the risk of developing skeletal metastases or local recurrence.

Transcription

1 Original Article SERUM PSA AND CLINICAL RECURRENCE AFTER RRP FOR LOCALIZED PROSTATE CANCER HAUKAAS et al. Is preoperative serum prostate-specific antigen level significantly related to clinical recurrence after radical retropubic prostatectomy for localized prostate cancer? SVEIN A. HAUKAAS, OLE J. HALVORSEN*, LARS DAEHLIN, JENS HOSTMARK and LARS A. AKSLEN* Department of Surgical Sciences, University of Bergen, *Section for Pathology, The Gade Institute, and Section of Urology, Haukeland University Hospital, Bergen, Norway Accepted for publication 12 August 2005 OBJECTIVE To evaluate the influence of preoperative serum prostate-specific antigen (PSA) level and other clinicopathological variables on the probability of biochemical failure and clinical recurrence after radical prostatectomy (RP) for localized prostate cancer. PATIENTS AND METHODS The study was a retrospective survival analysis in 211 patients undergoing retropubic RP for clinically localized prostate cancer in the period Survival was estimated using the Kaplan-Meier method; survival endpoints were biochemical failure, defined as a PSA level of ng/ml or clinical recurrence consisting of palpable tumours in the prostatic fossa or distant metastases. In 58 patients with biochemical failure after surgery, we assessed the impact of the doubling time of serum PSA level (PSADT) on the risk of developing skeletal metastases or local recurrence. RESULTS The median (range) observation period was 66 (9 160) months. Biochemical failure occurred in 92 patients (44%) of whom 39 (42%) had local recurrence or skeletal metastases. There was a highly significant association (P < 01) between clinical T stage, histological grade, capsular penetration, surgical margin status, seminal vesicle invasion, preoperative serum PSA level and the probability of biochemical failure-free survival. By contrast there was no statistically significant association between preoperative serum PSA level, clinical T stage, surgical margin status, and clinical recurrence. There was a significant relationship between age (P = 21), histological grade (P = 25), capsular penetration (P = 18), seminal vesicle invasion (P = 0014), and clinical recurrence. Cox regression analysis showed that only histological grade and seminal vesicle invasion were independent predictors of clinical recurrence. In a subgroup of 58 patients with a rising serum PSA level after RP, a PSADT of 12.8 months conferred a significantly higher risk (P = 15) of developing skeletal metastases than a PSADT of >12.8 months. CONCLUSION In the present patients undergoing RP the preoperative serum PSA level was not associated with the clinical outcome, whereas it was significantly related to biochemical failure rate. The probability of skeletal metastases was significantly associated with the PSADT after biochemical failure. KEYWORDS prostate cancer progression, radical prostatectomy, PSA, PSA doubling time INTRODUCTION Radical prostatectomy (RP) is a treatment option for clinically localized carcinoma of the prostate in patients with a life-expectancy of years. When counselling men with prostate cancer whether to undergo surgery, the serum PSA level is important in the decision because it can be used to predict pathological stage and biochemical failure after RP [1 3]. There are conflicting data implying only a weak relationship between morphological features of prostate cancer and preoperative serum PSA level [4]. In large volume cancers, preoperative serum PSA level has failed to predict biochemical failure [5]. Since the advent of PSA testing there has been a tendency towards lower serum PSA levels in patients advised to undergo RP. However, up to about half of men treated by RP for clinically localized prostate cancer will eventually have biochemical failure [6]. Although a detectable serum PSA level after surgery is commonly taken as an early indication of recurrent or persistent cancer, the disease course after biochemical failure is largely unknown [6 9]. Thus in the present study we assessed the influence of preoperative serum PSA level and other prognostic factors on the clinical outcome of localized prostate cancer after RP. PATIENTS AND METHODS The study was designed as a retrospective review of 211 patients treated with RP for clinically localized prostate cancer in the period at Haukeland University Hospital, Bergen, Norway. The patient series was consecutive, albeit that since the mid- 1990s at least half of the patients, who were good candidates for surgery, opted for external beam radiotherapy (EBRT). Patient age at surgery, clinical stage (2002 TNM classification) and serum PSA level before and after surgery were obtained from hospital records. The median (range) patient age was 61 (40 72) years. Before RP the median (range) serum PSA level was 10 ( 70) ng/ml. The histological grade (WHO system) and pathological stage, including seminal vesicle invasion, capsular penetration and surgical margins, were included. The pathology after RP was based 2006 BJU INTERNATIONAL 97, doi:1111/j x x 51

2 HAUKAAS ET AL. on totally embedded surgical specimens. One pathologist specialized in uropathology (O.J.H.) examined all the surgical specimens. Clinical stage T1/T2 disease, a negative bone scan and general good health with years of life-expectancy were the prerequisites for offering patients retropubic RP. Whenever possible a nerve-sparing operation was attempted, but 63% of the tumours were clinically stage T2 and the neurovascular bundle on one or both sides had to be removed in an effort to achieve tumour-free surgical margins. The outcome data were collected from the outpatient office charts, with attention to serum PSA level and local or distant recurrence. At the follow-up, serum PSA level was determined every third month in the first year, semi-annually for another 2 years and annually thereafter, when the serum PSA level remained undetectable. Biochemical failure was defined as persistent or rising serum PSA level of ng/ml. A palpable tumour in the prostatic fossa or evidence of distant disease on bone scan, X-ray or MRI was recorded as clinical recurrence. The doubling time of serum PSA level (PSADT) was calculated according to Patel et al. [10]. The PSADT was calculated using the two first detectable serum PSA values measured after surgery as: PSADT = time (months) log e (2)/ [log e (PSA2) log e (PSA1)] TABLE 1 Univariate analysis (Kaplan-Meier method) of time to biochemical failure and clinical recurrence in 211 patients with prostate cancer Biochemical failure Clinical recurrence Variables N Events P* Events P* Age, years > Clinical T stage <01 15 T1b T1c T2a T2b T2c Histological grade (WHO) <01 25 Well Moderate Poor Capsular penetration <01 18 Absent Present Surgical margins <01 70 Negative Positive Seminal vesicle invasion <01 14 Negative Positive Preoperative PSA, ng/ml < > *log-rank test; Upper quartile. PSADT could be calculated in 58 of 92 men with detectable levels of serum PSA after RP. During the follow-up, 47 of 92 patients (51%) received antiandrogen treatment; 27 of these patients were treated early after biochemical failure, whilst 20 received deferred treatment. Salvage EBRT was delivered to eight patients with local recurrence. The follow-up was complete until May The time to biochemical failure or clinical recurrence after surgery was calculated using the product-limit method (Kaplan-Meier), with the log-rank test for differences between categories of each variable. Variables were categorized by quartiles and tertiles. The Cox proportional hazards method was applied for multivariate survival analyses, including only significant variables (P = 5) from univariate analyses, and tested by the likelihood ratio test. Model assumptions were examined by log log plots. Variables with a significant outcome in the first multivariate model (P = ) were analysed in a second (final) multivariate model. RESULTS The median (range) observation time was 66 (9 160) months, during which there was a rising serum PSA level in 92 patients (44%), of whom 39 (42%) developed clinical recurrence. Biochemical failure occurred at a mean of 33 (1 112) months, whilst the mean time to clinical recurrence was 54 (7 138) months. Eighteen patients had metastases and 21 developed local recurrence as the sole clinical manifestation of recurrence. Ten patients died from prostate cancer and 19 from unrelated causes. There was a highly significant association between clinical T stage, histological grade, capsular penetration, surgical margin status, seminal vesicle invasion, preoperative serum PSA level, and the probability of biochemical failure-free survival (Table 1). Biochemical failure-free rates at 3 and 5 years in patients categorized according to various clinicopathological variables are shown in Table 2. Preoperative serum PSA level, clinical T stage and surgical margin status did not correlate significantly with clinical recurrence (Table 1). The 5- and 10-year freedom from clinical recurrence in patients stratified into different risk groups are also shown in Table 2. There was a 91% probability of clinical recurrence-free survival at 10 years for men with well-differentiated cancer, whilst those with poorly differentiated tumours only had a 47% probability of being free of clinical recurrence. Being younger at surgery was associated with a lower risk of clinical recurrence after RP than when older (Table 1) BJU INTERNATIONAL

3 SERUM PSA AND CLINICAL RECURRENCE AFTER RRP FOR LOCALIZED PROSTATE CANCER TABLE 2 The 3-and 5-year biochemical failure-free and the 5-and 10-year clinical recurrence-free survival after RP Biochemical failure-free rate, % Clinical recurrence-free rate, % Variables 3-year (95% CI) 5-year (95% CI) 5-year (95% CI) 10-year (95% CI) Histological grade (WHO) Well 97 (92 99) 85 (68 97) 91 (74 100) 91 (74 100) Moderate 71 (63 79) 64 (56 73) 87 (81 94) 72 (60 83) Poor 50 (34 66) 32 (16 49) 76 (62 91) 47 (26 68) Capsular penetration Absent 86 (79 92) 76 (65 86) 87 (79 96) 84 (74 95) Present 57 (47 67) 46 (36 56) 84 (76 91) 60 (48 72) Seminal vesicle invasion Absent 80 (74 87) 71 (62 79) 88 (75 94) 77 (68 87) Present 41 (31 55) 28 (14 41) 79 (67 91) 49 (30 68) Surgical margins Negative 83 (76 90) 77 (67 86) Positive 60 (51 70) 46 (36 57) FIG. 1. Survival free from: a, skeletal metastases, and b, local recurrence (Kaplan-Meier method) in 58 of 92 patients with a rising serum PSA level after RP. PSADT >12.8 (>upper tertile); PSADT 12.8 (<upper tertile). Number of events in parenthesis. a Proportion free from skeletal metastases b Proportion free from local recurrences 0 P = 15 Log rank test PSADT 12.8 n = 39 (15) P = 98 Log rank test PSADT > 12.8 n = 19 (0) PSADT 12.8 n = 39 (10) PSADT > 12.8 n = 19 (4) Months after PSA recurrence In the subset of patients with detectable levels of serum PSA who did not have early postoperative endocrine treatment or salvage EBRT due to adverse histological features, the FIG. 2. Cancer-specific and overall survival (Kaplan- Meier method) in 211 men after RP. Cumulative survival 0 20 Cancer-specific Overall Months after surgery median PSADT was 9.6 months. For the upper tertile (66.7%) and upper quartile (75%) the PSADT was 12.8 and 14.7 months, respectively. A PSADT of 12.8 months (upper tertile) after biochemical failure inferred a significantly higher risk of skeletal metastases than a PSADT of >12.8 months (Fig. 1a). There was a trend (P = 98) to a greater risk of developing local recurrence with a shorter than a longer PSADT after biochemical failure (Fig. 1b). The 5- and 10-year cancer-specific survival rates (95% CI) in the series were 98 (96 100)% and 89 (82 97)%, respectively, with an overall survival at 5 and 10 years of 94 (90 97)% and 75 (66 84)%, respectively (Fig. 2). In Cox regression analyses, histological grade (well-moderate vs poorly), capsular penetration, seminal vesicle invasion, surgical margins, clinical T stage (T 2 vs T 1) and preoperative serum PSA level (PSA > 20 vs <20 ng/ml) were included as covariates if there was a statistically significant prognostic influence in univariate analyses. The preoperative serum PSA levels were available for 205 patients. As shown in Table 3 all covariates, except capsular penetration, remained independent predictors of biochemical failure, whereas only histological grade and seminal vesicle invasion were identified as independent predictors of clinical recurrence. DISCUSSION PSA has become an important marker in the diagnosis, staging and follow-up of prostate cancer. PSA recurrence-free survival is in general use as a surrogate endpoint for outcome after RP. The use of PSA testing started in Norway in the mid-1990s, and most of the cancers in the present series were clinical stage T2 and only a third stage T1c. Consequently, the prevalence of adverse prognostic factors like capsular penetration, positive surgical margins and invasion of the seminal vesicles is higher than in more contemporary series. The present study corroborates the prognostic importance of such risk factors for biochemical failure after RP. In all, 44% of the patients had biochemical failure at a mean follow-up of 2.8 years after RP. The high proportion of patients with a detectable serum PSA level after RP reflects the high prevalence of adverse prognostic factors in the present series (Table 1). In comparison, 879 (31%) of 2809 men undergoing RP for localized prostate cancer in the period had detectable serum PSA levels at a median of 2.9 years after surgery [11]. The preoperative serum PSA level is generally held to be an important prognostic factor either alone or combined with other factors of biochemical failure-free survival after RP [1,12]. In large-volume cancers ( 6 ml) the preoperative serum PSA level did not correlate well with biochemical failure-free survival rates [5]. Interestingly, when biochemical failure has occurred, further progression of prostate cancer seems less dependent on the preoperative serum PSA level [10,11]. Jhavieri et al. [13] found in 1132 consecutive patients that the 10-year overall survival for those with high serum PSA levels after RP was equivalent to those with no increasing serum PSA level. This seemingly contradictory 2006 BJU INTERNATIONAL 53

4 HAUKAAS ET AL. TABLE 3 Multivariate Cox regression analysis of time to biochemical failure and clinical recurrence after RP in 211 patients Biochemical failure Clinical recurrence Variables/categories HR (95% CI) P* HR (95% CI) P* N Histological grade (WHO) Well/moderately Poorly differentiated 2.2 ( ) 2.0 ( 4.0) Seminal vesicle invasion Absent Present 2.2 ( ) 2.0 ( 3.8) Surgical margins Negative 49 Positive 1.6 ( 2.5) Clinical T stage T1 09 T2 2.1 ( ) Preoperative PSA, ng/ml > ( ) HR, hazard ratio; *Likelihood ratio test. Information on preoperative serum PSA levels not available for six patients. Although the present median follow-up was only 5.5 years, interestingly the median 89 (82 97)% 10-year cancer-specific survival rate is comparable to that reported after RP for clinically localized prostate cancer [3,18,19]. In conclusion, in the present study the preoperative serum PSA level was not significantly associated with clinical recurrence after RP, whilst it was significantly related to the biochemical failure rate. However, the probability of skeletal metastases was significantly associated with the PSADT after RP. Morphological features such as histological grade and pathological stage were significantly related to the risk of clinical progression, but there is an urgent need for better preoperative prognostic variables to predict the long-term outcome after RP. Molecular markers obtained from the biopsy could be incorporated into prognostic models to improve the ability to predict the long-term risk of clinical progression after radical treatment of localized prostate cancer. CONFLICT OF INTEREST observation may be a result of the putative innocuous nature of detectable serum PSA level in some patients in whom there is no further increase in serum PSA level even in the presence of adverse prognostic factors [6 8]. In a series of 1997 patients undergoing RP [14], 315 (15%) had biochemical failure after surgery; only a third developed distant metastases after a median follow-up of 8 years from the time when serum PSA became detectable. In the present study the 10 men who developed metastases did so at a median of 8 years, which is comparable with the time to distant metastases in the earlier study [14]. The mean follow-up in both studies was only 5 years, and biochemical failure and clinical recurrence may occur even after 5 years free of disease [9]. Emerging data support the prognostic importance of PSADT for metastasis-free survival following biochemical failure after RP [9,10,11,15]. In a multivariate analysis the PSADT performed better as a predictor of time to clinical recurrence than preoperative serum PSA level, pathological stage and Gleason score [10,16]. The likelihood of a patient developing metastases was significantly related to the following factors; <2 years from surgery to the initial increase in serum PSA level, Gleason score >7 and PSADT <10 months [14]. In the present univariate analysis, metastasis-free survival was significantly different in patients with a PSADT of 12.8 months than in those with a PSADT of >12.8 months (Fig. 1a). None of the patients with a PSADT of >12.8 months developed metastases during the follow-up, whilst 10 patients with a PSADT of 12.8 months had distant metastases at a median (95% CI) of 8 (7 10) years. The median PSADT of 9.6 months in the present study is accordance with the median PSADT of 9 months reported by Pruthi et al. [15]. The risk of developing metastases with highgrade tumours is corroborated in the present study, as the probability of 10-year clinical recurrence-free survival was 91% for welldifferentiated tumours, in contrast to 47% for poorly differentiated prostate cancer (Table 2). The influence of age on the behaviour of localized prostate cancer is controversial. In a recent review, age had no prognostic significance for the outcome of localized prostate cancer [17]. However, in the present study of patients treated by RP, a more favourable prognosis was suggested for men aged 57 years (upper quartile) at RP than for men aged >57 years (Table 1). None declared. REFERENCES 1 Kupelian P, Katcher J, Levin H, Zippe C, Klein E. Correlation of clinical and pathologic factors with rising prostatespecific antigen profiles after radical prostatectomy for clinically localised prostate cancer. Urology 1996; 48: Han M, Partin AW, Zahurak M, Piantadosi S, Epstein JI, Walsh PC. Biochemical (prostate specific antigen) recurrence probability following radical prostatectomy for clinically localized prostate cancer. J Urol 2003; 169: Hull GW, Rabbani F, Abbas F, Wheeler TM, Kattan MW, Scardino PT. Cancer control with prostatectomy alone in 1,000 consecutive patients. J Urol 2002; 167: Stamey TA, Johnstone IM, McNeal JE, Lu AY, Yemoto CM. Preoperative serum prostate specific antigen between 2 and 22 ng./ml. correlate poorly with post-radical prostatectomy cancer morphology: prostate specific antigen cure rates appear constant between 2 and 9 ng./ml. J Urol 2002; 167: Noguchi M, Stamey TA, McNeal JE, BJU INTERNATIONAL

5 SERUM PSA AND CLINICAL RECURRENCE AFTER RRP FOR LOCALIZED PROSTATE CANCER Yemoto CM. Preoperative serum prostate specific antigen does not reflect biochemical failure rates after radical prostatectomy in men with large volume cancers. J Urol 2000; 164: Djavan B, Moul JW, Zlotta A, Remzi M, Ravery V. PSA progression following radical prostatectomy and radiation therapy: new standards in the new millennium. Eur Urol 2003; 43: Moul JW. Prostate specific antigen only progression of prostate cancer. J Urol 2000; 163: Shinghal R, Yemoto C, McNeal JE, Brooks JD. Biochemical recurrence without PSA progression characterizes a subset of patients after radical prostatectomy. Urology 2003; 61: Ward JF, Blute ML, Slezak J, Bergstralh EJ, Zincke H. The long-term clinical impact of biochemical recurrence of prostate cancer 5 or more years after radical prostatectomy. J Urol 2003; 170: Patel A, Dorey F, Franklin J, dekernion JB. Recurrence patterns after radical retropubic prostatectomy: clinical usefulness of prostate specific antigen doubling times and log slope prostate specific antigen. J Urol 1997; 158: Roberts SG, Blute ML, Bergstralh EJ, Slezak JM, Zincke H. PSA doubling time as a predictor of clinical progression after biochemical failure following radical prostatectomy for prostate cancer. Mayo Clin Proc 2001; 76: D Amico AV, Whittington R, Malkowicz SB et al. Predicting prostate specific antigen outcome preoperatively in the prostate specific antigen era. J Urol 2001; 166: Jhaveri FM, Zippe CD, Klein EA, Kupelian PA. Biochemical failure does not predict overall survival after radical prostatectomy for localized prostate cancer: 10-year results. Urology 1999; 54: Pound CR, Partin AW, Eisenberger MA, Chan DW, Pearson JD, Walsh PC. Natural history of progression after PSA elevation following radical prostatectomy. JAMA 1999; 281: Pruthi RS, Johnstone I, Tu IP, Stamey TA. Prostate-specific antigen doubling times in patients who have failed radical prostatectomy. correlation with histologic characteristics of the primary cancer. Urology 1997; 49: Cannon GM Jr, Walsh PC, Partin AW, Pound CR. Prostate-specific antigen doubling time in the identification of patients at risk for progression after treatment and biochemical recurrence for prostate cancer. Urology 2003; 62 (Suppl. 1): Parker CC, Gospodarowicz M, Warde P. Does age influence the behavior of localized prostate cancer. BJU Int 2001; 87: Lu-Yao GL, Yao SL. Population based study of long-term survival in patients with clinically localised prostate cancer. Lancet 1997; 349: Barry MJ, Albertsen PC, Bagshaw MA et al. Outcomes for men with clinically non-metastatic prostate carcinoma managed with radical prostatectomy, external beam radiotherapy, or expectant management: a retrospective analysis. Cancer 2001; 91: Correspondence: Svein A. Haukaas, Haraldsplass Deaconess Hospital, Ulriksdal 8, N-5009 Bergen, Norway. svein.haukaas@kir.uib.no Abbreviations: RP, radical prostatectomy; EBRT, external beam radiotherapy; PSADT, PSA doubling time BJU INTERNATIONAL 55