|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2002 American Society for Clinical Oncology Analysis of Clinical Stage T2 Prostate Cancer: Do Current Subclassifications Represent an Improvement?ByFrom the Departments of Urology and Biostatistics, Memorial Sloan-Kettering Cancer Center, New York, NY; Louisiana State University Health Science Center, Shreveport, LA; Baylor College of Medicine, Houston, TX; and University Hospital Hamburg-Eppendorf, Hamburg, Germany. Address reprint requests to Michael W. Kattan, PhD, Departments of Urology and Biostatistics, Memorial Sloan-Kettering Cancer Center, 1275 York Ave, New York, NY 10021; email: kattanm{at}mskcc.org
PURPOSE: The purpose of this study was to determine whether the extent of palpable cancer within the prostate predicts outcome after radical prostatectomy. PATIENTS AND METHODS: We combined prospectively collected data on 1,755 consecutive clinical stage T2 patients treated with radical prostatectomy alone at four institutions. According to the 1992 American Joint Committee on Cancer tumor-node-metastasis system, 645 (37%) were T2a, 758 (43%) were T2b, and 352 (20%) were T2c. Kaplan-Meier and proportional hazards regression analyses were performed on the 1992 and 1997 T2 subclassifications. After controlling for the effects of prostate-specific antigen (PSA) and biopsy Gleason sum, the two staging systems were compared for their ability to predict recurrence-free survival (RFS). Adjusted RFS curves were constructed using the corrected group prognosis method. RESULTS: Follow-up ranged from 1 to 166 months (median, 26 months). Cancer recurred in 417 (24%) of the T2 patients. The 1992 (P = .005) but not the 1997 (P = .100) T2 subclassification predicted outcome after controlling for PSA and Gleason sum. After covariate adjustment, RFS was 7% higher at 5 years in the 1992 T2a subcategory relative to the T2b subcategory. CONCLUSION: The 1992 American Joint Committee on Cancer system is superior to the 1997 system, and the former adds prognostic information to a model containing pretreatment PSA and Gleason sum. These results suggest that 1992 T2 subclassification derived from palpable findings improves prognostication over the 1997 subclassification.
CLINICAL STAGING SYSTEMS are used to describe the anatomic extent of disease and estimate prognosis, thus guiding treatment. In 1956, Whitmore1 was the first to categorize prostate cancer into stages. His ABCD system remained the international standard for many years. Beginning in the late 1970s there was increasing interest to shift to a tumor-node-metastasis (TNM) system. However, it was not until 1992 that the American Joint Committee on Cancer (AJCC) and the International Union Against Cancer adopted the fourth revision of the TNM system.2 A further revision was published by the AJCC in 1997.3 Although the general classification remained constant, the 1997 revision attempted to simplify clinical stage by merging palpable tumors occupying less than half of one lobe with larger tumors in one lobe into a new category, T2a, and changing the designation of palpable tumor in both lobes from T2c to T2b (Table 1). The 1997 staging system has, however, not gained widespread popularity and is felt by some to be inferior in its ability to predict treatment outcome.4,5
Although the 1992 clinical stage T2 subcategories do reflect differences in pathologic stage and prognosis when used alone, it is not clear that these substage designations actually reflect cancers with different prognoses when the pretreatment level of serum prostate-specific antigen (PSA) and biopsy Gleason grade are considered.4,5 Inclusion and adjustment for these variables is important because they are used in conjunction with clinical stage to gauge the patient’s prognosis.6,7 On the basis of this premise, we investigated the ability of T2 subclassification to predict outcome after radical prostatectomy after controlling for PSA and biopsy Gleason sum.
To obtain a large series of patients with clinical stage T2 prostate cancer, we combined data from four institutional series of patients treated with radical prostatectomy. In each institution, all consecutive patients between the dates listed in Table 2 were included. Pretreatment serum PSA, clinical stage (using the 1992 AJCC guidelines), and biopsy Gleason sum were recorded for each patient. Clinical stage was assigned by the attending surgeon or residents at the participating institution on the basis of digital rectal examination, regardless of the results of ultrasonography or other imaging techniques. Of the 4,187 patients, 888 were excluded from the study because of incomplete clinical information (n = 234) or treatment with neoadjuvant androgen ablation (n = 654), leaving 3,299 assessable patients. Of these, 1,755 (53%) were clinical stage T2. Using the 1992 AJCC substages, 645 patients (37%) were classified as T2a, 758 (43%) were T2b, and 352 (20%) were T2c. Average age at the time of surgery of all patients was 62.6 ± 6.6 years (range, 38 to 81 years). The clinical characteristics of the patients with clinical stage T2 disease are listed in Table 3.
Postoperatively, patients were followed clinically and with serial serum PSA measurements. The Hybritech PSA assay (Hybritech Tandem R, San Diego, CA) was used principally but not exclusively for both pretreatment evaluation and disease follow-up. PSA failures were defined by each center individually and ranged from 0.1 ng/mL to 0.4 ng/mL and rising. Frequency of PSA determinations varied at the different institutions. Patients were generally followed every 3 months for the first year, every 6 months for the second year, and then annually. The administration of adjuvant therapy in advance of PSA increase was considered as treatment failure at the initiation of the adjuvant therapy. Patients in whom radical prostatectomy was aborted because of positive intraoperative lymph node frozen section were considered immediate failures. Patients who did not relapse were censored at the date of their last undetectable PSA.
Statistical Analysis
For the 1,755 stage T2 patients, follow-up ranged from 1 to 166 months (median, 25.6 months). Biochemical disease recurred in 368 patients. Three patients had clinical failure before PSA failure. An additional 27 T2 patients had radical prostatectomy aborted because of positive lymph nodes on frozen section and were considered immediate failures (Table 3). Nineteen patients underwent adjuvant radiotherapy before a PSA increase and were considered to have failed at the initiation of radiation. Death without disease occurred in 32 patients. Follow-up was 5 years or longer for 364 patients (21%). As a measure of loss to follow-up, 556 patients (32%) were not evaluated within the last year of data collection. The actuarial recurrence-free survival (RFS) for the clinical stage T2 patients was 68% (95% confidence interval, 0.65 to 0.70) and 49% (95% confidence interval, 0.43 to 0.55) at 5 and 10 years, respectively. Figures 1, 2, and 3 show the Kaplan-Meier curves illustrating actuarial RFS for all T2 patients as well as for the 1992 and 1997 AJCC T2 substages. The median RFS of all clinical stage T2 patients was 118 months. In a univariable Cox model, RFS was different for the 1992 AJCC substages T2a versus T2b (P < .001) but not for T2b versus T2c (P = .107). RFS was also significantly different for the 1997 AJCC substages T2a and T2b (P < .001, univariable Cox model).
Table 4 lists the results of the Cox multivariable analyses. Two models were constructed: one containing the 1992 TNM clinical staging system, and the other containing the 1997 TNM staging system. Both models included pretreatment PSA and biopsy Gleason sum. The 1992 TNM staging system was associated with RFS (P = .005), whereas the 1997 staging system was not (P = .096). When restricted to the 1992 T2a and T2b patients, clinical substage was associated with RFS after controlling for pretreatment PSA and biopsy Gleason sum (P = .004, likelihood ratio test). The adjusted survival curves for these two substages appear in Fig 4. The curves suggest that the long-term difference between the 1992 T2a and T2b RFS approaches 10% after adjustment for differences in pretreatment PSA and biopsy Gleason sum.
In the present study, we examined whether the clinical T2 substages had different prognoses after radical prostatectomy. Differences in outcomes for T2 substages of both the 1992 and 1997 TNM systems were demonstrated in univariable analysis. In multivariable analysis, these differences dissipated for the 1997 classification, whereas the 1992 system continued to be strongly associated with outcome. It seems that the 1992 T2a versus T2b distinction is important, and that combining these subclassifications undermines the usefulness of T2 substaging. There are some limitations to our study. Our length of follow-up appears somewhat short, with a median of 25.6 months. However, 364 patients (21%) were followed for 5 years or longer, and there were 449 events, which together are more informative measures of data maturity than is median follow-up. Furthermore, other studies have shown that the largest hazard rates for biochemical recurrence occur within the first 2 years and the majority of recurrences occur within the first 5 years after radical prostatectomy.9,10 This suggests that longer follow-up may not affect our findings greatly but is still needed. Another limitation is that we are using a single modeling methodology, the Cox proportional hazards regression model, to combine the information contained in PSA, Gleason sum, and clinical stage. Alternative modeling approaches (eg, regression trees) may be able to exploit information contained in clinical stage in a different manner. However, the methodology for significance testing of variables in this type of analysis remains work in progress. In our study, clinical stage was determined exclusively on the basis of digital rectal examination findings. The use of ultrasound or other imaging modalities such as endorectal magnetic resonance imaging could affect the results. Another limitation is that the end point of our analysis is PSA recurrence. Although this is the earliest and a highly sensitive and specific indicator of treatment failure, it precedes evidence of disease by many years. Thus, not all patients who experience PSA recurrence will live long enough to experience metastasis or death from prostate cancer. Previous studies have focused on clinical staging subclassifications. Stamey et al4 examined clinically localized prostate cancer, stratified according to the 1992 AJCC clinical stage. In their analysis, T2a tumors were similar to T1c tumors when comparing their pathologic features and RFS. They also found that T2a and T1c tumors were significantly different from T2b and T2c cancers. The authors therefore argued against combining 1992 stage T2a with 1992 stage T2b according to the 1997 AJCC TNM system. However, this analysis was performed without consideration of other prognostic factors such as biopsy Gleason grade and PSA. An evaluation of the 1992 and 1997 AJCC TNM clinical staging systems for prostate cancer was recently performed by investigators from Johns Hopkins.5 They reported that 1992 AJCC T2b and T2c cancers had similar freedom from biochemical recurrence after radical prostatectomy, which was significantly worse than for T2a cancers. Therefore, they criticized the modifications in the TNM staging system made in 1997 and suggested grouping T2b and T2c instead. Although this analysis did not consider other prognostic factors, our adjusted analysis agrees with their findings. Others have conducted multivariable analysis of the 1992 TNM T2 substaging system. Iyer et al11 found that T2 substage was associated with RFS in patients receiving external-beam irradiation therapy. Analyses of brachytherapy and watchful waiting for patient outcomes are needed. Staging of localized disease according to palpable findings may be highly subjective because important variability has been shown when palpable findings of the prostate are compared between urologists.12 Although there is considerable noise in the 1992 T2 substage variable, there is apparently valuable signal. Nonetheless, substaging of clinical stage T2 incorporating serum marker results (as is done for testicular cancer) and tumor grade may better stratify patients into different prognostic groups. Furthermore, including detailed information from systematic biopsy specimens or imaging modalities may better identify the extent of cancer and location of extracapsular extension, aiding in treatment planning. Many feel that the goal of a clinical staging system should be to group patients whose outcome is more homogeneous within groups than between groups. Although this is the prevailing opinion, we and others feel that the ideal clinical staging system is that which maximizes prognostic accuracy.13 This may or may not occur by grouping patients, but oftentimes, it will not. Limited studies comparing continuous estimates of risk of relapse have shown better discrimination than those systems that group patients.14,15 We believe in accurate characterization of risk, first and foremost, and thus we were interested in the prognostic value of each staging system in the context of a multivariable prediction model. The 1997 AJCC TNM classification of prostate cancer has been criticized.4,5,11 Our study suggests that the 1992 AJCC system is indeed superior to the 1997 revision by remaining strongly associated with RFS after adjustment for the effects of pretreatment PSA and biopsy Gleason sum.
Supported by grant no. RPG-00-202-01-CCE from the American Cancer Society, a grant from the Deutsche Krebshilfe, and grant no. GR 1866/1-1 from the Deutsche Forschungsgemeinschaft. P.I.K. was partially supported by the American Foundation for Urologic Diseases, the National Cancer Institute of Canada, and the Medical Research Council of Canada. We thank an anonymous reviewer for helpful suggestions on a previous draft of this manuscript.
1. Whitmore WFJ: Hormone therapy in prostate cancer. Am J Med 21: 697-713, 1956[CrossRef][Medline] 2. Beahrs OH, Henson DE, Hutter RVP, et al: American Joint Committee on Cancer Staging Manual, ed 4 . Philadelphia, PA, Lippincott, 1992, pp 181-186 3. Fleming ID, Cooper JS, Henson DE, et al: American Joint Committee on Cancer Staging Manual, ed 5 . Philadelphia, PA, Lippincott, 1997, pp 219-222 4. Stamey TA, Sözen 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[CrossRef][Medline] 5. 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[CrossRef][Medline]
6.
Kattan MW, Eastham JA, Stapleton AMF, et al: A preoperative nomogram for disease recurrence following radical prostatectomy for prostate cancer. J Natl Cancer Inst 90: 766-771, 1998
7.
Partin AW, Kattan MW, Subong ENP, et al: Combination of prostate-specific antigen, clinical stage, and Gleason score to predict pathological stage in men with localized prostate cancer: A multi-institutional update. JAMA 277: 1445-1451, 1997
8.
Ghali W, Quan H, Brant R, et al: Comparison of 2 methods for calculating adjusted survival curves from proportional hazards models. JAMA 286: 1494-1497, 2001 9. Dillioglugil O, Leibman BD, Kattan MW, et al: Hazard rates for progression after radical prostatectomy for clinically localized prostate cancer. Urology 50: 93-99, 1997[CrossRef][Medline] 10. Amling CL, Blute ML, Bergstralh EJ, et al: Long-term hazard of progression after radical prostatectomy for clinically localized prostate cancer: Continued risk of biochemical failure after 5 years. J Urol 164: 101-105, 2000[CrossRef][Medline] 11. Iyer RV, Hanlon AL, Pinover WH, et al: Outcome evaluation of the 1997 American Joint Committee on Cancer staging systems for prostate carcinoma treated by radiation therapy. Cancer 85: 1816-1821, 1999[CrossRef][Medline] 12. Smith DS, Catalona WJ: Interexaminer variability of digital rectal examination in detecting prostate cancer. Urology 45: 70-74, 1995[CrossRef][Medline] 13. Begg CB, Cramer LD, Venkatraman ES, et al: Comparing tumor staging and grading systems: A case study and a review of the issues. Stat Med 19: 1997-2014, 2000[CrossRef][Medline]
14.
Kattan MW, Zelefsky MJ, Kupelian PA, Scardino PT, Fuks Z, Leibel SA: Pretreatment nomogram for predicting the outcome of three-dimensional conformal radiotherapy in prostate cancer. J Clin Oncol 18: 3352-3359, 2000 15. Kattan MW, Reuter V, Motzer RJ, Katz J, Russo P: A postoperative prognostic nomogram for renal cell carcinoma. J Urol 166: 63-67, 2001[CrossRef][Medline] Submitted May 29, 2001; accepted January 15, 2002.
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2002 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
