Evaluation of a new, rapid, qualitative, one-step PSA Test for prostate cancer screening: the PSA RapidScreen test

Transcription

1 Evaluation of a new, rapid, qualitative, one-step PSA Test for prostate cancer screening: the PSA RapidScreen test R Miano 1 *, GO Mele 2, S Germani 1, P Bove 1, S Sansalone 1, PF Pugliese 2 & F Micali 1 1 Department of Urology, Policlinico Tor Vergata, University of Rome Tor Vergata, Rome, Italy; and 2 Italian Tumour League, Provincial Section of Cosenza, Italy (2005) 8, & 2005 Nature Publishing Group All rights reserved /05 $ To increase the acceptance rate and reduce the cost of the screening programme for prostate cancer, a new qualitative and one-step test for prostate-specific antigen (PSA), called PSA RapidScreen, has been evaluated. PSA RapidScreen test is a chromatographic lateral flow immunoassay, which generates a positive or negative result for PSA values Xor o4 ng/ml, respectively. Capillary blood samples from 188 men were evaluated. Two independent observers interpreted the test at 10, 15, 20 and 25 min. A total of 10 women were tested as controls. Parallel serum samples were simultaneously collected and tested with an ordinary quantitative assay (Elecsys s 2010, Roche). Sensitivity, specificity, accuracy, negative and positive predictive values of the test were 97.6, 90.4, 94, 98 and 89%, respectively. PSA RapidScreen tests on female capillary samples were negative. Reproducibility of the test was 99.5%, while interobserver variation was 5%. Specificity of the test was altered by variations in the reading time. Quantitative assessment of the intensity of the band correlated with the PSA value (r ¼ 0.87; t ¼ 23.97; Po0.001). PSA RapidScreen is a rapid, simple and reproducible one-step test. The low cost and the speed of the test make it a powerful and convenient tool for prostate cancer screening programmes. (2005) 8, doi: /sj.pcan ; published online 17 May 2005 Keywords: prostate-specific antigen; screening; prostate cancer Introduction Prostate cancer (CaP) remains the most common cancer in men with an estimated new cases in 2004 in the United States. CaP represents approximately 30% of all newly diagnosed cancers. The widespread use of the prostate-specific antigen (PSA) test is thought to be responsible for the rapid increase in CaP diagnosis between 1988 and 1992 and the decline that has recently followed. 1 Few topics in CaP are more controversial than the issue of screening. The potential benefits and harm continue to be debated among health professionals. *Correspondence: R Miano, Via Giulio Curioni, 131, Rome, Italy. mianor@virgilio.it Received 27 February 2005; accepted 24 March 2005; published online 17 May 2005 The high prevalence of the disease and the availability of a test (PSA) capable of detecting cancer at an early presymptomatic stage are important criteria required to support mass screening. The significant increase in the diagnosis of the organ-confined prostatic tumours and the reduction (more than 50%) of metastatic disease further supports CaP screening. 2 4 Moreover, recent reports seem to indicate a decline in CaP mortality rates, probably due to a combination of factors, including early detection through PSA-based screening. 5 8 These data, if confirmed in the future with further evidence, could represent one of the most persuasive arguments in support of screening. Two major controversies on CaP screening still have to be solved: (1) The possibility of over-detection, by diagnosing some clinically insignificant tumours that would probably never lead to morbidity or mortality.

2 220 (2) The economics of CaP screening, including not only the cost of detection, but of the treatment and of the treatment-related complications also. Optenberg and Thompson 9 estimated that the cost of the screening could be as high as $25 billion annually if all men y of age in the United States participate in a screening programme. In this study, we evaluated a new, qualitative, rapid, onestep and low-cost PSA screening test on capillary whole blood, called PSA RapidScreen test (Craig Medical Distribution, Vista, CA, USA), in order to reduce the cost of the screening and to increase the acceptance rate of an early detection programme. Materials and methods Patients and serum samples From May 2001 to January 2002, 188 male patients (mean age 65 y, range 40 88), with or without prostate disease were analysed. All men were investigated at the Policlinico Tor Vergata in Rome. All the patients underwent a double evaluation with PSA RapidScreen test for a total of 376 tests performed; 10 min was the time interval between the tests in each patient. For comparative purpose, a blood sample from each patient was collected immediately before the test for the quantitative determination of PSA. The blood was allowed to clot for 1 h at room temperature before being centrifuged. The sera were immediately analysed by Elecsys s 2010 (Roche) PSA assay. Female capillary whole blood samples (n ¼ 10) were collected to perform PSA RapidScreen test as a control. Principles of PSA RapidScreen test PSA RapidScreen test presents as a special cassette, containing two different regions: the sample well and the test area. It is a chromatographic lateral flow immunoassay, containing a filter membrane coated with PSA specific antibodies and coloured gold colloidal reagents labelled with PSA-specific antibodies. The PSA cutoff point of the test is 4 ng/ml. When PSA concentration levels are X4 ng/ml, a burgundy-coloured band should develop on the test region (Figure 1a). Conversely, PSA levels below 4 ng/ml should not develop a test line and the test region should be clear. The control (C) line will always develop in the control (C) region of the test panel regardless of the presence of PSA and serves as an internal quality control, indicating that the test is working properly (Figure 1b). Test procedure The PSA test cassette is provided with a pipette and a buffer/wash solution. The fingertip should be swabbed with an alcohol swab and 30 s should be allowed for the alcohol to dry. Then the fingertip is pricked with a sterile lancet to obtain one or two drops of blood; a pipette is provided to draw up the blood sample and then to Figure 1 PSA RapidScreen test. (a) Positive test. A coloured band develops on the test region (b) Negative test. No band develops on the test region. transfer it to the sample well on the PSA cassette. After s, two drops of a buffer/wash solution are added to the sample well to allow the blood to migrate. If visible migration across the test area does not start, an additional drop of a buffer/wash solution can be added to the sample well. Interpretation of the test must be done after 20 min. Evaluation procedure Two independent urologists interpreted the test. The observers were unaware of the measured PSA values and performed the test interpretation independently from one another to avoid bias. The observers were also unaware that the two different tests were being performed on the same patients to avoid bias. To assess the effect of reading time on the results, every test was read at four different reading times (10, 15, 20, 25 min); the reading results of the observers were compared to determine the inter-observer variation. The tests were interpreted as negative if no band appeared the test region; if a band developed the tests were judged as weak positive ( þ ) or strong positive ( þþ) according to the intensity of the colour reaction. To validate the interpretation of the different band intensity in order to understand when a test has to be interpreted as weak or strong positive, a specific trial, with 100 PSA RapidScreen test performed, was provided before the start of the study. Positive and negative results of each test were classified as true positive or negative and false negative or positive by comparing them with the quantitative PSA results, measured by Elecsys s 2010 (Roche) PSA assay, using the threshold value of 4 ng/ml.

3 Statistical analysis Statistical analysis was performed using the commercial software MedCalc. Sensitivity, specificity, accuracy, positive and negative predictive values of the PSA Rapid- Screen test were calculated. Sensitivity was defined as the ratio between the true positives results and the sum of the true positive and the false negative results. Specificity was defined as the ratio between the true negative results and the sum of the true negative and the false positive results. Positive predictive value was defined as the ratio between the true positive results and the sum of all positive results. Negative predictive value was defined as the ratio between the true negative results and the sum of all negative results. Accuracy was defined as the ratio between the sum of the true observations (positive and negative) and the total observations. A Box-and-Whisker plot was used to show the correlation between the PSA quantitative value and the intensity of the colour reaction of the band that appeared on PSA RapidScreen tests. The Pearson s product correlation test was used to correlate the two variables described above. Results Table 1 Reading time and PSA RapidScreen test Reading time (min) Sensitivity (%) Specificity (%) Patients were divided into two groups on the basis of the PSA quantitative value. Out of 188, 104 patients (55%) had a PSA value o4 ng/ml (median PSA 0.9 ng/ml; range ) and 84/188 (45%) patients presented with a PSA value 44 ng/ml (median PSA 8.51 ng/ml; range ). A total of 54 (29%) patients had a PSA value between 4 and 10 ng/ml. Results of the test were especially affected by variations in the reading time; less was the time interval between test and reading (10 min), the lower was the sensitivity. At a reading time of 25 min, the number of false positives increased, decreasing the specificity of the test (92% at 15 min reading time; 90% at 20 min and 86% at 25 min). The optimal reading time was at 20 min (Table 1). PSA RapidScreen test showed 97.6% (82/84) of correct positive results (corresponding to diagnostic sensitivity of the test). A total of two (2.4%) tests out of 84 were misjudged (PSA quantitative value was 5.29 and 7.69 ng/ml, respectively). Considering the 54 men with a PSA quantitative value between 4 and 10 ng/ml, the sensitivity of the test was 96.3%. In the specific PSA range 4 5 ng/ml, sensitivity was 100% (14/14). Among the 104 patients with PSA value o4 ng/ml, 94 were correctly interpreted as negative and 10 were positive with PSA RapidScreen test, determining the specificity of the test (94/104 ¼ 90.4%). Out of 10 false positive results, 5/10 (50%) were in the PSA concentration range of 3.0 o 4.0 ng/ml (respectively, 3.53, 3.58, 3.62, 3.7 and 3.8 ng/ml), while the other five were o3 ng/ml (respectively, 0.1, 1.16, 1.5, 2.2 and 2.48 ng/ml); the false positive results were all judged as mild positive. Specificity of the test in the PSA range 3 4 ng/ml was 58% (7/12). Accuracy, negative and positive predictive values of PSA RapidScreen test were 94, 98 and 89%, respectively. All of PSA RapidScreen tests performed on female samples were correctly judged as negative. Analysis of the interpretation of the tests revealed a high level of reproducibility of this test (99.5%). Interobserver variation was 5%, with an overall concordance rate between the observers of 95%. The Pearson s product correlation test revealed a nonzero (positive) correlation between the intensity of the colour reaction of the band that appeared on the test region and the PSA quantitative value: (r ¼ 0.87; t ¼ 23.97; Po ) (Figure 2a and b). Out of 82 correct positive results, 52 patients had a PSA value between 4 and 10 ng/ml; 35/52 (67%) were interpreted as weak positive by PSA RapidScreen test, while 17/52 (33%) were judged as strong positive. A total of 30 patients presented a PSA value 410 ng/ml; 26/30 (87%) were judged as strong positive, while 4/30 (13%) as weak positive (Table 2). a 500 PSA (ng/ml) b PSA (ng/ml) Negative Weak Strong Negative Weak Strong Figure 2 Box-and-Whisker plot. PSA quantitative values were plotted according to the intensity of the colour reaction of the band appearing on PSA RapidScreen tests. The central box represents the values from the lower to upper quartile (25 75 percentile). The middle line represents the median. (a) Total samples. (b) Samples in the PSA range 0 20 ng/ml. 221

4 222 Table 2 Correlation between PSA quantitative value and PSA RapidScreen test interpretation (Pearson product correlation test: r ¼ 0.87; t ¼ 23.97; Po ) PSA value (ng/ml) n PSA RapidScreen ( ) PSA RapidScreen (+) PSA RapidScreen (++) Concordance rate (%) o Table 3 Performance of various PSA one-step tests reported in the literature Author PSA test Sample Sensitivity Specificity Jung et al 17 Chembio Serum Jung et al 17 Medpro Serum Jung et al 17 Syntron Serum Jung et al 17 Seratec Serum Dok An et al 16 One Step PSAt Serum Madersbacher et al 18 Oncoscreen s Serum Berg et al 15 Uralen s Capillary whole blood Miano et al PSA RapidScreen Capillary whole blood Discussion Although screening for CaP is still one of the most controversial issue in oncology, it is well recognised that digital rectal examination (DRE) and the PSA test, used together, are the most useful and effective tools to diagnose prostatic carcinoma. 3,10 11 In particular, the widespread use of PSA testing is considered one of the factors that has led to a significant increase in the diagnosis of organ-confined tumours and a decline in CaP mortality rates, as some recently reported. 5 8 If the other controversies on CaP screening (high cost, possibility of over-detection, consensus on the best treatment modalities) could be solved, early detection programmes would be more acceptable worldwide. The acceptance rates of screening and medical checkup programmes for the diagnosis of CaP has varied greatly; the participation rate has been between 22.7 and 74% Difficult bureaucratic and technical procedures that prevent the final screening response to be given before several days, is one of the reasons of the low acceptance rate. A screening programme for any disease should be acceptable by the patients if it is simple, non-invasive and one-day ; acceptance by National Health Services should be obtained with two specific characteristics: low cost and effectiveness. In a recent study by Berg et al, 15 a PSA one-step test showed a high acceptance rate due to a well-conducted publicity campaign and because it proved to be a fast, easy to perform and inexpensive test. 15 In our study, PSA RapidScreen test has confirmed all of the above-mentioned characteristics: (1) Simple. It is very easy to use and it does not need costly additional equipment; it only needs an initial and brief training to interpret the test correctly. It would be useful in the general practitioner or office urologist setting. (2) Noninvasive. It uses capillary whole blood that is more patient-friendly, avoiding the traditional venepuncture. This is the second report on a one-step test for PSA in capillary whole blood. Berg et al 15 described a very similar test with a lower performance rate (sensitivity 91%; specificity 84%). Other PSA rapid tests, previously described in the literature, used serum samples (3) One-day screening. Using capillary whole blood rather than a serum sample, saves time (a standard laboratory process to obtain serum needs 1-h for blood clotting after collection and min to separate serum from blood by a centrifuge) and consequently produces the final screening response in about 30 min. (4) Low-cost. The charge for a PSA RapidScreen test is significantly lower than a quantitative serum PSA determination (about 3.00 vs 7.00 euro). Considering the population that could be involved in a screening programme for CaP, savings should be very interesting for a National Health Service. Moreover, savings are much more clear if we consider the high percentage of PSA negative results (about 90%) in a population-based CaP screening programme, 14 so that, using the PSA one-step test, a significant number of quantitative PSA determinations could be spared in a screening population. (5) Effectiveness. Given the design criteria of this test (initial prostate cancer screening programme), sensitivity is more important than specificity. The PSA RapidScreen test has showed a very high sensitivity (97.6%), also in the PSA range near the cutoff value of 4 ng/ml. While total specificity was 90.4%, it suffers in the PSA range 3 4 ng/ml where the false-positive results are quite common (42% in our series). Accuracy is very high too (93.6%). The results of PSA RapidScreen test are comparable with those of the other PSA one-step tests described in the literature, performed both in serum or in capillary blood (Table 3). A positive membrane test result would then signal the need for an accurate quantitative measurement of PSA. However, in the presence of a negative test result, this test should be always integrated with digital rectal examination to exclude a potentially significant CaP. 19

5 The main limitations of this test appear to be the poor specificity in the range 3 4 ng/ml where the possibility of CaP detection is about 25% 20 and more precision is needed to obtain the lowest number of false positive results. In our opinion, future technology will focus on the development of a disposable quantitative PSA rapid test, similar to the Haemo Gluco test used in the follow-up of diabetes, to make a precise, easily available and selfadministrated PSA evaluation. Conclusions The PSA RapidScreen has been shown to be a simple, feasible and reproducible one-step PSA test. The cost lower than the ordinary quantitative PSA kits, its easy handling and the fast procedure make this test useful in the general practitioner or office urologist setting for an initial prostate cancer screening programme. Acknowledgements This study was supported by the Italian Tumour League, Provincial Section of Cosenza, Italy. We are grateful to all the staff and in particular to Avv. Franco Martire and Prof. Luigi Pugliese, President and Vice-President of the Provincial Section of Cosenza respectively, for their kind and essential support. References 1 American Cancer Society, Inc. Cancer Facts and Figures American Cancer Society, Inc: Atlanta, Ga, 2004, pp Crawford ED. Prostate cancer awareness week: September 22 28, CA Cancer J Clin 1997; 47: Catalona WJ et al. Comparison of digital rectal examination and serum prostate specific antigen in the early detection of prostate cancer: results of a multicenter clinical trial of 6,630 men. JUrol 1994; 151: Stephenson RA, Stanford JL. Population-based prostate cancer trends in the United States: patterns of change in the era of prostate-specific antigen. World J Urol 1997; 15: Merrill RM, Stephenson RA. Trends in mortality rates inpatients with prostate cancer during the era of prostate specific antigen screening. JUrol2000; 163: Bartsch G et al. Prostate cancer mortality after the introduction of prostate-specific antigen mass screening in the Federal State of Tyrol, Austria. Urology 2001; 58: Meyer F, Moore L, Bairati I, Fradet Y. Downward trend in prostate cancer mortality in Quebec and Canada. JUrol1999; 161: Tarone RE, Chu KC, Brawley OW. Implications of stage-specific survival rates in assessing recent declines in prostate cancer mortality rates. Epidemiology 2000; 11: Optenberg SA, Thompson IM. Economics of screening for carcinoma of the prostate. Urol Clin North Am 1990; 17: Littrup PJ et al. Cost-effective prostate cancer detection. Cancer 1994; 74: Potter SR et al. Age, prostate-specific antigen, and digital rectal examination as determinants of the probability of having prostate cancer. Urology 2001; 57: Martin E et al. Final results of a screening campaign for prostate cancer. Eur Urol 1999; 35: Gustafsson O et al. Diagnostic methods in the detection of prostate cancer. A study of a randomly selected population of 2,400 men. JUrol1992; 148: Reissigl A et al. Prostate carcinoma screening in the county of Tyrol, Austria: experience and results. Cancer 1997; 80: Berg W, Linder C, Eschholz G, Schubert J. Pilot study of the practical relevance of a one-step test for prostate-specific antigen in capillary blood to improve the acceptance rate in the early detection program of prostate carcinoma. Int Urol Nephrol 2001; 32: Dok An C, Yoshiki T, Lee G, Okada Y. Evaluation of a rapid qualitative prostate specific antigen assay, the One Step PSAt test. Cancer Lett 2001; 162: Jung K et al. Rapid detection of elevated prostate-specific antigen levels in blood: performance of various membrane strip tests compared. Urology 1999; 53: Madersbacher S, Mian C, Maier U, Simak R. Validation of a 10-minute dipstick test for serum prostate-specific antigen. Eur Urol 1996; 30: Carvalhal GF et al. Digital rectal examination for detecting prostate cancer at prostate specific antigen levels of 4 ng/ml or less. JUrol1999; 161: Roehl KA, Antenor JA, Catalona WJ. Robustness of free prostate specific antigen measurements to reduce unnecessary biopsies in the ng/ml range. JUrol2002; 168: