,SeitzM,AlbingerW, TilingR,ScherrM,BeckerHC, et al. Valueof 11C-choline PET and PET/CT in patients with suspected prostatecancer. Eur J Nucl Med Mol Imaging 2007;34:45–53.

37. BeheshtiM, Imamovic L, BroingerG, Vali R,Waldenberger P, Stoiber F,et al. 18FcholinePET/CT in thepreoperative stagingof prostate cancerin patients with intermediate or high risk of extracapsular disease: aprospective study of 130 patients. Radiology 2010;254:925–33.

38. Hacker A, Jeschke S, Leeb K, Prammer K, Ziegerhofer J, SegaW, et al.Detection of pelvic lymph node metastases in patients with clinicallylocalized prostate cancer: comparison of [18F]fluorocholine positronemission tomography-computerized tomography and laparoscopicradioisotope guided sentinel lymph node dissection. J Urol2006;176:2014–8; discussion 8–9.

39. Husarik DB, Miralbell R, Dubs M, John H, Giger OT, Gelet A, et al.Evaluation of [(18)F]-choline PET/CT for staging and restaging ofprostate cancer. Eur J Nucl Med Mol Imaging 2008;35:253–63.

40. Hricak H, Dooms GC, Jeffrey RB, Avallone A, Jacobs D, Benton WK,et al. Prostatic carcinoma: staging by clinical assessment, CT, andMRimaging. Radiology 1987;162:331–6.

41. Clements R. Imaging in prostate cancer. 2010 [cited; Available from:http://emedicine.medscape.com/article/379996-overview.

42. Al-Saeedi F, Welch AE, Smith TA. [methyl-3H]Choline incorporationinto MCF7 tumor cells: correlation with proliferation. Eur J Nucl MedMol Imaging 2005;32:660–7.

43. Katz-Brull R, Seger D, Rivenson-Segal D, Rushkin E, Degani H.Metabolic markers of breast cancer: enhanced choline metabolismand reduced choline-ether-phospholipid synthesis. Cancer Res2002;62:1966–70.

44. Ramirez de Molina A, Banez-Coronel M, Gutierrez R, Rodriguez-Gonzalez A, Olmeda D, Megias D, et al. Choline kinase activation isa critical requirement for the proliferation of primary human mammaryepithelial cells and breast tumor progression. Cancer Res 2004;64:6732–9.

45. Piert M, Park H, Khan A, Siddiqui J, Hussain H, Chenevert T, et al.Detection of aggressive primary prostate cancer with 11C-cholinePET/CT using multimodality fusion techniques. J Nucl Med 2009;50:1585–93.






2011;17:7673-7683. Published OnlineFirst October 28, 2011.Clin Cancer Res Kaiyumars Contractor, Amarnath Challapalli, Tara Barwick, et al. with Choline Kinase ExpressionPelvic Lymph Node Status from Prostate Cancer and Relationship

C]Choline PET-CT as a Noninvasive Method for Detecting11Use of [

Updated version

10.1158/1078-0432.CCR-11-2048doi:

Access the most recent version of this article at:

Material

Supplementary

http://clincancerres.aacrjournals.org/content/suppl/2011/10/27/1078-0432.CCR-11-2048.DC1

Access the most recent supplemental material at:

Cited articles

http://clincancerres.aacrjournals.org/content/17/24/7673.full#ref-list-1

This article cites 43 articles, 13 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/17/24/7673.full#related-urls

This article has been cited by 12 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/17/24/7673To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-11-2048

http://clincancerres.aacrjournals.org/content/suppl/2011/10/27/1078-0432.CCR-11-2048.DC1

http://clincancerres.aacrjournals.org/content/17/24/7673.full#ref-list-1

http://clincancerres.aacrjournals.org/content/17/24/7673.full#related-urls

http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/17/24/7673


Use of [ C]Choline PET-CT as a Noninvasive Method …clincancerres.aacrjournals.org/content/clincanres/17/24/7673.full.pdf · Imaging, Diagnosis, Prognosis Use of [11C]Choline PET-CT

Documents