Peripheral Blood RT-PCR Assays for Detection and Prognosis of Prostate Cancer

The application of molecular biology in medicine has led to new, potentially powerful modalities for the diagnosis and staging of cancer. For example, reverse transcriptase polymerase chain reaction (RT-PCR) has been investigated as a potentially new diagnostic and staging modality for prostate cancer. The RT-PCR assay is a relatively simple and rapid, yet sensitive and specific technique that amplifies a targeted, tissue-specific messenger ribonucleic acid (mRNA) sequence. RT-PCR makes possible the detection of small quantities of cells that are currently undetectable in peripheral blood or within other body fluids or tissues. Since Moreno and associates¹ and Vessella and associates² pioneered the use of RT-PCR for the detection of circulating occult prostate cells, there has been a rapid increase in the number of reports in the literature about the use of RT-PCR for detecting cells expressing prostate-specific antigen (PSA) and other prostatic markers, such as prostate-specific membrane antigen (PSMA) and human glandular kallikrein (hK2). This new assay technology has the potential to allow development and assessment of new therapeutic strategies and to help monitor disease activity more accurately. In cases of localized prostate cancer, it could help identify patients who harbor occult micrometastases, sparing them the morbidity of a radical prostatectomy or radiation therapy that would be ineffective, and/or help select those patients best suited for clinical trials of neo-adjuvant and/or adjuvant therapy.

However, while foci of metastatic prostate cancer detectable by conventional modalities (eg, histology, bone scan, or computed tomography scan) are almost always associated with biologically significant advanced disease, the presence of prostate cells in the circulation does not necessarily indicate that clinically significant metastasis has occurred. Therefore, RT-PCR technology places the burden squarely on investigators to demonstrate the clinical and biological significance of a positive result. Some studies have demonstrated a significant predictive value in RT-PCR for PSA or PSMA for both the pathologic stage of prostate cancer^3–6 and progression-free survival.^7–9 However, the majority of studies have failed to demonstrate any clinically significant role for RT-PCR for PSA or PSMA.^10–14 The limited follow-up time and formidable technical problems regarding sample processing and handling may underlie these conflicting results. Recently, Klein and associates (article 1) and Slawin and associates (article 2) reported on the absence of any association between RT-PCR for PSA and/or PSMA and clinical and pathologic characteristics and biochemical progression in patients who underwent radical prostatectomy for clinically localized disease and who had long-term prospective follow-up (median of 31 and 53 months, respectively). Interestingly, Slawin and his team did find that RT-PCR detection of the native hK2 protein was associated with features of biologically aggressive prostate cancer and clinical outcome (article 2). These studies further emphasize the importance of targeting prostate-specific markers that are associated with biologically and clinically aggressive disease.

One possible limitation of most studies to date is that they have relied on the simple qualitative detection of PSA-expressing cells or on semiquantitative techniques, even if measures were taken to increase the sensitivity by using nested PCR techniques or radioactive detection methods. Because target amplification by PCR is exponential in nature, these approaches show no linearity of reaction to the amount of prostatic target mRNA applied. Moreover, virtually all quantitative detection methods published thus far have the disadvantage of not being easily adapted for standardized clinical application in many centers. To overcome these limitations, Ylikoski and associates recently developed highly sensitive, quantitative, internally standardized RT-PCR assays for PSA¹⁵ and hK2,¹⁶ and evaluated their utility for diagnosis and staging of prostate cancer (article 3).

In view of the findings of the reviewed articles, and the literature in general, the consensus is that at the present time RT-PCR technology is not ready for use in the clinical management of patients with prostate cancer. However, this molecular test is still highly promising. Technological advances such as immunomagnetic enrichment and digital microscopy, among others, may improve the assays and lead to a quantitative, reproducible assay that can specifically detect those select prostate cancer cells that are associated with the development of clinically evident distant metastases, thereby making them more useful in the clinical setting.

Preoperative Combined Nested Reverse Transcriptase Polymerase Chain Reaction for Prostate-Specific Antigen and Prostate-Specific Membrane Antigen Does Not Correlate with Pathologic State of Biochemical Failure in Patients with Localized Prostate Cancer Undergoing Radical Prostatectomy.

Thomas J, Gupta M, Grasso Y, et al.

J Clin Oncol. 2002;20:3213–3218.

In 1997⁵ and 1998,⁶ Klein and associates developed nested RT-PCR assays for PSA and PSMA and reported on their preliminary results. They tested the hypothesis that the combination of these two markers would increase the sensitivity of their assays by partially overcoming the problem of tumor cell heterogeneity. They found that their RT-PCR assay for PSMA was consistently more sensitive than that for PSA, yielding more than 1.8–3 times the number of positive results. However, the combination of both assays together (PSA and/or PSMA positive vs both negative) was superior to either assay alone in sensitivity and specificity. Like most reported assays, the lower end of detection of their assays was one LNCaP cell diluted in 10⁶ mononuclear cells. The clinical specificity of their combined assay was 100%, yielding no false-positives in normal subjects (n = 15), female subjects (n = 5), and men with confirmed benign prostatic hyperplasia (n = 15). The clinical sensitivity of their combined assay was 100% in patients with clinically evident distant metastases (n = 11) and yielded no false-negative results. Here again the RT-PCR-PSA assay performed suboptimally, yielding only 64% positivity when used alone, while RT-PCR-PSMA detected all except one patient. In both of their initial studies, Klein and associates found that a combined, nested preoperative RT-PCR assay for both PSA and PSMA was an independent predictor of extracapsular extension on radical prostatectomy specimens.^5,6

In a well-designed and executed follow-up study, the team at the Cleveland Clinic Foundation evaluated the association of preoperative peripheral blood RT-PCR for PSA and PSMA with pathologic stage and, more importantly, with biochemical progression in 141 patients undergoing radical prostatectomy for clinically localized disease. For quality control they duplicated the assay in the first 25% of patients and found less than 5% discordance in assay results. Overall, 52% of the patients had a positive assay result for PSA and/or PSMA, a rate that is higher than noted in their previous reports and higher than reported in most published studies. In concordance with their previous reports, the large majority of patients were positive for PSMA only, with 5% of patients having a positive assay result for both RT-PCR-PSA and RT-PCR-PSMA and 3% for RT-PCR-PSA only. In contrast with their initial studies, in this larger set of more contemporary patients, the authors failed to find an association between RT-PCR status and pathologic stage. In addition, with a median follow-up of 31 months, RT-PCR status was not associated with disease progression after surgery, even in the subset of patients with non-organ-confined disease. These findings did not change when patients with preoperative androgen-deprivation therapy were excluded (n = 27). The authors concluded that preoperative combined nested RT-PCR for PSA and PSMA has no clinical utility in patients undergoing radical prostatectomy for clinically localized disease.

Preoperative Blood RT-PCR Assays for Prostate-Specific Antigen and Human Glandular Kallikrein for Prediction of Prostate Cancer Progression After Radical Prostatectomy.

Shariat SF, Gottenger E, Nguyen C, et al.

Cancer Res. 2002;62:5974–5979.

The questionable specificity of current markers for prostate cancer cells and the potentially variable biologic and clinical potential of the cells detected by ultrasensitive assays have discouraged the use of PSA and PSMA as targets for RT-PCR assays in the clinical setting.^17–20 Therefore, we developed a highly sensitive and specific RT-PCR assay for hK2 mRNA²¹ that was designed to differentially identify two previously described splice variants of the hKLK2 gene.²² Our primers differentiated between amplification of the native hk2 transcript (hK2-L) that encodes for the full-length hK2 protein, and an alternate spliced transcript (hK2-U) that contains an additional 37 nucleotides downstream from the native splice donor site in intron IV.²² This larger, alternatively spliced mRNA is predicted to encode a truncated and, presumably, nonfunctional version of the hK2 protein. Our assay provided maximum sensitivity and specificity for hK2 without any cross-reactivity to the closely related kallikreins, hK1 and PSA. In addition, our assay was highly sensitive, as demonstrated by a reliable detection of five copies of hK2 cDNA and at least one LNCaP cell in 10⁹ cultured lymphoblasts. When evaluated in peripheral blood specimens obtained from 14 men at low risk for harboring prostate cancer, 14% of the control specimens tested positive for hK2-L. On the other hand, the clinical sensitivity was superior to that reported in the literature for RT-PCR-hK2 assays with 86% of the 7 patients with untreated metastatic prostate cancer yielding a positive assay result.^23,24 When the assay was performed on peripheral blood of patients with clinically localized prostate cancer prior to radical prostatectomy, the native hK2-amplified fragment, but not the splice variant fragment (hK2-U), was an independent predictor of metastasis to regional lymph nodes.²¹ On the basis of these observations, we hypothesized that RT-PCR for hK2-L assays would detect circulating cells that might indicate the presence of occult metastatic disease in patients undergoing radical prostatectomy and would therefore be associated with prostate cancer that progressed despite effective control of local disease. In addition, to determine the relationship between pre-operative peripheral blood RT-PCR for PSA and the risk for prostate cancer progression, we studied a large (n = 224) cohort of consecutive patients with clinically localized prostate cancer who underwent radical prostatectomy and who had long-term follow-up (median follow-up of 53 months, range 1.3–73).

Our assay for PSA was highly sensitive, as demonstrated by a reliable detection of five copies of PSA cDNA and at least one LNCaP cell in 10⁶ cultured lymphoblasts, but less sensitive than our assay for hK2. Our RT-PCR assay was clinically more sensitive for detecting PSA-expressing cells in patients with metastatic disease (seven of eight patients positive, 88%) than the average of peripheral blood assays used in previous studies (53%, range 13%–100%).²⁵ Only one of 14 control specimens (7%) obtained from men at low risk for harboring prostate cancer tested positive for PSA, which is in the range of false positives reported in the literature.²⁵

Of the peripheral blood specimens obtained from 224 patients with clinically localized prostate cancer, RT-PCR assays for PSA, hK2-L, and hK2-U were positive in 24%, 25%, and 26%, respectively. RT-PCR-hK2-L positivity was associated with higher pathologic Gleason score but not organ-confined disease. This confirmed our previous finding that preoperative peripheral blood RT-PCR-hK2-L is an independent predictor of metastases to regional lymph nodes. While these associations with pathologic features are important, an association with occult metastases that can lead to disease progression in patients treated effectively for clinically localized disease would be more useful for managing patients with prostate cancer. Although in univariable analysis, preoperative RT-PCR-hK2-L was a predictor of disease progression after surgery, when adjusted for the effects for preoperative PSA level, biopsy Gleason sum, and clinical stage, this association was no longer significant. However, the relatively low progression rate (21%) and the high rate of patients with features of non-aggressive failure (59%) in this data set might have limited the statistical power of our study for detecting a predictive effect of RT-PCR-hK2-L on overall prostate cancer progression. An improved ability to predict the likelihood of clinical disease progression to metastases would have a greater clinical impact on managing prostate cancer patients. We found that preoperative RT-PCR-hK2-L was associated with features of aggressive prostate cancer progression, defined by a PSA doubling time of less than 10 months,^26,27 the failure to respond to salvage local radiation therapy,²⁸ and/or a positive metastatic work-up. This suggests an association of RT-PCR-hK2-L with occult metastatic disease present at the time of radical prostatectomy.

Neither RT-PCR for PSA nor hK2-U was associated with any clinical or pathologic characteristics or clinical outcome of patients with clinically localized disease undergoing radical prostatectomy and long-term follow-up. RT-PCR assays for PSA or hK2-U are not useful staging tools for guiding therapy or predicting outcome in patients with clinically localized prostate cancer, and, therefore, they have limited clinical utility.

Simultaneous Quantification of Prostate-Specific Antigen and Human Glandular Kallikrein 2 mRNA in Blood Samples From Patients with Prostate Cancer and Benign Disease.

Ylikoski A, Pettersson K, Nurmi J et al.

Clin Chem. 2002;48:1265–1271.

The above reviewed articles support the need for a more objective assay (ie, real-time standardized, quantitative PCR format) that analyzes product quantity during the logarithmic phase of the reaction, before the plateau, and ideally includes internal controls that co-amplify with the same efficiency as the target molecule. Ylikoski and associates have recently developed the first multiplexed, rapid, and reproducible method that allows simultaneous exact linear quantitation of PSA and hK2 gene expression from the same peripheral blood samples.¹⁶ The method uses two target-molecule-like, exogenous, internal standard mRNAs that differ only by two base pairs deleted from their target mRNAs. The similarity in sequence of the internal standards and the targets themselves (PSA and hK2) allow their coamplification with the same primers but still permit distinction in the target amplification product. In addition, the internal standards match the targets in amplification efficiency and product size. A known quantity of synthetic internal standard mRNAs are added into each cell pellet at the beginning of the mRNA extraction and are co-amplified in the RT-PCR reaction with the target mRNAs in the sample, providing a standardized correction for any variations allowing reproducible quantification of the targets. PSA and hK2 and their respective internal standards amplification products are detected with fluorescent europium chelate labels and with detection probes in a microtitration, well-based hybridization assay. The quantification of PSA and hK2 mRNAs is based on the use of an external calibration curve that compares the target with the internal standard fluorescence ratio in the sample to the ratio in the calibration curve. The assay detected 50 copies of hK2 and PSA mRNA and at least one PSA-expressing and 10 hK2-expressing LNCaP cells in 2.5 × 10⁶ hK2- and PSA-negative cells.¹⁶ The functional detection limit of the RT-PCR assay was 100 copies of hK2 or PSA mRNA with an interassay coefficient of variation of less than 23%. Preliminary evaluation in peripheral blood showed that the number of hK2 mRNA copies was significantly higher than the number of PSA mRNA copies in prostate cancer patients with biochemical progressive disease (n = 4), and with locally advanced and metastatic disease to regional lymph nodes and bones (n = 13).

In a cross-sectional study, Ylikoski and associates further evaluated the staging ability of their assay in heterogeneous groups of prostate cancer patients and men with benign disease. The clinical specificity of the assay for detection of prostate cancer was 95%, with 1 of 19 men having benign disease (5%) that yielded a false-positive assay result (Table 1). However, two of the men with benign disease had a prostate biopsy for suspected prostate cancer; one had high-grade intraepithelial neoplasia, and one had prostatitis. The man with the positive PSA and hK2 mRNA assay result had a very low free PSA, for which he had twice undergone biopsy. PSA and hK2 mRNA significantly discriminated between prostate cancer and benign disease (Table 1). Sixty percent of patients with organ-confined cancer (n = 10) had a positive assay for PSA and hK2 mRNA. This is the highest rate reported in the literature for patients with organ-confined prostate cancer.²⁵ Within patients with organ-confined disease, PSA and hK2 mRNA positivity was significantly associated with a higher Gleason score. This association was, however, not significant when tested in all prostate cancer patients. PSA and hK2 mRNA was detected in all non-organ-confined (n = 6) and pelvic lymph node-positive (n = 8) patients. In addition, PSA and hK2 mRNA were detected in 83% and 100%, respectively, of patients with distant, hormone-responsive bone metastases (n = 6). In patients with hormone-independent cancer (n = 11), PSA and hK2 mRNAs were each detected in 82% of patients, whereas only 64% were positive for both. There was no association between PSA or hK2 mRNA copy numbers and different disease stages. However, the small sample size of the study limited the statistical power for detecting such association. In addition, the study design and the absence of follow-up do not allow conclusions about the clinical utility of this promising new assay. We are awaiting prospective evaluation of the clinical utility of this assay in a consecutive patient cohort with long-term follow-up.

Table 1.

Number of Men with Positive RT-PCR Assay Results

	Benign	Prostate Cancer
	Disease
				Non-	Lymph	Hormone	Hormone
			Organ	Organ	Node	Responsive	Independent	Clinically
	Total	Total	Confined	Confined	Involvement	Bone Metastases	Bone Metastases	Staged
	(n = 19)	(n = 51)	(n = 10)	(n = 6)	(n = 8)	(n = 6)	(n = 11)	(n = 10)
RT-PCR Positive (No. pts, %)
PSA	1 (5)	41 (80)	6 (60)	6 (100)	8 (100)	5 (83)	9 (82)	7 (70)
hK2	1 (5)	43 (84)	6 (60)	6 (100)	8 (100)	6 (100)	9 (82)	8 (80)
PSA and hK2	1 (5)	38 (75)	5 (50)	6 (100)	8 (100)	5 (83)	7 (64)	7 (70)
PSA or hK2	1 (5)	46 (90)	7 (70)	6 (100)	8 (100)	6 (100)	11 (100)	8 (80)

PSA, prostate-specific antigen; RT-PCR, reverse transcription polymerase chain reaction; hK2, human glandular kallikrein.

Data from Ylikoski et al.¹⁶