PCA3: A Genetic Marker of Prostate Cancer
PCRI Insights August, 2003 vol. 6, no. 3
By Alejandra B.Torres and Leonard S.Marks, MD*
Urological Sciences Research Foundation (USRF)
*also: David Geffen School of Medicine at UCLA,
Department of Urology
The advent of molecular diagnostics has brought the promise of a specific test for prostate cancer PC, the urinary PCA3 gene test. Widespread testing with prostate-specific antigen (PSA) has increased the numbers of prostate biopsies to perhaps one million annually in the U.S. However, serum PSA levels are not specific for PC. Thus, of approximately four men with elevated PSA levels who undergo prostate biopsies, only one will be found to have the disease. Moreover, some cancers in men with “normal” PSA levels escape detection with the PSA measurement. Another marker is needed, and the urinary PCA3 gene test may well be that marker. Early studies indicate this new marker has a much greater degree of PC specificity than PSA testing.
Limitations of PSA Testing for Prostate Cancer
Within the prostate gland, benign prostatic hyperplasia (BPH) cells contain a concentration of PSA several-fold higher than that of adjacent cancer cells.1 This seriously undermines the theoretical basis of PC testing with PSA. With data from the Prostate Cancer Prevention Trial, where biopsies were obtained irrespective of PSA levels, Thompson has shown that “There is no cutpoint of PSA with simultaneous high sensitivity and high specificity for monitoring healthy men for prostate cancer, but rather a continuum of prostate cancer risk at all values of PSA (Table 1)2.”
And Stamey, an early advocate of PSA testing, has declared, “Serum PSA levels are no longer related to prostate cancer, but only to the volume of BPH present3.”Why? Because the disease has changed! Nowadays, instead of finding large primary cancers in the prostate such as was seen 20 years ago, the usual findings are multiple small lesions, where the serum PSA contribution to the prostate cancer is overwhelmed by the BPH contribution (Figure 1). Despite these changes, nearly 30,000 men will still die of PC this year, so an accurate test for the disease is an urgent priority. The major foibles of PSA testing for PC were recently detailed in a USRF Web site posting (http://www.usrf.org/news/10Foibles_of_PSA/index.htm).
|Figure 1. Diagram showing effect of PSA testing, which began in the mid-1980s, on volume of index (largest) cancer in the prostate over ensuing decades. The index cancer volume has progressively decreased. In data from Stanford University, volume of the average index cancer found at radical prostatectomy has decreased from 5.3 cc to 2.4 cc over the past 20 years, mainly because of increasingly early diagnosis.3|
Discovery of the PCA3 Gene
In the early 1990s, at about the same time that PSA testing was starting to gain widespread adoption, a young molecular biologist from Holland began post-doctoral work at The Brady Urological Institute of Johns Hopkins University. There, in the laboratory of William B. Isaacs, Marion Bussemakers performed studies on human prostate tissue using the technique of differential display, a then newly described method to identify gene expression in different tissues. During this series of experiments, an mRNA was discovered that appeared to be highly specific for prostate cancer (Figure 2)4.
|Figure 2. Northern blot analysis using probes for DD3,now called PCA3 (upper lane) and PSA (middle lane),with rRNA (28S) as a control (lower lane).Ten loaded per lane. Numbers 1-14 refer to different patients whose mRNA was studied in the analysis. Tissues studied: T=tumor, B=benign prostatic hyperplasia, N=normal, N/T=normal + 10% tumor cells, and M=metastasis. Over-expression is determined by comparing the intensity of the bands. Note that the PSA mRNA does not distinguish normal from tumor tissues, whereas DD3 (PCA3) is clearly over expressed in tumors. Reproduced from M.J.Bussemakers, et al,Cancer Res. 59: 5975, 19994 (with permission).|
This gene could not be found in any of the existing gene databases. Bussemakers and Isaacs called their new gene DD3, referring to its appearance in the display, and they concluded that it “might be useful in prostate cancer detection. ”The gene was ultimately found to be over-expressed in 53 of 56 prostate cancers and absent from 18 other normal human tissues. Further study revealed the new gene to be a noncoding RNA, which could be mapped to chromosome 9q21-22 (Figure 3).
Figure 3. Hybridization of metaphase chromosomes of human lymphocytes, using a DD3-specific probe, shows that DD3 is mapped to chromosome 9q21-22 (arrowhead)4.
DD3 was initially described in 1994 at the Congress of the European Society for Urolological Oncology and Endocrinology in Berne, Switzerland. Further development of PCA3 was performed in the laboratories of Jack A. Schalken, Bussemakers’ supervising professor at University Hospital, Nijmegen, the Netherlands. Among the important contributions from Nijmegen were the first clinical demonstration of the specificity of PCA3, its measurability in urine, and the importance of denoting PCA3 expression vis-Ã¡-vis a background of normal prostate epithelial genetic material.5,6 Interest in urinary prostate cells, which had been generally abandoned years before, was then resurrected, and urinary PCA3 research studies were soon instituted by Yves Fradet in Laval University in Canada. A prototype urine assay known as uPM3 was developed at DiagnoCure. During this time, the nomenclature for DD3 was formally changed to PCA3.
Why Molecular Markers?
Molecular biology may be defined as the branch of biology focused on the formation, structure, and function of DNA, RNA and proteins, and their roles in the transmission of genetic information. The central theme of molecular biology is as follows: Information encoded in a sequence of the DNA strand passes to molecules of RNA through a process called transcription. The RNA acts as a messenger (mRNA) to pass the information to proteins through a process called translation. The message transcribed from the gene is therefore translated into a protein product that is specialized for a particular function based on the instruction stored in the gene. With the sequencing of the human genome, molecular biologists faced another hurdle: determining the function of individual genes and their protein products. Knowing the function of each gene is essential to biotechnology, which is a branch of engineering that focuses on using such knowledge for the development of molecular markers and treatments for diseases such as prostate cancer. Thus, a gene is the fundamental unit of storage and transmission of cell biology. To know the genetic make-up of a biological unit is to know the potential direction of its development. How genetic information is passed and how cancer may develop when this process goes awry is shown in online videos from the National Cancer Institute
Initial Clinical Experience with PCA3
While the PCA3 gene was clearly discovered in Isaacs’s lab at Johns Hopkins, the Netherlands is where the gene was initially translated from lab to clinic.5,7 The earlier work of Bussemakers and Isaacs was confirmed and expanded at Schalken’s institution in Nijmegen. A method to accurately quantify the gene in urine was developed, using the RTqPCR gene amplification method. Receiver operating characteristics for PCA3 (tumor vs. benign cells) were shown to be remarkable, with a 0.985 area under the curve (AUC), i.e. the accuracy of the test at the cellular level was nearly perfect. The median upregulation of PCA3 from normal to tumor tissue was found to be 34- fold5, increasing to 66-fold in tumor tissue containing more than 10% cancer cells7.This upregulation in cancer tissues provided a theoretical basis for detecting the presence of the gene in tissues containing only a small number of cancer cells, against a background of low expression by many normal or BPH prostate cells, i.e.,“…in tissue biopsies and bodily fluids5.”Thus, the importance of denoting PCA3 as a ratio with PSA mRNA (a surrogate for background prostate epithelial cell nuclear material) was established. Equally important, a practical application was confirmed: the PCA3 ratio determined in voided urine, especially after light prostatic massage, or ”˜attentive’ digital rectal exam, was shown to be a sensitive and specific test for PC in the host.7
In recent clinical trials from Canada8 and Austria9, the potential diagnostic value of the PCA3 urine test was soon established. In these two trials, more than 700 men who were undergoing prostate biopsy donated urine after attentive digital rectal exams. When the urinary sediment contained enough prostate epithelial nuclear material to be evaluated, the PCA3-to-PSA mRNA urinary levels exhibited a 66-82% sensitivity and 76-89% specificity for cancer. Both values compare quite favorably with PSA accuracy. However, using the early assay method, approximately 15%-20% of PCA3 samples were deemed “non-evaluable” because the urine did not contain a sufficient quantity of PSA mRNA to allow detection of background genetic material.
Evolution to Present-Day Test
In the clinical trials cited above, gene testing was performed at DiagnoCure, a Canadian biotech company founded by Dr. Yves Fradet. Fradet had obtained the PCA3 patent from the group at Nijmegen. The gene was then known as uPM3, and the test was a qualitative assay. In November 2003, Gen-Probe, Inc of San Diego, CA acquired from DiagnoCure exclusive worldwide diagnostic rights to this new prostate cancer gene, which is now known, according to standard nomenclature, as PCA3. Gen-Probe soon developed a quantitative PCA3 molecular assay employing the technologies of Target Capture, Transcription Mediated Amplification (TMA), and Hybridization Protection (HPA). (Figure 4).10 In collaboration with Urological Sciences Research Foundation (USRF) of Culver City, CA, clinical testing of the Gen-Probe assay began in early 2004, and the first presentation of data from that work was made at the Gordon Research Conference on Biomarkers in January, 2005.
Figure 4. Distinguishing features of the PCA3 assay (Gen-Probe, Inc.) are shown in sequence. In step No.1 (top), target capture of the mRNA is performed, using magnetic bead (purple).
In step No. 2, the captured gene is amplified using Transcription-Mediated Amplification, a process that generates some 10 billion copies of PCA3 in one hour.
In step No. 3, the Hybridization Protection Assay is performed using DNA probes tagged with a chemiluminescent substance that is activated upon contact with detection reagents.Details of the assay are described in a recent publication10.
The PCA3 test is actually a dual assay in which both PCA3 and PSA mRNA are separately quantified and the ratio of the two, the PCA3 Score, is determined. The ratio is used because the denominator, PSA mRNA, establishes the amount of prostate-specific nuclear material in the specimen. A low level of PCA3 is expressed by normal prostate cells, and if absolute concentration of PCA3 were used, a high Score might be obtained from a specimen rich only in normal prostate cells. Thus, the PCA3 Score tells the expression of PCA3 corrected for the background of normal or BPH epithelial cells present in the specimen (Figure 5). In early clinical testing, it was soon determined that the higher the urinary PCA3 Score, the greater the likelihood of prostate cancer (Figure 6).
Figure 5. Diagram showing that a background low level of PCA3 expression is present from the benign prostate cells in urine (left). On right, a single cancer cell is shown to greatly over-express the gene, allowing detection in a urine specimen of an abnormal quantity of PCA3 relative to the normal background.
|Figure 6. Chart showing the higher the PCA3 Score (PCA3/PSAmRNA) (horizontal axis), the greater the likelihood of cancer (vertical axis). From L.S. Marks, et al, 2006 AUA Meeting, Atlanta, GA.|
In addition to normalizing the PCA3 signal, measurement of PSA mRNA also serves to confirm that the yield of prostate specific RNA is sufficient to generate a valid or “informative” test. Without a certain minimum amount of prostate-specific genetic material in the sample, the test is deemed “non-informative”. An attentive digital rectal exam (three sweeps on each side of the prostate), performed just prior to urine specimen collection, improves the informative rate from approximately 80% to greater than 95%. It is likely that the informative rate now being obtained with the new assay is attributable to both the attentive DRE and the increased sensitivity of the new assay technologies explained above.
Current Use and Availability of PCA3 Testing
In presentations at the 2006 American Urological Association meeting (J.Urol., 175: 174-6 (S), 2006), in recent data gathered on approximately 1000 men, the Gen-Probe PCA3 test was shown to exhibit a high degree of sensitivity and specificity for prostate cancer. For cancer vs. non-cancer, a specificity of 76% at 50% sensitivity (PCA3 cutoff = 35 copies/copy of PSA mRNA), with an area under the ROC curve (AUC) of 0.680, was reported by Fradet’s group. By comparison, serum tPSA specificity was only 22% for the same men. In addition, the quantitative PCA3 score correlated with the probability of positive biopsy in this population: at low PCA3 Scores (< 5) the biopsy positive rate was only 20%, while at PCA3 scores > 100 the risk of positive biopsy was 67%. A suggestion was presented in Schalken’s recent data that some correlation with Gleason grade and cancer volume may also be present. In data from USRF, almost no overlap was seen in PCA3 scores from men with cancer and men with only BPH, confirming the specificity of the test. PCA3 RNA is uniformly undetectable in urine from post-radical prostatectomy patients, even following attentive DRE.
A particularly important role of the new marker appears to be in men with persistently elevated serum PSA levels, but a negative initial biopsy. In such men, who constitute a large problematic group, the odds ratio for the PCA3 test to predict cancer upon re-biopsy is 3.6, compared to only 1.2 for serum PSA testing10 (Table 2).
PCA3 testing is highly dependent on the cutoff score used to determine a “positive” or “negative” test, because sensitivity and specificity vary reciprocally with the score. The higher the cutoff, the greater the specificity and the lower the sensitivity; the lower the cutoff, the greater the sensitivity and the lower the specificity. Thus, although the test is now available commercially, physicians must be cautious in interpreting the lab report. They should know the performance characteristics of the assay before decisions are based on a “positive” or “negative” test result. U.S. laboratories currently offering the PCA3 test commercially include Bostwick Laboratories, Richmond, VA (http://www.bostwicklaboratories.com/about/PCA3.html) and AmeriPath Laboratories, Palm Beach Gardens, FL (http://www.ameripathgu.com/PCA3/). The test is not currently approved by the US FDA. The method of specimen collection is shown in Figure 7.
Figure 7. Diagram showing PCA3 specimen collection.The procedure begins with an attentive digital rectal exam (3 sweeps on each side of the prostate). First voided urine is then collected and sent to laboratory for analysis.
PCA3 Score vs. PSA Testing
In comparison with serum levels of PSA, the urinary PCA3 score appears to be highly specific for prostate cancer (Figure 8). While serum PSA levels are known to be influenced by volume of BPH tissue, age, inflammation, trauma, and use of 5 alpha-reductase inhibitors (finasteride, dutasteride), preliminary data indicate that these factors do not appear to influence PCA3 scores. For example, standard teaching is to draw blood for PSA levels before a DRE, for fear the exam might cause spurious elevations in serum PSA. However, an attentive DRE actually increases the “informative” rate of PCA3 determinations and is, in fact, recommended.
Figure 8. Diagram showing urinary PCA3 (lower arrow) vs serum PSA (upper arrow).Whereas PSA is a glycoprotein that may enter the bloodstream, PCA3 is a gene that exists in the nuclear material of prostate epithelial cells which may be shed into the urine. Those cells, if cancerous, over-express the gene. That over-expression, which may be many times that found in benign prostate cells, is detected by the assay. Importantly, PCA3 expression is normalized against a background of prostate specific nuclear material (PSAmRNA), yielding a PCA3 score.The PCA3 score is much more cancer-specific than serum PSA levels,which are confounded by factors such as prostate volume, age, trauma, and certain drugs.
In Figure 9, the effect of prostate volume is shown on both PSA and PCA3 in the same group of adult men. Clearly, PSA is directly related to prostate volume, while PCA3 is not. Unpublished data from USRF indicate that the same is likely to be true for age and use of 5ARI drugs. Thus, with the caveat that data are limited, the urinary PCA3 score appears to offer a great specificity advantage over serum PSA levels in the early diagnosis of prostate cancer.
Figure 9. Charts showing relationship of prostate volume to serum PSA levels (right) and PCA3 score (left).While PSA is directly related to prostate volume, the PCA3 score appears to have no relationship to prostate volume. Thus, prostate volume, one of the primary factors influencing serum PSA levels, is not a confound of the PCA3 score. Improved cancer specificity is a major advantage of PCA3 testing. From L.S.Marks, et al, 2006 AUA Meeting, Atlanta, GA.
Conclusion and Future Directions
The PCA3 gene, a noncoding segment of mRNA located on chromosome 9, is over-expressed by prostate cancer cells in comparison with all other cells studied. The differential expression is great, permitting detection of the gene in nuclear material from cancer cells shed into urine after an attentive DRE. Thus, urinary PCA3 appears useful as a highly specific marker for prostate cancer. However, while the early data look promising, the PCA3 test must still be regarded as a “work in progress”, from several perspectives. PCA3 expression is denoted against a background of prostate-specific genetic material, a PCA3 score (i.e. a ratio of PCA3 to PSA mRNA), and normative values have only been defined in a preliminary fashion. Factors regulating PCA3 gene expression are not yet clearly defined, but the great confounds of serum PSA levels (prostate volume, age, trauma) appear to affect PCA3 to a much lesser degree than PSA. Additional clinical research trials, now in an organizational phase, should provide further guidelines for widespread application of the urinary PCA3 score.
1.Magklara A, Scorilas A, Stephan C, Kristiansen GO, Hauptmann S, Jung K and Diamandis EP: Decreased concentrations of prostate-specific antigen and human glandular kallikrein 2 in malignant versus nonmalignant prostatic tissue. Urology. 56: 527-32, 2000.
2. Thompson IM, Ankerst DP, Chi C, Lucia MS, Goodman PJ, Crowley JJ, Parnes HL and Coltman CA, Jr.: Operating characteristics of prostate-specific antigen in men with an initial PSA level of 3.0 ng/ml or lower. JAMA. 294: 66-70, 2005.
3. Stamey TA, Caldwell M, McNeal JE, Nolley R, Hemenez M and Downs J: The prostate specific antigen era in the United States is over for prostate cancer: what happened in the last 20 years? J Urol. 172: 1297-301, 2004.
4. Bussemakers MJ, van Bokhoven A, Verhaegh GW, Smit FP, Karthaus HF, Schalken JA, Debruyne FM, Ru N and Isaacs WB: DD3: a new prostate-specific gene, highly over expressed in prostate cancer. Cancer Res. 59: 5975-9, 1999.
5. de Kok JB, Verhaegh GW, Roelofs RW, Hessels D, Kiemeney LA, Aalders TW, Swinkels DW and Schalken JA: DD3(PCA3), a very sensitive and specific marker to detect prostate tumors. Cancer Res. 62: 2695-8, 2002.
6. Schalken JA, Hessels D and Verhaegh G: New targets for therapy in prostate cancer: differential display code 3 (DD3(PCA3)), a highly prostate cancer-specific gene.Urology. 62: 34-43, 2003.
7. Hessels D,Klein Gunnewiek JM, van Oort I,Karthaus
HF, van Leenders GJ, van Balken B, Kiemeney LA,
Witjes JA and Schalken JA: DD3 (PCA3)-based
molecular urine analysis for the diagnosis of prostate
cancer. Eur Urol. 44: 8-15; discussion 15-6, 2003.
8. Fradet Y, Saad F, Aprikian A, Dessureault J, Elhilali M,
Trudel C, Masse B, Piche L and Chypre C: uPM3, a
new molecular urine test for the detection of prostate
cancer. Urology. 64: 311-5; discussion 315-6, 2004.
9. Tinzl M, Marberger M, Horvath S and Chypre C:
DD3PCA3 RNA analysis in urine–a new perspective
for detecting prostate cancer. Eur Urol.46: 182-6; discussion
10. Groskopf J, Aubin SM, Deras IL, Blase A, Bodrug S,
Clark C, Brentano S, Mathis J, Pham J, Meyer T et al.:
APTIMA PCA3 molecular urine test: development of
a method to aid in the diagnosis of prostate cancer.
Clin Chem. 52: 1089-95, 2006.
11. Marks LS, Fradet Y, Deras IL, Blase A, Mathis J, Aubin
SMJ, Cancio AT, Desaulniers M, Ellis WJ, Rittenhouse
HG, Groskopf J: Prostate cancer specificity of PCA3
Urinary Gene Test. In Press, Urology, 2006.