|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2001 American Society for Clinical Oncology Prognostic Significance of Pathologic Features in Localized Prostate Cancer Treated With Radical Prostatectomy: Implications for Staging Systems and Predictive ModelsByFrom the Cancer Research Program, Garvan Institute of Medical Research and Departments of Urology, Anatomical Pathology, and Medical Oncology, St Vincent’s Hospital, Darlinghurst; and Douglass Hanly Moir Pathology and Sydney Diagnostic Services, North Ryde, Australia. Address correspondence to Phillip D. Stricker, MD, Department of Urology, St Vincent’s Clinic, Victoria St, Darlinghurst, New South Wales 2010, Australia; email: stricker{at}ozemail.com.au
PURPOSE: Although predicting outcome for men with clinically localized prostate cancer (PC) has improved, the staging system and nomograms used to do this are based on results from the North American health system. To be internationally applicable, these models require testing in cohorts from a variety of different health systems based on the predominant PC case identification methods used. PATIENTS AND METHODS: We studied 732 men with localized PC treated with radical prostatectomy and no preoperative therapy between 1986 and 1999 at one Australian institution to determine the effect of clinicopathologic features on disease-free survival.
RESULTS: Preoperative serum prostate-specific antigen (PSA) concentration, Gleason score, pathologic stage, and year of surgery were independent predictors of outcome. Although margin status demonstrated only a trend toward significance in multivariate modeling overall, it proved to be independent in subgroups based on later year of surgery (1986 to 1994 v 1995 to 1998), preoperative PSA of less than 10 ng/mL, and Gleason score CONCLUSION: This work confirms the prognostic significance of pathologic stage, Gleason score, and preoperative serum PSA. In the context of a contemporaneous screening effect in Australia, these findings may have implications for methods that predict outcome following surgery as screening becomes more prevalent in a population. The independent prognostic effect of margin status may alter with an increase in the proportion of screening-identified PCs. Staging systems and nomograms that predict outcome following surgery require validation in cohorts with different health practices before being universally applied.
PROSTATE CANCER (PC) is a major international health problem.1-4 The benefit of screening and early treatment of PC is controversial.5-8 Despite this, a large number of men undergo investigations directed at early diagnosis and subsequently have therapy for localized PC. Decision-making for these men and their clinicians is hampered by a deficit of evidence on the outcome for PC. Several North American groups have published data on cancer control following radical prostatectomy (RP). These cohorts now form the basis of recent work on predicting adverse pathologic features based on pretreatment factors,9,10 outcome following surgery based on pretreatment and RP specimen pathologic factors,11-14 and the natural history of patients having prostate-specific antigen (PSA) relapse following RP.15 The predictive importance of pathologic substages A, B, and C within pT2 and A and B within pT3 subgroups of the tumor-node-metastasis (TNM) classification is debated.16,17 Data from different populations have potential value in modifying the staging system to ensure it is simple yet prognostic and internationally applicable.18,19 Staging systems are of great utility in planning therapy for patients. However, there are few studies of the parameters used within these systems in large cohort studies from outside the United States. This has potential importance because the United States has a unique health system that includes advocated PC screening20 and annual medical evaluation. The presentation and clinical outcome for patients outside the United States and in countries where screening is not advocated and annual checks are not considered as the standard of care may differ from those within such systems on which predictive nomograms and staging systems are based. Recent evidence suggests that there are significant practice differences between the United States and Canada that are likely to alter the selection for therapy and outcome for PC,21 whereas factors predicting outcome in smaller European RP series may vary in impact when compared with data from the United States.22 Hence, sizable international cohorts that test the value of the parameters used in prognostic systems generated from patients studied within the United States are needed. Individual studies have examined outcomes in primary treatment cohorts based on prognostic pathologic features.23-26 However, none have evaluated the relative effect of predictive factors in subgroups delineated by clinicopathologic features across one cohort at the same time. In addition, there are to date no published studies that examine changes in factors predictive of outcome with the onset of PSA-based case finding, although several North American groups have evaluated change in pathologic features over time.27-30 We report on a cohort of 732 patients with clinically localized adenocarcinoma of the prostate treated with retropubic RP and no preoperative therapy in an Australian university hospital. A focus of this study was to test the previously defined predictive clinicopathologic factors, pretreatment serum PSA, Gleason score, and stage, as well as other pathologic features, such as surgical margin status, capsular invasion without penetration, perineural invasion (PNI), and vascular invasion. In addition, subgroups based on pathologic features were evaluated for factors predictive of outcome with the aim of providing data for decision-making in these groups and to raise questions that may warrant examination in prospective trials. Outcome and prognostic variables in year-based cohorts were evaluated to delineate changes in the prevalence of PSA-based case finding in the population over time. The fact that the cohort has been assembled outside of North America means that similarities and differences delineated have implications for the broader applicability of previously published work, especially for countries that do not have a PC screening program or culture of annual health checks like that of the United States.
Study Population The study proceeded with the approval of the St Vincent’s Campus Research Ethics Committee. Between December 1986 and December 1999, 834 men with clinically localized cancer of the prostate underwent surgery for planned RP on the St Vincent’s Hospital campus. Preoperative work-up in addition to serum PSA, digital rectal examination (DRE), and histologic diagnosis included pelvic computed tomography, transrectal prostatic ultrasound, and technetium bone scan. From February 1996 onward, patients with serum PSA concentrations below 10 ng/mL were not routinely subjected to bone scan. Only two patients in the entire cohort did not undergo pelvic computed tomography scanning. Of the original cohort, the following 102 patients were excluded from this analysis: 83 men treated with neoadjuvant hormonal therapy, eight patients in whom RP did not proceed because of pelvic lymph node metastases detected intraoperatively, four with histologic diagnoses other than acinar adenocarcinoma (prostatic ductal predominant carcinoma, three; sarcoma, one), two men treated with perineal RP, and five men in whom, despite having a preoperative diagnosis of PC, their RP specimen contained no identifiable cancer (the so-called "vanishing cancer syndrome").31 The cohort for analysis consisted of 732 men with acinar adenocarcinoma of the prostate treated with retropubic RP and no preoperative treatment with a final censor date of December 31, 1999. Surgery was undertaken by one of six urologists (Phillip C. Brenner, MD, David Golovsky, MD, Raji Kooner, MD, Gordon F. O’Neill, MD, Phillip D. Stricker, MD, or the late J. David Wilson, MD) using a similar retropubic prostatectomy technique. The urologists performed 69, 103, 24, 22, 300, and 214 RPs, respectively, within the study. Relapse-free survival rates were not significantly different between surgeons with known prognostic factors taken into account (P values range from .23 to .98 when each surgeon was compared with all others and modeled with multivariate factors found to be independent. More than 97% of the patients were white. Patients given adjuvant radiation therapy (n = 44) were treated with a four-field technique delivering 49.6 to 66.9 Gy in 25 to 30 fractions commencing 2.5 to 5.6 months after surgery. Serum PSA testing levels corrected for age vary significantly across Australia. Our cohort was predominantly drawn from New South Wales, where in 1996 the age-standardized PSA testing rate was 5,790 per 100,000 males, which is similar to Australia as a whole, which has a rate of 5,674 per 100,000 males.32
Tissue Handling and Pathology Reporting
Follow-Up and Data Acquisition
Outcome Measures
Statistical Analysis
Demographic, clinical, pathologic, and outcome data for the cohort of 732 patients are summarized in Tables 1 and 2. The mean follow-up of the cohort, measured from the date of surgery, was 41.1 months (range, 1.0 to 167.7 months; median, 39.3 months). The mean time to relapse was 19.2 months (range, 0.2 to 79.4 months; median, 14.0 months). The overall relapse rate was 20.4% with an actuarial disease-free survival rate at 5 years after RP of 72.7% (95% confidence interval [CI], 68.6% to 76.7%) and at 7 years, 68.4% (95% CI, 63.7% to 73.2%). Most relapses (123 of 149 relapses, 82.5%) occurred within 3 years of surgery. Of 149 patients to relapse, 101 did so with a increase in PSA alone, 27 had local recurrence and PSA increase, four had distant metastases with PSA increase, one had a local recurrence with no change in PSA, three had orchidectomy, and 13 commenced indefinite hormonal therapy. In comparing preoperative factors in those patients who commenced indefinite postoperative hormonal therapy with those who relapsed on the basis of PSA and/or local recurrence, serum PSA (P = .63) and biopsy Gleason score (P = .38) did not distinguish between the two groups. However the two groups differed, with a higher clinical stage in the hormonal therapy group (P = .02). In comparing pathologic features between these two relapsed groups, patients receiving indefinite hormonal therapy had higher RP Gleason scores (mean, 7.81 v 6.87; P = .01), overall pathologic stage (P = .001), SVI rates (11 of 16 [69%] v 41 of 133 [31%]; P = .003), lymph node (LN) metastases (four of 16 [25%] v eight of 133 [6%], P = .01), and margin involvement rates (15 of 16 [94%] v 48 of 133 [36%], P = .007) than those with local or biochemical disease recurrence. This suggests that the institution of indefinite hormonal therapy could not be easily predicted preoperatively and that the presence of adverse prognostic features would have meant the patients had a high risk for relapse in the absence of hormonal therapy. The small number of deaths in the cohort meant that absolute and cancer-specific survival analyses were of limited value at this length of follow-up (Table 1).
Significant predictors of relapse in univariate analysis for the entire cohort were preoperative serum PSA, Gleason score, overall pathologic stage, surgical margin involvement, extraprostatic extension, PNI, vascular invasion, SVI, LN metastases, and year of RP (1986 to 1994 v 1995 to 1999) (Table 3, Figs 1 and 2). In examining Gleason 7 cases (n = 251), primary grade 4 patients were more likely to relapse than primary grade 3 patients (log-rank P = .013). In multivariate analysis for Gleason 7 cases only, primary grade 4 versus 3 was not statistically significant (P = .056) whereas SVI, LN metastases, and margin status were independent predictors of outcome in the model. Patients undergoing any form of adjuvant therapy had a significantly worse prognosis compared with patients treated with RP alone (Table 3). This was due to the selection of patients for adjuvant therapy who had adverse features (data not shown). However, the exclusion of patients given adjuvant therapy did not alter any of the interrelationships or outcome parameters described. In multivariate analysis, independent predictors of outcome were the following, in descending order of predictive magnitude as determined by  2 statistic: LN metastases, pT2 versus pT3, SVI, year of RP, Gleason score, and preoperative PSA, whereas vascular invasion, PNI, margin status, and adjuvant therapy use were not (Table 4). Once again, the exclusion from analysis of the 83 patients who received adjuvant therapy yielded an identical ranking based on the 2 statistic in multivariate analysis. The use of different years to dichotomize the cohort resulted in the year of RP being an independent variable, using the end of 1993, 1994, and 1996 as cutoffs, with later groups having better outcome in all cases. Year of surgery was an independent predictor of outcome when used as a continuous variable (hazards ratio [HR], 0.82; 95% CI, 0.75 to 0.89; P < .0001), with the chance of relapse decreasing 18% for every year later that surgery was undertaken. Similar results were obtained from multivariate analysis using PSA dichotomized at 10 ng/mL or as a continuous variable, Gleason score dichotomized at  6 and  7 or as a continuous variable, and year of surgery dichotomized at the end of 1994 or as a continuous variable.
Subgroups Based on Staging and Other Pathologic Features pT2/capsule-confined cancer. Pathologic T2 disease was present in 55.5% of the cohort (Table 2). Of these 406 patients, 113 (27.8%) had at least one positive surgical margin. The presence of surgical margin involvement in pT2 disease, where the PC would otherwise have been organ-confined, is adversely prognostic (HR, 1.97; 95% CI, 1.01 to 3.87; P = .05). pT2 substages (A, B, or C) did not have a prognostic impact overall (P = .22), but the difference between pT2A and pT2B combined with pT2C suggested a trend toward significance (HR, 2.48; 95% CI, 0.88 to 7.01; P = .09, log-rank P = .08) (Fig 3A). Although the extent of capsular invasion and penetration correlated with outcome overall (Fig 2A, Table 3), within the group of patients with pT2 disease the presence of capsular invasion compared with noninvolvement of the capsule demonstrated only a trend toward significance (log-rank P = .11). In multivariate analysis, independent predictors of relapse at a P < .05 level within this subgroup were RP Gleason score and preoperative PSA 10 ng/mL.
Extracapsular extension. Extracapsular extension (ECE) was present in 42.8% of the cohort. ECE was associated with SVI (P < .0001), LN metastases (P = .004), and RP Gleason score 7 (P < .0001). SVI and/or LN metastases occurred in the absence of extracapsular extension in 14 patients (1.9%); 10 had SVI, four had LN metastases, and one had both. Within cases staged pT3A or B, stages A and B did not have prognostic importance (log-rank P = .61) (Fig 3B). In multivariate analysis for the group staged pT3A or pT3B, independent predictors of relapse at a P < .05 level were LN metastases, SVI, RP Gleason score, and pretreatment PSA 10 ng/mL. Importantly, the presence of established ECE as opposed to focal ECE did not have prognostic value within the pT3A/B group in either univariate or multivariate models. SVI. SVI was present in 13.1% of the cohort. SVI correlated strongly with all standard preoperative and postoperative prognostic indicators. SVI was a strong predictor of outcome in univariate analysis (Table 3) and an independent predictor of outcome in multivariate analysis. In cases with SVI, a number of factors predicted relapse, including margin status. Outcome was independently predicted, in descending order of effect, by LN metastases, margin status, RP Gleason score, and pretreatment PSA but not adjuvant therapy, vascular invasion, or PNI.
LN metastases.
LN metastases were reported in 2.3% of the cohort. LN status correlated significantly with Gleason score (P < .0001), P stage (< 0.0001), ECE (P = .004), PNI (P = .01), and vascular invasion (P = .02) as well as pretreatment serum PSA as a continuous variable (P = .01). Significant thresholds for LN metastases occurred at RP Gleason score Surgical margin involvement. Positive surgical margins were present in 45.9% of the entire cohort with multiple margins present in 18.2%. Margin status was a predictor of outcome in univariate analysis (Fig 2B, Table 3). The presence of positive margins correlated significantly with all postoperative parameters of univariate significance. The presence of a single apical margin was of the same prognostic significance as single margins at other anatomic sites. In multivariate analyses for the whole cohort, the prognostic effect of positive margins was lost. However, margin status was prognostic in patients with pT2 PC and an independent predictor of outcome in those patients with SVI. There was a trend toward adjuvant radiation therapy altering the outcome of patients with positive margins as a whole group, but this did not reach statistical significance (P = .18). However, of 133 patients with multiple margins, postoperative RT was given to 30, and these patients had significantly better outcomes compared with the 103 not given such therapy (HR, 2.99; 95% CI, 1.27 to 7.06; P = .01; log-rank P = .008) (Fig 4). In patients with multiple positive margins, independent predictors of outcome were, in descending order of effect, LN metastases, SVI, adjuvant radiation therapy, Gleason score, and preoperative PSA, whereas pT2/pT3, PNI, and vascular invasion were not.
Perineural and vascular invasion. PNI was reported in 51.3% of RP specimens (Table 2). The presence and extent of PNI predicted relapse (Table 3, Fig 2C). RP PNI correlated with ECE (P < .0001), SVI (P < .0001), LN metastases (P = .01), positive margins (P < .0001), RP Gleason score (P < .0001), and vascular invasion (P < .0001). The relationship between RP PNI and pretreatment PSA was relatively weak (P > .02). Within the RP specimen, lymphatic invasion was present in 36 cases, venous invasion in eight, and five patients exhibited both, yielding a total of 38 patients with vascular invasion, comprising 5.2% of the total cohort. Vascular invasion correlated with RP Gleason score, ECE, positive surgical margins, PNI, and SVI (all P < .0001). Relationships between vascular invasion and pretreatment PSA (P > .01) and LN metastases (P = .02) were relatively weak. Vascular invasion was a univariate predictor of outcome (Table 3). Neither perineural nor vascular invasion was an independent predictor of outcome in the whole cohort (Table 4).
Year of surgery.
When the cohort was divided based on the year of surgery into two groups at the end of 1993, 1994, 1995, or 1996, the more recent groups demonstrated a better survival rate than the former (as illustrated in Fig 5 and Table 3 for 1986 to 1994 v 1995 to 1998). In multivariate analyses, dichotomizing the cohort at the end of 1993, 1994, or 1996 or using the year of surgery as a continuous variable resulted in an effect on outcome that was independent of preoperative serum PSA, pathologic stage variables, and Gleason score. In comparing the two groups, 1986 to 1994 and 1995 to 1998 (Table 5), there was a statistically nonsignificant increase in patients with no palpable clinical abnormality (T1C) (P = .12) and significantly lower preoperative serum PSA concentration (P < .0001) in the latter group. In comparing staging variables based on TNM classification, including SVI and LN involvement, there was no significant change between the two groups. There was an increase in Gleason score (P < .0001) and a fall in the use of adjuvant therapy (P < .0001), two factors unlikely to contribute to improved outcome alone. In further assessing the changes in Gleason score between the groups, there was a significant progressive increase in cancers with Gleason grade 6 or 7 and a corresponding reduction in Gleason grade 4 or 5 tumors and those with Gleason score
When univariate analysis was undertaken in each of the groups based on year of surgery, all variables significant in the entire cohort predicted outcome (data not shown). However, in multivariate analysis, independent predictors varied with time so that whereas margin status was not an independent predictor of outcome in the early part of the cohort undergoing surgery from 1986 to 1994, it became so in that part of the cohort receiving RP from 1995 to 1998 (Table 6). Given that significant changes in preoperative serum PSA concentration and Gleason score were evident between the two cohorts (Table 5), multivariate analyses were undertaken in subgroups of patients with preoperative PSA less than or 10 ng/mL and Gleason score 6 or 7, respectively (Tables 6 and 7). The result was that margin status proved independently prognostic in the subgroup in whom the preoperative PSA was less than 10 ng/mL but not the subgroup of cases in whom it was 10 ng/mL (Table 7). Similarly, margin status independently predicted outcome in patients with a Gleason score 7 but not in the subgroup with Gleason score 6 (Table 8).
This study confirms a number of key relationships between prognostic parameters and relapse demonstrated in previously published large North American cohorts, that is, inflection points for prognosis in pretreatment serum PSA at 4, 10, and 20 ng/mL; in stage at ECE (pT2/pT3), SVI, and LN metastases and in Gleason score at 4, 6, 7, and 811,12,41; and the independence of serum PSA, ECE (pT2/pT3), SVI, LN metastases, and Gleason score in predicting disease relapse.11,12,42 These data show the generalizability of these inflection points and their geographical transportability to an Australian setting.43 Furthermore, the data from our cohort demonstrate the following: (1) Gleason score, preoperative PSA, and the presence of either or both LN metastases and SVI are consistent prognostic factors in most subgroups based on pathologic features; (2) the difference in outcome between pT2 substages based on the 1992 TNM classification does not reach statistical significance; (3) there is no difference in outcome between pT3A and pT3B cases based on the 1992 TNM classification and neither the extent of extraprostatic extension nor margin status has any prognostic significance within this group; (4) margin status showed only a trend toward an independent prognostic effect in the overall cohort but had prognostic effect in pT2 stage disease and an independent prognostic effect in patients with SVI. The independent effect of margin status in multivariate analysis was time-dependent, preoperative serum PSA concentration–dependent, and Gleason score–dependent. Specifically, analyses of subgroups undergoing surgery after 1995 with preoperative serum PSA concentration of less than 10 ng/mL or Gleason scores 7 demonstrate an independent prognostic effect whereas analyses of groups before 1995 with preoperative serum PSA concentration of 10 ng/mL or with Gleason score of 6 did not; (5) adjuvant radiation therapy reduces relapse in patients with multiple positive margins; (6) men undergoing RP between 1986 and 1999 have become progressively younger, with lower preoperative serum PSA concentrations and a higher likelihood of impalpable disease similar to that described in the United States,29 and (7) during the same time period, there has been a reduction in the number of patients with positive surgical margins and established extracapsular carcinoma, and an improvement in relapse-free survival rates. This improvement in outcome has occurred without significant change to pathologic staging parameters, including ECE and SVI. Assessment of other potential predictive factors, such as PNI and vascular invasion, proved not to be of independent prognostic significance. A limitation of the study is that we did not test for new prognostic inflection points, such as different preoperative serum PSA concentrations, but this was not the main objective of the study and, along with calibration exercises, would be best undertaken in large multi-institutional or population-based cohorts.43 A further limitation of this study is its reliance on subgroup multivariate analyses to explain differences in the prognostic effect of margin status based on Gleason score, preoperative PSA level, and year of surgery. Although such subsets are never strictly comparable, interaction analysis of the four variables concerned demonstrates complex relationships between each of them. In this context, dichotomizing the variables at predetermined independent prognostic inflection points allows comparison of the subsets and suggests the need for further analysis with nomogram models that can include or exclude margin status. In comparison to large contemporary series emanating from the United States, this cohort has higher preoperative serum PSA concentrations and a higher incidence of clinically palpable, extracapsular extension, and margin-positive tumors. In many respects it displays similar characteristics to series accumulated soon after the introduction of PSA testing in the United States.41,44 However, these characteristics have changed markedly with time so that our more recent patients undergoing RP are similar in many respects to those described in later North American cohorts.38 The epidemiology of PC in Australia during the period from which our cohort derives has also changed markedly. PC incidence in Australia increased substantially between 1988 and 1994 and decreased in 1995 and 1996.1 This mirrors patterns seen in the United States, albeit with a 3-year lag,7,45 and corresponds with the increasing use of serum PSA testing.46 A recent decrease in PC mortality in the United States may also be evident in mortality data for Australia in 1997 and 1998.47
In summary, Australia has evolved toward a PC-screened population similar to the United States, albeit several years later. When changes in the relative effect of prognostic parameters are considered in the context of this evolution, it may have significant implications for staging systems as well as predictive nomograms that use staging system variables. Our cohort is characterized by a progressive decline in preoperative serum PSA concentrations, an increase in Gleason scores, and concurrent development of an independent prognostic effect of margin status. Further subgroup analysis in the same cohort demonstrates that margin status is only independently prognostic in groups with Gleason The data presented in this study support the merger of pT2A with pT2B and pT3A with pT3B stages in the 1997 revision of the TNM staging system. No pathologic parameters, including margin status, divided this pT3 subgroup in a prognostically significant way. However, the presence of positive surgical margins was prognostic within the pT2 group, although this was not an independent prognostic factor in multivariate analysis. Within the pT2 group, the stratification of patients by substages did result in prognostic ordering of the A, B, and C subgroups so that each higher subgroup had a successively poorer outcome (Fig 3A). However, prognostic differences between each of the successive groups failed to reach statistical significance. The presence or absence of spread into the capsule in pT2 cancers was only definitely reported in 339 of 407 patients, with the remaining 78 not defined in the contemporaneous pathology report. When those of unknown status were excluded, there was a trend toward patients with spread into the capsule having a poorer outcome than those who had no capsular involvement (P = .06), a finding consistent with a previous report from another group demonstrating this status to be prognostic in pT2 cancers.37 The utility of substages, margin status, and capsular invasion within the pT2 group may be evident with longer follow-up in our cohort but will ultimately require multicenter pooled cohort analysis for resolution. For consistent application and ease of use, TNM staging systems used to classify neoplasms clinically and pathologically should ideally be parallel if not identical. Two recent studies by other groups have demonstrated that the merger of cT2A92 and cT2B92 cases to one stage (cT2A97) in the transition from 1992 to 1997 TNM systems results in a loss of prognostic effect, whereas, conversely, there is no statistical difference in outcome between cT2B92 and cT2C92 cases.48,49 It has been suggested that the clinical staging system might be modified by returning to the 1992 definition for cT2A and/or merging 1992 cT2B and cT2C groups. Such a change translated into the pathologic staging system has the potential to obscure a difference in prognostic effect between PC involving both lobes of the prostate but not extending through the capsule (pT2C92 or pT2B97) and that involving one lobe or less (pT2A/B92 and pT2A97) suggested by our data. In this regard, uniformity in clinical and pathologic T2 PC staging may be better served by returning to the A, B, C subgroup classification originally used in the 1992 TNM system. The fact that margin positivity and the presence of extensive extracapsular disease are not designated by either 1992 or 1997 TNM staging systems may be important deficiencies in these systems if they are to be applicable internationally. Margin status was prognostic in two pathologically defined subgroups, pT2 and SVI. Analysis of a large series from the Mayo clinic suggested that margin involvement is adversely prognostic in pT2 disease, and the authors recommended that pT2/margin-positive status be incorporated into the TNM staging system.23 In a recently published study, Epstein et al26 found that margin involvement, Gleason score, and vascular invasion independently predict outcome in patients with SVI. Multiple margin involvement was a major indication for use of adjuvant radiation therapy (RT) and the use of adjuvant RT in patients with multiple margins improved outcome and was an independent prognostic factor in this subgroup of our cohort. This provides further evidence for efficacy of adjuvant RT in the margin-positive group50 and emphasizes the need for randomized trials to further define the role of RT in margin-positive patients. Several groups have reported a fall in positive margin and extracapsular disease rates over similar time courses to ours.28-30 These groups are all based in the United States, where PSA screening is recommended by several national bodies, most notably the American Cancer Society,20 and annual medical checks, including DRE and serum PSA estimation, are common practice. In Australia, although national screening programs are in place for cervical and breast cancer, there is no national screening program for PC, and no national body has recommended population-based screening. Despite this, many men undergo de novo screening either on the basis of lower urinary tract symptoms or concern about PC, and a smaller number have serum PSA estimations performed as part of a yearly medical assessment.51-53 The use of PSA testing has increased markedly over the past decade,46 and more than 50% of such tests undertaken in the late 1990s were for screening in asymptomatic men.54,55 This has resulted in more men being diagnosed, diagnosis at a younger age, and more men undergoing RP with no palpable abnormality on DRE.46 Although two other Australian groups have reported selection criteria for surgery and postoperative morbidity in cohorts of men undergoing RP,56,57 this study is the first to report cancer control outcome. The improved relapse-free survival rate seen in patients treated after 1994 may be a result of a number of factors, including lead-time bias, lag-time bias, and/or differential therapeutic benefit from surgical intervention.58 A progressive reduction in preoperative PSA with time favors a change in selection of patients for surgery, possibly facilitated by a screening effect. The increase in the number of patients with T1C disease and a fall in age at surgery also favor a screening effect in our RP population. The lack of a significant decline in pathologic stage and in particular in the numbers of patients with SVI in our cohort, a finding consistent with one large United States series28 but inconsistent with another,29 reemphasizes the need for reproducible preoperative predictors of pathologic stage if such patients are to be spared surgery or stratified for more aggressive combination therapy. A decline in margin rates, particularly multiple margin rates, and the incidence of extracapsular disease can be at least partially explained by patient selection with lower clinical stage and preoperative serum PSA concentrations. However, a learning effect leading to better surgical techniques may have contributed significantly to reduced positive-margin rates. This latter assertion is difficult to prove but is suggested by the fact that, in multivariate analysis, successive year-based cohort division was an independent predictor of outcome. The inference is that improvement in and modification of surgical technique in each subsequent time period contributed to better outcome, as described by others.59 In conclusion, this large single-institution Australian series of PC treated with RP provides evidence to support recent changes in the TNM pathologic staging system and the use of parameters incorporated in recent postoperative nomograms.12,14 The international utility of such nomograms will require validation in a variety of therapeutic scenarios and health systems. The potential importance of margin status in patients with pT2 disease or with SVI and of intracapsular invasion in patients with pT2 disease requires further evaluation for their potential to improve the current system. Recently, improved patient selection, in part as a result of a screening effect in the treated population, has contributed to better disease-free outcomes for patients undergoing RP. The finding that adjuvant RT significantly improved outcome in patients with multiple surgical margin involvement provides evidence for the use of this therapy in these patients but requires a randomized, stratified, control trial to further delineate its effect.
1. Coates M, Tracey EA: Cancer in New South Wales: Incidence and Mortality 1997. Sydney, Australia, NSW Central Cancer Registry, 2000
2.
Parkin DM, Pisani P, Ferlay J: Global cancer statistics. CA Cancer J Clin 49: 33-64, 1999 3. Pisani P, Parkin DM, Bray F, et al: Estimates of the worldwide mortality from 25 cancers in 1990. Int J Cancer 83: 18-29, 1999[Medline] 4. Hsing AW, Tsao L, Devesa SS: International trends and patterns of prostate cancer incidence and mortality. Int J Cancer 85: 60-67, 2000[Medline]
5.
Coley CM, Barry MJ, Fleming C, et al: Early detection of prostate cancer: II. Estimating the risks, benefits, and costs—American College of Physicians. Ann Intern Med 126: 468-479, 1997
6.
Boyle P: Prostate specific antigen (PSA) testing as screening for prostate cancer: The current controversy. Ann Oncol 9: 1263-1264, 1998
7.
Hankey BF, Feuer EJ, Clegg LX, et al: Cancer surveillance series: Interpreting trends in prostate cancer—I: Evidence of the effects of screening in recent prostate cancer incidence, mortality, and survival rates. J Natl Cancer Inst 91: 1017-1024, 1999 8. Horninger W, Reissigl A, Rogatsch H, et al: Prostate cancer screening in the Tyrol, Austria: Experience and results. Eur J Cancer 36: 1322-1335, 2000 9. Partin AW, Kattan MW, Subong EN, et al: Combination of prostate-specific antigen, clinical stage, and Gleason score to predict pathological stage of localized prostate cancer: A multi-institutional update. JAMA 277: 1445-1451, 1997[Abstract] 10. Kattan MW, Stapleton AM, Wheeler TM, et al: Evaluation of a nomogram used to predict the pathologic stage of clinically localized prostate carcinoma. Cancer 79: 528-537, 1997[Medline]
11.
Kattan MW, Eastham JA, Stapleton AM, et al: A preoperative nomogram for disease recurrence following radical prostatectomy for prostate cancer. J Natl Cancer Inst 90: 766-771, 1998
12.
Kattan MW, Wheeler TM, Scardino PT: Postoperative nomogram for disease recurrence after radical prostatectomy for prostate cancer. J Clin Oncol 17: 1499-1507, 1999
13.
Stamey TA, McNeal JE, Yemoto CM, et al: Biological determinants of cancer progression in men with prostate cancer. JAMA 281: 1395-1400, 1999 14. Stamey TA, Yemoto CM, McNeal JE, et al: Prostate cancer is highly predictable: A prognostic equation based on all morphological variables in radical prostatectomy specimens. J Urol 163: 1155-1160, 2000[Medline]
15.
Pound CR, Partin AW, Eisenberger MA, et al: Natural history of progression after PSA elevation following radical prostatectomy. JAMA 281: 1591-1597, 1999 16. Grignon DJ, Sakr WA: Pathologic staging of prostate carcinoma: What are the issues? Cancer 78: 337-340, 1996[Medline] 17. Yarbro JW, Page DL, Fielding LP, et al: American Joint Committee on Cancer prognostic factors consensus conference. Cancer 86: 2436-2446, 1999[Medline] 18. Chisholm GD, Hedlund PO, Adolfsson J, et al: The TNM system of 1992: Comments from the TNM working group. Consensus Conference on Diagnosis and Prognostic Parameters in Localized Prostate Cancer. Stockholm, Sweden, May 12-13, 1993. Scand J Urol Nephrol Suppl 162:107-114; discussion 115-127, 1994 19. Hermanek P, Sobin LH, Wittekind C: How to improve the present TNM staging system. Cancer 86: 2189-2191, 1999[Medline] 20. von Eschenbach A, Ho R, Murphy GP, et al: American Cancer Society guidelines for the early detection of prostate cancer: Update, June 10:1997. Cancer 80: 1805-1807, 1997[Medline] 21. Fleshner N, Rakovitch E, Klotz L: Differences between urologists in the United States and Canada in the approach to prostate cancer. J Urol 163: 1461-1466, 2000[Medline] 22. van den Ouden D, Hop WC, Kranse R, et al: Tumour control according to pathological variables in patients treated by radical prostatectomy for clinically localized carcinoma of the prostate. Br J Urol 79: 203-211, 1997[Medline] 23. Blute ML, Bostwick DG, Seay TM, et al: Pathologic classification of prostate carcinoma: The impact of margin status. Cancer 82: 902-908, 1998[Medline] 24. Tefilli MV, Gheiler EL, Tiguert R, et al: Should Gleason score 7 prostate cancer be considered a unique grade category? Urology 53: 372-377, 1999[Medline] 25. Tefilli MV, Gheiler EL, Tiguert R, et al: Role of radical prostatectomy in patients with prostate cancer of high Gleason score. Prostate 39: 60-66, 1999[Medline] 26. Epstein JI, Partin AW, Potter SR, et al: Adenocarcinoma of the prostate invading the seminal vesicle: Prognostic stratification based on pathologic parameters. Urology 56: 283-288, 2000[Medline] 27. Walsh PC, Partin AW, Epstein JI: Cancer control and quality of life following anatomical radical retropubic prostatectomy: Results at 10 years. J Urol 152: 1831-1836, 1994[Medline] 28. Soh S, Kattan MW, Berkman S, et al: Has there been a recent shift in the pathological features and prognosis of patients treated with radical prostatectomy? J Urol 157: 2212-2218, 1997[Medline] 29. Stamey TA, Donaldson AN, Yemoto CE, et al: Histological and clinical findings in 896 consecutive prostates treated only with radical retropubic prostatectomy: Epidemiologic significance of annual changes. J Urol 160: 2412-2417, 1998[Medline]
30.
Jhaveri FM, Klein EA, Kupelian PA, et al: Declining rates of extracapsular extension after radical prostatectomy: Evidence for continued stage migration. J Clin Oncol 17: 3167-3172, 1999 31. Goldstein NS, Begin LR, Grody WW, et al: Minimal or no cancer in radical prostatectomy specimens: Report of 13 cases of the "vanishing cancer phenomenon." Am J Surg Pathol 19: 1002-1009, 1995[Medline] 32. Smith D, Armstrong B, Saunders R: Patterns of Prostate-Specific Antigen Testing in Australia, 1992 to 1996: An Examination of Medicare Data. Sydney, Australia, NSW Cancer Council, 1998 33. True LD: Surgical pathology examination of the prostate gland: Practice survey by American Society of Clinical Pathologists. Am J Clin Pathol 102: 572-579, 1994 (published erratum appears in Am J Clin Oncol 103:376, 1995)[Medline] 34. Gleason DF: Histologic grading of prostate cancer: A perspective. Hum Pathol 23: 273-279, 1992[Medline] 35. Epstein JI: The diagnosis and reporting of adenocarcinoma of the prostate in core needle biopsy specimens. Cancer 78: 350-356, 1996[Medline] 36. Schroder FH, Hermanek P, Denis L, et al: The TNM classification of prostate cancer. Prostate 4: 129-138, 1992 37. Wheeler TM, Dillioglugil O, Kattan MW, et al: Clinical and pathological significance of the level and extent of capsular invasion in clinical stage T1-2 prostate cancer. Hum Pathol 29: 856-862, 1998[Medline] 38. Catalona WJ, Smith DS: Cancer recurrence and survival rates after anatomic radical retropubic prostatectomy for prostate cancer: Intermediate-term results. J Urol 160: 2428-2434, 1998[Medline] 39. Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53: 457-481, 1958 40. Cox DR: Regression models and life tables (life tables). J R Stat Soc 34: 187-189, 1972 41. Stamey TA, Kabalin JN, McNeal JE, et al: Prostate specific antigen in the diagnosis and treatment of adenocarcinoma of the prostate: II. Radical prostatectomy–treated patients. J Urol 141: 1076-1083, 1989[Medline] 42. Catalona WJ, Smith DS: 5-Year tumor recurrence rates after anatomical radical retropubic prostatectomy for prostate cancer. J Urol 152: 1837-1842, 1994[Medline]
43.
Justice AC, Covinsky KE, Berlin JA: Assessing the generalizability of prognostic information. Ann Intern Med 130: 515-524, 1999 44. Ackerman DA, Barry JM, Wicklund RA, et al: Analysis of risk factors associated with prostate cancer extension to the surgical margin and pelvic node metastasis at radical prostatectomy. J Urol 150: 1845-1850, 1993[Medline] 45. Wingo PA, Ries LA, Rosenberg HM, et al: Cancer incidence and mortality, 1973-1995: A report card for the U.S. Cancer 82: 1197-1207, 1998[Medline] 46. Smith DP, Armstrong BK: Prostate-specific antigen testing in Australia and association with prostate cancer incidence in New South Wales. Med J Aust 169: 17-20, 1998[Medline] 47. Anonymous ABS Causes of Death Australia 1998. Canberra, Australia, Australian Bureau of Statistics, 1999 48. Stamey TA, Sozen TS, Yemoto CM, et al: Classification of localized untreated prostate cancer based on 791 men treated only with radical prostatectomy: Common ground for therapeutic trials and TNM subgroups. J Urol 159: 2009-2012, 1998[Medline] 49. Han M, Walsh PC, Partin AW, et al: Ability of the 1992 and 1997 American Joint Committee on Cancer staging systems for prostate cancer to predict progression-free survival after radical prostatectomy for stage T2 disease. J Urol 164: 89-92, 2000[Medline] 50. Leibovich BC, Engen DE, Patterson DE, et al: Benefit of adjuvant radiation therapy for localized prostate cancer with a positive surgical margin. J Urol 163: 1178-1182, 2000[Medline] 51. Ward J, Young J, Sladden M: Australian general practitioners’ views and use of tests to detect early prostate cancer. Aust N Z J Public Health 22: 374-380, 1998[Medline] 52. Pinnock CB, Weller DP, Marshall VR: Self-reported prevalence of prostate-specific antigen testing in South Australia: A community study. Med J Aust 169: 25-28, 1998[Medline] 53. Weller D, Pinnock C, Silagy C, et al: Prostate cancer testing in SA men: Influence of sociodemographic factors, health beliefs and LUTS. Aust N Z J Public Health 22: 400-402, 1998[Medline] 54. Ward JE, Gupta L, Taylor NJ: Do general practitioners use prostate-specific antigen as a screening test for early prostate cancer? Med J Aust 169: 29-31, 1998[Medline] 55. Perkins JJ, Sanson-Fisher RW, Clarke SJ, et al: An exploration of screening practices for prostate cancer and the associated community expenditure. Br J Urol 82: 524-529, 1998[Medline] 56. Heathcote PS, Mactaggart PN, Boston RJ, et al: Health-related quality of life in Australian men remaining disease-free after radical prostatectomy. Med J Aust 168: 483-486, 1998[Medline] 57. Frydenberg M, Giles GG, Mameghan H, et al: Prostate cancer in Victoria in 1993: Patterns of reported management. Med J Aust 172: 270-274, 2000[Medline] 58. Brawley OW: Prostate carcinoma incidence and patient mortality: The effects of screening and early detection. Cancer 80: 1857-1863, 1997[Medline] 59. Villers A, Stamey TA, Yemoto C, et al: Modified extrafascial radical retropubic prostatectomy technique decreases frequency of positive surgical margins in T2 cancers < 2 cm. Eur Urol 38: 64-73, 2000[Medline] Submitted December 27, 2000; accepted May 8, 2001. This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
INTRODUCTION
