| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Molecular Oncology, Markers, Clinical Correlates |
1 Department of Urology and 2 Department of Radiooncology, Offenbach, Germany; 3 Department of Anatomy and Cell Biology, Philipps-University, Marburg, Germany; and 4 Human Biology Division, Fred Hutchinson Cancer Research Center, Seattle, Washington
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Experimental Design: Peripheral blood from 358 patients was obtained before radical prostatectomy. Corresponding obturatory lymph node samples were collected from 153 of these patients. Nested RT-PCR for prostate-specific antigen (PSA), human kallikrein 2 (hK2), and prostate-specific membrane antigen (PSMA) were performed on cDNA from peripheral blood. The lymph node cDNA was analyzed for PSA und hK2 expression.
Results: RT-PCR in peripheral blood was positive in 124 (34.6%) of 358 samples for PSA, 215 (60.1%) of 358 for PSMA, and 97 (27.1%) of 358 for hK2. Comparison of positive RT-PCR rates of pT2 and pT3 tumors in corresponding peripheral blood for PSA, PSMA, and hK2 were 31.9 and 40.0%, 58.8 and 62.5%, and 26.9 and 27.5%, respectively. Histopathologically, cancer-free lymph node samples were positive in RT-PCR for PSA and hK2 in 70 (49.6%) of 141 and 89 (63.2%) of 141 of cases. All histologically positive lymph node samples (n = 12, pN+) were positive for PSA RT-PCR. PSA RT-PCR alone, as well as combined PSA/PSMA RT-PCR evaluation, in peripheral blood showed a significant association with grading. PSA RT-PCR lymph node-negative samples were significantly less likely positive in their corresponding peripheral blood RT-PCR sample.
Conclusions Although the preoperative PSA RT-PCR in peripheral blood correlated with the grading of prostate cancer, no combination of RT-PCR results using "triple" markers (PSA, hK2, PSMA) in peripheral blood and/or lymph nodes yielded additional preoperative staging information.
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Despite extensive research, it remains difficult to discriminate preoperatively between patients with locally confined prostate cancer and patients with extraprostatic disease. This is demonstrated by the fact that more than one third of men with clinically localized prostate cancer have extraprostatic disease at the time of surgery (2) . Extensive screening using serum prostate-specific antigen (PSA) has led to earlier detection of the disease, and most tumors are now recognized at an organ-confined stage (3) . To date, a combination of digital rectal examination, PSA serum levels, and needle biopsy are the basis of preoperative staging and can help to decrease the incidence of pathologic upstaging (4) . However, significant inaccuracies in preoperative staging remain. It has been known that hematogeneous dissemination of tumor cells occurs in cancer as well as in in vivo models (5) . Efforts to develop molecular approaches to detect disseminating tumor cells of solid tumors in peripheral blood or bone marrow, and to correlate the detection with respect to diagnosis, prognostic value, and clinical monitoring, are under investigation (6) .
Two groups initially reported the use of reverse transcription polymerase chain reaction (RT-PCR) to study prostate cancer micrometastasis; tissue sources for PCR analysis included peripheral blood, bone marrow, and pelvic lymph nodes (7 , 8) . RT-PCR of histologically cancer-free pelvic lymph nodes from patients undergoing radical lymphadenectomy seems to be particularly promising. Potential molecular markers are PSA, prostate-specific membrane antigen (PSMA), and human kallikrein-2 (hK2). Unfortunately, the published studies have yielded conflicting and controversial results (9, 10, 11) .
A major argument for the use of RT-PCR is the fact that clinically and pathologically localized prostate cancer has a significant chance of biochemical failure after radical retropubic prostatectomy, even for low-grade cancer with all margins and lymph nodes histopathologically negative. RT-PCR might detect truly clinically occult metastases and may account for disease recurrence of patients initially diagnosed with localized cancer. A recent study failed to correlate preoperative RT-PCR for PSA and PSMA results with biochemical failure or pathologic stage in 141 patients with localized prostate cancer (12) .
In our study, we examined the expression of all three established molecular markers (PSA, PSMA, and hK2) in the peripheral blood of 358 consecutive patients for preoperative staging and correlated the results with clinical data. In addition, RT-PCR for PSA and hK2 was performed in corresponding 153 lymph node samples. As the hypogastric obturatory lymph nodes are believed to be the first lymph nodes involved in metastatic prostate cancer, we examined this lymph node station regarding micrometastatic spread (13) . We did not include PSMA RT-PCR in our lymph node analysis because it is not exclusively expressed in the prostate (14) . Specifically, we will address whether all three markers are superior versus a single marker, whether lymph node RT-PCR results correlate with clinical variables more accurately than do peripheral blood RT-PCR data, and whether there is any association between peripheral blood and lymph node results.
|
PATIENTS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In a subgroup of 153 patients, lymph node samples of the obturatory lymph nodes were obtained during radical lymphadenectomy before radical retropubic prostatectomy and were cut in half. One part was snap-frozen in liquid nitrogen, and the other was used for histopathologic evaluation. Prostatic tumor tissue from 20 patients was collected during surgery for sensitivity experiments and were used as control.
Pathologic stages were determined following the TNM (tumor-node-metastasis) classification system. Grading was performed according to WHO guidelines. Roughly, grade 1 tumors relate to Gleason score 2–4, grade 2 tumors to Gleason 5–6/7 and grade 3 tumors to Gleason 8–10. Detailed pathologic data of all of the patients included in this study are shown in Table 1
.
|
Isolation of Mononuclear Cells from Peripheral Blood and RNA Extraction.
The nucleated cell fraction was isolated from 5 ml of whole blood in EDTA with a Percoll gradient (Pharmacia, Freiburg, Germany) centrifugation according to the manufacturer’s description. Total RNA was isolated with the RNeasy kit (Qiagen, Hilden, Germany) according to the manufacturer’s description, immediately snap-frozen, and stored at –70°C until further processing. Total RNA from snap-frozen lymph node samples was extracted with TRIzol reagent (Invitrogen, Carlsbad, CA).
Reverse Transcription.
Total RNA (1.5 µg) and 0.5 µg of oligo d(T) (Promega, Heidelberg, Germany) in a total volume of 11 µL were incubated for 10 min at 70°C and were immediately cooled on ice. To each RNA sample, 4 µL of reaction buffer (Promega) consisting of 250 mmol/L Tris-HCl (pH 7.5), 358 mmol/L KCl, 15 mmol/L MgCl2, 50 mmol/L dithiothreonine (DTT), 2 µL of distilled water, 1 µL of dNTP (10 mmol/L each) (Boehringer Mannheim, Mannheim, Germany), 1 µL (10 units) of RNase inhibitor (Boehringer Mannheim), and 1 µL of M-MLV reverse transcriptase (Promega) were added, shortly centrifuged and subsequently incubated at 37°C for 90 minutes. Final cDNA was stored at –20°C or was immediately used for amplification by PCR.
PCR and Nested PCR.
All of the primers used in this study were custom designed by MWG-Biotech (Ebersberg, Germany). Intron spanning primer pairs specific for human PSA (NM_001648) were sense, 5'-GCG TGA TCT TGC TGG GTC GG-3', and antisense, 5'-CCT TCT GAG GGT GAA CTT GCG-3'. The antisense primer was replaced by 5'-AGA ATC ACC CGA GCA GGT GCT-3' for nested PCR. Intron spanning primer pairs specific for human PSMA (NM_004476) were for sense, 5'-TCA CCG GGA CTC ATG GGT GT-3', and antisense, 5'-GCC TGA AGC AAT TCC AAG TCG-3'; sense (nested PCR), 5'-AAG GAA GGG TGG AGA CCT AG-3', and antisense (nested PCR), 5'-ACT GAA CTC TGG GGA AGG AC-3'). Intron spanning primer pairs specific for hK2 [Kawakami et al. (21)
, NM_005551], were sense, 5'-GGT GGC TGT GTA CAG TCA TGG AT-3', and antisense, 5'-ACA CAC TGA AGA CTC CTG GGG CG-3'; antisense primer was replaced for nested PCR by 5'-TGT CTT CAG GCT CAA ACA GGT TG-3'). Primers for amplification of the housekeeping gene glyceraldehyde-3-phosphate-dehydrogenase (GAPDH; NM_002046) were sense, 5'-CGT CTT CAC CAC CAT GGA GA-3', and antisense, 5'-CGG CCA TCA CGC CAC AGT TT-3'); and they served as internal quality control to assess RNA integrity. PCR reactions were performed in a total volume of 50 µL containing 1 µL of reverse transcription product, 5 µL of 10x PCR-reaction buffer [200 mmol/L Tris-HCl (pH 8.4) and 500 mmol/L KCl], 0.1 µmol/L sense, and 0.1 µmol/L antisense primer, 0.2 mmol/L of each dNTP, 1 unit Taq-DNA polymerase (PAN Biotech, Aidenbach, Germany) using a tube-controlled thermal cycler (AGS-Hybaid, Heidelberg, Germany). Amplification for PSA and hK2 was performed as follows: 1 minute at 98°C, followed by two cycles at 98°C for 30 seconds, 55°C for 1 minute, and 72°C for 1 minute. Subsequently, there were 28 cycles at 94°C for 30 seconds, 55°C for 1 minute, and 72°C for 1 minute, and a final extension at 72°C for 5 minutes. One µL of PCR product was further amplified for an additional 20 cycles, with the application of the same conditions but replacing antisense primer by nested antisense primer.
For PSMA and GAPDH, the following PCR profile was used: 4 minutes at 95°C, 94°C for 30 seconds, 60°C for 30 seconds, 72°C for 1 minute for 30 cycles. Thereafter, 1 µL of a 1:100 dilution in H2O of the PSMA-RT-PCR product was used as a template in the nested RT-PCR reaction for an additional 15 cycles.
RT-PCR Assay Sensitivity.
RT-PCR assay sensitivity was determined by dilution of LNCaP cells with peripheral blood mononuclear cells of a negative control patient. Peripheral blood mononuclear cells (PBMCs) were isolated as described above and were counted. LNCaP cells were diluted as follows: in PBMCs at 0:107; at 1:107; at 10:107; at 100:107; and at 1000:107. RNA isolation and RT-PCR were performed as described above.
Analysis of PCR Products.
Ten µL of PCR product were mixed with 2 µL of 6x loading dye (MBI Fermentas, Heidelberg, Germany) and were assayed for 60 minutes on a 1.5% agarose gel in TBE buffer [0.1 mol/L Tris (pH 8.4), 90 mmol/L boric acid, and 1 mmol/L EDTA). The gel was stained with ethidium bromide, and the amplicons were visualized with a UV-transilluminator. Samples were subjected to Southern blot hybridization with digoxigenin-labeled probes to confirm product identity. Representative amplicons were commercially sequenced (SeqLab, Göttingen, Germany).
Statistical Analysis.
Kolmogorov–Smirnoff test was applied to test for normal distribution.5
P values were calculated by Mann–Whitney test, Student’s t test, or 
2 test.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In another control group of 10 patients with proven distant metastatic prostate cancer, 8 (80%), 7 (70%), and 9 (90%) were positive for PSA, hK2, and PSMA expression by RT-PCR in peripheral blood, respectively. Specificity testing of the markers used in the control group consisting of 45 healthy male volunteers showed false-positive results in peripheral blood RT-PCR for PSA in 3 (6.7%) of 45, for hK2 in 0 (0%) of 45, and for PSMA in 2 (4.5%) of 45.
RT-PCR Results in Peripheral Blood Samples.
An overview of the general positivity rates for PSA, PSMA, and hK2 is given in Table 2
. PSA-, hK2-, or PSMA-RT-PCR, as well as all possible permutations including the combination of all three markers, were not associated with pathologic T stage or N stage. For example, 76 (31.9%) of 238 patients with organ-confined T2 tumors and 48 (40%) of 120 patients with extraprostastic T3 or T4 tumors were PSA-RT-PCR positive (P = 0.13, 2 test), whereas 114 (34.3%) of 331 pN0 staged patients were PSA positive compared with 10 positive patients among a total of 27 pN+ patients (37.0%) in the cohort. For detailed results of all RT-PCR marker combinations in association with final T stage and N stage, see Table 3
.
|
|
However, when combining PSA and PSMA-RT-PCR (only double-positive patients in peripheral blood RT-PCR are counted as valid cases in statistical analysis) a statistically significant difference appears: 11 (17.5%) of 63 grade 1 tumors, 36 (17.2%) of 209 grade 2 tumors, and 24 (27.9%) of 86 of grade 3 tumors are double positive. The comparison of grade 1/grade 2 versus grade 3 results in a P value of 0.03 in 2 analysis.
The combination of PSA and hK2 did not show any statistical difference in the groups, whereas a triple marker analysis resulted in a P value of 0.053 in the 2 test [for grade 1, 3 (4.8%) of 63; for grade 2, 12 (5.7%) of 209; for grade 3, 10 (11,6%) of 86]. Table 4
shows the comprehensive result.
|
.
|
In this subgroup, the PSA-, hK2-, and PSA + hK2 RT-PCR-positivity rates of the patients with pT2 tumors (n = 100) versus the patients with pT3 or higher (n = 53) were 56 (56%) of 100 versus 26 (49%) of 53 for PSA, 68 (68%) of 100 versus 25 (47%) of 53 for hK2, and 48 (48%) of 100 versus 18 (34%) of 53 for combined PSA + hK2, respectively.
Association of Lymph Node and Peripheral Blood RT-PCR.
We were especially interested in a possible association between lymph node and blood RT-PCR results in the group of patients with pN0 disease. We, therefore, analyzed all of the possible marker combinations. The detailed results of this analysis, which leads to multiple permutations, are given in Table 6
. Interestingly, PSA-RT-PCR lymph node-positive patients were significantly more often positive in the blood PSA RT-PCR than were lymph node-negative patients (43.8 versus 27.9%; P = 0.05 in 2 analysis). A combined PSMA/PSA analysis yielded similar results: 28 (38.4%) of 73 lymph-node PSA-positive patients were positive for PSA and PSMA in their corresponding peripheral blood RT-PCR analysis, compared with 11 (16.2%) of 68 (P = 0.003, 2 analysis).
|
Preoperative serum PSA values were significantly different in patients with pT1–2 versus pT3–4 tumors [8.5 ng/mL versus 13.4 ng/mL, respectively (P < 0.001, Mann–Whitney U test)], pN0 versus pN1 tumors (9.0 ng/mL versus 16.4 ng/mL, respectively (P = 0.004, Mann–Whitney U test), and in low-grade (grade 1) versus high grade (grade 3) tumors (8.0 ng/mL versus 13.4 ng/mL, respectively; P = 0.001, Mann-Whitney U test). Preoperative prostate volume did not show any association with the above-mentioned staging and grading data.
Furthermore, no significant differences in pre- or postoperative serum PSA values, as well as in preoperative prostate volume, were seen in association with RT-PCR results in all possible permutations (data not shown).
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The use of RT-PCR for clinical monitoring of prostate cancer patients and detection of disseminating prostate cancer has been under investigation for almost 10 years, but the published results are still controversial. Different studies have shown that RT-PCR is able to detect circulating PSA-, hK2-, and PSMA-expressing cells, not only in patients with systemic disease but also in patients with clinically localized disease (19, 20, 21, 22) .
Few studies have shown a correlation between RT-PCR and pathologic stage or prognostic value, specifically survival duration (23, 24, 25, 26)
. In reports critically summarizing the literature on the use of RT-PCR to preoperatively predict extra prostatic disease and PSA recurrence after radical retropubic prostatectomy, the authors argue that additional studies regarding the clinical utility of RT-PCR in prostate cancer patients are needed (9
, 27) . Quantitative approaches to detect micrometastatic circulating cells, including expression levels, are thought to improve on the results (observed as simply positive or negative) obtained with standard RT-PCR assays (28)
. Despite a trend toward a higher PSA-RT-PCR positivity in peripheral blood in pT3 cancer patients (40.0 versus 31.9%; P = 0.13, 2 analysis), in concordance with the vast majority of studies published thus far, we were not able to show that RT-PCR analysis by triple markers or any possible permutations could stratify patients with localized cancers to patients with extraprostatic disease. The fact that PSA-RT-PCR is significantly more often positive in high-grade tumors (41%) than in the group of low- (32%) and intermediate-grade cancers (33%) as well as the increase of statistical significance when analyzing a combination of PSA and PSMA-RT-PCR results (17% double positivity in low and intermediate grade versus 28% in high-grade cancers) is promising. Interestingly, a study also found an improvement of RT-PCR significance when combining PSA and PSMA RT-PCR results of peripheral blood and extracapsular penetration (24)
.
Encouraging results using RT-PCR to detect micrometastasis in histologically cancer-free lymph nodes were published (29 , 30) . Deguchi et al. (29) pioneered the use of PSA RT-PCR on lymph nodes. The number of analyzed samples was small (22 patients), but the authors were able to show, as we do in the present analysis, that all nodes with histopathologically confirmed metastasis were positive in PSA RT-PCR (28) . Another study reported that 23 (79%) of 29 patients, pathologically free of micrometastatic cells, were positive by RT-PCR (30) . An earlier report addressed the same question and found, in a retrospective analysis, 16 (44%) of 57 of pathologically N0 staged patients with positive RT-PCR (31) . Our results fall in-between these last studies with 70 (49.6%) of 141 for PSA RT-PCR and 89 (63.2%) of 141 hK2 RT-PCR-positive pN0 lymph node samples.
A major criticism of all studies examining the use of RT-PCR in detection of occult metastases is that the expression of the message could result from basal illegitimate transcription of nonprostatic cells. An indicator that the high number of RT-PCR-positive lymph nodes in pN0 specimens does not result from illegitimate transcription is the fact that we found a correlation of lymph node and peripheral blood RT-PCR for PSA and combined PSA/PSMA RT-PCR. Only 11 (16.2%) of 68 of PSA RT-PCR-negative pN0 patients were double positive for PSA and PSMA in the corresponding peripheral blood sample, compared with 28 (38.4%) of 73 of lymph node-negative PSA RT-PCR patients.
This study is the first to analyze lymph node samples from a large defined patient cohort by RT-PCR for hK2 and PSA, associate these findings with clinical variables, and combine the results of three markers (PSA, PSMA, and hK2) in peripheral blood. Disappointingly, hK2 does not add any value to RT-PCR-based prostate cancer evaluation. As with all other results presented in this study, further follow-up of hK2 lymph node RT-PCR-positive patients has to be awaited. Additionally, the triple-marker evaluation in peripheral blood does not seem to be of value and revealed basically the same observation as the PSA/PSMA RT-PCR study presented recently (12) .
One could argue that the results presented in this study may have been compromised by the fact that patients received neoadjuvant antiandrogen therapy. It was found that patients who had pT2 and pT3 tumors and who were treated with neoadjuvant hormones were less likely to have circulating PSMA-positive cells in their peripheral blood, compared with those in the nontreated control group, before and after surgery (32) . In concordance with this finding, it was reported that the hormone-treated group showed a significantly lower incidence of preoperative PSA mRNA positivity than did the group undergoing surgery alone (20 versus 69%, P = 0.036; ref. 33 ). However, a recent study that compared hormone-treated versus untreated patients did not find any difference in RT-PCR positivity in peripheral blood or in bone marrow (34) . Our overall positivity rates for PSA, hK2, and PSMA do not differ much from results of other studies, in which patients have not been treated hormonally. In addition, a recent publication did not see significant changes in the Kaplan–Maier analysis for biochemical relapse-free survival excluding patients with preoperative androgen-deprivation therapy (12) . In summary, to date, studies that examine the influence of neoadjuvant hormonal therapy on RT-PCR testing are conflicting and not conclusive. Even if we presume a significant effect of hormonal ablation on RT-PCR studies, our study population represented a homogeneous group because all of the patients received hormonal ablation, and no systemic error in the analysis took place.
|
FOOTNOTES |
|---|
Requests for reprints: Ralf Kurek, Klinikum Offenbach, Department of Radiation Oncology, Starkenburgring 66, 63069 Offenbach, Germany. Phone: 49-0-69-8405-3335; Fax: 49-0-69-8405-3334; E-mail: rkurek{at}gmx.de
5 SPSS software version 7.5.2 from SPSS Inc., Chicago, IL, 1999. 
Received 7/ 2/03; revised 5/10/04; accepted 6/ 1/04.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  H. Schmidt, G. DeAngelis, E. Eltze, I. Gockel, A. Semjonow, and B. Brandt Asynchronous Growth of Prostate Cancer Is Reflected by Circulating Tumor Cells Delivered from Distinct, Even Small Foci, Harboring Loss of Heterozygosity of the PTEN Gene. Cancer Res., September 15, 2006; 66(18): 8959 - 8965. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. C. Ferrari, N. N. Stone, R. Kurek, E. Mulligan, R. McGregor, R. Stock, P. Unger, U. Tunn, A. Kaisary, M. Droller, et al. Molecular Load of Pathologically Occult Metastases in Pelvic Lymph Nodes Is an Independent Prognostic Marker of Biochemical Failure After Localized Prostate Cancer Treatment J. Clin. Oncol., July 1, 2006; 24(19): 3081 - 3088. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
