- Email: [email protected]
Prostate Specific Antigen Half-Time and Prostate Specific Antigen Doubling Time as Predictors of Response to Androgen Deprivation Therapy for Metastatic Prostate Cancer
Prostate Specific Antigen Half-Time and Prostate Specific Antigen Doubling Time as Predictors of Response to Androgen Deprivation Therapy for Metastatic Prostate Cancer Yong Hyun Park, In Sik Hwang, Chang Wook Jeong, Hyeon Hoe Kim, Sang Eun Lee and Cheol Kwak* From the Department of Urology, Seoul National University Hospital, Seoul, Korea
Abbreviations and Acronyms ADT ⫽ androgen deprivation therapy ECOG ⫽ Eastern Cooperative Oncology Group HRPC ⫽ hormone refractory prostate cancer LHRH ⫽ luteinizing hormonereleasing hormone mAb ⫽ maximum androgen blockade PSA ⫽ prostate specific antigen PSADTAN ⫽ prostate specific antigen doubling time after prostate specific antigen nadir PSAT1/2 ⫽ prostate specific antigen half-time Submitted for publication September 2, 2008. Nothing to disclose. * Correspondence: 28 Yeongeon-dong, Jongno-gu, Seoul 110-744, Korea (telephone: 822-2072-0817; FAX: 82-2-742-4665; e-mail: [email protected]).
Editor’s Note: This article is the fourth of 5 published in this issue for which category 1 CME credits can be earned. Instructions for obtaining credits are given with the questions on pages 2834 and 2835.
Purpose: We determined the clinical significance of prostate specific antigen half-time and prostate specific antigen doubling time after the prostate specific antigen nadir as predictors of the response to androgen deprivation therapy for metastatic prostate cancer. Materials and Methods: A total of 131 patients with metastatic prostate cancer treated with androgen deprivation were included in this analysis. Clinicopathological features and cancer specific survival were compared among the patients who were divided according to prostate specific antigen half-time and prostate specific antigen doubling time after the prostate specific antigen nadir. Results: Median followup was 53.0 months. Baseline and nadir prostate specific antigen did not differ between the patients with a short prostate specific antigen half-time (1 month or less) and those with a long prostate specific antigen half-time (longer than 1 month). Patients with a short prostate specific antigen half-time had a higher Gleason score, shorter nadir duration and shorter cancer specific survival. No differences were found between the patients with a short (6 months or less) and those with a long (longer than 6 months) prostate specific antigen doubling time after the prostate specific antigen nadir in terms of baseline prostate specific antigen, nadir prostate specific antigen, biopsy Gleason score and prostate specific antigen half-time. A short prostate specific antigen doubling time after the prostate specific antigen nadir was associated with shorter nadir duration and poorer median cancer specific survival. On multivariate analysis Gleason score, nadir prostate specific antigen and prostate specific antigen half-time remained independent predictors of an increase in prostate specific antigen after androgen deprivation therapy. Nadir prostate specific antigen, prostate specific antigen half-time and prostate specific antigen doubling time after the prostate specific antigen nadir were prognostic factors for cancer specific survival. Conclusions: The results of our study suggest that prostate specific antigen half-time and prostate specific antigen doubling time after the prostate specific antigen nadir are independent prognostic indicators for an increase in prostate specific antigen after androgen deprivation therapy and cancer related death in patients with metastatic prostate cancer treated with androgen deprivation. Key Words: prostatic neoplasms, prostate-specific antigen, drug therapy
2520
www.jurology.com
0022-5347/09/1816-2520/0 THE JOURNAL OF UROLOGY® Copyright © 2009 by AMERICAN UROLOGICAL ASSOCIATION
Vol. 181, 2520-2525, June 2009 Printed in U.S.A. DOI:10.1016/j.juro.2009.01.104
PROSTATE SPECIFIC ANTIGEN AS PREDICTOR OF TREATMENT SUCCESS
THE initial treatment of patients with metastatic prostate cancer is androgen deprivation. Although the majority of patients show a good clinical response after initiation of ADT, this effect is limited and HRPC inevitably develops. Until the publication of 2 pivotal trials (TAX 327 and Southwest Oncology Group 99-16) there was no proven therapy to improve the survival of patients with HRPC. However, the median improvement in survival with docetaxel is only 2 or 3 months and the survival of patients with HRPC remains poor.1,2 Thus, an improved ability to predict the response to ADT and individual survival would allow us to inform patients about prognosis, and aid in clinical decision making and interpretation of the results through proper stratification. In phases before and after the treatment of prostate cancer the change in PSA over time is considered an important parameter in assessing the progression of prostate cancer. Moreover the variable PSA kinetics based on PSA measurements were investigated, including PSA density, PSA doubling time and PSA velocity, with the purpose of improving the test accuracy. PSA kinetics are also easy to apply and are logical tools to use in clinical decision making during the treatment of prostate cancer. However, to our knowledge to date the most wellknown efforts to predict the response to ADT have focused on the analysis of clinically available variables such as age, performance status, Gleason score and the presence or absence of visceral disease.3,4 Therefore, we introduced the concept of the PSAT1/2, the time necessary for serum PSA to reach half of its baseline concentration during ADT, and PSADTAN, the PSA doubling time after the PSA nadir. We determined the clinical significance of PSAT1/2 and PSADTAN as predictors of the patient response to ADT for metastatic prostate cancer.
MATERIALS AND METHODS A total of 131 patients with metastatic prostate cancer (stage D2) treated with ADT from 1990 to 2004 were included in this analysis. All patients had metastatic disease at diagnosis and had not received any treatment such as prostatectomy or radiation therapy for the primary lesion. The patients were treated with a LHRH agonist or surgical castration with or without an antiandrogen. PSA was assayed every 3 months and more frequently when progression was evident. The PSAT1/2 was calculated as log 2 divided by the slope of the linear regression of log PSA vs time, using PSA values from the first decrease below pretreatment PSA to the nadir during ADT. The PSADTAN was calculated by the same method as the PSAT1/2, using PSA values from the first increase greater than the nadir PSA to the peak after ADT. To be eligible for calculation patients were required to have had a minimum of 2 PSA values separated by at least 3 months.
2521
Differences in demographics as well as clinical and pathological factors were examined using the Student t test, and chi-square test for continuous and categorical variables, respectively. Survival curves were estimated using the Kaplan-Meier method and compared using the log rank test. Prognostic factors were assessed using multivariate analysis by the Cox proportional hazards regression model in a forward stepwise regression. Risk factors in the multivariate analysis were age, ECOG performance status, baseline PSA, nadir PSA, biopsy Gleason score, PSAT1/2 and PSADTAN.
RESULTS Mean patient age was 68.5 (⫾8.7) and median followup was 53.0 (⫾8.5) months. Median PSA at baseline was 143.0 ng/dl (range 4.5 to 7,040), mean biopsy Gleason score was 7.8 (⫾1.5) and median nadir PSA was 0.9 ng/dl (range undetectable to 1,664). Of 131 patients 106 (80.9%) had an increase in PSA during followup and of the 69 who had adequate data to evaluate HRPC developed in 49 (71.0%). Of those 131 patients 76 died of prostate cancer and 13 had died of other causes at the time of the analysis. Patient demographics and disease characteristics as a function of PSAT1/2 categories are shown in table 1. The baseline and nadir PSAs did not differ between the patients with short (1 month or less) and long (longer than 1 month) PSAT1/2. However, the patients with a short PSAT1/2 had a higher biopsy Gleason score and a shorter duration of PSA nadir. Median time to increase in PSA was 10 months (95% CI 5.7–14.3) for the short PSAT1/2 and 24 months (95% CI 14.9 –33.1) for the long PSAT1/2 (log rank p ⬍0.001). Median cancer specific survival was 35 months (95% CI 23.3– 46.7) for the short PSAT1/2 and 95 months (95% CI 47.6 –142.4) for the long PSAT1/2 (log rank p ⬍0.001) (fig. 1). Table 2 lists the baseline characteristics of the 69 patients according to PSADTAN (6 months or less vs longer than 6 months). There were no differences between the patients with short and long PSADTAN in terms of baseline PSA, nadir PSA, biopsy Gleason score and PSAT1/2. The short PSADTAN was associated with a shorter duration of the nadir PSA and progression to HRPC. Median cancer specific survival was 42 months (95% CI 30.7–53.3) for the short PSADTAN and more than 180 months for the long PSADTAN (log rank p ⫽ 0.004) (fig. 2). Multivariate Cox proportional hazards regression models were constructed to analyze the relationship of PSAT1/2 and PSADTAN to the increase in the PSA and cancer specific survival (table 3). The entire population was analyzed for potential clinical, pathological and treatment related factors relative to the increase in PSA and cancer specific survival. A low biopsy Gleason score, low nadir PSA and short PSAT1/2 were significantly associated with an in-
2522
PROSTATE SPECIFIC ANTIGEN AS PREDICTOR OF TREATMENT SUCCESS
Table 1. Baseline characteristics of patients according to PSAT1/2 categories Short PSAT1/2 (1 mo or less) No. pts Mean ⫾SD age No. ECOG performance status (%): 0 1 2 Baseline PSA (ng/dl): Median (range) Mean ⫾SD Mean ⫾SD biopsy Gleason score Nadir PSA (ng/dl): Median (range) Mean ⫾SD Mean ⫾SD mos nadir No. later PSA increase (%) No. survival (%): Survival Death from prostate Ca Death from other cause
Long PSAT1/2 (greater than 1 mo)
68 67.4 ⫾ 8.5
63 69.7 ⫾ 8.9
26 (38.2) 27 (39.7) 13 (19.1)
32 (50.8) 21 (33.3) 10 (15.9)
p Value — 0.127 0.311
0.367 150 (9.4–5,390) 445.3 ⫾ 777.1 8.2 ⫾ 1.3
132.3 (4.5–7,040) 606.1 ⫾ 1,220.4 7.3 ⫾ 1.6
0.7 (undetectable ⬃ 1,664) 35.9 ⫾ 201.8 13.3 ⫾ 16.1 58 (85.3)
0.9 (undetectable ⬃ 95) 10.3 ⫾ 21.3 32.2 ⫾ 34.9 48 (76.2)
16 (23.5) 48 (70.6) 4 (5.9)
25 (39.7) 28 (44.4) 9 (14.3)
crease in PSA after ADT. In addition, a low nadir PSA, short PSAT1/2 and short PSADTAN were associated with prostate cancer specific mortality.
DISCUSSION Since the 1940s ADT has been the mainstay for the management of advanced prostate cancer.5 Generally an incomplete or sluggish response to ADT is considered evidence of a significant androgen refractory population associated with a poor prognosis. Arai et al studied 44 patients treated with ADT for advanced prostate cancer and reported that there
⬍0.001 0.303
⬍0.001 0.266 0.019
was a suggestion of longer progression-free survival when the interval to attain the PSA nadir was shorter.6 Another study supported this finding but these studies were limited by small sample sizes and short followup.7 The results of the current study demonstrated that a short PSAT1/2, in other words, a rapid decrease in PSA after ADT, was associated with a shorter time to increase in PSA and a poorer cancer specific survival. These findings are not consistent with those of prior studies. In our study the short PSAT1/2 was associated with a higher Gleason score and not with baseline or nadir PSA. Aihara et al
Figure 1. A, time to increase in PSA according to PSAT1/2 categories. B, cancer specific survival according to PSAT1/2 categories
PROSTATE SPECIFIC ANTIGEN AS PREDICTOR OF TREATMENT SUCCESS
2523
Table 2. Baseline characteristics of patients according to PSADTAN categories Short PSADTAN (6 mos or less) No. pts Mean ⫾SD age No. ECOG performance status (%): 0 1 2 Baseline PSA (ng/dl): Median (range) Mean ⫾SD Mean ⫾SD biopsy Gleason score Nadir PSA (ng/dl): Median (range) Mean ⫾SD Mean ⫾SD mos nadir Mean ⫾SD mos PSAT1/2 No. progression to HRPC (%) No. survival (%): Survival Death from prostate Ca Death from other cause
Long PSADTAN (greater than 6 mos)
49 67.4 ⫾ 8.5
20 67.3 ⫾ 10.0
18 (36.7) 18 (36.7) 12 (24.5)
11 (55.0) 7 (35.0) 2 (10.0)
p Value — 0.940 0.389
0.806 250 (4.6–7,040) 619.9 ⫾ 1,117.5 8.3 ⫾ 1.0
150 (9.4–5,219) 697.5 ⫾ 1,344.9 8.1 ⫾ 1.4
0.9 (undetectable–95) 13.0 ⫾ 24.7 13.2 ⫾ 11.7 1.8 ⫾ 2.4 39 (79.6)
0.8 (undetectable–80.3) 7.0 ⫾ 17.9 20.5 ⫾ 14.5 1.7 ⫾ 1.8 6 (30.0)
11 (22.4) 37 (75.5) 1 (2.0)
reported that there was a strong inverse correlation between Gleason grade and PSA level of serum and prostate cancer tissue.8 Berruti et al demonstrated that the extent of neuroendocrine differentiation was significantly associated with a poor prognosis and with a relatively low serum PSA.9 Therefore, we can speculate that in the patients with a short PSAT1/2 there might be neuroendocrine differentiation that could induce poorly differentiated cancer and counteract the expression of PSA. One of the main mechanisms of action of ADT is androgen deprivation, which counteracts the phe-
0.578 0.330
0.033 0.945 ⬍0.001 0.012
8 (40.0) 8 (40.0) 3 (15.0)
nomenon of androgen mediated resistance to apoptosis and, thus, inhibits tumor growth.10 A slow decrease in PSA and long time to the PSA nadir are explained by the fact that, unlike ablative treatment modalities such as radical prostatectomy, it does not cause immediate cell death. It is similar to radiation therapy in terms of the induction of apoptosis. Double strand breaks of nuclear DNA, caused by radiation, lead to the irreversible loss of the reproductive integrity of the cell and eventual cell death.11 In their retrospective study of 871 patients treated with external beam radiation therapy Kestin et al reported that the 5-year biochemical recurrence-free survival of patients who reached their nadir at less than 1.0, 1.0 to 1.9, 2.0 to 2.9 and 3.0 or more years was 30%, 52%, 64% and 92%, respectively.12 More recently Ray et al studied 4,839 patients treated with external beam radiation therapy and reported that a longer time to the nadir PSA was significantly associated with improved biochemical recurrencefree survival and metastasis-free survival regardless of the nadir PSA.13 Several studies support the
Table 3. Multivariate analysis of factors predicting an increase in PSA and cancer specific survival Factors
Figure 2. Cancer specific survival according to PSADTAN categories.
Increase in PSA after ADT: Biopsy Gleason score Nadir PSA PSAT1/2 Ca-specific survival: Nadir PSA PSAT1/2 PSADTAN
HR
95% CI
p Value
1.249 1.022 2.316
1.052–1.482 1.014–1.031 1.476–3.632
0.011 ⬍0.001 ⬍0.001
1.038 3.021 2.919
1.022–1.054 1.281–7.126 1.093–7.797
⬍0.001 0.012 0.033
2524
PROSTATE SPECIFIC ANTIGEN AS PREDICTOR OF TREATMENT SUCCESS
possibility that a slow decrease in PSA implies more favorable outcomes.14 –16 The importance of the study results is that the data help define the ideal patient population for aggressive followup and early systemic therapy. There were 76 prostate cancer specific mortalities during the study, 48 (63.2%) in the short PSAT1/2 group and 38 (50.0%) in the short PSADTAN group. Therefore, an alternative means of treatment which may decrease mortality might be considered earlier for this population of patients in whom the disease is biologically aggressive. The potential limitations of this study should be considered. The major limitation was the nonrandomized, retrospective nature of this study design. However, many factors that could influence the results were well controlled in our analysis. It is unlikely that the apparent survival benefit associated with a long PSAT1/2 and PSADTAN was an artifact of an imbalance of the prognostic factors. Another potential limitation is that the median followup during this study was only 53.0 months. Data from Pound
et al indicated that the actuarial median time to death after the development of metastatic disease was slightly less than 5 years.17 Because followup in this study was less than 5 years, the results may change with more data during a longer period. Finally a potential confounding point regarding the ADT used was that some patients (7.2%) received a LHRH agonist while others (92.8%) received a LHRH agonist and antiandrogen (mAb). If the mAb was shown to be superior to the LHRH agonist monotherapy, our study should be reanalyzed by the type of ADT. However, a large number of studies have shown no additional benefit from the mAb.18 –20
CONCLUSIONS The results of this study provide evidence to support that the PSAT1/2 and PSADTAN are independent prognostic factors for an increase in PSA and prostate cancer specific mortality in patients with metastatic prostate cancer. For these patients an alternative means of treatment might be considered.
REFERENCES 1. Tannock IF, de Wit R, Berry WR, Horti J, Pluzanska A, Chi KN et al: Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer. N Engl J Med 2004; 351: 1502. 2. Petrylak DP, Tangen CM, Hussain MH, Lara PN Jr, Jones JA, Taplin ME et al: Docetaxel and estramustine compared with mitoxantrone and prednisone for advanced refractory prostate cancer. N Engl J Med 2004; 351: 1513. 3. Smaletz O, Scher HI, Small EJ, Verbel DA, McMillan A, Regan K et al: Nomogram for overall survival of patients with progressive metastatic prostate cancer after castration. J Clin Oncol 2002; 20: 3972. 4. Halabi S, Small EJ, Kantoff PW, Kattan MW, Kaplan EB, Dawson NA et al: Prognostic model for predicting survival in men with hormonerefractory metastatic prostate cancer. J Clin Oncol 2003; 21: 1232. 5. Huggins C and Hodges CV: Studies on prostatic cancer. I. The effect of castration, of estrogen and androgen injection on serum phosphatases in metastatic carcinoma of the prostate. CA Cancer J Clin 1972; 22: 232. 6. Arai Y, Yoshiki T and Yoshida O: Prognostic significance of prostate specific antigen in endocrine treatment for prostatic cancer. J Urol 1990; 144: 1415. 7. Cooper EH, Armitage TG, Robinson MR, Newling DW, Richards BR, Smith PH et al: Prostatic specific antigen and the prediction of prognosis in
metastatic prostatic cancer. Cancer 1990; 66: 1025. 8. Aihara M, Lebovitz RM, Wheeler TM, Kinner BM, Ohori M and Scardino PT: Prostate specific antigen and Gleason grade: an immunohistochemical study of prostate cancer. J Urol 1994; 151: 1558. 9. Berruti A, Mosca A, Tucci M, Terrone C, Torta M, Tarabuzzi R et al: Independent prognostic role of circulating chromogranin A in prostate cancer patients with hormone-refractory disease. Endocr Relat Cancer 2005; 12: 109. 10. Coffey RN, Watson RW, O’Neill AJ, Mc Eleny K and Fitzpatrick JM: Androgen-mediated resistance to apoptosis. Prostate 2002; 53: 300. 11. Shinomiya N: New concepts in radiation-induced apoptosis: ‘premitotic apoptosis’ and ‘postmitotic apoptosis’. J Cell Mol Med 2001; 5: 240. 12. Kestin LL, Vicini FA, Ziaja EL, Stromberg JS, Frazier RC and Martinez AA: Defining biochemical cure for prostate carcinoma patients treated with external beam radiation therapy. Cancer 1999; 86: 1557. 13. Ray ME, Thames HD, Levy LB, Horwitz EM, Kupelian PA, Martinez AA et al: PSA nadir predicts biochemical and distant failures after external beam radiotherapy for prostate cancer: a multiinstitutional analysis. Int J Radiat Oncol Biol Phys 2006; 64: 1140. 14. Lee WR, Hanlon AL and Hanks GE: Prostate specific antigen nadir following external beam
radiation therapy for clinically localized prostate cancer: the relationship between nadir level and disease-free survival. J Urol 1996; 156: 450. 15. Hanlon AL, Diratzouian H and Hanks GE: Posttreatment prostate-specific antigen nadir highly predictive of distant failure and death from prostate cancer. Int J Radiat Oncol Biol Phys 2002; 53: 297. 16. Pollack A, Zagars GK, Starkschall G, Antolak JA, Lee JJ, Huang E et al: Prostate cancer radiation dose response: results of the M. D. Anderson phase III randomized trial. Int J Radiat Oncol Biol Phys 2002; 53: 1097. 17. Pound CR, Partin AW, Eisenberger MA, Chan DW, Pearson JD and Walsh PC: Natural history of progression after PSA elevation following radical prostatectomy. JAMA 1999; 281: 1591. 18. Maximum androgen blockade in advanced prostate cancer: an overview of 22 randomised trials with 3283 deaths in 5710 patients. Prostate Cancer Trialists’ Collaborative Group. Lancet 1995; 346: 265. 19. Maximum androgen blockade in advanced prostate cancer: an overview of the randomised trials. Prostate Cancer Trialists’ Collaborative Group. Lancet 2000; 355: 1491. 20. Samson DJ, Seidenfeld J, Schmitt B, Hasselblad V, Albertsen PC, Bennett CL et al: Systematic review and meta-analysis of monotherapy compared with combined androgen blockade for patients with advanced prostate carcinoma. Cancer 2002; 95: 361.
PROSTATE SPECIFIC ANTIGEN AS PREDICTOR OF TREATMENT SUCCESS
2525
EDITORIAL COMMENTS The search for biomarkers predictive of the clinical course of prostate cancer continues. Park et al reviewed 131 patients with metastatic prostate cancer treated with ADT and correlated the outcome to PSA kinetics during treatment, namely PSAT1/2 and PSADTAN. A short PSAT1/2 was associated with prostate cancer specific mortality as was a short PSADTAN. The thrust of this article is in identifying men at higher risk for failure, thus allowing earlier chemotherapy. Although significance was achieved in the study for PSAT1/2 and PSADTAN, it appears that
Gleason score remains correlated with PSAT1/2 and predictive of an increase in PSA after nadir from ADT. The independence of these markers from Gleason score in predicting cancer specific survival may be a movement away from static values based on histology toward more physiological models of tumor kinetics in planning therapy for these individuals.
Although this study has a small sample size (131), it provides initial evidence for predicting response to androgen deprivation therapy, the initial treatment for metastatic prostate cancer. The authors demonstrate fitting multivariable Cox proportional hazards regression models that longer values of 2 relatively uncorrelated measures of PSA levels, PSAT1/2 and PSADTAN, lead to longer cancer specific survival. The study only considered patients with a minimum of 2 PSA values separated by at least 3 months. This criterion led to the exclusion of 34 patients including 4 deceased, 18 who missed laboratory results, 4 who transferred to other hospitals
and 8 who were lost to followup. This may cause differential missingness and affect the analysis. Thus, the authors note this as a limitation of their retrospective study. Nonetheless this study provides an important first step in predicting the response to androgen deprivation therapy and should help better inform patients with metastatic prostate cancer in the future.
James M. Cummings Division of Urology St. Louis University St. Louis, Missouri
Heather J. Litman Department of Biostatistics New England Research Institutes Watertown, Massachusetts
Abbreviations and Acronyms ADT ⫽ androgen deprivation therapy ECOG ⫽ Eastern Cooperative Oncology Group HRPC ⫽ hormone refractory prostate cancer LHRH ⫽ luteinizing hormonereleasing hormone mAb ⫽ maximum androgen blockade PSA ⫽ prostate specific antigen PSADTAN ⫽ prostate specific antigen doubling time after prostate specific antigen nadir PSAT1/2 ⫽ prostate specific antigen half-time Submitted for publication September 2, 2008. Nothing to disclose. * Correspondence: 28 Yeongeon-dong, Jongno-gu, Seoul 110-744, Korea (telephone: 822-2072-0817; FAX: 82-2-742-4665; e-mail: [email protected]).
Editor’s Note: This article is the fourth of 5 published in this issue for which category 1 CME credits can be earned. Instructions for obtaining credits are given with the questions on pages 2834 and 2835.
Purpose: We determined the clinical significance of prostate specific antigen half-time and prostate specific antigen doubling time after the prostate specific antigen nadir as predictors of the response to androgen deprivation therapy for metastatic prostate cancer. Materials and Methods: A total of 131 patients with metastatic prostate cancer treated with androgen deprivation were included in this analysis. Clinicopathological features and cancer specific survival were compared among the patients who were divided according to prostate specific antigen half-time and prostate specific antigen doubling time after the prostate specific antigen nadir. Results: Median followup was 53.0 months. Baseline and nadir prostate specific antigen did not differ between the patients with a short prostate specific antigen half-time (1 month or less) and those with a long prostate specific antigen half-time (longer than 1 month). Patients with a short prostate specific antigen half-time had a higher Gleason score, shorter nadir duration and shorter cancer specific survival. No differences were found between the patients with a short (6 months or less) and those with a long (longer than 6 months) prostate specific antigen doubling time after the prostate specific antigen nadir in terms of baseline prostate specific antigen, nadir prostate specific antigen, biopsy Gleason score and prostate specific antigen half-time. A short prostate specific antigen doubling time after the prostate specific antigen nadir was associated with shorter nadir duration and poorer median cancer specific survival. On multivariate analysis Gleason score, nadir prostate specific antigen and prostate specific antigen half-time remained independent predictors of an increase in prostate specific antigen after androgen deprivation therapy. Nadir prostate specific antigen, prostate specific antigen half-time and prostate specific antigen doubling time after the prostate specific antigen nadir were prognostic factors for cancer specific survival. Conclusions: The results of our study suggest that prostate specific antigen half-time and prostate specific antigen doubling time after the prostate specific antigen nadir are independent prognostic indicators for an increase in prostate specific antigen after androgen deprivation therapy and cancer related death in patients with metastatic prostate cancer treated with androgen deprivation. Key Words: prostatic neoplasms, prostate-specific antigen, drug therapy
2520
www.jurology.com
0022-5347/09/1816-2520/0 THE JOURNAL OF UROLOGY® Copyright © 2009 by AMERICAN UROLOGICAL ASSOCIATION
Vol. 181, 2520-2525, June 2009 Printed in U.S.A. DOI:10.1016/j.juro.2009.01.104
PROSTATE SPECIFIC ANTIGEN AS PREDICTOR OF TREATMENT SUCCESS
THE initial treatment of patients with metastatic prostate cancer is androgen deprivation. Although the majority of patients show a good clinical response after initiation of ADT, this effect is limited and HRPC inevitably develops. Until the publication of 2 pivotal trials (TAX 327 and Southwest Oncology Group 99-16) there was no proven therapy to improve the survival of patients with HRPC. However, the median improvement in survival with docetaxel is only 2 or 3 months and the survival of patients with HRPC remains poor.1,2 Thus, an improved ability to predict the response to ADT and individual survival would allow us to inform patients about prognosis, and aid in clinical decision making and interpretation of the results through proper stratification. In phases before and after the treatment of prostate cancer the change in PSA over time is considered an important parameter in assessing the progression of prostate cancer. Moreover the variable PSA kinetics based on PSA measurements were investigated, including PSA density, PSA doubling time and PSA velocity, with the purpose of improving the test accuracy. PSA kinetics are also easy to apply and are logical tools to use in clinical decision making during the treatment of prostate cancer. However, to our knowledge to date the most wellknown efforts to predict the response to ADT have focused on the analysis of clinically available variables such as age, performance status, Gleason score and the presence or absence of visceral disease.3,4 Therefore, we introduced the concept of the PSAT1/2, the time necessary for serum PSA to reach half of its baseline concentration during ADT, and PSADTAN, the PSA doubling time after the PSA nadir. We determined the clinical significance of PSAT1/2 and PSADTAN as predictors of the patient response to ADT for metastatic prostate cancer.
MATERIALS AND METHODS A total of 131 patients with metastatic prostate cancer (stage D2) treated with ADT from 1990 to 2004 were included in this analysis. All patients had metastatic disease at diagnosis and had not received any treatment such as prostatectomy or radiation therapy for the primary lesion. The patients were treated with a LHRH agonist or surgical castration with or without an antiandrogen. PSA was assayed every 3 months and more frequently when progression was evident. The PSAT1/2 was calculated as log 2 divided by the slope of the linear regression of log PSA vs time, using PSA values from the first decrease below pretreatment PSA to the nadir during ADT. The PSADTAN was calculated by the same method as the PSAT1/2, using PSA values from the first increase greater than the nadir PSA to the peak after ADT. To be eligible for calculation patients were required to have had a minimum of 2 PSA values separated by at least 3 months.
2521
Differences in demographics as well as clinical and pathological factors were examined using the Student t test, and chi-square test for continuous and categorical variables, respectively. Survival curves were estimated using the Kaplan-Meier method and compared using the log rank test. Prognostic factors were assessed using multivariate analysis by the Cox proportional hazards regression model in a forward stepwise regression. Risk factors in the multivariate analysis were age, ECOG performance status, baseline PSA, nadir PSA, biopsy Gleason score, PSAT1/2 and PSADTAN.
RESULTS Mean patient age was 68.5 (⫾8.7) and median followup was 53.0 (⫾8.5) months. Median PSA at baseline was 143.0 ng/dl (range 4.5 to 7,040), mean biopsy Gleason score was 7.8 (⫾1.5) and median nadir PSA was 0.9 ng/dl (range undetectable to 1,664). Of 131 patients 106 (80.9%) had an increase in PSA during followup and of the 69 who had adequate data to evaluate HRPC developed in 49 (71.0%). Of those 131 patients 76 died of prostate cancer and 13 had died of other causes at the time of the analysis. Patient demographics and disease characteristics as a function of PSAT1/2 categories are shown in table 1. The baseline and nadir PSAs did not differ between the patients with short (1 month or less) and long (longer than 1 month) PSAT1/2. However, the patients with a short PSAT1/2 had a higher biopsy Gleason score and a shorter duration of PSA nadir. Median time to increase in PSA was 10 months (95% CI 5.7–14.3) for the short PSAT1/2 and 24 months (95% CI 14.9 –33.1) for the long PSAT1/2 (log rank p ⬍0.001). Median cancer specific survival was 35 months (95% CI 23.3– 46.7) for the short PSAT1/2 and 95 months (95% CI 47.6 –142.4) for the long PSAT1/2 (log rank p ⬍0.001) (fig. 1). Table 2 lists the baseline characteristics of the 69 patients according to PSADTAN (6 months or less vs longer than 6 months). There were no differences between the patients with short and long PSADTAN in terms of baseline PSA, nadir PSA, biopsy Gleason score and PSAT1/2. The short PSADTAN was associated with a shorter duration of the nadir PSA and progression to HRPC. Median cancer specific survival was 42 months (95% CI 30.7–53.3) for the short PSADTAN and more than 180 months for the long PSADTAN (log rank p ⫽ 0.004) (fig. 2). Multivariate Cox proportional hazards regression models were constructed to analyze the relationship of PSAT1/2 and PSADTAN to the increase in the PSA and cancer specific survival (table 3). The entire population was analyzed for potential clinical, pathological and treatment related factors relative to the increase in PSA and cancer specific survival. A low biopsy Gleason score, low nadir PSA and short PSAT1/2 were significantly associated with an in-
2522
PROSTATE SPECIFIC ANTIGEN AS PREDICTOR OF TREATMENT SUCCESS
Table 1. Baseline characteristics of patients according to PSAT1/2 categories Short PSAT1/2 (1 mo or less) No. pts Mean ⫾SD age No. ECOG performance status (%): 0 1 2 Baseline PSA (ng/dl): Median (range) Mean ⫾SD Mean ⫾SD biopsy Gleason score Nadir PSA (ng/dl): Median (range) Mean ⫾SD Mean ⫾SD mos nadir No. later PSA increase (%) No. survival (%): Survival Death from prostate Ca Death from other cause
Long PSAT1/2 (greater than 1 mo)
68 67.4 ⫾ 8.5
63 69.7 ⫾ 8.9
26 (38.2) 27 (39.7) 13 (19.1)
32 (50.8) 21 (33.3) 10 (15.9)
p Value — 0.127 0.311
0.367 150 (9.4–5,390) 445.3 ⫾ 777.1 8.2 ⫾ 1.3
132.3 (4.5–7,040) 606.1 ⫾ 1,220.4 7.3 ⫾ 1.6
0.7 (undetectable ⬃ 1,664) 35.9 ⫾ 201.8 13.3 ⫾ 16.1 58 (85.3)
0.9 (undetectable ⬃ 95) 10.3 ⫾ 21.3 32.2 ⫾ 34.9 48 (76.2)
16 (23.5) 48 (70.6) 4 (5.9)
25 (39.7) 28 (44.4) 9 (14.3)
crease in PSA after ADT. In addition, a low nadir PSA, short PSAT1/2 and short PSADTAN were associated with prostate cancer specific mortality.
DISCUSSION Since the 1940s ADT has been the mainstay for the management of advanced prostate cancer.5 Generally an incomplete or sluggish response to ADT is considered evidence of a significant androgen refractory population associated with a poor prognosis. Arai et al studied 44 patients treated with ADT for advanced prostate cancer and reported that there
⬍0.001 0.303
⬍0.001 0.266 0.019
was a suggestion of longer progression-free survival when the interval to attain the PSA nadir was shorter.6 Another study supported this finding but these studies were limited by small sample sizes and short followup.7 The results of the current study demonstrated that a short PSAT1/2, in other words, a rapid decrease in PSA after ADT, was associated with a shorter time to increase in PSA and a poorer cancer specific survival. These findings are not consistent with those of prior studies. In our study the short PSAT1/2 was associated with a higher Gleason score and not with baseline or nadir PSA. Aihara et al
Figure 1. A, time to increase in PSA according to PSAT1/2 categories. B, cancer specific survival according to PSAT1/2 categories
PROSTATE SPECIFIC ANTIGEN AS PREDICTOR OF TREATMENT SUCCESS
2523
Table 2. Baseline characteristics of patients according to PSADTAN categories Short PSADTAN (6 mos or less) No. pts Mean ⫾SD age No. ECOG performance status (%): 0 1 2 Baseline PSA (ng/dl): Median (range) Mean ⫾SD Mean ⫾SD biopsy Gleason score Nadir PSA (ng/dl): Median (range) Mean ⫾SD Mean ⫾SD mos nadir Mean ⫾SD mos PSAT1/2 No. progression to HRPC (%) No. survival (%): Survival Death from prostate Ca Death from other cause
Long PSADTAN (greater than 6 mos)
49 67.4 ⫾ 8.5
20 67.3 ⫾ 10.0
18 (36.7) 18 (36.7) 12 (24.5)
11 (55.0) 7 (35.0) 2 (10.0)
p Value — 0.940 0.389
0.806 250 (4.6–7,040) 619.9 ⫾ 1,117.5 8.3 ⫾ 1.0
150 (9.4–5,219) 697.5 ⫾ 1,344.9 8.1 ⫾ 1.4
0.9 (undetectable–95) 13.0 ⫾ 24.7 13.2 ⫾ 11.7 1.8 ⫾ 2.4 39 (79.6)
0.8 (undetectable–80.3) 7.0 ⫾ 17.9 20.5 ⫾ 14.5 1.7 ⫾ 1.8 6 (30.0)
11 (22.4) 37 (75.5) 1 (2.0)
reported that there was a strong inverse correlation between Gleason grade and PSA level of serum and prostate cancer tissue.8 Berruti et al demonstrated that the extent of neuroendocrine differentiation was significantly associated with a poor prognosis and with a relatively low serum PSA.9 Therefore, we can speculate that in the patients with a short PSAT1/2 there might be neuroendocrine differentiation that could induce poorly differentiated cancer and counteract the expression of PSA. One of the main mechanisms of action of ADT is androgen deprivation, which counteracts the phe-
0.578 0.330
0.033 0.945 ⬍0.001 0.012
8 (40.0) 8 (40.0) 3 (15.0)
nomenon of androgen mediated resistance to apoptosis and, thus, inhibits tumor growth.10 A slow decrease in PSA and long time to the PSA nadir are explained by the fact that, unlike ablative treatment modalities such as radical prostatectomy, it does not cause immediate cell death. It is similar to radiation therapy in terms of the induction of apoptosis. Double strand breaks of nuclear DNA, caused by radiation, lead to the irreversible loss of the reproductive integrity of the cell and eventual cell death.11 In their retrospective study of 871 patients treated with external beam radiation therapy Kestin et al reported that the 5-year biochemical recurrence-free survival of patients who reached their nadir at less than 1.0, 1.0 to 1.9, 2.0 to 2.9 and 3.0 or more years was 30%, 52%, 64% and 92%, respectively.12 More recently Ray et al studied 4,839 patients treated with external beam radiation therapy and reported that a longer time to the nadir PSA was significantly associated with improved biochemical recurrencefree survival and metastasis-free survival regardless of the nadir PSA.13 Several studies support the
Table 3. Multivariate analysis of factors predicting an increase in PSA and cancer specific survival Factors
Figure 2. Cancer specific survival according to PSADTAN categories.
Increase in PSA after ADT: Biopsy Gleason score Nadir PSA PSAT1/2 Ca-specific survival: Nadir PSA PSAT1/2 PSADTAN
HR
95% CI
p Value
1.249 1.022 2.316
1.052–1.482 1.014–1.031 1.476–3.632
0.011 ⬍0.001 ⬍0.001
1.038 3.021 2.919
1.022–1.054 1.281–7.126 1.093–7.797
⬍0.001 0.012 0.033
2524
PROSTATE SPECIFIC ANTIGEN AS PREDICTOR OF TREATMENT SUCCESS
possibility that a slow decrease in PSA implies more favorable outcomes.14 –16 The importance of the study results is that the data help define the ideal patient population for aggressive followup and early systemic therapy. There were 76 prostate cancer specific mortalities during the study, 48 (63.2%) in the short PSAT1/2 group and 38 (50.0%) in the short PSADTAN group. Therefore, an alternative means of treatment which may decrease mortality might be considered earlier for this population of patients in whom the disease is biologically aggressive. The potential limitations of this study should be considered. The major limitation was the nonrandomized, retrospective nature of this study design. However, many factors that could influence the results were well controlled in our analysis. It is unlikely that the apparent survival benefit associated with a long PSAT1/2 and PSADTAN was an artifact of an imbalance of the prognostic factors. Another potential limitation is that the median followup during this study was only 53.0 months. Data from Pound
et al indicated that the actuarial median time to death after the development of metastatic disease was slightly less than 5 years.17 Because followup in this study was less than 5 years, the results may change with more data during a longer period. Finally a potential confounding point regarding the ADT used was that some patients (7.2%) received a LHRH agonist while others (92.8%) received a LHRH agonist and antiandrogen (mAb). If the mAb was shown to be superior to the LHRH agonist monotherapy, our study should be reanalyzed by the type of ADT. However, a large number of studies have shown no additional benefit from the mAb.18 –20
CONCLUSIONS The results of this study provide evidence to support that the PSAT1/2 and PSADTAN are independent prognostic factors for an increase in PSA and prostate cancer specific mortality in patients with metastatic prostate cancer. For these patients an alternative means of treatment might be considered.
REFERENCES 1. Tannock IF, de Wit R, Berry WR, Horti J, Pluzanska A, Chi KN et al: Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer. N Engl J Med 2004; 351: 1502. 2. Petrylak DP, Tangen CM, Hussain MH, Lara PN Jr, Jones JA, Taplin ME et al: Docetaxel and estramustine compared with mitoxantrone and prednisone for advanced refractory prostate cancer. N Engl J Med 2004; 351: 1513. 3. Smaletz O, Scher HI, Small EJ, Verbel DA, McMillan A, Regan K et al: Nomogram for overall survival of patients with progressive metastatic prostate cancer after castration. J Clin Oncol 2002; 20: 3972. 4. Halabi S, Small EJ, Kantoff PW, Kattan MW, Kaplan EB, Dawson NA et al: Prognostic model for predicting survival in men with hormonerefractory metastatic prostate cancer. J Clin Oncol 2003; 21: 1232. 5. Huggins C and Hodges CV: Studies on prostatic cancer. I. The effect of castration, of estrogen and androgen injection on serum phosphatases in metastatic carcinoma of the prostate. CA Cancer J Clin 1972; 22: 232. 6. Arai Y, Yoshiki T and Yoshida O: Prognostic significance of prostate specific antigen in endocrine treatment for prostatic cancer. J Urol 1990; 144: 1415. 7. Cooper EH, Armitage TG, Robinson MR, Newling DW, Richards BR, Smith PH et al: Prostatic specific antigen and the prediction of prognosis in
metastatic prostatic cancer. Cancer 1990; 66: 1025. 8. Aihara M, Lebovitz RM, Wheeler TM, Kinner BM, Ohori M and Scardino PT: Prostate specific antigen and Gleason grade: an immunohistochemical study of prostate cancer. J Urol 1994; 151: 1558. 9. Berruti A, Mosca A, Tucci M, Terrone C, Torta M, Tarabuzzi R et al: Independent prognostic role of circulating chromogranin A in prostate cancer patients with hormone-refractory disease. Endocr Relat Cancer 2005; 12: 109. 10. Coffey RN, Watson RW, O’Neill AJ, Mc Eleny K and Fitzpatrick JM: Androgen-mediated resistance to apoptosis. Prostate 2002; 53: 300. 11. Shinomiya N: New concepts in radiation-induced apoptosis: ‘premitotic apoptosis’ and ‘postmitotic apoptosis’. J Cell Mol Med 2001; 5: 240. 12. Kestin LL, Vicini FA, Ziaja EL, Stromberg JS, Frazier RC and Martinez AA: Defining biochemical cure for prostate carcinoma patients treated with external beam radiation therapy. Cancer 1999; 86: 1557. 13. Ray ME, Thames HD, Levy LB, Horwitz EM, Kupelian PA, Martinez AA et al: PSA nadir predicts biochemical and distant failures after external beam radiotherapy for prostate cancer: a multiinstitutional analysis. Int J Radiat Oncol Biol Phys 2006; 64: 1140. 14. Lee WR, Hanlon AL and Hanks GE: Prostate specific antigen nadir following external beam
radiation therapy for clinically localized prostate cancer: the relationship between nadir level and disease-free survival. J Urol 1996; 156: 450. 15. Hanlon AL, Diratzouian H and Hanks GE: Posttreatment prostate-specific antigen nadir highly predictive of distant failure and death from prostate cancer. Int J Radiat Oncol Biol Phys 2002; 53: 297. 16. Pollack A, Zagars GK, Starkschall G, Antolak JA, Lee JJ, Huang E et al: Prostate cancer radiation dose response: results of the M. D. Anderson phase III randomized trial. Int J Radiat Oncol Biol Phys 2002; 53: 1097. 17. Pound CR, Partin AW, Eisenberger MA, Chan DW, Pearson JD and Walsh PC: Natural history of progression after PSA elevation following radical prostatectomy. JAMA 1999; 281: 1591. 18. Maximum androgen blockade in advanced prostate cancer: an overview of 22 randomised trials with 3283 deaths in 5710 patients. Prostate Cancer Trialists’ Collaborative Group. Lancet 1995; 346: 265. 19. Maximum androgen blockade in advanced prostate cancer: an overview of the randomised trials. Prostate Cancer Trialists’ Collaborative Group. Lancet 2000; 355: 1491. 20. Samson DJ, Seidenfeld J, Schmitt B, Hasselblad V, Albertsen PC, Bennett CL et al: Systematic review and meta-analysis of monotherapy compared with combined androgen blockade for patients with advanced prostate carcinoma. Cancer 2002; 95: 361.
PROSTATE SPECIFIC ANTIGEN AS PREDICTOR OF TREATMENT SUCCESS
2525
EDITORIAL COMMENTS The search for biomarkers predictive of the clinical course of prostate cancer continues. Park et al reviewed 131 patients with metastatic prostate cancer treated with ADT and correlated the outcome to PSA kinetics during treatment, namely PSAT1/2 and PSADTAN. A short PSAT1/2 was associated with prostate cancer specific mortality as was a short PSADTAN. The thrust of this article is in identifying men at higher risk for failure, thus allowing earlier chemotherapy. Although significance was achieved in the study for PSAT1/2 and PSADTAN, it appears that
Gleason score remains correlated with PSAT1/2 and predictive of an increase in PSA after nadir from ADT. The independence of these markers from Gleason score in predicting cancer specific survival may be a movement away from static values based on histology toward more physiological models of tumor kinetics in planning therapy for these individuals.
Although this study has a small sample size (131), it provides initial evidence for predicting response to androgen deprivation therapy, the initial treatment for metastatic prostate cancer. The authors demonstrate fitting multivariable Cox proportional hazards regression models that longer values of 2 relatively uncorrelated measures of PSA levels, PSAT1/2 and PSADTAN, lead to longer cancer specific survival. The study only considered patients with a minimum of 2 PSA values separated by at least 3 months. This criterion led to the exclusion of 34 patients including 4 deceased, 18 who missed laboratory results, 4 who transferred to other hospitals
and 8 who were lost to followup. This may cause differential missingness and affect the analysis. Thus, the authors note this as a limitation of their retrospective study. Nonetheless this study provides an important first step in predicting the response to androgen deprivation therapy and should help better inform patients with metastatic prostate cancer in the future.
James M. Cummings Division of Urology St. Louis University St. Louis, Missouri
Heather J. Litman Department of Biostatistics New England Research Institutes Watertown, Massachusetts