Dr. de la Taille and colleagues from Columbia University provide an overview of the concept of molecular staging of prostate cancer using reverse transcriptasepolymerase chain reaction (RT-PCR). They do an admirable job of summarizing all of the currently available data on the results of this assay in the clinical staging of prostate cancer. As they note, only their group and one other have been able to demonstrate that a positive assay correlates with final pathologic stage. A limited number of other studies have suggested that the RT-PCR assay can predict prostate-specific antigen (PSA) recurrence.
The authors correctly question the utility of RT-PCR assays in the molecular staging of prostate cancer. They first reported that the RT-PCR assay was superior to all other modalities in predicting the final pathologic stage of patients prior to radical prostatectomy. In fact, the authors made such a point of the utility of this new assay in 1994 that many clinicians began advising their patients to have an RT-PCR determination before undergoing radical prostatectomy. If the RT-PCR assay were positive for circulating cells, radical prostatectomy was not to be performed since this indicated advanced disease. With more studies now available from other centers, the utility of the assay as a molecular staging tool has not yet been validated.
This early, well-publicized ballyhoo in the absence of correlative studies cast great confusion over the entire field. Attempts were made to rapidly commercialize this assay based solely on preliminary studies.
Many clinicians today still have a dim view of the RT-PCR detection of micrometastasis because of the failure of this assay to meet the lofty clinical expectations put forth by our colleagues at Columbia University. However, based on our present level of understanding, this blanket criticism of molecular techniques in the detection of circulating micro-metastasis is probably unjustified. The role of RT-PCR and the entire field of the detection of micrometastasis of solid tumors, such as prostate cancer, deserve continued intense investigation.
Biology of Prostate Cancer Metastasis
The mechanism of the spread of prostate cancer micrometastasis remains unknown, with limited experimental data available on the process of tumor dissemination.[reviewed in reference 2] The earliest research published in 1940 suggested that hematogenous micrometastasis spread to the bone through Batsons pre-vertebral venous plexus. Other investigators suggested that metastatic spread occurred first through lymphatic channels and then to the hematogenous system.
In 1973, Whitmore proposed that the progression of prostate cancer is variable, with some cancers disseminating by the lymphatic route and others spreading through the venous system. Since bone metastasis is almost always the cause of death in patients with prostate cancer, it is likely that the hemato-genous route of metastasis is of clinical significance.
Key concepts in the study of metastasis of any solid tumor derive from extensive animal data. Studies have shown that normal epithelial cells and the majority of cancer cells are rapidly removed from the peripheral circulation in animal models. In these experimental solid tumor systems, approximately 0.01% of circulating cancer cells can eventually create a single metastasis.
Using these animal model as a basis, a single prostatic bone metastasis in man would have to be generated by 10,000 circulating cancer cells, or the equivalent of 2 cells per milliliter of blood. It is in this regard that technologies such as RT-PCR may allow for the detection of these very early metastatic events.
Detection of Circulating Cancer Cells
Flow cytometric analyses have previously demonstrated the presence of PSA-expressing cells in the peripheral blood of patients with metastatic prostate cancer. As early as 1961, there were published reports of the transient dissemination of prostate cancer cells into the vena cava as detected by Papanicolaou stain in patients undergoing prostatic manipulation. These cells reached peak detectable levels in 10 minutes and cleared after 30 minutes. Studies of other circulating cancer cells, such as colon cancer cells, using gross cytologic detection methods, also suggested that cells with a malignant phenotype are cleared from the circulation.
However, these studies were not widely embraced as being clinically significant. In a 1969 review, a National Cancer Institute (NCI) study group concluded that there was limited utility to the detection of micrometastasis in the peripheral blood by cytologic methods. In 1975, a 20-year literature review expressed doubt that the detection of circulating cancer cells would carry any prognostic value. Hence, this field subsequently remained relatively dormant until the advent of molecular PCR and RT-PCR technologies.
In 1992, our group at Thomas Jefferson University published the first clinical study demonstrating that circulating prostate cells could be detected using molecular biological methods. Using RT-PCR targeted at PSA messenger RNA (mRNA), we collected peripheral blood from men with newly diagnosed metastatic prostate cancer and showed that malignant cells in the circulation could be detected by molecular methods. Further studies demonstrated that circulating prostate cancer cells are malignant.
Dozens of centers have now studied the RT-PCR assay in blood, bone marrow, and nodal tissue, with no definitive consensus reached on the clinical utility of this assay. Technical aspects of this intense assay provide the most plausible explanation for the reported differences in its prognostic ability. This renewed interest has spawned other allied technologies, such as the magnetic bead and nanoparticle enhancement of the sample, aimed at improving the detection of circulating cells.[6,7] These cell extraction technologies represent the next major avenue of investigation into the detection of micrometastatic cells, either alone or in combination with RT-PCR.
Technical and Biological Issues Related to RT-PCR
The enormous technical challenges raised by this line of investigation should not be underestimated. These factors are highlighted by de la Tallie and associates in Table 4 of their article and have also been summarized elsewhere.[2,8] These include such diverse considerations as patient populations, statistical considerations of sample size, specimen handling (eg, timing of phlebotomy, containers used, shipping time and temperature), cell separation, cell lysis, RT-PCR and PCR assay conditions, and primer set, to name just a few.
Looming in the background are the uncertain implications of the biology of the cells detected by this technology. Tumor shedding after biopsy of the prostate has been detected by RT-PCR, suggesting that the timing of phlebotomy may be critical. However, if and when these cells clear from the circulation is uncertain. These technical and biological issues must and will continue to be addressed as we learn more about how to optimize the RT-PCR assay as a clinical tool.
There is widespread interest in the detection of micrometastasis using RT-PCR technologies. These assays are being used in the study of dozens of solid tumors, including breast cancer, colon cancer, and melanoma, and have proven clinical utility in determining recurrence risk in bone marrow studies of hematologic tumors.
We firmly believe that RT-PCR and related assays that detect the presence of cancer cells outside the target organ have the potential to be the optimal prognostic tool in the management of solid tumors. Current prognostic factors for tumor recurrence are derived from surrogate features of the primary tumor, such as grade, stage, oncogene expression, and microvessel density. Therefore, the presence or absence of micrometastatic cells should equate to the optimal prognostic tool to determine the risk of tumor progression or recurrence. Reverse transcriptasepolymerase chain reaction has and will continue to make a contribution to our understanding of the biology of prostate cancer and other solid tumors.
Reverse transcriptasepolymerase chain reaction can detect PSA-expressing cells that are undetectable by other means in patients with localized and metastatic cancers. This assay is highly specific, as PSA-expressing cells are undetectable in the peripheral circulation of patients without prostate cancer in almost all studies.
Our colleagues from Columbia University have had the best results with their modifications of our originally reported RT-PCR assay. With their group continuing to report successful molecular staging, the subtleties of their assay should be studied closely by all those who wish to follow this line of research. It should be stressed, however, that while molecular staging may not prove to be the best use of the assay, RT-PCR may improve the detection of prostate cancer lymph node metastasis over immunohistochemistry and may have prognostic utility when the bone marrow is involved. As de la Taille and colleagues note, several prognostic studies have been able to associate a positive RT-PCR assay with PSA recurrence.
This test has been and remains a powerful research tool in the study of the biology and behavior of prostate cancer. The clinical utility of RT-PCR as a diagnostic and prognostic tool in prostate cancer is just beginning to be determined. Reverse transcriptasepolymerase chain reaction detection of these cells has refocused prostate cancer research on the identification of micrometastatic disease.
Drs. de la Taille title their article, Staging for Prostate Cancer: Dream or Reality? In our view, the reality of RT-PCR as a useful clinical tool in prostate cancer remains a dream well worth pursuing.