- Email: [email protected]
Use of 111In-Capromab Pendetide Immunoscintigraphy to Image Localized Prostate Cancer Foci Within the Prostate Gland
Use of 111In-Capromab Pendetide Immunoscintigraphy to Image Localized Prostate Cancer Foci Within the Prostate Gland Vladimir Mouraviev, John F. Madden, Gloria Broadwater, Janice M. Mayes,* James L. Burchette, Frank Schneider, Jill Smith, Matvey Tsivian, Terence Wong and Thomas J. Polascik†,‡ From the Duke Prostate Center and Division of Urologic Surgery, Department of Surgery (VM, JMM, JS, MT, TJP), Department of Pathology (JFM, JLB, FS), Cancer Center Biostatistics (GB) and Department of Radiology/Nuclear Medicine (TW), Duke University Medical Center, Durham, North Carolina
Abbreviations and Acronyms CT ⫽ computerized tomography FDA ⫽ Food and Drug Administration HGPIN ⫽ high grade prostatic intraepithelial neoplasia IHC ⫽ immunohistochemistry PCa ⫽ prostate cancer PSA ⫽ prostate specific antigen PSMA ⫽ prostate specific membrane antigen PTI ⫽ percent tumor involvement RP ⫽ radical prostatectomy RSI ⫽ radiology signal intensity SPECT ⫽ single photon emission CT Submitted for publication February 9, 2009. Study received internal review board approval. Supported by Cytogen Corp. * Financial interest and/or other relationship with Galil Medical. † Correspondence: Department of Urology, Duke University Medical Center, Box 2804, Yellow Zone, Durham, North Carolina 27710 (telephone: 919-684-4946; FAX: 919-684-5220; e-mail: [email protected]). ‡ Financial interest and/or other relationship with Cytogen.
Purpose: We compared the results of a preoperative 111In-capromab pendetide scan co-registered with computerized tomography with pathological findings in the surgically excised prostate to determine whether the scan can efficiently detect cancer in the prostate. Materials and Methods: This prospective trial included 25 hormone naïve men with clinically localized prostate cancer who underwent 111In-capromab pendetide single photon emission computerized tomography/computerized tomography as part of the preoperative evaluation. In addition to routine histological analysis, representative prostate sections were stained for prostate specific membrane antigen using the same antibody used in the scan. A pathologist and a radiologist were blinded to pathology and imaging findings, respectively. Prostate specific membrane antigen immunohistochemistry was correlated with the 3-dimensional location of the prostate specific membrane antigen signal detected by scan. Results: Scan sensitivity was 37% to 87% for 4 quadrants (right vs left and apical vs basal) with 0% to 50% specificity, as validated by final pathological assessment of the same quadrants. Stratifying positive scan signal strength did not statistically improve specificity (p ⫽ 0.35). There was no significant correlation between prostate specific membrane antigen over expression and tumor stage distribution (p ⫽ 0.23). Conclusions: The scan did not localize prostate cancer to a particular quadrant based on comparison with radical prostatectomy specimen pathology. The antibody used has affinity for benign and malignant prostatic glands in excised, formalin fixed prostate tissue, which may contribute to low scan specificity in vivo. The scan cannot be used to reliably detect or image cancer foci in the prostate. Key Words: prostate; prostatic neoplasms; tomography, emission-computed, single-photon; tomography, x-ray computed; capromab pendetide
For another article on a related topic see page 1186.
938
www.jurology.com
111
CAPROMAB pendetide is an agent used to detect soft tissue metastasis in patients with PCa that was approved for clinical use in 1996 by the FDA. A scintigraphic
In radiolabeled murine IgG1 monoclonal antibody reactive with PSMA is the basis for ProstaScint® (111In-capromab pendetide) imaging. 111In-ca-
0022-5347/09/1823-0938/0 THE JOURNAL OF UROLOGY® Copyright © 2009 by AMERICAN UROLOGICAL ASSOCIATION
Vol. 182, 938-948, September 2009 Printed in U.S.A. DOI:10.1016/j.juro.2009.05.047
111
IN-CAPROMAB PENDETIDE IMMUNOSCINTIGRAPHY FOR PROSTATE CANCER
promab pendetide is the murine monoclonal IgG1k subclass antibody 7E11-C5.3 conjugated to the linkerchelator glycyl-tyrosyl-lysyl-diethylenetetramine pentaacetic acid. The PSMA epitope recognized by mAb 7E11-C5.3 is located in the cytoplasmic domain and is highly specific for PCa. The scan has high sensitivity for detecting metastatic PCa and is capable of imaging metastatic lesions as small as 5 mm.1–5 In recent years earlier PCa diagnosis has led to a shift in the patient population toward men with lower tumor stage and limited tumor volume.6 This provoked interest in gland preserving, focal targeted ablation for treatment, which led in turn to a search for imaging methods that provide accurate preoperative spatial localization of primary PCa in the gland.7,8 Recently hybrid SPECT/CT scanners have become available that combine ␥ camera SPECT with CT. These scanners allow co-registered 111Incapromab pendetide images to be obtained with CT. This provides anatomical localization for the tracer accumulation and could improve the accuracy of this technique for localizing focal PCa in the gland. SPECT images can be reconstructed and interactive display software enables the display of fused CT and SPECT images along with review of the reconstructed images in arbitrary planes. We hypothesized that using advanced pretreatment 111In-capromab pendetide immunoscintigraphy fused with SPECT and CT images we might visualize and localize small volume PCa lesion(s) that were still organ confined. In this prospective clinical trial we evaluated the role of 111 In-capromab pendetide SPECT/CT to localize prostate tumor(s) in the prostate in patients electing RP with subsequent correlation and validation of image findings by final pathological assessment of prostatectomy specimens.
939
The radiologist assigned a semiquantitative immunoscintigraphy signal (none, low, moderate or high) to each of the 4 quadrants. The prostatectomy specimen was processed for histopathological examination, and hematoxylin and eosin stained sections were analyzed with particular attention given to tumor distribution in each of the 4 prostate quadrants, as described. In addition, representative sections with the largest cancer foci in each positive quadrant were stained for PSMA using the same antibody used in the 111In-capromab pendetide scan. A uropathologist with expertise in step-section processing and reading final histological slides, and a radiologist specializing in cross-sectional imaging and nuclear radiology were blinded to the results of imaging and pathological findings, respectively, when completing their assessments. PSMA IHC was correlated with the 3-dimensional location of the PSMA signal detected on the 111In-capromab pendetide/CT co-registered scan.
Co-Registered Prostate Imaging Our current 111In-capromab pendetide immunoscintigraphy technique was previously described in detail.9,10 Briefly, 180 to 220 MBq (5 to 6 mCi) 111In-capromab pendetide were administered by slow intravenous injection. The patient was asked to void before imaging and no Foley catheter was used. Imaging was performed 4 days after injection on a dual head Discovery VH scanner (GE Healthcare, Chicago, Illinois) with an integrated CT scanner built onto the same rotating gantry as the camera heads. CT of the pelvis was done in 10 minutes with the patient breathing quietly. While CT quality was degraded in the upper abdomen due to respiratory motion, these artifacts were not generally evident in the pelvis and the prostate gland could be clearly delineated. No intravenous or oral contrast medium was used. SPECT images were reconstructed using 2 iterations of ordered subsets expectation maximization with CT based attenuation correction and a final Butterworth filter (tenth order with cutoff at 0.26 Nyquist frequency). The interactive display software allows reconstruction and review of the images in the axial, coronal and sagittal planes, and display of the fused CT and SPECT images.
Specific Uptake Interpretation
MATERIALS AND METHODS Between January 2005 and April 2007, 25 hormone naïve men with clinically localized PCa (PSA less than 20 ng/ml, clinical stage less than T3, biopsy Gleason score 5– 8 and negative bone scan) underwent 111In-capromab pendetide scan with CT co-registration at our institution. Patients were scheduled for SPECT an average ⫾ SD of 73 ⫾ 24 days after prostate biopsy to avoid the influence of biopsy artifact on scan interpretation. We assumed that this term would be long enough to spontaneously resolve any artifact, although we found no evidence supporting this in the literature since to our knowledge it has not been studied. Within 2 months after the scan the men underwent RP. Since the resolution of the co-registered scan is not as detailed as pathological mapping of the RP specimen, we divided the prostate into 4 equal-sized quadrants (left and right base, and left and right apex) for analysis purposes.
Capromab pendetide SPECT scans were analyzed after overlay fusion on CT scans. CT images were used to localize regions of the prostate gland for evaluation and the corresponding capromab pendetide accumulation was scored on a quadrant by quadrant basis. Sites identified in the original SPECT report were reevaluated by a single experienced nuclear imager (TW). RSI was assigned to each of the 4 prostate quadrants and presented to an unbiased observer for statistical analysis.
RP Specimen Pathological Processing Each RP specimen was weighed, measured, inked and fixed in 10% neutral formalin. After fixation the apical and basal margins were thinly shaved and the bulk of the gland was sectioned at 3 mm intervals perpendicular to the rectal surface. Sections were formalin postfixed and paraffin embedded in the usual fashion, and 5 m hematoxylin and eosin stained sections were prepared. Routine pathological parameters were assessed, including pT
111
940
IN-CAPROMAB PENDETIDE IMMUNOSCINTIGRAPHY FOR PROSTATE CANCER
stage, Gleason score, PTI, extracapsular extension, positive surgical margins and seminal vesicle invasion. In addition, cancer localization by right-left/apical-basal quadrant was recorded. The apical-basal axis was considered an imaginary line parallel to the rectal surface of the gland that passed through the apical urethral orifice. The anatomical landmark considered to separate the apical from the basal portion of the gland was the verumontanum. Pathological interpretation was done by a genitourinary pathologist (JFM).
Table 2. PTI in each of 4 quadrants per patients in 25 patients
IHC Staining Sections (5 ) of formalin fixed, paraffin embedded tissue were placed on positively charged glass slides, dried and deparaffinized in the usual fashion. Endogenous peroxidase activity was quenched with methanolic H2O2 and the slides were hydrated with water. Heat induced epitope retrieval was performed by placing the tissue slides in a preheated solution of 10.0 mM tris and 1.0 mM ethylenediaminetetraacetic acid (pH 9.0) with heating for 20 minutes in a 100C water bath, followed by cooling for 20 minutes at room temperature. Sections were rinsed in D20 and placed in tris buffered saline before applying the primary antibody. P504S/racemase (Dako, Carpinteria, California) was diluted 1:100 and CYT-351 (clone 7E11 C3.5) monoclonal antibody (Cytogen®) was diluted 1:400 and 1:800 in 1% bovine serum albumin/tris buffered saline. The 2 antibodies were incubated for 60 minutes at room temperature. The bound primary antibody was detected with an application of horseradish peroxidase labeled EnVision™ Plus for rabbit and mouse primary antibodies for 30 minutes. DAB⫹ (Dako) was used to visualize the formed immune complex. A light hematoxylin counter stain was applied, and the sections were dehydrated, cleared and permanently mounted. Stained sections were examined by conventional light microscopy.
IHC Interpretation Staining intensity and localization pattern was semiquantitatively assessed using a published scheme.11 The distribution of staining in PCa cells was graded as focal— 10% or less of all PCa tumor cells on the selected slide, Table 1. Patient demographics and tumor characteristics Parameters Median age (range) No. race (%): Black White ⫹ other Median ng/ml diagnostic PSA (range) No. diagnostic PSA category (%): 10.00 ng/ml or Less Greater than 10.01 ng/ml No. clinical T stage (%): T1c T2a T2b T2c No. biopsy Gleason score (%): Less than 7 7 Greater than 7
61.0
% PTI
No. Quadrants
0 1 2 3 5 10 15 20 25 30 35 50 60 70 95
19 4 6 5 20 16 5 5 8 3 2 2 1 2 2
Total
100
regional—11% to 50% of PCa cells or diffuse— greater than 50% of PCa cells. Cytoplasmic staining intensity was subjectively graded as weak, moderate or strong. Cases in which the staining patterns were categorized as intense, moderate or strong (diffuse, intense regional or moderately diffuse) were considered positive for PSMA expression. The experimental design and controls for IHC PSMA antibody staining were repeated in the manner described by Ross et al.11 Pathological data were entered and presented to a neutral observer for statistical analysis.
Statistical Methods Descriptive statistics were performed for patient and tumor characteristics. We compared PTI with the 4 radiological signals while correcting for interclass correlation among the quadrants. PSMA expression identified in PCa quadrants was compared to that in benign quadrants using the Pearson chi-square test. The Wilcoxon rank sum test was used to compare Gleason score central tendencies in nonexpressed and expressed groups. The association between PSMA expression and tumor stage was compared using the chi-square test. Statistical analysis was performed by statisticians using SAS® 9.1. This study was internal review board approved (No. 7404-06-7R0ER).
(47–73)
1 (4) 24 (96) 5.0 (1.2–13.3) 21 4
(84) (16)
17 6 1 1
(68) (24) (4) (4)
12 8 5
(48) (32) (20)
RESULTS Table 1 lists patient demographics and tumor characteristics. Of the patients 80% had preoperative low and moderate risk features of PCa according to the D’Amico classification.12
Table 3. PTI vs RSI in quadrants RSI
No. quadrants
% Median PTI
None Low Moderate High
8 37 51 4
5 5 5 15
111
IN-CAPROMAB PENDETIDE IMMUNOSCINTIGRAPHY FOR PROSTATE CANCER
Table 6. PCa in quadrants by Gleason score
Table 4. No vs low/moderate/high RSI sensitivity and specificity by quadrant with PCa as gold standard target
No. Quadrant (%)
Quadrant
% Sensitivity
% Specificity
Lt apical Lt basal Rt apical Rt basal
87 57 89 37
0 50 17 50
Table 2 shows pathologically detected PTI in all 100 examined quadrants. Table 3 shows RSI stratified as negative, low, moderate or high. On pathological evaluation 19 of the 100 examined quadrants were cancer-free, while 35, 26 and 20 had less than 10%, 10% to less than 25% and 25% or greater tumor involvement with PTI, respectively. Table 3 shows RSI, graded negative, low, moderate or high, in the 100 examined quadrants and median PTI in the corresponding subset of quadrants. Low or moderate RSI was seen in 88 of the 100 quadrants examined. Only 4 of 100 quadrants showed negative RSI. Median PTI was identical in the negative, low and moderate RSI groups. The high RSI group, comprising only 4 of the 100 examined quadrants, showed slightly higher median PTI. Of these 4 quadrants representing 3 patients, only 1 patient with tumor in 2 quadrants had preoperative high risk features such as Gleason score greater than 8 by the D’Amico definition12 but in all 4 quadrants PTI was less than 50%. Statistically there were no significant differences in PTI among the 4 RSIs (none, low, moderate and high) when controlling for the interclass correlation among PTI measures in the 4 quadrants (p ⫽ 0.35, table 3). In any given quadrant the sensitivity and specificity of the co-registered scan for PCa was low (table 4). Sensitivity was 37% to 87% for the 4 quadrants with 0% to 50% specificity, correlating with pathologically confirmed cancer in the same quadrant in the resected gland. RSI correlated poorly with the mean total Gleason score of PCa foci (p ⫽ 0.44, table 5). Tumors were more frequently pathologically identified in the apical area, that is 23 of 25 patients (92%) had tumors in the right apex and 92% had tumors in the left apex. In contrast, the rate was 19 of 25 cases (75%) in the left basal quadrant and 17 of 25 (68%) in the right basal quadrant. Therefore, the Table 5. Total Gleason score vs RSI in areas with PCa on pathological evaluation RSI
No. PCa
Mean ⫾ SD Gleason Score
None Slight Moderate Marked p Value
7 31 40 4
7.29 ⫾ 0.756 6.90 ⫾ 0.746 7.08 ⫾ 0.997 6.50 ⫾ 0.577 0.442
941
Gleason Score
Lt Apical
Lt Basal
Rt Apical
Rt Basal
Total No. (%)
6 7 8 9
8 (34.8) 11 (47.8) 2 (8.7) 2 (8.7)
4 (21.1) 11 (57.9) 2 (10.5) 2 (10.5)
7 (30.4) 12 (52.2) 2 (8.7) 2 (8.7)
4 (23.5) 10 (58.8) 1 (5.9) 2 (11.8)
23 (27.1) 45 (52.9) 7 (8.2) 10 (11.8)
Totals
23 (100)
19 (100)
23 (100)
17 (100)
85 (100)
high tumor prevalence in the apical area accounts for the high scan sensitivity. Table 6 lists Gleason scores in the prostate quadrants with almost an equal ratio in the apical and basal quadrants. Moreover, a significant correlation was not found between PTI (greater or less than 50%) and SPECT/CT by quadrant (data not shown). IHC staining of formalin fixed, paraffin embedded prostate sections revealed PSMA staining of malignant foci in 85% of cases but also staining of benign tissue regions in 76% (p ⫽ 0.16). PSMA expression in malignant foci was more common in cases with a higher total Gleason score (p ⫽ 0.03, table 7). There was no significant association between PSMA expression strength and tumor stage (p ⫽ 0.23, data not shown). Figures 1 to 6 show the weak correlation between RSI and PSMA expression by PCa tissue. High RSI was registered on the scan in the left base but only 1% PCa was found at final pathological assessment with equal staining of micro PCa and benign tissue (figs. 1 and 2). Moderate RSI was seen in each apical quadrant but PTI in the left apex was 60% and only 15% in the right apical quadrant (figs. 3 and 4). Furthermore, PCa in the left apex was unstained while benign tissue intensively stained with 7E11C5.3 PSMA antibody. Mild RSI was documented on the scan in the 2 basal quadrants while only 1 focus of micropapillary pattern HGPIN was identified in the right base (figs. 5 and 6).
DISCUSSION Currently there is a need to develop imaging modalities capable of localizing PCa foci in the prostate. This would enable image directed biopsy for diagnoTable 7. Gleason score and PCa cell PSMA expression in areas with PCa on pathological evaluation PSMA Intensity Weak Moderate Strong Total p Value
No. PCa
Mean ⫾ SD Gleason Score
3 18 61 82
6.67 ⫾ 0.577 6.56 ⫾ 0.511 7.15 ⫾ 0.928 7.00 ⫾ 0.875 0.031
942
111
IN-CAPROMAB PENDETIDE IMMUNOSCINTIGRAPHY FOR PROSTATE CANCER
Figure 1. Patient 19. Co-registered capromab pendetide scan shows high RSI in prostate left base (arrows)
sis, image guidance for focal therapy and image based surveillance strategies after treatment. Such an advance might have far-reaching implications for our approach to PCa management. Although the 111 In-capromab pendetide scan is not FDA approved for this indication, we theorized that if this immunoscintigraphy scan was co-registered with crosssectional CT, it might delineate cancer foci in the prostate. In fact, FDA approved indications include only high risk features or cases suspicious for recurrent disease after primary definitive therapy.12 The rationale on which positron emission tomography/CT is based (combining the functional features of positron emission tomography with the anatomical detail of CT) provides many advantages that may be easily transferable to SPECT.13–15 Our efforts focused on applying fused 111In-capromab pendetide SPECT/CT to identify PCa inside the prostate by radio immunoscintigraphy. Our choice of specimen pathological preparation deserves mention. At our institution we use routine
close (every 3 mm) step sectioning of RP specimens. While there are advantages of a complete embedding technique with whole mounted processing to match with imaging, studies such as those by Hollenbeck16 and Desai17 et al do not show a significant difference in pathological outcome between the 2 techniques when using multivariate logistics to control for preoperative variables. In fact, to our knowledge the problem of ideal preoperative imaging superimposition with a whole mounted slide remains to be solved. Therefore, we think that using quadrant involvement may be a reasonable option for correlation, although this approach may also have limitations. Quadrant evaluation was also ideal to prevent bias in the results. Alterations in gland geometry by fixation and sectioning do not extend to the conflating right and left or anterior and posterior. We were confident that we could maintain registration on radiology and pathology images between right and left, and between anterior and posterior. By claim-
111
IN-CAPROMAB PENDETIDE IMMUNOSCINTIGRAPHY FOR PROSTATE CANCER
ing only quadrant registration we set the threshold for success as low as possible for the 111In-capromab pendetide method. If it could not even localize disease by quadrant, it could certainly not localize disease by sextant or octant. Therefore, to claim that we could guarantee registration over small volumes would have been more and not less stringent toward the 111In-capromab pendetide method. The diagnostic value of 111In-capromab pendetide scan to detect extraprostatic PCa is well studied. Many groups have noted that the technique is valuable when formulating a treatment plan for salvage radiation therapy after primary RP and predicting the outcome after salvage treatment.11,14,18 –21 However, other groups could not corroborate these findings.9,15,22,23 Reading and interpreting 111In-capromab pendetide scans, especially by young radiologists without sufficient experience, remain subjective but coregistration with CT may allow better anatomical correlation regarding tumor location. In addition, the high volume of 111In-capromab pendetide scans performed at centers of excellence may decrease interpretation and conclusion errors. The fused 111Incapromab/CT technology evolves constantly. As further advances are made, the technology will improve, allowing more accurate diagnoses. In particular improvement in CT image quality and SPECT image processing would potentially enable more accurate evaluation of the prostatic and regional lymph node beds. To date no accurate imaging modality is available that provides precise visualization of small, prostate confined PCa. To our knowledge we present the first study in the literature to combine 111In-capromab pendetide with CT to image PCa in the gland, as validated by final pathology assessment of RP specimens. We could not determine the efficacy of fused 111 In-capromab pendetide SPECT/CT to image localized PCa in the gland, as validated by the results of final pathology assessment of RP specimens. Analysis of IHC stained RP specimens using the same antibody to PSMA as in the 111In-capromab pendetide scan did not show any correlation between staining and PTI, tumor stage and primary or secondary Gleason score. When comparing the RSI of the co-registered scan to the pathological IHC of the RP specimen using the same 7E11-C5.3 PSMA antibody as the 111In-capromab pendetide scan, no significant correlation was found. We discerned no particular subset of patients in our group in whom the co-registered scan correlated with RP pathology when evaluating PTI. In the literature we found an assessment scale for 7E11 antibody staining. IHC staining interpretation was done according to semiquantitative assessment of IHC results for the 7E11 antibody previously re-
943
Figure 2. Patient 19. IHC of corresponding quadrant (fig. 1) reveals tiny PCa focus (arrows) stained with weak-moderate intensity and surrounding benign glandular tissue with weak, moderate and strong staining. PSMA antibody (7E11-C5.3) staining, reduced from ⫻4.
ported by Ross et al.11 The advantage of this interpretation is that it incorporates the staining distribution in focal, regional or diffuse fashion along with the intensity of cytoplasmic staining as weak, moderate or intense, respectively. All prostatic conditions were examined as cancer vs no cancer or benign lesions, including inflammatory foci, atrophy etc, according to the aim of the trial. In regard to the literature we are unaware of published findings showing the usefulness of 111indiumcapromab imaging for localized PCa in the prostate gland, as validated by IHC results using 7E11C5.3 antibody on closed step-sectioning of RP specimens. A few studies describe IHC staining differences between benign and malignant prostatic glands.11,24 –26 Our findings differ from those in these previous series. Our IHC results reveal significant PSMA immunoreactivity in benign prostatic epithelium as well as heterogeneous PSMA immunoreactivity in a significant proportion of prostatic adenocarcinoma foci, similar to results in the recent study by Mannweiler et al.27 Our findings support the conclusion that IHC using standard techniques with 7E11-C5.3 antibody in formalin fixed, paraffin embedded sections has poor specificity for PCa since the antibody revealed significant staining of normal prostatic epithelium. This finding is not surprising since benign and malignant prostate epithelial cells express PSMA.11,28,29 Also, in RP specimens we found no association between the intensity of 7E11-C5.3 PSMA antibody staining with variables such as primary and secondary Gleason grade, tumor stage and PTI. The reason that 111In-capromab pendetide immunoscintigraphy
944
111
IN-CAPROMAB PENDETIDE IMMUNOSCINTIGRAPHY FOR PROSTATE CANCER
Figure 3. Patient 25. Co-registered capromab pendetide scan demonstrates moderate RSI in prostate right and left apical quadrants (arrows).
Figure 4. Patient 25. IHC (fig 3). A, prostate left apex shows strong staining of benign tissue (thick arrows) and weak staining of PCa (thin arrows). B, moderate-strong staining of right apex benign tissue with weak staining of atrophic area (arrows). PSMA antibody (7E11-C5.3) staining, reduced from ⫻4.
111
IN-CAPROMAB PENDETIDE IMMUNOSCINTIGRAPHY FOR PROSTATE CANCER
945
Figure 5. Patient 22. Co-registered capromab pendetide scan demonstrates mild RSI in 2 prostate basal quadrants (arrows)
is useful for extraprostatic disease, eg lymph node metastasis, is because only malignant elements metastasize or invade beyond the capsule and, thus, there are no benign prostate elements to distinguish. There are several limitations of this study. 1) The size of the prospective cohort of 25 patients was restricted. However, statistically significant findings were identified as reported despite the sample size. 2) The resolution of these imaging techniques continues to improve and SPECT/CT technology continues to evolve. It is possible that with a better monoclonal antibody or improved imaging PCa foci may be imaged in the prostate using a similar technique. However, given these limitations, and based on the commercially available antibody and available imaging equipment at our centers between 2005 and 2007 when this study was performed, coregistered 111In-capromab pendetide SPECT/CT could not reliably image intraprostatic PCa foci due to low specificity of the 7E11-C5.3 antibody for PCa.
3) The fundamental biological principle of how an antibody can be delivered into a cancer cell to interact with an intracellular epitope deserves mention. Following the basic principle of receptor mediated signaling and/or endocytosis, the Ig molecule, like any other large hydrophilic protein, does not cross the lipophilic plasma membrane without specifically binding to an extracellular receptor. However, this commercially available 7E11-C5.3 antibody has the capromab epitope intracellularly and, therefore, it is unclear how this specific antibody binds to its target mechanistically. 4) Recent data suggest that PSMA over expression may be more common in high vs low grade PCa, which may make it a nonspecific tool, especially for early stage, localized PCa.26,30 Furthermore, changes in PSMA expression were noted in other benign and malignant tumors, such as schwannoma and endometrial adenocarcinoma, respectively.31,32 Therefore, these cases should be clinically considered in the differential diagnosis of a lesion that is positive on PSMA radio immunoscin-
946
111
IN-CAPROMAB PENDETIDE IMMUNOSCINTIGRAPHY FOR PROSTATE CANCER
Figure 6. Patient 22. IHC staining of prostate right base (fig. 5). A, highly specific marker reveals HGPIN and PCa. P504S/racemase, reduced from ⫻4. B, strongly stained benign and HGPIN areas. PSMA antibody (7E11-C5.3), reduced from ⫻4.
tigraphy. Ultimately we understand that the scan may be more beneficial in patients with high risk features, in whom PSMA expression could be greater. 5) Finally, the exact clinical implication of 111 In-capromab pendetide scan results remains concerning when defining the spatial tumor distribution inside the prostate, especially for early stage, localized PCa. The recent emergence of high resolution imaging tools coupled with advances in computerized modeling software should be used in the near future to provide alternative treatment options in men with localized, early stage cancer. This approach is an important step in our quest for better ways to diagnose and treat the disease while maintaining good quality of life in our patients.
CONCLUSIONS Diagnostic 111In-capromab pendetide scan with CT co-registration imaging of the prostate gland did not correlate with the pathologically assessed presence of prostate confined PCa in RP specimens. IHC staining with the same antibody (7E11C5.3) as the
111
In-capromab pendetide scan showed staining of benign and malignant prostatic epithelium, suggesting that lack of specificity for malignant prostatic glands may be a cause of the poor correlation. Further investigation using fused 111In-capromab pendetide scan with CT or magnetic resonance imaging is needed to define the possible role of this scan in treatment diagnostics.
ACKNOWLEDGMENTS Dr. Bercedis Peterson, Department of Biostatistics and Bioinformatics provided statistical analysis. Steven W. Shipes, Chief Technologist, Department of Radiology/Nuclear Medicine, Duke University Medical Center collected and analyzed imaging data. Duke is a designated Partners in Excellence center of Cytogen. Each Partners in Excellence Site receives rigorous training, undergoes proficiency testing and is subject to ongoing quality assurance protocols. To qualify as a Partners in Excellence Site each center must be certified as proficient in the interpretation of ProstaScint scans by the American College of Nuclear Physicians.
REFERENCES 1. Sodee DB, Ellis RJ, Samuels MA et al: Prostate cancer and prostate bed SPECT imaging with ProstaScint: semiquantitative correlation with prostatic biopsy results. Prostate 1998; 37: 140. 2. Raj GV, Partin AW and Polascik TJ: Clinical utility of indium 111-capromab pendetide immunoscintigraphy in the detection of early, recurrent prostate carcinoma after radical prostatectomy. Cancer 2002; 94: 987. 3. Polascik TJ, Manyak MJ, Haseman MK et al: Comparison of clinical staging algorithms and 111indium-capromab pendetide immunoscintigra-
phy in the prediction of lymph node involvement in high risk prostate carcinoma patients. Cancer 1999; 85: 1586. 4. Manyak MJ, Hinkle GH, Olsen JO et al: Immunoscintigraphy with indium-111-capromab pendetide: evaluation before definitive therapy in patients with prostate cancer. Urology 1999; 54: 1058. 5. Ellis RJ, Kim EY, Conant R et al: Radioimmunoguided imaging of prostate cancer foci with histopathological correlation. Int J Radiat Oncol Biol Phys 2001; 49: 1281.
6. Polascik TJ, Mayes JM, Sun L et al: Pathologic stage T2a and T2b prostate cancer in the recent prostate-specific antigen era: implications for unilateral ablative therapy. Prostate 2008; 68: 1380. 7. Polascik TJ and Mouraviev V: Focal therapy for prostate cancer. Curr Opin Urol 2008; 18: 269. 8. Ahmed HU and Emberton M: Active surveillance and radical therapy in prostate cancer: can focal therapy offer the middle way? World J Urol 2008; 26: 457. 9. Koontz BF, Mouraviev V, Johnson JL et al: Use of local (111)in-capromab pendetide scan results to
111
IN-CAPROMAB PENDETIDE IMMUNOSCINTIGRAPHY FOR PROSTATE CANCER
predict outcome after salvage radiotherapy for prostate cancer. Int J Radiat Oncol Biol Phys 2008; 71: 358. 10. Wong TZ, Turkington TG, Polascik TJ et al: ProstaScint (capromab pendetide) imaging using hybrid gamma camera-CT technology. AJR Am J Roentgenol 2005; 184: 676. 11. Ross JS, Sheehan CE, Fisher HA et al: Correlation of primary tumor prostate-specific membrane antigen expression with disease recurrence in prostate cancer. Clin Cancer Res 2003; 9: 6357. 12. D’Amico AV, Moul J, Carroll PR et al: Cancerspecific mortality after surgery or radiation for patients with clinically localized prostate cancer managed during the prostate-specific antigen era. J Clin Oncol 2003; 21: 2163. 13. Ellis RJ and Kaminsky DA: Fused radioimmunoscintigraphy for treatment planning. Rev Urol, suppl., 2006; 1: S11. 14. Sodee DB, Sodee AE and Bakale G: Synergistic value of single-photon emission computed tomography/computed tomography fusion to radioimmunoscintigraphic imaging of prostate cancer. Semin Nucl Med 2007; 37: 17. 15. Nagda SN, Mohideen N, Lo SS et al: Long-term follow-up of 111In-capromab pendetide (ProstaScint) scan as pretreatment assessment in patients who undergo salvage radiotherapy for rising prostate-specific antigen after radical prostatectomy for prostate cancer. Int J Radiat Oncol Biol Phys 2007; 67: 834. 16. Hollenbeck BK, Bassily N, Wei JT et al: Whole mounted radical prostatectomy specimens do not increase detection of adverse pathological features. J Urol 2000; 164: 1583.
17. Desai A, Wu H, Sun L et al: Complete embedding and close step-sectioning of radical prostatectomy specimens both increase detection of extra-prostatic extension, and correlate with increased disease-free survival by stage of prostate cancer patients. Prostate Cancer Prostatic Dis 2002; 5: 212. 18. Ellis RJ, Zhou EH, Fu P et al: Single photon emission computerized tomography with capromab pendetide plus computerized tomography image set co-registration independently predicts biochemical failure. J Urol 2008; 179: 1768.
947
nign and malignant prostate tissues. Urol Oncol 1995; 1: 18. 25. Silver DA, Pellicer I, Fair WR et al: Prostatespecific membrane antigen expression in normal and malignant human tissues. Clin Cancer Res 1997; 3: 81. 26. Perner S, Hofer MD, Kim R et al: Prostate-specific membrane antigen expression as a predictor of prostate cancer progression. Hum Pathol 2007; 38: 696.
19. Keane TE, Rosner IL, Wingo MS et al: The emergence of radioimmunoscintigraphy for prostate cancer. Rev Urol, suppl., 2006; 1: S20.
27. Mannweiler S, Amersdorfer P, Trajanoski S et al: Heterogeneity of prostate-specific membrane antigen (PSMA) expression in prostate carcinoma with distant metastasis. Pathol Oncol Res 2008; 15: 167.
20. Sodee DB, Malguria N, Faulhaber P et al: Multicenter ProstaScint imaging findings in 2154 patients with prostate cancer. The ProstaScint Imaging Centers. Urology 2000; 56: 988.
28. Holmes EH: PSMA specific antibodies and their diagnostic and therapeutic use. Expert Opin Investig Drugs 2001; 10: 511.
21. Proano JM, Sodee DB, Resnick MI et al: The impact of a negative (111)indium-capromab pendetide scan before salvage radiotherapy. J Urol 2006; 175: 1668.
29. Sweat SD, Pacelli A, Murphy GP et al: Prostatespecific membrane antigen expression is greatest in prostate adenocarcinoma and lymph node metastases. Urology 1998; 52: 637.
22. Thomas CT, Bradshaw PT, Pollock BH et al: Indium-111-capromab pendetide radioimmunoscintigraphy and prognosis for durable biochemical response to salvage radiation therapy in men after failed prostatectomy. J Clin Oncol 2003; 21: 1715.
30. Bostwick DG, Pacelli A, Blute M et al: Prostate specific membrane antigen expression in prostatic intraepithelial neoplasia and adenocarcinoma: a study of 184 cases. Cancer 1998; 82: 2256.
23. Kahn D, Williams RD, Manyak MJ et al: 111Indiumcapromab pendetide in the evaluation of patients with residual or recurrent prostate cancer after radical prostatectomy. The ProstaScint Study Group. J Urol 1998; 159: 2041. 24. Wright GL, Haley C, Beckett ML et al: Expression of prostate-specific membrane antigen in normal, be-
31. Wang W, Tavora F, Sharma R et al: PSMA expression in Schwannoma: a potential clinical mimicker of metastatic prostate carcinoma. Urol Oncol, 2008. 32. Mhawech-Fauceglia P, Smiraglia DJ, Bshara W et al: Prostate-specific membrane antigen expression is a potential prognostic marker in endometrial adenocarcinoma. Cancer Epidemiol Biomarkers Prev 2008; 17: 571.
EDITORIAL COMMENTS I concur with these authors that the ability to identify cancer in the prostate gland would have far-reaching implications. A key question is whether one may reasonably expect to specifically image PCa using an antibody to PSMA in a gland that also contains PSMA positive normal glands. Surprisingly perhaps the answer is yes. The reason is that beyond the requisite expression of antigen several other factors determine antibody localization and, therefore, imaging or therapy. One of these factors relates to the accessibility of the antigen to administered antibody. In normal prostate glands PSMA is expressed on the luminal surface of the secretory epithelium. This site is not exposed to circulating antibody due to the intervening basal cell layer as well as to epithelial tight junctions. Conversely in cases of invasive PCa PSMA loses its polarized luminal expression, and the basal cell layer and tight junction barriers are
gone, exposing PSMA to circulating antibody. How significant is this barrier to antibody penetration in normal glands? These barriers account for the 1,000,000-fold concentration gradient of PSA between normal prostatic ducts and plasma. By extension normal glandular PSMA would be exposed to a 1,000,000-fold lower concentration of antiPSMA antibody as would invasive PCa. Given what appears to be a feasible setting for PSMA antibody to specifically target PCa, why did capromab not succeed? Because capromab has an additional barrier that is often cited but rarely heeded. Capromab binds to the intracellular region of the PSMA molecule, which is hidden from antibody view by the intact cell membrane.1 As a result, capromab cannot see any PSMA regardless of whether it is expressed by a normal cell or a cancer cell.2,3 Only when the cell membrane is disrupted, as in dying cells, can capromab bind its
948
111
IN-CAPROMAB PENDETIDE IMMUNOSCINTIGRAPHY FOR PROSTATE CANCER
target.2,4 However, PSMA antibodies are available that bind the PSMA external domain, avoiding this capromab related barrier. One such antibody (J591) is currently in clinical trials and has shown substantially improved imaging of metastatic PCa relative to capromab,5,6 although not in a head-tohead comparison. This suggests that the potential exists to ultimately identify PCa in the prostate
and achieve the far-reaching implications alluded to by these authors. Neil H. Bander Department of Urology Weill/Cornell Medical College Memorial Sloan-Kettering Cancer Center New York, New York
REFERENCES 1. Troyer JK, Beckett ML and Wright GL: Location of prostate-specific membrane antigen in the LNCaP prostate carcinoma cell line. Prostate 1997; 30: 232.
monoclonal antibodies specific for the extracellular domain of prostate-specific membrane antigen. Cancer Res 2000; 60: 5237.
2. Liu H, Moy P, Kim S et al: Monoclonal antibodies to the extracellular domain of prostate-specific membrane antigen also react with tumor vascular endothelium. Cancer Res 1997; 57: 3629.
4. Smith-Jones PM, Vallabhajosula S, Navarro V et al: Radiolabeled monoclonal antibodies specific to the extracellular domain of prostate-specific membrane antigen: preclinical studies in nude mice bearing LNCaP human prostate tumor. J Nucl Med 2003; 44: 610.
3. Smith-Jones PM, Vallabahajosula S, Goldsmith SJ et al: In vitro characterization of radiolabeled
These authors studied a small, nonconsecutive, lower risk cohort with absent followup, resulting in findings that raise more questions than answers. In contrast, we reported long-term followup in a clinical series of 239 patients demonstrating the abililty of capromab pendetide to predict biochemical failure in those at intermediate risk and improve biochemical disease-free survival with dose escalation to SPECT/CT defined target volumes (reference 18 in article).1 In our protocol radiology sets ␥ contrast using positive findings on diagnostic biopsy histopathology. The parameters of the current study (selecting lower risk patients and blinding radiology to diagnostic pathology) effectively biased study results. This potential systematic error can be observed in the figures, in which bone marrow inten-
5. Milowsky MI, Nanus DM, Kostaglou L et al: Phase I trial of 90Y-labeled anti-PSMA monoclonal antibody J591 for androgen-independent prostate cancer. J Clin Oncol 2004; 22: 2522. 6. Bander NH, Milowsky MI, Nanus DM et al: Phase I trial of 177lutetium-labeled J591, a monoclonal antibody to prostate-specific membrane antigen, in patients with androgen-independent prostate cancer. J Clin Oncol 2005; 23: 4591.
sity appears to directly correlate with examples of high to mild tracer uptake in tumor. An improved study design for histopathology confirmation of intraprostatic imaging should incorporate in vivo fiducial marker correlation with diagnostic biopsy pathology and hybrid molecular image sets (reference 13 in article). Rodney J. Ellis Radiation Oncology Aultman Hospital Canton, Ohio Department of Radiology Northeastern Ohio Universities College of Medicine Rootstown, Ohio Department of Urology Case Western Reserve University School of Medicine Cleveland, Ohio
REFERENCE 1. Ellis RJ, Zhou H, Kim EY et al: Biochemical disease-free survival rates following definitive low-dose-rate prostate brachytherapy with dose escalation to biologic target volumes identified with SPECT/CT capromab pendetide. J Brachyther 2007; 6: 16.
Abbreviations and Acronyms CT ⫽ computerized tomography FDA ⫽ Food and Drug Administration HGPIN ⫽ high grade prostatic intraepithelial neoplasia IHC ⫽ immunohistochemistry PCa ⫽ prostate cancer PSA ⫽ prostate specific antigen PSMA ⫽ prostate specific membrane antigen PTI ⫽ percent tumor involvement RP ⫽ radical prostatectomy RSI ⫽ radiology signal intensity SPECT ⫽ single photon emission CT Submitted for publication February 9, 2009. Study received internal review board approval. Supported by Cytogen Corp. * Financial interest and/or other relationship with Galil Medical. † Correspondence: Department of Urology, Duke University Medical Center, Box 2804, Yellow Zone, Durham, North Carolina 27710 (telephone: 919-684-4946; FAX: 919-684-5220; e-mail: [email protected]). ‡ Financial interest and/or other relationship with Cytogen.
Purpose: We compared the results of a preoperative 111In-capromab pendetide scan co-registered with computerized tomography with pathological findings in the surgically excised prostate to determine whether the scan can efficiently detect cancer in the prostate. Materials and Methods: This prospective trial included 25 hormone naïve men with clinically localized prostate cancer who underwent 111In-capromab pendetide single photon emission computerized tomography/computerized tomography as part of the preoperative evaluation. In addition to routine histological analysis, representative prostate sections were stained for prostate specific membrane antigen using the same antibody used in the scan. A pathologist and a radiologist were blinded to pathology and imaging findings, respectively. Prostate specific membrane antigen immunohistochemistry was correlated with the 3-dimensional location of the prostate specific membrane antigen signal detected by scan. Results: Scan sensitivity was 37% to 87% for 4 quadrants (right vs left and apical vs basal) with 0% to 50% specificity, as validated by final pathological assessment of the same quadrants. Stratifying positive scan signal strength did not statistically improve specificity (p ⫽ 0.35). There was no significant correlation between prostate specific membrane antigen over expression and tumor stage distribution (p ⫽ 0.23). Conclusions: The scan did not localize prostate cancer to a particular quadrant based on comparison with radical prostatectomy specimen pathology. The antibody used has affinity for benign and malignant prostatic glands in excised, formalin fixed prostate tissue, which may contribute to low scan specificity in vivo. The scan cannot be used to reliably detect or image cancer foci in the prostate. Key Words: prostate; prostatic neoplasms; tomography, emission-computed, single-photon; tomography, x-ray computed; capromab pendetide
For another article on a related topic see page 1186.
938
www.jurology.com
111
CAPROMAB pendetide is an agent used to detect soft tissue metastasis in patients with PCa that was approved for clinical use in 1996 by the FDA. A scintigraphic
In radiolabeled murine IgG1 monoclonal antibody reactive with PSMA is the basis for ProstaScint® (111In-capromab pendetide) imaging. 111In-ca-
0022-5347/09/1823-0938/0 THE JOURNAL OF UROLOGY® Copyright © 2009 by AMERICAN UROLOGICAL ASSOCIATION
Vol. 182, 938-948, September 2009 Printed in U.S.A. DOI:10.1016/j.juro.2009.05.047
111
IN-CAPROMAB PENDETIDE IMMUNOSCINTIGRAPHY FOR PROSTATE CANCER
promab pendetide is the murine monoclonal IgG1k subclass antibody 7E11-C5.3 conjugated to the linkerchelator glycyl-tyrosyl-lysyl-diethylenetetramine pentaacetic acid. The PSMA epitope recognized by mAb 7E11-C5.3 is located in the cytoplasmic domain and is highly specific for PCa. The scan has high sensitivity for detecting metastatic PCa and is capable of imaging metastatic lesions as small as 5 mm.1–5 In recent years earlier PCa diagnosis has led to a shift in the patient population toward men with lower tumor stage and limited tumor volume.6 This provoked interest in gland preserving, focal targeted ablation for treatment, which led in turn to a search for imaging methods that provide accurate preoperative spatial localization of primary PCa in the gland.7,8 Recently hybrid SPECT/CT scanners have become available that combine ␥ camera SPECT with CT. These scanners allow co-registered 111Incapromab pendetide images to be obtained with CT. This provides anatomical localization for the tracer accumulation and could improve the accuracy of this technique for localizing focal PCa in the gland. SPECT images can be reconstructed and interactive display software enables the display of fused CT and SPECT images along with review of the reconstructed images in arbitrary planes. We hypothesized that using advanced pretreatment 111In-capromab pendetide immunoscintigraphy fused with SPECT and CT images we might visualize and localize small volume PCa lesion(s) that were still organ confined. In this prospective clinical trial we evaluated the role of 111 In-capromab pendetide SPECT/CT to localize prostate tumor(s) in the prostate in patients electing RP with subsequent correlation and validation of image findings by final pathological assessment of prostatectomy specimens.
939
The radiologist assigned a semiquantitative immunoscintigraphy signal (none, low, moderate or high) to each of the 4 quadrants. The prostatectomy specimen was processed for histopathological examination, and hematoxylin and eosin stained sections were analyzed with particular attention given to tumor distribution in each of the 4 prostate quadrants, as described. In addition, representative sections with the largest cancer foci in each positive quadrant were stained for PSMA using the same antibody used in the 111In-capromab pendetide scan. A uropathologist with expertise in step-section processing and reading final histological slides, and a radiologist specializing in cross-sectional imaging and nuclear radiology were blinded to the results of imaging and pathological findings, respectively, when completing their assessments. PSMA IHC was correlated with the 3-dimensional location of the PSMA signal detected on the 111In-capromab pendetide/CT co-registered scan.
Co-Registered Prostate Imaging Our current 111In-capromab pendetide immunoscintigraphy technique was previously described in detail.9,10 Briefly, 180 to 220 MBq (5 to 6 mCi) 111In-capromab pendetide were administered by slow intravenous injection. The patient was asked to void before imaging and no Foley catheter was used. Imaging was performed 4 days after injection on a dual head Discovery VH scanner (GE Healthcare, Chicago, Illinois) with an integrated CT scanner built onto the same rotating gantry as the camera heads. CT of the pelvis was done in 10 minutes with the patient breathing quietly. While CT quality was degraded in the upper abdomen due to respiratory motion, these artifacts were not generally evident in the pelvis and the prostate gland could be clearly delineated. No intravenous or oral contrast medium was used. SPECT images were reconstructed using 2 iterations of ordered subsets expectation maximization with CT based attenuation correction and a final Butterworth filter (tenth order with cutoff at 0.26 Nyquist frequency). The interactive display software allows reconstruction and review of the images in the axial, coronal and sagittal planes, and display of the fused CT and SPECT images.
Specific Uptake Interpretation
MATERIALS AND METHODS Between January 2005 and April 2007, 25 hormone naïve men with clinically localized PCa (PSA less than 20 ng/ml, clinical stage less than T3, biopsy Gleason score 5– 8 and negative bone scan) underwent 111In-capromab pendetide scan with CT co-registration at our institution. Patients were scheduled for SPECT an average ⫾ SD of 73 ⫾ 24 days after prostate biopsy to avoid the influence of biopsy artifact on scan interpretation. We assumed that this term would be long enough to spontaneously resolve any artifact, although we found no evidence supporting this in the literature since to our knowledge it has not been studied. Within 2 months after the scan the men underwent RP. Since the resolution of the co-registered scan is not as detailed as pathological mapping of the RP specimen, we divided the prostate into 4 equal-sized quadrants (left and right base, and left and right apex) for analysis purposes.
Capromab pendetide SPECT scans were analyzed after overlay fusion on CT scans. CT images were used to localize regions of the prostate gland for evaluation and the corresponding capromab pendetide accumulation was scored on a quadrant by quadrant basis. Sites identified in the original SPECT report were reevaluated by a single experienced nuclear imager (TW). RSI was assigned to each of the 4 prostate quadrants and presented to an unbiased observer for statistical analysis.
RP Specimen Pathological Processing Each RP specimen was weighed, measured, inked and fixed in 10% neutral formalin. After fixation the apical and basal margins were thinly shaved and the bulk of the gland was sectioned at 3 mm intervals perpendicular to the rectal surface. Sections were formalin postfixed and paraffin embedded in the usual fashion, and 5 m hematoxylin and eosin stained sections were prepared. Routine pathological parameters were assessed, including pT
111
940
IN-CAPROMAB PENDETIDE IMMUNOSCINTIGRAPHY FOR PROSTATE CANCER
stage, Gleason score, PTI, extracapsular extension, positive surgical margins and seminal vesicle invasion. In addition, cancer localization by right-left/apical-basal quadrant was recorded. The apical-basal axis was considered an imaginary line parallel to the rectal surface of the gland that passed through the apical urethral orifice. The anatomical landmark considered to separate the apical from the basal portion of the gland was the verumontanum. Pathological interpretation was done by a genitourinary pathologist (JFM).
Table 2. PTI in each of 4 quadrants per patients in 25 patients
IHC Staining Sections (5 ) of formalin fixed, paraffin embedded tissue were placed on positively charged glass slides, dried and deparaffinized in the usual fashion. Endogenous peroxidase activity was quenched with methanolic H2O2 and the slides were hydrated with water. Heat induced epitope retrieval was performed by placing the tissue slides in a preheated solution of 10.0 mM tris and 1.0 mM ethylenediaminetetraacetic acid (pH 9.0) with heating for 20 minutes in a 100C water bath, followed by cooling for 20 minutes at room temperature. Sections were rinsed in D20 and placed in tris buffered saline before applying the primary antibody. P504S/racemase (Dako, Carpinteria, California) was diluted 1:100 and CYT-351 (clone 7E11 C3.5) monoclonal antibody (Cytogen®) was diluted 1:400 and 1:800 in 1% bovine serum albumin/tris buffered saline. The 2 antibodies were incubated for 60 minutes at room temperature. The bound primary antibody was detected with an application of horseradish peroxidase labeled EnVision™ Plus for rabbit and mouse primary antibodies for 30 minutes. DAB⫹ (Dako) was used to visualize the formed immune complex. A light hematoxylin counter stain was applied, and the sections were dehydrated, cleared and permanently mounted. Stained sections were examined by conventional light microscopy.
IHC Interpretation Staining intensity and localization pattern was semiquantitatively assessed using a published scheme.11 The distribution of staining in PCa cells was graded as focal— 10% or less of all PCa tumor cells on the selected slide, Table 1. Patient demographics and tumor characteristics Parameters Median age (range) No. race (%): Black White ⫹ other Median ng/ml diagnostic PSA (range) No. diagnostic PSA category (%): 10.00 ng/ml or Less Greater than 10.01 ng/ml No. clinical T stage (%): T1c T2a T2b T2c No. biopsy Gleason score (%): Less than 7 7 Greater than 7
61.0
% PTI
No. Quadrants
0 1 2 3 5 10 15 20 25 30 35 50 60 70 95
19 4 6 5 20 16 5 5 8 3 2 2 1 2 2
Total
100
regional—11% to 50% of PCa cells or diffuse— greater than 50% of PCa cells. Cytoplasmic staining intensity was subjectively graded as weak, moderate or strong. Cases in which the staining patterns were categorized as intense, moderate or strong (diffuse, intense regional or moderately diffuse) were considered positive for PSMA expression. The experimental design and controls for IHC PSMA antibody staining were repeated in the manner described by Ross et al.11 Pathological data were entered and presented to a neutral observer for statistical analysis.
Statistical Methods Descriptive statistics were performed for patient and tumor characteristics. We compared PTI with the 4 radiological signals while correcting for interclass correlation among the quadrants. PSMA expression identified in PCa quadrants was compared to that in benign quadrants using the Pearson chi-square test. The Wilcoxon rank sum test was used to compare Gleason score central tendencies in nonexpressed and expressed groups. The association between PSMA expression and tumor stage was compared using the chi-square test. Statistical analysis was performed by statisticians using SAS® 9.1. This study was internal review board approved (No. 7404-06-7R0ER).
(47–73)
1 (4) 24 (96) 5.0 (1.2–13.3) 21 4
(84) (16)
17 6 1 1
(68) (24) (4) (4)
12 8 5
(48) (32) (20)
RESULTS Table 1 lists patient demographics and tumor characteristics. Of the patients 80% had preoperative low and moderate risk features of PCa according to the D’Amico classification.12
Table 3. PTI vs RSI in quadrants RSI
No. quadrants
% Median PTI
None Low Moderate High
8 37 51 4
5 5 5 15
111
IN-CAPROMAB PENDETIDE IMMUNOSCINTIGRAPHY FOR PROSTATE CANCER
Table 6. PCa in quadrants by Gleason score
Table 4. No vs low/moderate/high RSI sensitivity and specificity by quadrant with PCa as gold standard target
No. Quadrant (%)
Quadrant
% Sensitivity
% Specificity
Lt apical Lt basal Rt apical Rt basal
87 57 89 37
0 50 17 50
Table 2 shows pathologically detected PTI in all 100 examined quadrants. Table 3 shows RSI stratified as negative, low, moderate or high. On pathological evaluation 19 of the 100 examined quadrants were cancer-free, while 35, 26 and 20 had less than 10%, 10% to less than 25% and 25% or greater tumor involvement with PTI, respectively. Table 3 shows RSI, graded negative, low, moderate or high, in the 100 examined quadrants and median PTI in the corresponding subset of quadrants. Low or moderate RSI was seen in 88 of the 100 quadrants examined. Only 4 of 100 quadrants showed negative RSI. Median PTI was identical in the negative, low and moderate RSI groups. The high RSI group, comprising only 4 of the 100 examined quadrants, showed slightly higher median PTI. Of these 4 quadrants representing 3 patients, only 1 patient with tumor in 2 quadrants had preoperative high risk features such as Gleason score greater than 8 by the D’Amico definition12 but in all 4 quadrants PTI was less than 50%. Statistically there were no significant differences in PTI among the 4 RSIs (none, low, moderate and high) when controlling for the interclass correlation among PTI measures in the 4 quadrants (p ⫽ 0.35, table 3). In any given quadrant the sensitivity and specificity of the co-registered scan for PCa was low (table 4). Sensitivity was 37% to 87% for the 4 quadrants with 0% to 50% specificity, correlating with pathologically confirmed cancer in the same quadrant in the resected gland. RSI correlated poorly with the mean total Gleason score of PCa foci (p ⫽ 0.44, table 5). Tumors were more frequently pathologically identified in the apical area, that is 23 of 25 patients (92%) had tumors in the right apex and 92% had tumors in the left apex. In contrast, the rate was 19 of 25 cases (75%) in the left basal quadrant and 17 of 25 (68%) in the right basal quadrant. Therefore, the Table 5. Total Gleason score vs RSI in areas with PCa on pathological evaluation RSI
No. PCa
Mean ⫾ SD Gleason Score
None Slight Moderate Marked p Value
7 31 40 4
7.29 ⫾ 0.756 6.90 ⫾ 0.746 7.08 ⫾ 0.997 6.50 ⫾ 0.577 0.442
941
Gleason Score
Lt Apical
Lt Basal
Rt Apical
Rt Basal
Total No. (%)
6 7 8 9
8 (34.8) 11 (47.8) 2 (8.7) 2 (8.7)
4 (21.1) 11 (57.9) 2 (10.5) 2 (10.5)
7 (30.4) 12 (52.2) 2 (8.7) 2 (8.7)
4 (23.5) 10 (58.8) 1 (5.9) 2 (11.8)
23 (27.1) 45 (52.9) 7 (8.2) 10 (11.8)
Totals
23 (100)
19 (100)
23 (100)
17 (100)
85 (100)
high tumor prevalence in the apical area accounts for the high scan sensitivity. Table 6 lists Gleason scores in the prostate quadrants with almost an equal ratio in the apical and basal quadrants. Moreover, a significant correlation was not found between PTI (greater or less than 50%) and SPECT/CT by quadrant (data not shown). IHC staining of formalin fixed, paraffin embedded prostate sections revealed PSMA staining of malignant foci in 85% of cases but also staining of benign tissue regions in 76% (p ⫽ 0.16). PSMA expression in malignant foci was more common in cases with a higher total Gleason score (p ⫽ 0.03, table 7). There was no significant association between PSMA expression strength and tumor stage (p ⫽ 0.23, data not shown). Figures 1 to 6 show the weak correlation between RSI and PSMA expression by PCa tissue. High RSI was registered on the scan in the left base but only 1% PCa was found at final pathological assessment with equal staining of micro PCa and benign tissue (figs. 1 and 2). Moderate RSI was seen in each apical quadrant but PTI in the left apex was 60% and only 15% in the right apical quadrant (figs. 3 and 4). Furthermore, PCa in the left apex was unstained while benign tissue intensively stained with 7E11C5.3 PSMA antibody. Mild RSI was documented on the scan in the 2 basal quadrants while only 1 focus of micropapillary pattern HGPIN was identified in the right base (figs. 5 and 6).
DISCUSSION Currently there is a need to develop imaging modalities capable of localizing PCa foci in the prostate. This would enable image directed biopsy for diagnoTable 7. Gleason score and PCa cell PSMA expression in areas with PCa on pathological evaluation PSMA Intensity Weak Moderate Strong Total p Value
No. PCa
Mean ⫾ SD Gleason Score
3 18 61 82
6.67 ⫾ 0.577 6.56 ⫾ 0.511 7.15 ⫾ 0.928 7.00 ⫾ 0.875 0.031
942
111
IN-CAPROMAB PENDETIDE IMMUNOSCINTIGRAPHY FOR PROSTATE CANCER
Figure 1. Patient 19. Co-registered capromab pendetide scan shows high RSI in prostate left base (arrows)
sis, image guidance for focal therapy and image based surveillance strategies after treatment. Such an advance might have far-reaching implications for our approach to PCa management. Although the 111 In-capromab pendetide scan is not FDA approved for this indication, we theorized that if this immunoscintigraphy scan was co-registered with crosssectional CT, it might delineate cancer foci in the prostate. In fact, FDA approved indications include only high risk features or cases suspicious for recurrent disease after primary definitive therapy.12 The rationale on which positron emission tomography/CT is based (combining the functional features of positron emission tomography with the anatomical detail of CT) provides many advantages that may be easily transferable to SPECT.13–15 Our efforts focused on applying fused 111In-capromab pendetide SPECT/CT to identify PCa inside the prostate by radio immunoscintigraphy. Our choice of specimen pathological preparation deserves mention. At our institution we use routine
close (every 3 mm) step sectioning of RP specimens. While there are advantages of a complete embedding technique with whole mounted processing to match with imaging, studies such as those by Hollenbeck16 and Desai17 et al do not show a significant difference in pathological outcome between the 2 techniques when using multivariate logistics to control for preoperative variables. In fact, to our knowledge the problem of ideal preoperative imaging superimposition with a whole mounted slide remains to be solved. Therefore, we think that using quadrant involvement may be a reasonable option for correlation, although this approach may also have limitations. Quadrant evaluation was also ideal to prevent bias in the results. Alterations in gland geometry by fixation and sectioning do not extend to the conflating right and left or anterior and posterior. We were confident that we could maintain registration on radiology and pathology images between right and left, and between anterior and posterior. By claim-
111
IN-CAPROMAB PENDETIDE IMMUNOSCINTIGRAPHY FOR PROSTATE CANCER
ing only quadrant registration we set the threshold for success as low as possible for the 111In-capromab pendetide method. If it could not even localize disease by quadrant, it could certainly not localize disease by sextant or octant. Therefore, to claim that we could guarantee registration over small volumes would have been more and not less stringent toward the 111In-capromab pendetide method. The diagnostic value of 111In-capromab pendetide scan to detect extraprostatic PCa is well studied. Many groups have noted that the technique is valuable when formulating a treatment plan for salvage radiation therapy after primary RP and predicting the outcome after salvage treatment.11,14,18 –21 However, other groups could not corroborate these findings.9,15,22,23 Reading and interpreting 111In-capromab pendetide scans, especially by young radiologists without sufficient experience, remain subjective but coregistration with CT may allow better anatomical correlation regarding tumor location. In addition, the high volume of 111In-capromab pendetide scans performed at centers of excellence may decrease interpretation and conclusion errors. The fused 111Incapromab/CT technology evolves constantly. As further advances are made, the technology will improve, allowing more accurate diagnoses. In particular improvement in CT image quality and SPECT image processing would potentially enable more accurate evaluation of the prostatic and regional lymph node beds. To date no accurate imaging modality is available that provides precise visualization of small, prostate confined PCa. To our knowledge we present the first study in the literature to combine 111In-capromab pendetide with CT to image PCa in the gland, as validated by final pathology assessment of RP specimens. We could not determine the efficacy of fused 111 In-capromab pendetide SPECT/CT to image localized PCa in the gland, as validated by the results of final pathology assessment of RP specimens. Analysis of IHC stained RP specimens using the same antibody to PSMA as in the 111In-capromab pendetide scan did not show any correlation between staining and PTI, tumor stage and primary or secondary Gleason score. When comparing the RSI of the co-registered scan to the pathological IHC of the RP specimen using the same 7E11-C5.3 PSMA antibody as the 111In-capromab pendetide scan, no significant correlation was found. We discerned no particular subset of patients in our group in whom the co-registered scan correlated with RP pathology when evaluating PTI. In the literature we found an assessment scale for 7E11 antibody staining. IHC staining interpretation was done according to semiquantitative assessment of IHC results for the 7E11 antibody previously re-
943
Figure 2. Patient 19. IHC of corresponding quadrant (fig. 1) reveals tiny PCa focus (arrows) stained with weak-moderate intensity and surrounding benign glandular tissue with weak, moderate and strong staining. PSMA antibody (7E11-C5.3) staining, reduced from ⫻4.
ported by Ross et al.11 The advantage of this interpretation is that it incorporates the staining distribution in focal, regional or diffuse fashion along with the intensity of cytoplasmic staining as weak, moderate or intense, respectively. All prostatic conditions were examined as cancer vs no cancer or benign lesions, including inflammatory foci, atrophy etc, according to the aim of the trial. In regard to the literature we are unaware of published findings showing the usefulness of 111indiumcapromab imaging for localized PCa in the prostate gland, as validated by IHC results using 7E11C5.3 antibody on closed step-sectioning of RP specimens. A few studies describe IHC staining differences between benign and malignant prostatic glands.11,24 –26 Our findings differ from those in these previous series. Our IHC results reveal significant PSMA immunoreactivity in benign prostatic epithelium as well as heterogeneous PSMA immunoreactivity in a significant proportion of prostatic adenocarcinoma foci, similar to results in the recent study by Mannweiler et al.27 Our findings support the conclusion that IHC using standard techniques with 7E11-C5.3 antibody in formalin fixed, paraffin embedded sections has poor specificity for PCa since the antibody revealed significant staining of normal prostatic epithelium. This finding is not surprising since benign and malignant prostate epithelial cells express PSMA.11,28,29 Also, in RP specimens we found no association between the intensity of 7E11-C5.3 PSMA antibody staining with variables such as primary and secondary Gleason grade, tumor stage and PTI. The reason that 111In-capromab pendetide immunoscintigraphy
944
111
IN-CAPROMAB PENDETIDE IMMUNOSCINTIGRAPHY FOR PROSTATE CANCER
Figure 3. Patient 25. Co-registered capromab pendetide scan demonstrates moderate RSI in prostate right and left apical quadrants (arrows).
Figure 4. Patient 25. IHC (fig 3). A, prostate left apex shows strong staining of benign tissue (thick arrows) and weak staining of PCa (thin arrows). B, moderate-strong staining of right apex benign tissue with weak staining of atrophic area (arrows). PSMA antibody (7E11-C5.3) staining, reduced from ⫻4.
111
IN-CAPROMAB PENDETIDE IMMUNOSCINTIGRAPHY FOR PROSTATE CANCER
945
Figure 5. Patient 22. Co-registered capromab pendetide scan demonstrates mild RSI in 2 prostate basal quadrants (arrows)
is useful for extraprostatic disease, eg lymph node metastasis, is because only malignant elements metastasize or invade beyond the capsule and, thus, there are no benign prostate elements to distinguish. There are several limitations of this study. 1) The size of the prospective cohort of 25 patients was restricted. However, statistically significant findings were identified as reported despite the sample size. 2) The resolution of these imaging techniques continues to improve and SPECT/CT technology continues to evolve. It is possible that with a better monoclonal antibody or improved imaging PCa foci may be imaged in the prostate using a similar technique. However, given these limitations, and based on the commercially available antibody and available imaging equipment at our centers between 2005 and 2007 when this study was performed, coregistered 111In-capromab pendetide SPECT/CT could not reliably image intraprostatic PCa foci due to low specificity of the 7E11-C5.3 antibody for PCa.
3) The fundamental biological principle of how an antibody can be delivered into a cancer cell to interact with an intracellular epitope deserves mention. Following the basic principle of receptor mediated signaling and/or endocytosis, the Ig molecule, like any other large hydrophilic protein, does not cross the lipophilic plasma membrane without specifically binding to an extracellular receptor. However, this commercially available 7E11-C5.3 antibody has the capromab epitope intracellularly and, therefore, it is unclear how this specific antibody binds to its target mechanistically. 4) Recent data suggest that PSMA over expression may be more common in high vs low grade PCa, which may make it a nonspecific tool, especially for early stage, localized PCa.26,30 Furthermore, changes in PSMA expression were noted in other benign and malignant tumors, such as schwannoma and endometrial adenocarcinoma, respectively.31,32 Therefore, these cases should be clinically considered in the differential diagnosis of a lesion that is positive on PSMA radio immunoscin-
946
111
IN-CAPROMAB PENDETIDE IMMUNOSCINTIGRAPHY FOR PROSTATE CANCER
Figure 6. Patient 22. IHC staining of prostate right base (fig. 5). A, highly specific marker reveals HGPIN and PCa. P504S/racemase, reduced from ⫻4. B, strongly stained benign and HGPIN areas. PSMA antibody (7E11-C5.3), reduced from ⫻4.
tigraphy. Ultimately we understand that the scan may be more beneficial in patients with high risk features, in whom PSMA expression could be greater. 5) Finally, the exact clinical implication of 111 In-capromab pendetide scan results remains concerning when defining the spatial tumor distribution inside the prostate, especially for early stage, localized PCa. The recent emergence of high resolution imaging tools coupled with advances in computerized modeling software should be used in the near future to provide alternative treatment options in men with localized, early stage cancer. This approach is an important step in our quest for better ways to diagnose and treat the disease while maintaining good quality of life in our patients.
CONCLUSIONS Diagnostic 111In-capromab pendetide scan with CT co-registration imaging of the prostate gland did not correlate with the pathologically assessed presence of prostate confined PCa in RP specimens. IHC staining with the same antibody (7E11C5.3) as the
111
In-capromab pendetide scan showed staining of benign and malignant prostatic epithelium, suggesting that lack of specificity for malignant prostatic glands may be a cause of the poor correlation. Further investigation using fused 111In-capromab pendetide scan with CT or magnetic resonance imaging is needed to define the possible role of this scan in treatment diagnostics.
ACKNOWLEDGMENTS Dr. Bercedis Peterson, Department of Biostatistics and Bioinformatics provided statistical analysis. Steven W. Shipes, Chief Technologist, Department of Radiology/Nuclear Medicine, Duke University Medical Center collected and analyzed imaging data. Duke is a designated Partners in Excellence center of Cytogen. Each Partners in Excellence Site receives rigorous training, undergoes proficiency testing and is subject to ongoing quality assurance protocols. To qualify as a Partners in Excellence Site each center must be certified as proficient in the interpretation of ProstaScint scans by the American College of Nuclear Physicians.
REFERENCES 1. Sodee DB, Ellis RJ, Samuels MA et al: Prostate cancer and prostate bed SPECT imaging with ProstaScint: semiquantitative correlation with prostatic biopsy results. Prostate 1998; 37: 140. 2. Raj GV, Partin AW and Polascik TJ: Clinical utility of indium 111-capromab pendetide immunoscintigraphy in the detection of early, recurrent prostate carcinoma after radical prostatectomy. Cancer 2002; 94: 987. 3. Polascik TJ, Manyak MJ, Haseman MK et al: Comparison of clinical staging algorithms and 111indium-capromab pendetide immunoscintigra-
phy in the prediction of lymph node involvement in high risk prostate carcinoma patients. Cancer 1999; 85: 1586. 4. Manyak MJ, Hinkle GH, Olsen JO et al: Immunoscintigraphy with indium-111-capromab pendetide: evaluation before definitive therapy in patients with prostate cancer. Urology 1999; 54: 1058. 5. Ellis RJ, Kim EY, Conant R et al: Radioimmunoguided imaging of prostate cancer foci with histopathological correlation. Int J Radiat Oncol Biol Phys 2001; 49: 1281.
6. Polascik TJ, Mayes JM, Sun L et al: Pathologic stage T2a and T2b prostate cancer in the recent prostate-specific antigen era: implications for unilateral ablative therapy. Prostate 2008; 68: 1380. 7. Polascik TJ and Mouraviev V: Focal therapy for prostate cancer. Curr Opin Urol 2008; 18: 269. 8. Ahmed HU and Emberton M: Active surveillance and radical therapy in prostate cancer: can focal therapy offer the middle way? World J Urol 2008; 26: 457. 9. Koontz BF, Mouraviev V, Johnson JL et al: Use of local (111)in-capromab pendetide scan results to
111
IN-CAPROMAB PENDETIDE IMMUNOSCINTIGRAPHY FOR PROSTATE CANCER
predict outcome after salvage radiotherapy for prostate cancer. Int J Radiat Oncol Biol Phys 2008; 71: 358. 10. Wong TZ, Turkington TG, Polascik TJ et al: ProstaScint (capromab pendetide) imaging using hybrid gamma camera-CT technology. AJR Am J Roentgenol 2005; 184: 676. 11. Ross JS, Sheehan CE, Fisher HA et al: Correlation of primary tumor prostate-specific membrane antigen expression with disease recurrence in prostate cancer. Clin Cancer Res 2003; 9: 6357. 12. D’Amico AV, Moul J, Carroll PR et al: Cancerspecific mortality after surgery or radiation for patients with clinically localized prostate cancer managed during the prostate-specific antigen era. J Clin Oncol 2003; 21: 2163. 13. Ellis RJ and Kaminsky DA: Fused radioimmunoscintigraphy for treatment planning. Rev Urol, suppl., 2006; 1: S11. 14. Sodee DB, Sodee AE and Bakale G: Synergistic value of single-photon emission computed tomography/computed tomography fusion to radioimmunoscintigraphic imaging of prostate cancer. Semin Nucl Med 2007; 37: 17. 15. Nagda SN, Mohideen N, Lo SS et al: Long-term follow-up of 111In-capromab pendetide (ProstaScint) scan as pretreatment assessment in patients who undergo salvage radiotherapy for rising prostate-specific antigen after radical prostatectomy for prostate cancer. Int J Radiat Oncol Biol Phys 2007; 67: 834. 16. Hollenbeck BK, Bassily N, Wei JT et al: Whole mounted radical prostatectomy specimens do not increase detection of adverse pathological features. J Urol 2000; 164: 1583.
17. Desai A, Wu H, Sun L et al: Complete embedding and close step-sectioning of radical prostatectomy specimens both increase detection of extra-prostatic extension, and correlate with increased disease-free survival by stage of prostate cancer patients. Prostate Cancer Prostatic Dis 2002; 5: 212. 18. Ellis RJ, Zhou EH, Fu P et al: Single photon emission computerized tomography with capromab pendetide plus computerized tomography image set co-registration independently predicts biochemical failure. J Urol 2008; 179: 1768.
947
nign and malignant prostate tissues. Urol Oncol 1995; 1: 18. 25. Silver DA, Pellicer I, Fair WR et al: Prostatespecific membrane antigen expression in normal and malignant human tissues. Clin Cancer Res 1997; 3: 81. 26. Perner S, Hofer MD, Kim R et al: Prostate-specific membrane antigen expression as a predictor of prostate cancer progression. Hum Pathol 2007; 38: 696.
19. Keane TE, Rosner IL, Wingo MS et al: The emergence of radioimmunoscintigraphy for prostate cancer. Rev Urol, suppl., 2006; 1: S20.
27. Mannweiler S, Amersdorfer P, Trajanoski S et al: Heterogeneity of prostate-specific membrane antigen (PSMA) expression in prostate carcinoma with distant metastasis. Pathol Oncol Res 2008; 15: 167.
20. Sodee DB, Malguria N, Faulhaber P et al: Multicenter ProstaScint imaging findings in 2154 patients with prostate cancer. The ProstaScint Imaging Centers. Urology 2000; 56: 988.
28. Holmes EH: PSMA specific antibodies and their diagnostic and therapeutic use. Expert Opin Investig Drugs 2001; 10: 511.
21. Proano JM, Sodee DB, Resnick MI et al: The impact of a negative (111)indium-capromab pendetide scan before salvage radiotherapy. J Urol 2006; 175: 1668.
29. Sweat SD, Pacelli A, Murphy GP et al: Prostatespecific membrane antigen expression is greatest in prostate adenocarcinoma and lymph node metastases. Urology 1998; 52: 637.
22. Thomas CT, Bradshaw PT, Pollock BH et al: Indium-111-capromab pendetide radioimmunoscintigraphy and prognosis for durable biochemical response to salvage radiation therapy in men after failed prostatectomy. J Clin Oncol 2003; 21: 1715.
30. Bostwick DG, Pacelli A, Blute M et al: Prostate specific membrane antigen expression in prostatic intraepithelial neoplasia and adenocarcinoma: a study of 184 cases. Cancer 1998; 82: 2256.
23. Kahn D, Williams RD, Manyak MJ et al: 111Indiumcapromab pendetide in the evaluation of patients with residual or recurrent prostate cancer after radical prostatectomy. The ProstaScint Study Group. J Urol 1998; 159: 2041. 24. Wright GL, Haley C, Beckett ML et al: Expression of prostate-specific membrane antigen in normal, be-
31. Wang W, Tavora F, Sharma R et al: PSMA expression in Schwannoma: a potential clinical mimicker of metastatic prostate carcinoma. Urol Oncol, 2008. 32. Mhawech-Fauceglia P, Smiraglia DJ, Bshara W et al: Prostate-specific membrane antigen expression is a potential prognostic marker in endometrial adenocarcinoma. Cancer Epidemiol Biomarkers Prev 2008; 17: 571.
EDITORIAL COMMENTS I concur with these authors that the ability to identify cancer in the prostate gland would have far-reaching implications. A key question is whether one may reasonably expect to specifically image PCa using an antibody to PSMA in a gland that also contains PSMA positive normal glands. Surprisingly perhaps the answer is yes. The reason is that beyond the requisite expression of antigen several other factors determine antibody localization and, therefore, imaging or therapy. One of these factors relates to the accessibility of the antigen to administered antibody. In normal prostate glands PSMA is expressed on the luminal surface of the secretory epithelium. This site is not exposed to circulating antibody due to the intervening basal cell layer as well as to epithelial tight junctions. Conversely in cases of invasive PCa PSMA loses its polarized luminal expression, and the basal cell layer and tight junction barriers are
gone, exposing PSMA to circulating antibody. How significant is this barrier to antibody penetration in normal glands? These barriers account for the 1,000,000-fold concentration gradient of PSA between normal prostatic ducts and plasma. By extension normal glandular PSMA would be exposed to a 1,000,000-fold lower concentration of antiPSMA antibody as would invasive PCa. Given what appears to be a feasible setting for PSMA antibody to specifically target PCa, why did capromab not succeed? Because capromab has an additional barrier that is often cited but rarely heeded. Capromab binds to the intracellular region of the PSMA molecule, which is hidden from antibody view by the intact cell membrane.1 As a result, capromab cannot see any PSMA regardless of whether it is expressed by a normal cell or a cancer cell.2,3 Only when the cell membrane is disrupted, as in dying cells, can capromab bind its
948
111
IN-CAPROMAB PENDETIDE IMMUNOSCINTIGRAPHY FOR PROSTATE CANCER
target.2,4 However, PSMA antibodies are available that bind the PSMA external domain, avoiding this capromab related barrier. One such antibody (J591) is currently in clinical trials and has shown substantially improved imaging of metastatic PCa relative to capromab,5,6 although not in a head-tohead comparison. This suggests that the potential exists to ultimately identify PCa in the prostate
and achieve the far-reaching implications alluded to by these authors. Neil H. Bander Department of Urology Weill/Cornell Medical College Memorial Sloan-Kettering Cancer Center New York, New York
REFERENCES 1. Troyer JK, Beckett ML and Wright GL: Location of prostate-specific membrane antigen in the LNCaP prostate carcinoma cell line. Prostate 1997; 30: 232.
monoclonal antibodies specific for the extracellular domain of prostate-specific membrane antigen. Cancer Res 2000; 60: 5237.
2. Liu H, Moy P, Kim S et al: Monoclonal antibodies to the extracellular domain of prostate-specific membrane antigen also react with tumor vascular endothelium. Cancer Res 1997; 57: 3629.
4. Smith-Jones PM, Vallabhajosula S, Navarro V et al: Radiolabeled monoclonal antibodies specific to the extracellular domain of prostate-specific membrane antigen: preclinical studies in nude mice bearing LNCaP human prostate tumor. J Nucl Med 2003; 44: 610.
3. Smith-Jones PM, Vallabahajosula S, Goldsmith SJ et al: In vitro characterization of radiolabeled
These authors studied a small, nonconsecutive, lower risk cohort with absent followup, resulting in findings that raise more questions than answers. In contrast, we reported long-term followup in a clinical series of 239 patients demonstrating the abililty of capromab pendetide to predict biochemical failure in those at intermediate risk and improve biochemical disease-free survival with dose escalation to SPECT/CT defined target volumes (reference 18 in article).1 In our protocol radiology sets ␥ contrast using positive findings on diagnostic biopsy histopathology. The parameters of the current study (selecting lower risk patients and blinding radiology to diagnostic pathology) effectively biased study results. This potential systematic error can be observed in the figures, in which bone marrow inten-
5. Milowsky MI, Nanus DM, Kostaglou L et al: Phase I trial of 90Y-labeled anti-PSMA monoclonal antibody J591 for androgen-independent prostate cancer. J Clin Oncol 2004; 22: 2522. 6. Bander NH, Milowsky MI, Nanus DM et al: Phase I trial of 177lutetium-labeled J591, a monoclonal antibody to prostate-specific membrane antigen, in patients with androgen-independent prostate cancer. J Clin Oncol 2005; 23: 4591.
sity appears to directly correlate with examples of high to mild tracer uptake in tumor. An improved study design for histopathology confirmation of intraprostatic imaging should incorporate in vivo fiducial marker correlation with diagnostic biopsy pathology and hybrid molecular image sets (reference 13 in article). Rodney J. Ellis Radiation Oncology Aultman Hospital Canton, Ohio Department of Radiology Northeastern Ohio Universities College of Medicine Rootstown, Ohio Department of Urology Case Western Reserve University School of Medicine Cleveland, Ohio
REFERENCE 1. Ellis RJ, Zhou H, Kim EY et al: Biochemical disease-free survival rates following definitive low-dose-rate prostate brachytherapy with dose escalation to biologic target volumes identified with SPECT/CT capromab pendetide. J Brachyther 2007; 6: 16.