- Email: [email protected]
PSA Velocity Is Associated With Gleason Score in Radical Prostatectomy Specimen: Marker for Prostate Cancer Aggressiveness
Oncology PSA Velocity Is Associated With Gleason Score in Radical Prostatectomy Specimen: Marker for Prostate Cancer Aggressiveness Stacy Loeb, Douglas E. Sutherland, Anthony V. D’Amico, Kimberly A. Roehl, and William J. Catalona OBJECTIVES
METHODS
RESULTS
CONCLUSIONS
Conflicting evidence has been reported on the association of prostate-specific antigen velocity (PSAV) with Gleason score in prostate needle biopsy specimens. The Gleason score is an important prognostic indicator for men with prostate cancer, and, in modern practice, it frequently affects treatment decisions. To our knowledge, the relationship between preoperative PSAV and Gleason score in the radical prostatectomy specimen has not been formally demonstrated. A total of 1049 men treated with radical prostatectomy had data on PSAV and Gleason score. Statistical analysis was performed to examine the relationship between the preoperative PSAV and the prostatectomy Gleason score and other adverse tumor features. The median preoperative PSAV was 0.84, 0.97, and 1.39 ng/mL/y in men with a Gleason score of 6, 7, and 8-10, respectively (P ⫽ .05). A PSAV greater than 2 ng/mL/y was significantly associated with a prostatectomy Gleason score of 7 or greater on univariate and multivariate analysis. In addition, the preoperative PSAV was significantly lower in men with organ-confined disease (0.82 vs 1.17 ng/mL/y, respectively, P ⫽ .002). Our results have further validated PSAV as a marker for prostate cancer aggressiveness. The preoperative PSAV was a significant independent predictor of the Gleason score and non– organconfined disease in the radical prostatectomy specimen. Thus, PSAV could be useful in treatment decision-making and in assessing the likelihood of long-term cancer control in men with prostate cancer. UROLOGY 72: 1116 –1120, 2008. © 2008 Elsevier Inc.
I
t has been estimated that 218 890 new cases of prostate cancer (CaP) will have been diagnosed in 2007.1 The prevalence of CaP as a result of prostatespecific antigen (PSA)-based screening has led to considerable investigation into clinical parameters that could help predict the aggressiveness of disease.2 Such markers would be useful for patient counseling and clinical decision-making, such as the choice between immediate intervention and active monitoring. The Gleason score is an important variable for treatment planning and prognostication in men with newly diagnosed CaP, regardless of the form of treatment under
This study was supported in part by the Urological Research Foundation and Beckman Coulter, Inc., Fullerton, CA. From the Department of Urology, Johns Hopkins School of Medicine, Baltimore, Maryland; Department of Urology, George Washington University School of Medicine, Washington, DC; Department of Radiation Oncology, Brigham and Women’s Hospital and Dana-Farber Cancer Institute, Boston, Massachusetts; Department of Psychiatry, Washington University School of Medicine, St. Louis, Missouri; and Department of Urology, Northwestern University Feinberg School of Medicine, Chicago, Illinois Reprint requests: William J. Catalona, M.D., Department of Urology, Northwestern University Feinberg School of Medicine, 675 North Saint Clair Street, Suite 20-150, Chicago, IL 60611. E-mail: [email protected] Submitted: September 3, 2007, accepted (with revisions): January 10, 2008
1116
© 2008 Elsevier Inc. All Rights Reserved
consideration. For men considering radical prostatectomy (RP), the Gleason score has been used in several nomograms to predict the likelihood of adverse pathologic features and biochemical progression postoperatively.3,4 Similarly, the Gleason score at diagnosis is a significant predictor of relapse-free survival after brachytherapy and external beam radiotherapy.5 Furthermore, the Gleason score is a useful indicator of disease aggressiveness for patients and physicians considering conservative management. For example, Albertsen et al.6 reported on 767 men with CaP who elected watchful waiting. After more than 20 years of follow-up, they reported only six CaP deaths per 1000 person-years for men with a Gleason score of 2-4. In contrast, men with Gleason score 8-10 tumors had 121 deaths per 1000 person-years within 10 years of diagnosis. Although there is little debate that the Gleason score is linked to treatment outcomes, the search continues for additional variables to more accurately identify those cancers that pose the greatest threat. Furthermore, the biopsy Gleason score is subject to sampling errors, and a considerable proportion of patients treated with RP have their score upgraded in the RP specimen. Thus, addi0090-4295/08/$34.00 doi:10.1016/j.urology.2008.01.082
tional preoperative markers for CaP aggressiveness could provide a useful adjunct to the biopsy Gleason score. Evidence is increasing of a link between PSAV and CaP aggressiveness. Specifically, a rapidly increasing PSA level appears to be an adverse prognostic factor for men with CaP. Studies have shown that PSAV is independently associated with survival after RP, external beam radiotherapy, and brachytherapy.7–11 Less is known about the precise relationship between the PSAV and the Gleason score. The purpose of the present study was, therefore, to analyze the association of a rapidly increasing PSA level and the Gleason score in the RP specimen to determine its utility as a surrogate marker for the biologic aggressiveness of CaP.
MATERIAL AND METHODS From 1989 to 2001, approximately 36 000 men participated in a community-based CaP screening study, as previously described.12 Serial PSA measurements and digital rectal examinations were performed at 6- to 12-month intervals. Until 1995, quadrant prostate biopsy was performed for a PSA level greater than 4.0 ng/mL or suspicious digital rectal examination findings. In May 1995, the PSA threshold for biopsy was reduced to 2.5 ng/mL, and at least sextant biopsies were performed. From the screening study, 1062 men underwent RP with sufficient preoperative PSA measurements to enable a PSAV calculation. Of the 1062 men, data on the “final” Gleason score in the RP specimen was missing in 13, resulting in a final study population of 1049 men. Ten of these men had missing data on Gleason score in the biopsy specimen, but all were included in all the other analyses. Also, 2 patients had a final Gleason score of 7 but incomplete data on the primary and secondary Gleason pattern. The PSAV was calculated by linear regression analysis of the PSA values from the year before diagnosis, as previously reported.7 The PSAV calculation included two PSA measurements in 909 (87%) and three or more PSA values in 140 (13%). Biochemical progression was defined as a postoperative PSA level greater than 0.2 ng/mL and confirmed by a second measurement. The Wilcoxon rank sum and Kruskall-Wallis tests were used to compare the median PSAV among the Gleason score strata. We compared the biopsy and RP Gleason scores using the kappa statistic, which is similar to a correlation coefficient, except that it also corrects for chance.13 Kappa values of less than 0, 0-0.2, 0.21-0.40, 0.41-0.60, 0.61-0.80, and 0.81-1.0 represent a poor, slight, fair, moderate, substantial, and almost perfect strength of agreement between two parameters beyond chance, respectively. We then used the 2 test and Fisher’s exact test to compare the proportion of men at greater than or less than each PSAV threshold who had high-grade disease, defined as a Gleason score of 7 or greater. Finally, we used multivariate logistic regression analysis to report the odds ratios (ORs) and 95% confidence intervals (CIs) for the prediction of high-grade disease in the RP specimen. The base model included the following covariates: PSAV, clinical stage, and race. We also performed a separate multivariate model that included the total PSA level. UROLOGY 72 (5), 2008
Table 1. Clinical and pathologic characteristics Characteristic Age Mean ⫾ SD Range Race (n) White Black Other PSA (ng/mL) Median Range Clinical stage (n) T1 T2 PSAV (ng/mL/y) Median Range PSA values for PSAV (n) Median Range Biopsy Gleason score (n) ⱕ6 7 8-10 RP Gleason score (n) ⱕ6 3⫹4⫽7 4⫹3⫽7 8-10 Organ-confined disease (n) ECE (n) Positive margins (n) SVI (n) Positive LNs (n)
Value 66 ⫾ 5.9 42-83 987 (95) 44 (4) 7 (1) 4.4 0.2-45.3 704 (67) 341 (33) 0.9 ⫺20.4 to 29.2 2 2-4 902 (87) 93 (9) 44 (4) 790 (75) 166 (16) 45 (4) 48 (5) 784 (75) 117 (11) 225 (22) 30 (3) 6 (1)
SD ⫽ standard deviation; PSA ⫽ prostate-specific antigen; PSAV ⫽ PSA velocity; RP ⫽ radical prostatectomy; ECE ⫽ extracapsular extension; SVI ⫽ seminal vesicle invasion; LNs ⫽ lymph nodes. Data in parentheses are percentages.
All statistical analysis was performed using Statistical Analysis Systems, version 8.2, for Linux (SAS Institute, Cary, NC).
RESULTS The study population included 1049 men treated with RP who had information on the final Gleason score and preoperative PSAV. Table 1 lists the clinicopathologic characteristics of our cohort. The mean age was 66 years, and most patients were white, with a median preoperative PSA level of 4.4 ng/mL. The mean preoperative PSAV in the overall cohort was 1.34 ng/mL/y (median 0.90). Most men had a biopsy Gleason score of 6 or less and clinical Stage T1c CaP. In the RP specimen, 75% of men had organ-confined disease, and 25% had a final Gleason score greater than 6 (Table 1). Table 2 compares the biopsy and RP Gleason scores. The kappa statistic was 0.3 (P ⬍ .0001), indicating only a fair strength of agreement between the parameters. Of the men with a Gleason score of 6 or less on biopsy, 125 (14%) were upgraded to 3⫹4⫽7, 24 (3%) to 4⫹3⫽7, and 14 (2%) to a Gleason score of 8-10 in the RP specimen. 1117
Table 2. Comparison between biopsy and RP Gleason scores
Biopsy Gleason Score ⱕ6 7 8-10
ⱕ6
RP Gleason Score 7 8-10
739 (82) 32 (35) 11 (25)
149 (16) 45 (48) 15 (34)
14 (2) 16 (17) 18 (41)
RP ⫽ radical prostatectomy. Data in parentheses are percentages. Kappa ⫽ 0.30 (0.24, 0.36); P ⬍ .0001.
The preoperative PSAV had a significant direct association with the final Gleason score. For example, the median preoperative PSAV was 0.84 ng/mL/y for men with a prostatectomy Gleason score of 6 or less compared with 1.11 ng/mL/y for those with a Gleason score of 7-10 (P ⫽ .04). Stratified further, the median preoperative PSAV was 0.97 ng/mL/y in men with Gleason score 7 CaP vs 1.39 ng/mL/y for those with a final Gleason score of 8-10 (P ⫽ .05). Even among the men with a Gleason score of 7, the preoperative PSAV was 0.94 ng/mL/y for those with 3⫹4⫽7 vs 1.04 ng/mL/y for those with 4⫹3⫽7. In the 140 men with a PSAV calculation determined using at least three PSA values, a significant association remained between the preoperative PSAV and the RP Gleason score. In this subset, the median PSAV was 0.6 ng/mL/y in men with a final Gleason score of 6 or less and was 1.3 ng/mL/y for those with a Gleason score of 7-10 (P ⫽ .01). A final Gleason score of 7 or greater was present in 112 men (23%) with a PSAV of 0.75 ng/mL/y or less compared with 147 men (26%) with a greater PSAV (P ⫽ .2). Stratifying by the median PSAV of 0.895 ng/mL/y, a Gleason score of 7 or greater was found in 118 men (22%) with a lower PSAV compared with 141 (27%) with a greater PSAV (P ⫽ .1). A Gleason score of 7 or greater was found in 123 men (22%) with a PSAV of 1 ng/mL/y or less vs 136 men (28%) with a PSAV greater than 1 ng/mL/y (P ⫽ .04). Finally, a Gleason score of 7 or greater was found in 177 men (23%) with a PSAV of 2 ng/mL/y or less vs 82 men (29%) with a PSAV greater than 2 ng/mL/y (P ⫽ .04). This cutpoint was associated with a sensitivity, specificity, positive predictive value, and negative predictive value of 32%, 75%, 29%, and 77%, respectively, for the prediction of high-grade disease. On multivariate analysis to predict a Gleason score of 7 or greater, a preoperative PSAV greater than 2 ng/mL/y (OR 1.4, 95% CI 1.1-2.0, P ⫽ .02) was a stronger predictor in this cohort than either clinical stage (OR 1.1, 95% CI 0.8-1.5, P ⫽ .54) or African-American heritage (OR 1.2, 95% CI 0.6-2.4, P ⫽ .55). In a separate model that included the total PSA level, both PSAV and PSA remained significant independent predictors of high-grade disease (P ⬍ .05 for both). Of the 1043 men with complete pathologic stage information, 259 (25%) had non– organ-confined disease. 1118
Specifically, extracapsular tumor extension was present in 117 (11%), seminal vesicle invasion in 30 (3%), lymph node metastases in 6 (1%), and positive surgical margins in 225 men (22%). The preoperative PSAV was 0.81 ng/mL/yr in men with organ-confined disease compared with 1.17 ng/mL/y in men with non– organ-confined disease (P ⫽ .001). Excluding men with positive surgical margins alone from the classification of non– organ-confined disease, the preoperative PSAV was 0.85 and 1.22 ng/mL/y in men with and without organ-confined disease, respectively (P ⫽ .002). At a mean follow-up of 85 months (median 85), 146 men (14%) had biochemical progression. The Gleason score among men with progression was 6 or less in 87 (59%), 3⫹4⫽7 in 35 (24%), 4⫹3⫽7 in 8 (5%), and 8-10 in 16 (11%). Including only men with progression in the analysis, a significant relationship was found between an increasing PSAV and high-grade disease (0.64 vs 1.48 ng/mL/y for Gleason score 6 vs 7-10, P ⫽ .02).
COMMENT The concept of using the PSAV to predict treatment outcomes has received considerable attention. In 2004, our research group, in collaboration with D’Amico et al.,7 reported on the association between preoperative PSAV and CaP outcomes in 1095 men treated with RP. Specifically, men with a preoperative PSAV greater than 2 ng/mL/y were 9.8 times more likely to die of CaP than men with a lower PSAV (P ⬍ .001). In a subsequent study, D’Amico et al.8 reported on the association between the preoperative PSAV and treatment outcomes in 358 men with localized CaP treated by external beam radiotherapy. A pretreatment PSAV greater than 2 ng/mL/y was associated with a significantly shorter time to CaP-specific mortality (adjusted hazard ratio 12.0, P ⫽ .001). Most recently, our group validated the association between the pretreatment PSAV and the outcome of primary radiotherapy in 83 men treated with external beam radiotherapy and 47 men treated with brachytherapy.11 At a mean follow-up of 64 months, the men with a pretreatment PSAV greater than 2 ng/mL/y had a fourfold increased risk of biochemical progression. The 6-year progression-free survival rate was 57% and 82% for men with a pretreatment PSAV of 2 ng/mL/y or greater and less than 2 ng/mL/y, respectively (P ⬍ .001). Relatively little published information is available about the specific relationship between the PSAV and the Gleason score. A few previous studies have examined the association between PSAV and the findings on prostate biopsy. For example, Schroder et al.14 recently reported on the PSAV in 588 men from the European Randomized Study of Screening for Prostate Cancer. Each of these men underwent biopsy after the second screening visit for an increase in PSA to a level greater than 4.0 ng/mL within the 4-year interval after a negative initial screening. The PSAV was calculated by subUROLOGY 72 (5), 2008
tracting the difference between the first and second PSA values and dividing by the 4-year interval between them. Men were classified as having “aggressive” disease if they had clinical Stage T1c or greater tumor with a Gleason score of 7-10. As the PSAV increased in the ranges of greater than 0.25, greater than 0.50, greater than 0.75, and greater than 1 ng/mL/y, the proportion of men with these “aggressive” preoperative features progressively increased (55%, 60%, 65%, and 65%, respectively). Thompson et al.15 reported on 5519 men from the placebo group of the Prostate Cancer Prevention Trial who had undergone biopsy during the study or at the end of the 7-year study period. They calculated the PSAV by linear regression analysis involving all PSA values within the 3 years before the biopsy. They reported that an increasing PSAV was associated with a significantly greater risk of high-grade (Gleason score 7 or greater) disease on biopsy (OR 8.93, 95% CI 5.71-13.97, P ⬍ .001). However, on multivariate analysis with other clinical variables, PSAV lost its statistical significance (OR 0.82, 95% CI 0.44-1.53, P ⫽ .54). Finally, Berger et al.16 reported on the PSAV in the 6 years before CaP diagnosis in men from the screening population in Tyrol, Austria. Among men with a biopsy Gleason score of 6 or less, 7, and 8-10, the median PSAV was 0.34, 0.46, and 0.74 ng/mL/y, respectively (P ⬍ .001). Thus, evidence is conflicting regarding the relationship between the PSAV and Gleason score. Nevertheless, these analyses were based on the biopsy report, which does not always correspond to the final Gleason score at RP. Chun et al.17 studied the correlation between biopsy and RP Gleason score in 4789 men and found significant upgrading in 1349 patients (28.2%). To our knowledge, the present study is the first to report specifically on the association between the preoperative PSAV and the pathologic Gleason score in a large screening population. Our findings of a strong association between the PSAV and the pathologic Gleason score should therefore be pertinent to men with newly diagnosed CaP, as they are confronted with the numerous treatment options. For example, our data suggest that a man with a biopsy Gleason score of 6 and a PSAV greater than 2 ng/mL/y is likely to have higher grade disease in the RP specimen. Such an individual might be at greater risk of disease progression in the absence of definitive therapy. Future prospective studies are warranted to address the role, if any, of PSAV in active monitoring protocols. Several limitations of our study deserve mention. First, data on the RP Gleason score and sufficient PSA measurements to calculate the PSAV were not available for all participants in the screening study. However, the median follow-up, biopsy Gleason score, and RP Gleason score were similar between the included and excluded men (data not shown), suggesting that this was unlikely to have materially affected the results. Second, the agreeUROLOGY 72 (5), 2008
ment between biopsy and RP Gleason scores would likely have been better with a more extended initial biopsy protocol. Nevertheless, our limited biopsy protocol should not have affected the primary outcome of this analysis, which was the utility of the preoperative PSAV to predict the RP Gleason score. That notwithstanding, that our analysis only included men who were treated with RP introduced a selection bias. In addition, many men in the screening study were diagnosed with CaP at the initial visit; thus, it was only possible to calculate the PSAV in men diagnosed through serial screening. Few men with cancer detected by serial screening had highgrade disease, limiting our ability to make statistical comparisons between groups. One strength of our study was that the annual PSA measurements were made using the same assay platform. Although this represents the ideal situation, in the real world setting differences between assays can confound the use of PSA kinetics. Another limitation was that the PSAV was calculated using PSA values from a 1-year period, because our group, and others, have previously shown that PSAV calculated in this manner correlates with cancer-specific survival after treatment. Nevertheless, longitudinal measurements for a longer interval are also valuable. Finally, the PSAV was not itself an indication for biopsy in our screening protocol. Thus, our results need to be validated in a prospective study.
CONCLUSIONS The results of our study have shown that in men with newly diagnosed CaP, a high PSAV is associated with a significantly greater risk of high-grade disease in the RP specimen. The strong association between the pretreatment PSAV and the pathologic Gleason score could be useful for patient counseling and treatment decisions. In particular, men with a PSAV greater than 2 ng/mL/y, if not caused by confounding from prostatic inflammation, assay variability, or random biologic variation, are significantly more likely to have adverse pathologic features, and should be treated accordingly. These results are in agreement with previous studies suggesting that the PSAV is a marker for CaP aggressiveness. References 1. American Cancer Society. Cancer Facts and Figures 2007. Available from: www.cancer.org/downloads/STT/CAFF2007PWSecured.pdf. Accessed January 28, 2007. 2. Epstein JI, Walsh PC, Carmichael M, et al. Pathologic and clinical findings to predict tumor extent of nonpalpable (stage T1c) prostate cancer. JAMA. 1994;271:368-374. 3. Kattan MW, Eastham JA, Stapleton AM, et al. A preoperative nomogram for disease recurrence following radical prostatectomy for prostate cancer. J Natl Cancer Inst. 1998;90:766-771. 4. Partin AW, Yoo J, Carter HB, et al. The use of prostate specific antigen, clinical stage and Gleason score to predict pathological stage in men with localized prostate cancer. J Urol. 1993;150:110114.
1119
5. Sylvester JE, Blasko JC, Grimm PD, et al. Ten-year biochemical relapse-free survival after external beam radiation and brachytherapy for localized prostate cancer: The Seattle experience. Int J Radiat Oncol Biol Phys. 2003;57:944-952. 6. Albertsen PC, Hanley JA, Fine J. 20-year outcomes following conservative management of clinically localized prostate cancer. JAMA. 2005;293:2095-2101. 7. D’Amico AV, Chen MH, Roehl KA, et al. Preoperative PSA velocity and the risk of death from prostate cancer after radical prostatectomy. N Engl J Med. 2004;351:125-135. 8. D’Amico AV, Renshaw AA, Sussman B, et al. Pretreatment PSA velocity and risk of death from prostate cancer following external beam radiation therapy. JAMA. 2005;294:440-447. 9. Sengupta S, Myers RP, Slezak JM, et al. Preoperative prostate specific antigen doubling time and velocity are strong and independent predictors of outcomes following radical prostatectomy. J Urol. 2005;174:2191-2196. 10. Patel DA, Presti JC Jr, McNeal JE, et al. Preoperative PSA velocity is an independent prognostic factor for relapse after radical prostatectomy. J Clin Oncol. 2005;23:6157-6162. 11. Eggener SE, Roehl KA, Yossepowitch O, et al. Prediagnosis prostate specific antigen velocity is associated with risk of prostate cancer progression following brachytherapy and external beam radiation therapy. J Urol. 2006;176:1399-1403. 12. Smith DS, Catalona WJ. Rate of change in serum prostate specific antigen levels as a method for prostate cancer detection. J Urol. 1994;152:1163-1167. 13. Landis JR, Koch GG. An application of hierarchical kappa-type statistics in the assessment of majority agreement among multiple observers. Biometrics. 1977;33:363-374. 14. Schroder FH, Roobol MJ, van der Kwast TH, et al. Does PSA velocity predict prostate cancer in pre-screened populations? Eur Urol. 2006;49:460-465. 15. Thompson IM, Ankerst DP, Chi C, et al. Assessing prostate cancer risk: Results from the Prostate Cancer Prevention Trial. J Natl Cancer Inst. 2006;98:529-534. 16. Berger AP, Deibl M, Strasak A, et al. Large-scale study of clinical impact of PSA velocity: Long-term PSA kinetics as method of differentiating men with from those without prostate cancer. Urology. 2007;69:134-138. 17. Chun FK, Briganti A, Shariat SF, et al. Significant upgrading affects a third of men diagnosed with prostate cancer: predictive nomogram and internal validation. BJU Int. 2006;98:329-334.
from CaP in the decade after CaP surgery.1 Compared with men with a PSAV of 2.0 ng/mL/y or less, those with a PSAV greater than 2.0 ng/mL/y in the year before diagnosis had a 10-fold increased risk of CaP death in the decade after treatment. Therefore, it seems reasonable to conclude that PSAV is associated with a more aggressive CaP phenotype. The question remaining is how to use this information in clinical practice to inform patients about treatment or the need for a prostate biopsy. As the authors point out, men with a Gleason score of 6 are not infrequently upgraded at RP to a higher Gleason score (Table 2), and a rapidly increasing PSA level provides evidence that a patient could be harboring a more aggressive CaP (despite the absence of high-grade disease on biopsy) for which surveillance might be inappropriate. In their report, even a PSAV greater than 1.0 ng/mL/y was significantly associated with a Gleason score of 7 or greater. On the basis of the work of these authors and others, I would be concerned about the safety of surveillance in a man with a consistently increasing PSA level, especially if the PSAV were more than 1 ng/mL/y, even though some have advocated that a PSA doubling time of more than 3 years is a safe threshold for entering a surveillance program.2,3 Given that no PSA level exists below which we can guarantee a man that he does not have lethal CaP, the greatest value of the PSAV could be among men who have PSA levels that have not reached a threshold level considered to be concerning. I believe that the man whose PSA level increases from 0.5 to 1.5 ng/ml during 1 year (a rapid increase) is at a greater risk of harboring lethal CaP than the man whose PSA level remains at 1.5 ng/mL from year to year.4 Although some believe that the PSAV has no clinical usefulness in terms of diagnosis,5 I believe that a rapid increase in PSA level should trigger a prostate biopsy in the absence of evidence of prostatic inflammation. Whether this approach will improve overall health outcomes is as yet unproved. H. Ballentine Carter, M.D., Department of Urology, Johns Hopkins University School of Medicine, James Buchanan Brady Urological Institute, Johns Hopkins Hospital, Baltimore, Maryland
EDITORIAL COMMENT Most urologists believe that early detection of CaP with PSA testing saves lives and uncovers some cases of CaP that would never have caused harm—a tradeoff that is unavoidable for now. Thus, the identification before treatment of those cancers that pose a threat to life and distinguishing them from those that do not is an important goal. In their study, Loeb et al. have shown that the PSAV (the rate of change in PSA) in the year before diagnosis is directly and independently associated with the Gleason score in the final pathologic specimen after RP for CaP and significantly associated with the disease extent (organ confined vs nonorgan confined). A PSAV greater than 2.0 ng/mL/y was a stronger predictor of high-grade disease (Gleason score of 7 or greater) than was clinical stage and black race and remained a significant predictor when accounting for PSA level. These data are consistent with their previous work demonstrating that the PSAV in the year before diagnosis was associated with death
1120
References 1. D’Amico AV, Chen MH, Roehl KA, et al. Preoperative PSA velocity and the risk of death from prostate cancer after radical prostatectomy. N Engl J Med. 2004;351:125-135. 2. van den Bergh RC, Roemeling S, Roobol MJ, et al. Prospective validation of active surveillance in prostate cancer: The PRIAS study. Eur Urol. 2007;52:1560-1563. 3. Klotz L. Active surveillance for favorable-risk prostate cancer: Who, how and why? Nat Clin Pract Oncol. 2007;4:692-698. 4. Carter HB, Ferrucci L, Kettermann A, et al. Detection of lifethreatening prostate cancer with prostate-specific antigen velocity during a window of curability J Natl Cancer Inst. 2006;98:1521-1527. 5. Etzioni RD, Ankerst DP, Weiss NS, et al. Is prostate-specific antigen velocity useful in early detection of prostate cancer? A critical appraisal of the evidence. J Natl Cancer Inst. 2007;99:1510-1515.
doi:10.1016/j.urology.2008.02.069 UROLOGY 72: 1116 –1120, 2008. © 2008 Elsevier Inc.
UROLOGY 72 (5), 2008
METHODS
RESULTS
CONCLUSIONS
Conflicting evidence has been reported on the association of prostate-specific antigen velocity (PSAV) with Gleason score in prostate needle biopsy specimens. The Gleason score is an important prognostic indicator for men with prostate cancer, and, in modern practice, it frequently affects treatment decisions. To our knowledge, the relationship between preoperative PSAV and Gleason score in the radical prostatectomy specimen has not been formally demonstrated. A total of 1049 men treated with radical prostatectomy had data on PSAV and Gleason score. Statistical analysis was performed to examine the relationship between the preoperative PSAV and the prostatectomy Gleason score and other adverse tumor features. The median preoperative PSAV was 0.84, 0.97, and 1.39 ng/mL/y in men with a Gleason score of 6, 7, and 8-10, respectively (P ⫽ .05). A PSAV greater than 2 ng/mL/y was significantly associated with a prostatectomy Gleason score of 7 or greater on univariate and multivariate analysis. In addition, the preoperative PSAV was significantly lower in men with organ-confined disease (0.82 vs 1.17 ng/mL/y, respectively, P ⫽ .002). Our results have further validated PSAV as a marker for prostate cancer aggressiveness. The preoperative PSAV was a significant independent predictor of the Gleason score and non– organconfined disease in the radical prostatectomy specimen. Thus, PSAV could be useful in treatment decision-making and in assessing the likelihood of long-term cancer control in men with prostate cancer. UROLOGY 72: 1116 –1120, 2008. © 2008 Elsevier Inc.
I
t has been estimated that 218 890 new cases of prostate cancer (CaP) will have been diagnosed in 2007.1 The prevalence of CaP as a result of prostatespecific antigen (PSA)-based screening has led to considerable investigation into clinical parameters that could help predict the aggressiveness of disease.2 Such markers would be useful for patient counseling and clinical decision-making, such as the choice between immediate intervention and active monitoring. The Gleason score is an important variable for treatment planning and prognostication in men with newly diagnosed CaP, regardless of the form of treatment under
This study was supported in part by the Urological Research Foundation and Beckman Coulter, Inc., Fullerton, CA. From the Department of Urology, Johns Hopkins School of Medicine, Baltimore, Maryland; Department of Urology, George Washington University School of Medicine, Washington, DC; Department of Radiation Oncology, Brigham and Women’s Hospital and Dana-Farber Cancer Institute, Boston, Massachusetts; Department of Psychiatry, Washington University School of Medicine, St. Louis, Missouri; and Department of Urology, Northwestern University Feinberg School of Medicine, Chicago, Illinois Reprint requests: William J. Catalona, M.D., Department of Urology, Northwestern University Feinberg School of Medicine, 675 North Saint Clair Street, Suite 20-150, Chicago, IL 60611. E-mail: [email protected] Submitted: September 3, 2007, accepted (with revisions): January 10, 2008
1116
© 2008 Elsevier Inc. All Rights Reserved
consideration. For men considering radical prostatectomy (RP), the Gleason score has been used in several nomograms to predict the likelihood of adverse pathologic features and biochemical progression postoperatively.3,4 Similarly, the Gleason score at diagnosis is a significant predictor of relapse-free survival after brachytherapy and external beam radiotherapy.5 Furthermore, the Gleason score is a useful indicator of disease aggressiveness for patients and physicians considering conservative management. For example, Albertsen et al.6 reported on 767 men with CaP who elected watchful waiting. After more than 20 years of follow-up, they reported only six CaP deaths per 1000 person-years for men with a Gleason score of 2-4. In contrast, men with Gleason score 8-10 tumors had 121 deaths per 1000 person-years within 10 years of diagnosis. Although there is little debate that the Gleason score is linked to treatment outcomes, the search continues for additional variables to more accurately identify those cancers that pose the greatest threat. Furthermore, the biopsy Gleason score is subject to sampling errors, and a considerable proportion of patients treated with RP have their score upgraded in the RP specimen. Thus, addi0090-4295/08/$34.00 doi:10.1016/j.urology.2008.01.082
tional preoperative markers for CaP aggressiveness could provide a useful adjunct to the biopsy Gleason score. Evidence is increasing of a link between PSAV and CaP aggressiveness. Specifically, a rapidly increasing PSA level appears to be an adverse prognostic factor for men with CaP. Studies have shown that PSAV is independently associated with survival after RP, external beam radiotherapy, and brachytherapy.7–11 Less is known about the precise relationship between the PSAV and the Gleason score. The purpose of the present study was, therefore, to analyze the association of a rapidly increasing PSA level and the Gleason score in the RP specimen to determine its utility as a surrogate marker for the biologic aggressiveness of CaP.
MATERIAL AND METHODS From 1989 to 2001, approximately 36 000 men participated in a community-based CaP screening study, as previously described.12 Serial PSA measurements and digital rectal examinations were performed at 6- to 12-month intervals. Until 1995, quadrant prostate biopsy was performed for a PSA level greater than 4.0 ng/mL or suspicious digital rectal examination findings. In May 1995, the PSA threshold for biopsy was reduced to 2.5 ng/mL, and at least sextant biopsies were performed. From the screening study, 1062 men underwent RP with sufficient preoperative PSA measurements to enable a PSAV calculation. Of the 1062 men, data on the “final” Gleason score in the RP specimen was missing in 13, resulting in a final study population of 1049 men. Ten of these men had missing data on Gleason score in the biopsy specimen, but all were included in all the other analyses. Also, 2 patients had a final Gleason score of 7 but incomplete data on the primary and secondary Gleason pattern. The PSAV was calculated by linear regression analysis of the PSA values from the year before diagnosis, as previously reported.7 The PSAV calculation included two PSA measurements in 909 (87%) and three or more PSA values in 140 (13%). Biochemical progression was defined as a postoperative PSA level greater than 0.2 ng/mL and confirmed by a second measurement. The Wilcoxon rank sum and Kruskall-Wallis tests were used to compare the median PSAV among the Gleason score strata. We compared the biopsy and RP Gleason scores using the kappa statistic, which is similar to a correlation coefficient, except that it also corrects for chance.13 Kappa values of less than 0, 0-0.2, 0.21-0.40, 0.41-0.60, 0.61-0.80, and 0.81-1.0 represent a poor, slight, fair, moderate, substantial, and almost perfect strength of agreement between two parameters beyond chance, respectively. We then used the 2 test and Fisher’s exact test to compare the proportion of men at greater than or less than each PSAV threshold who had high-grade disease, defined as a Gleason score of 7 or greater. Finally, we used multivariate logistic regression analysis to report the odds ratios (ORs) and 95% confidence intervals (CIs) for the prediction of high-grade disease in the RP specimen. The base model included the following covariates: PSAV, clinical stage, and race. We also performed a separate multivariate model that included the total PSA level. UROLOGY 72 (5), 2008
Table 1. Clinical and pathologic characteristics Characteristic Age Mean ⫾ SD Range Race (n) White Black Other PSA (ng/mL) Median Range Clinical stage (n) T1 T2 PSAV (ng/mL/y) Median Range PSA values for PSAV (n) Median Range Biopsy Gleason score (n) ⱕ6 7 8-10 RP Gleason score (n) ⱕ6 3⫹4⫽7 4⫹3⫽7 8-10 Organ-confined disease (n) ECE (n) Positive margins (n) SVI (n) Positive LNs (n)
Value 66 ⫾ 5.9 42-83 987 (95) 44 (4) 7 (1) 4.4 0.2-45.3 704 (67) 341 (33) 0.9 ⫺20.4 to 29.2 2 2-4 902 (87) 93 (9) 44 (4) 790 (75) 166 (16) 45 (4) 48 (5) 784 (75) 117 (11) 225 (22) 30 (3) 6 (1)
SD ⫽ standard deviation; PSA ⫽ prostate-specific antigen; PSAV ⫽ PSA velocity; RP ⫽ radical prostatectomy; ECE ⫽ extracapsular extension; SVI ⫽ seminal vesicle invasion; LNs ⫽ lymph nodes. Data in parentheses are percentages.
All statistical analysis was performed using Statistical Analysis Systems, version 8.2, for Linux (SAS Institute, Cary, NC).
RESULTS The study population included 1049 men treated with RP who had information on the final Gleason score and preoperative PSAV. Table 1 lists the clinicopathologic characteristics of our cohort. The mean age was 66 years, and most patients were white, with a median preoperative PSA level of 4.4 ng/mL. The mean preoperative PSAV in the overall cohort was 1.34 ng/mL/y (median 0.90). Most men had a biopsy Gleason score of 6 or less and clinical Stage T1c CaP. In the RP specimen, 75% of men had organ-confined disease, and 25% had a final Gleason score greater than 6 (Table 1). Table 2 compares the biopsy and RP Gleason scores. The kappa statistic was 0.3 (P ⬍ .0001), indicating only a fair strength of agreement between the parameters. Of the men with a Gleason score of 6 or less on biopsy, 125 (14%) were upgraded to 3⫹4⫽7, 24 (3%) to 4⫹3⫽7, and 14 (2%) to a Gleason score of 8-10 in the RP specimen. 1117
Table 2. Comparison between biopsy and RP Gleason scores
Biopsy Gleason Score ⱕ6 7 8-10
ⱕ6
RP Gleason Score 7 8-10
739 (82) 32 (35) 11 (25)
149 (16) 45 (48) 15 (34)
14 (2) 16 (17) 18 (41)
RP ⫽ radical prostatectomy. Data in parentheses are percentages. Kappa ⫽ 0.30 (0.24, 0.36); P ⬍ .0001.
The preoperative PSAV had a significant direct association with the final Gleason score. For example, the median preoperative PSAV was 0.84 ng/mL/y for men with a prostatectomy Gleason score of 6 or less compared with 1.11 ng/mL/y for those with a Gleason score of 7-10 (P ⫽ .04). Stratified further, the median preoperative PSAV was 0.97 ng/mL/y in men with Gleason score 7 CaP vs 1.39 ng/mL/y for those with a final Gleason score of 8-10 (P ⫽ .05). Even among the men with a Gleason score of 7, the preoperative PSAV was 0.94 ng/mL/y for those with 3⫹4⫽7 vs 1.04 ng/mL/y for those with 4⫹3⫽7. In the 140 men with a PSAV calculation determined using at least three PSA values, a significant association remained between the preoperative PSAV and the RP Gleason score. In this subset, the median PSAV was 0.6 ng/mL/y in men with a final Gleason score of 6 or less and was 1.3 ng/mL/y for those with a Gleason score of 7-10 (P ⫽ .01). A final Gleason score of 7 or greater was present in 112 men (23%) with a PSAV of 0.75 ng/mL/y or less compared with 147 men (26%) with a greater PSAV (P ⫽ .2). Stratifying by the median PSAV of 0.895 ng/mL/y, a Gleason score of 7 or greater was found in 118 men (22%) with a lower PSAV compared with 141 (27%) with a greater PSAV (P ⫽ .1). A Gleason score of 7 or greater was found in 123 men (22%) with a PSAV of 1 ng/mL/y or less vs 136 men (28%) with a PSAV greater than 1 ng/mL/y (P ⫽ .04). Finally, a Gleason score of 7 or greater was found in 177 men (23%) with a PSAV of 2 ng/mL/y or less vs 82 men (29%) with a PSAV greater than 2 ng/mL/y (P ⫽ .04). This cutpoint was associated with a sensitivity, specificity, positive predictive value, and negative predictive value of 32%, 75%, 29%, and 77%, respectively, for the prediction of high-grade disease. On multivariate analysis to predict a Gleason score of 7 or greater, a preoperative PSAV greater than 2 ng/mL/y (OR 1.4, 95% CI 1.1-2.0, P ⫽ .02) was a stronger predictor in this cohort than either clinical stage (OR 1.1, 95% CI 0.8-1.5, P ⫽ .54) or African-American heritage (OR 1.2, 95% CI 0.6-2.4, P ⫽ .55). In a separate model that included the total PSA level, both PSAV and PSA remained significant independent predictors of high-grade disease (P ⬍ .05 for both). Of the 1043 men with complete pathologic stage information, 259 (25%) had non– organ-confined disease. 1118
Specifically, extracapsular tumor extension was present in 117 (11%), seminal vesicle invasion in 30 (3%), lymph node metastases in 6 (1%), and positive surgical margins in 225 men (22%). The preoperative PSAV was 0.81 ng/mL/yr in men with organ-confined disease compared with 1.17 ng/mL/y in men with non– organ-confined disease (P ⫽ .001). Excluding men with positive surgical margins alone from the classification of non– organ-confined disease, the preoperative PSAV was 0.85 and 1.22 ng/mL/y in men with and without organ-confined disease, respectively (P ⫽ .002). At a mean follow-up of 85 months (median 85), 146 men (14%) had biochemical progression. The Gleason score among men with progression was 6 or less in 87 (59%), 3⫹4⫽7 in 35 (24%), 4⫹3⫽7 in 8 (5%), and 8-10 in 16 (11%). Including only men with progression in the analysis, a significant relationship was found between an increasing PSAV and high-grade disease (0.64 vs 1.48 ng/mL/y for Gleason score 6 vs 7-10, P ⫽ .02).
COMMENT The concept of using the PSAV to predict treatment outcomes has received considerable attention. In 2004, our research group, in collaboration with D’Amico et al.,7 reported on the association between preoperative PSAV and CaP outcomes in 1095 men treated with RP. Specifically, men with a preoperative PSAV greater than 2 ng/mL/y were 9.8 times more likely to die of CaP than men with a lower PSAV (P ⬍ .001). In a subsequent study, D’Amico et al.8 reported on the association between the preoperative PSAV and treatment outcomes in 358 men with localized CaP treated by external beam radiotherapy. A pretreatment PSAV greater than 2 ng/mL/y was associated with a significantly shorter time to CaP-specific mortality (adjusted hazard ratio 12.0, P ⫽ .001). Most recently, our group validated the association between the pretreatment PSAV and the outcome of primary radiotherapy in 83 men treated with external beam radiotherapy and 47 men treated with brachytherapy.11 At a mean follow-up of 64 months, the men with a pretreatment PSAV greater than 2 ng/mL/y had a fourfold increased risk of biochemical progression. The 6-year progression-free survival rate was 57% and 82% for men with a pretreatment PSAV of 2 ng/mL/y or greater and less than 2 ng/mL/y, respectively (P ⬍ .001). Relatively little published information is available about the specific relationship between the PSAV and the Gleason score. A few previous studies have examined the association between PSAV and the findings on prostate biopsy. For example, Schroder et al.14 recently reported on the PSAV in 588 men from the European Randomized Study of Screening for Prostate Cancer. Each of these men underwent biopsy after the second screening visit for an increase in PSA to a level greater than 4.0 ng/mL within the 4-year interval after a negative initial screening. The PSAV was calculated by subUROLOGY 72 (5), 2008
tracting the difference between the first and second PSA values and dividing by the 4-year interval between them. Men were classified as having “aggressive” disease if they had clinical Stage T1c or greater tumor with a Gleason score of 7-10. As the PSAV increased in the ranges of greater than 0.25, greater than 0.50, greater than 0.75, and greater than 1 ng/mL/y, the proportion of men with these “aggressive” preoperative features progressively increased (55%, 60%, 65%, and 65%, respectively). Thompson et al.15 reported on 5519 men from the placebo group of the Prostate Cancer Prevention Trial who had undergone biopsy during the study or at the end of the 7-year study period. They calculated the PSAV by linear regression analysis involving all PSA values within the 3 years before the biopsy. They reported that an increasing PSAV was associated with a significantly greater risk of high-grade (Gleason score 7 or greater) disease on biopsy (OR 8.93, 95% CI 5.71-13.97, P ⬍ .001). However, on multivariate analysis with other clinical variables, PSAV lost its statistical significance (OR 0.82, 95% CI 0.44-1.53, P ⫽ .54). Finally, Berger et al.16 reported on the PSAV in the 6 years before CaP diagnosis in men from the screening population in Tyrol, Austria. Among men with a biopsy Gleason score of 6 or less, 7, and 8-10, the median PSAV was 0.34, 0.46, and 0.74 ng/mL/y, respectively (P ⬍ .001). Thus, evidence is conflicting regarding the relationship between the PSAV and Gleason score. Nevertheless, these analyses were based on the biopsy report, which does not always correspond to the final Gleason score at RP. Chun et al.17 studied the correlation between biopsy and RP Gleason score in 4789 men and found significant upgrading in 1349 patients (28.2%). To our knowledge, the present study is the first to report specifically on the association between the preoperative PSAV and the pathologic Gleason score in a large screening population. Our findings of a strong association between the PSAV and the pathologic Gleason score should therefore be pertinent to men with newly diagnosed CaP, as they are confronted with the numerous treatment options. For example, our data suggest that a man with a biopsy Gleason score of 6 and a PSAV greater than 2 ng/mL/y is likely to have higher grade disease in the RP specimen. Such an individual might be at greater risk of disease progression in the absence of definitive therapy. Future prospective studies are warranted to address the role, if any, of PSAV in active monitoring protocols. Several limitations of our study deserve mention. First, data on the RP Gleason score and sufficient PSA measurements to calculate the PSAV were not available for all participants in the screening study. However, the median follow-up, biopsy Gleason score, and RP Gleason score were similar between the included and excluded men (data not shown), suggesting that this was unlikely to have materially affected the results. Second, the agreeUROLOGY 72 (5), 2008
ment between biopsy and RP Gleason scores would likely have been better with a more extended initial biopsy protocol. Nevertheless, our limited biopsy protocol should not have affected the primary outcome of this analysis, which was the utility of the preoperative PSAV to predict the RP Gleason score. That notwithstanding, that our analysis only included men who were treated with RP introduced a selection bias. In addition, many men in the screening study were diagnosed with CaP at the initial visit; thus, it was only possible to calculate the PSAV in men diagnosed through serial screening. Few men with cancer detected by serial screening had highgrade disease, limiting our ability to make statistical comparisons between groups. One strength of our study was that the annual PSA measurements were made using the same assay platform. Although this represents the ideal situation, in the real world setting differences between assays can confound the use of PSA kinetics. Another limitation was that the PSAV was calculated using PSA values from a 1-year period, because our group, and others, have previously shown that PSAV calculated in this manner correlates with cancer-specific survival after treatment. Nevertheless, longitudinal measurements for a longer interval are also valuable. Finally, the PSAV was not itself an indication for biopsy in our screening protocol. Thus, our results need to be validated in a prospective study.
CONCLUSIONS The results of our study have shown that in men with newly diagnosed CaP, a high PSAV is associated with a significantly greater risk of high-grade disease in the RP specimen. The strong association between the pretreatment PSAV and the pathologic Gleason score could be useful for patient counseling and treatment decisions. In particular, men with a PSAV greater than 2 ng/mL/y, if not caused by confounding from prostatic inflammation, assay variability, or random biologic variation, are significantly more likely to have adverse pathologic features, and should be treated accordingly. These results are in agreement with previous studies suggesting that the PSAV is a marker for CaP aggressiveness. References 1. American Cancer Society. Cancer Facts and Figures 2007. Available from: www.cancer.org/downloads/STT/CAFF2007PWSecured.pdf. Accessed January 28, 2007. 2. Epstein JI, Walsh PC, Carmichael M, et al. Pathologic and clinical findings to predict tumor extent of nonpalpable (stage T1c) prostate cancer. JAMA. 1994;271:368-374. 3. Kattan MW, Eastham JA, Stapleton AM, et al. A preoperative nomogram for disease recurrence following radical prostatectomy for prostate cancer. J Natl Cancer Inst. 1998;90:766-771. 4. Partin AW, Yoo J, Carter HB, et al. The use of prostate specific antigen, clinical stage and Gleason score to predict pathological stage in men with localized prostate cancer. J Urol. 1993;150:110114.
1119
5. Sylvester JE, Blasko JC, Grimm PD, et al. Ten-year biochemical relapse-free survival after external beam radiation and brachytherapy for localized prostate cancer: The Seattle experience. Int J Radiat Oncol Biol Phys. 2003;57:944-952. 6. Albertsen PC, Hanley JA, Fine J. 20-year outcomes following conservative management of clinically localized prostate cancer. JAMA. 2005;293:2095-2101. 7. D’Amico AV, Chen MH, Roehl KA, et al. Preoperative PSA velocity and the risk of death from prostate cancer after radical prostatectomy. N Engl J Med. 2004;351:125-135. 8. D’Amico AV, Renshaw AA, Sussman B, et al. Pretreatment PSA velocity and risk of death from prostate cancer following external beam radiation therapy. JAMA. 2005;294:440-447. 9. Sengupta S, Myers RP, Slezak JM, et al. Preoperative prostate specific antigen doubling time and velocity are strong and independent predictors of outcomes following radical prostatectomy. J Urol. 2005;174:2191-2196. 10. Patel DA, Presti JC Jr, McNeal JE, et al. Preoperative PSA velocity is an independent prognostic factor for relapse after radical prostatectomy. J Clin Oncol. 2005;23:6157-6162. 11. Eggener SE, Roehl KA, Yossepowitch O, et al. Prediagnosis prostate specific antigen velocity is associated with risk of prostate cancer progression following brachytherapy and external beam radiation therapy. J Urol. 2006;176:1399-1403. 12. Smith DS, Catalona WJ. Rate of change in serum prostate specific antigen levels as a method for prostate cancer detection. J Urol. 1994;152:1163-1167. 13. Landis JR, Koch GG. An application of hierarchical kappa-type statistics in the assessment of majority agreement among multiple observers. Biometrics. 1977;33:363-374. 14. Schroder FH, Roobol MJ, van der Kwast TH, et al. Does PSA velocity predict prostate cancer in pre-screened populations? Eur Urol. 2006;49:460-465. 15. Thompson IM, Ankerst DP, Chi C, et al. Assessing prostate cancer risk: Results from the Prostate Cancer Prevention Trial. J Natl Cancer Inst. 2006;98:529-534. 16. Berger AP, Deibl M, Strasak A, et al. Large-scale study of clinical impact of PSA velocity: Long-term PSA kinetics as method of differentiating men with from those without prostate cancer. Urology. 2007;69:134-138. 17. Chun FK, Briganti A, Shariat SF, et al. Significant upgrading affects a third of men diagnosed with prostate cancer: predictive nomogram and internal validation. BJU Int. 2006;98:329-334.
from CaP in the decade after CaP surgery.1 Compared with men with a PSAV of 2.0 ng/mL/y or less, those with a PSAV greater than 2.0 ng/mL/y in the year before diagnosis had a 10-fold increased risk of CaP death in the decade after treatment. Therefore, it seems reasonable to conclude that PSAV is associated with a more aggressive CaP phenotype. The question remaining is how to use this information in clinical practice to inform patients about treatment or the need for a prostate biopsy. As the authors point out, men with a Gleason score of 6 are not infrequently upgraded at RP to a higher Gleason score (Table 2), and a rapidly increasing PSA level provides evidence that a patient could be harboring a more aggressive CaP (despite the absence of high-grade disease on biopsy) for which surveillance might be inappropriate. In their report, even a PSAV greater than 1.0 ng/mL/y was significantly associated with a Gleason score of 7 or greater. On the basis of the work of these authors and others, I would be concerned about the safety of surveillance in a man with a consistently increasing PSA level, especially if the PSAV were more than 1 ng/mL/y, even though some have advocated that a PSA doubling time of more than 3 years is a safe threshold for entering a surveillance program.2,3 Given that no PSA level exists below which we can guarantee a man that he does not have lethal CaP, the greatest value of the PSAV could be among men who have PSA levels that have not reached a threshold level considered to be concerning. I believe that the man whose PSA level increases from 0.5 to 1.5 ng/ml during 1 year (a rapid increase) is at a greater risk of harboring lethal CaP than the man whose PSA level remains at 1.5 ng/mL from year to year.4 Although some believe that the PSAV has no clinical usefulness in terms of diagnosis,5 I believe that a rapid increase in PSA level should trigger a prostate biopsy in the absence of evidence of prostatic inflammation. Whether this approach will improve overall health outcomes is as yet unproved. H. Ballentine Carter, M.D., Department of Urology, Johns Hopkins University School of Medicine, James Buchanan Brady Urological Institute, Johns Hopkins Hospital, Baltimore, Maryland
EDITORIAL COMMENT Most urologists believe that early detection of CaP with PSA testing saves lives and uncovers some cases of CaP that would never have caused harm—a tradeoff that is unavoidable for now. Thus, the identification before treatment of those cancers that pose a threat to life and distinguishing them from those that do not is an important goal. In their study, Loeb et al. have shown that the PSAV (the rate of change in PSA) in the year before diagnosis is directly and independently associated with the Gleason score in the final pathologic specimen after RP for CaP and significantly associated with the disease extent (organ confined vs nonorgan confined). A PSAV greater than 2.0 ng/mL/y was a stronger predictor of high-grade disease (Gleason score of 7 or greater) than was clinical stage and black race and remained a significant predictor when accounting for PSA level. These data are consistent with their previous work demonstrating that the PSAV in the year before diagnosis was associated with death
1120
References 1. D’Amico AV, Chen MH, Roehl KA, et al. Preoperative PSA velocity and the risk of death from prostate cancer after radical prostatectomy. N Engl J Med. 2004;351:125-135. 2. van den Bergh RC, Roemeling S, Roobol MJ, et al. Prospective validation of active surveillance in prostate cancer: The PRIAS study. Eur Urol. 2007;52:1560-1563. 3. Klotz L. Active surveillance for favorable-risk prostate cancer: Who, how and why? Nat Clin Pract Oncol. 2007;4:692-698. 4. Carter HB, Ferrucci L, Kettermann A, et al. Detection of lifethreatening prostate cancer with prostate-specific antigen velocity during a window of curability J Natl Cancer Inst. 2006;98:1521-1527. 5. Etzioni RD, Ankerst DP, Weiss NS, et al. Is prostate-specific antigen velocity useful in early detection of prostate cancer? A critical appraisal of the evidence. J Natl Cancer Inst. 2007;99:1510-1515.
doi:10.1016/j.urology.2008.02.069 UROLOGY 72: 1116 –1120, 2008. © 2008 Elsevier Inc.
UROLOGY 72 (5), 2008