|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 22, No 16 (August 15), 2004: pp. 3316-3322 © 2004 American Society of Clinical Oncology. DOI: 10.1200/JCO.2004.09.104 Use of the University of California Los Angeles Integrated Staging System to Predict Survival in Renal Cell Carcinoma: An International Multicenter StudyFrom the Department of Urology, Rennes University Hospital, Rennes; Department of Urology, CHU Henri Mondor, Créteil; Department of Urology, North Hospital, CHU of Saint Etienne, France; Department of Urology, David Geffen School of Medicine at University of California Los Angeles, CA; Department of Urology, University of Verona, Verona; Department of Urology, Medical School of University "Federico II," Naples, Italy; Department of Urology, The University of Texas M.D. Anderson Cancer Center, Houston, TX; Department of Urology, University Medical Center, Nijmegen, the Netherlands Address reprint requests to Arie S. Belldegrun, MD, Department of Urology, David Geffen School of Medicine at the University of California Los Angeles, CHS 66-118, Box 951738, 10833 Le Conte Ave, Los Angeles, CA 90095-1738; e-mail: abelldegrun{at}mednet.ucla.edu
PURPOSE: To evaluate ability of the University of California Los Angeles Integrated Staging System (UISS) to stratify patients with localized and metastatic renal cell carcinoma (RCC) into risk groups in an international multicenter study. PATIENTS AND METHODS: 4,202 patients from eight international academic centers were classified according to the UISS, which combines TNM stage, Fuhrman grade, and Eastern Cooperative Oncology Group performance status. Distribution of the UISS categories was assessed in the overall population and in each center. RESULTS: The UISS stratified both localized and metastatic RCC into three different risk groups (P < .001). For localized RCC, the 5-year survival rates were 92%, 67%, and 44% for low-, intermediate-, and high-risk groups, respectively. A trend toward a higher risk of death was observed in all centers for increasing UISS risk category. For metastatic RCC, the 3-year survival rates were 37%, 23%, and 12% for low-, intermediate-, and high-risk groups, respectively; in 6 of 8 centers, a trend toward a higher risk of death was observed for increasing UISS risk category. A greater variability in survival rates among centers was observed for high-risk patients. CONCLUSION: This study defines the general applicability of the UISS for predicting survival in patients with RCC. The UISS is an accurate predictor of survival for patients with localized RCC applicable to external databases. Although the UISS may be useful for patients with metastatic RCC, it may be less accurate in this subset of patients due to the heterogeneity of patients and treatments.
The wide variability in the clinical behavior of renal cell carcinoma (RCC) reflects the biologic complexity of this disease.1 Although recent advances have been made in understanding the molecular mechanisms of tumor progression,2 validated biomolecular markers for prognosis are not yet widely available for clinical use. Currently, tumor stage,3 histologic grade,4 and Eastern Cooperative Oncology Group (ECOG) performance status (PS) remain the most commonly used predictors of survival for patients with RCC. However, these predictors have varying degrees of accuracy, and new prognostic algorithms and nomograms have been recently proposed.5,6 The University of California Los Angeles Integrated Staging System (UISS) was designed to account for the complex interactions between 1997 TNM stage (I to IV), ECOG PS, and Fuhrman grade,7 with the ultimate goal of producing a simple and accurate prognostic system. The UISS was constructed to define subgroups of patients with different risks of death following nephrectomy. The original UISS was subsequently modified into a clinical outcome algorithm to stratify survival based on 1997 TNM stage, ECOG PS, and Fuhrman grade, resulting in three risk groups (low- [LR], intermediate- [IR], and high-risk [HR]) for localized and metastatic disease.8 We applied the modified UISS to patients with RCC from eight different institutions from four different countries, to better define the general applicability of the UISS. An accurate system for predicting prognosis that has been validated in independent patient populations will be useful for counseling patients, selecting patients for clinical trials, and customizing patient follow-up after surgery.
Patients and Tumors The study included patients from eight international academic centers: Rennes (France), Saint Etienne (France), Créteil (France), Naples (Italy), Verona (Italy), Nijmegen (Netherlands), M.D. Anderson (United States), and University of California Los Angeles (United States). Patient records were extracted from each institutional database. Data collected from each patient included TNM stage, Fuhrman grade, ECOG PS, and survival. Other variables analyzed included age, sex, tumor size, and histology. All data were labeled with their respective institution, and pooled. Patients included in the study were staged preoperatively with computed tomography (CT) scan of the abdomen and pelvis, chest CT or chest x-ray, serum electrolytes, and liver function tests. The specific treatments were based on the preferences of the treating physicians at their respective institutions. Patients with localized RCC were treated with either a partial or a radical nephrectomy. Patients with metastatic RCC had a nephrectomy, followed by interferon (IFN) or interleukin-2 (IL-2)-based immunotherapy. Postoperative follow-up was according to protocols established at each institution. Tumor stage was determined according to the 1997 Union Internationale Contre le Cancer (UICC) TNM classification of renal tumors.3 Tumor histology was classified according to the Heidelberg classification,9 and tumors were graded according to the Fuhrman grading scheme4 by the pathologists at each of the eight institutions. All histologic types of tumors were included in each of the databases. ECOG PS was assigned for each patient during the office visit before nephrectomy, according to the original criteria set forth by Oken et al.10 The UISS outcome algorithm was applied to a total 4,202 patients from eight institutions, who underwent radical or partial nephrectomy. As previously described,8 two decision boxes based on 1997 TNM stage, Fuhrman grade, and ECOG PS were used to categorize patients with localized disease and metastatic disease into LR, IR, or HR groups.
Statistical Analysis
Patients and Tumors A total of 4,202 patients with RCC from two centers in the United States (n = 1,463) and six centers in Europe (n = 2,739) were included in this study. Patient characteristics from each institution are presented in Table 1. The number of patients from each center ranged from 208 to 809. Mean (± standard deviation) patient age at diagnosis was 59.8 ± 12.2 years. Mean (± standard deviation) tumor size was 7.0 ± 3.8 cm. The distribution of TNM stages I, II, III, and IV was 38.9%, 12.3%, 26.1%, and 22.7%, respectively. Fuhrman grades were I, II, III, and IV in 20.2%, 36.2%, 34.6%, and 9% of cases, respectively. ECOG PS was evaluated as  1 in 34.6% of cases. Among 3,119 patients with localized RCC, UISS was classified as LR, IR, and HR in 36.6%, 52.5%, and 10.9% of cases, respectively. Among 1,083 patients with metastatic RCC (N1 or N2 or M1), UISS was classified as LR, IR, and HR in 24.9%, 63.5%, and 11.6% of cases, respectively. At last follow-up, 35.4% of the patients have died.
UISS and Survival In the 3,119 patients with localized disease, stratification by the UISS produced risk groups with statistically different survival rates (Fig 1A; log-rank P < .001). Similarly, when considering the 1,083 patients with metastatic disease, the UISS categories defined three groups with significantly different survivals (Fig 1B; log-rank P < .001). For localized RCC, the 5-year survival rates were 92%, 67%, and 44% for LR, IR, and HR, respectively. For metastatic RCC, the 3-year survival rates were 37%, 23%, and 12% for LR, IR, and HR, respectively.
The variability in the survival rates between the eight centers was examined by plotting the survival curves for each center separately for the three UISS categories for localized RCC and metastatic RCC (Figs 2A through F). Patients with localized RCC in the LR and IR groups had similar survival rates between each of the centers (Figs 2A and 2B), while there was more variability between the centers for patients in the HR group (Fig 2C). There was more variability between centers in all risk groups for patients with metastatic RCC (Figs 2D through F).
To summarize the survival rates at each institution, the number of deaths per 5 years of patient follow-up was calculated for each UISS category (Table 2). For localized disease, a trend toward a higher risk of death was observed in all centers for increasing UISS risk category. For metastatic disease, a similar trend toward a higher death rate was observed in six of eight centers as the UISS risk category increased.
In a similar analysis, the survival rates were assessed using Cox proportional hazard models. Each UISS category was modeled as a categorical variable (LR v IR and LR v HR; Table 3). The P value was statistically significant for all centers for localized disease, and in six of eight centers for metastatic disease. For localized disease, among the eight centers, hazard ratios ranged from 4.85 to 31.79, whereas for metastatic disease, hazard ratios ranged from 1.49 to 17.01.
Table 3 presents the results of the c statistics for each center stratified by the UISS. This was calculated to determine how accurate the UISS clinical algorithm was at placing patients into the correct risk group. Within the eight centers, this index ranged from 0.76 to 0.86 in localized disease, whereas it ranged from 0.58 to 0.77 in metastatic disease. Survival rates for each center according to the UISS stratification for patients with localized and metastatic disease are presented in Table 4.
Defining risk groups is useful for counseling patients, individualizing follow-up, choosing treatment options, and selecting patients for clinical trials. Recently, efforts have been made to better predict survival in RCC by developing prognostic models that use a combination of variables that have been shown to be independent predictors of prognosis. Early efforts to combine multiple variables into a risk stratification system include the work of Elson et al.12 In 1999, Motzer et al13 published a three-tiered system for stratifying patients with metastatic disease. More recently Motzer et al described a system that utilizes three risk categories that were identified on the basis of five prognostic factors for predicting survival in patients with advanced RCC.14,15 In 2001, the Memorial Sloan-Kettering Cancer Center introduced a postoperative nomogram for patients with localized RCC, which assigns points based on a combination of variables that include histology, tumor size, 1997 T stage, and symptoms at presentation.5 The Mayo Clinic introduced a prediction model based on 1997 TNM Stage, tumor Size, nuclear Grade, and tumor Necrosis (SSIGN).6 The SSIGN model divides patients into 10 different subgroups. While all of these systems have merit, none of these staging systems include both localized and metastatic RCC. The UISS was developed using the kidney cancer database from the University of California Los Angeles Kidney Cancer Program with the goal of providing a simple and accurate algorithm for predicting survival using variables that are available in any modern medical practice.7,8 The validity of any predictive model is best evaluated by applying it to a large independent patient population from multiple centers. Recently, the UISS model has been validated for localized tumors in a series of 1,060 patients from three referral centers.16 The purpose of the present study is to evaluate how well the system works in a large, multinational population with both localized and metastatic RCC. Although the populations from each of the eight centers were very heterogeneous, it appears that survival rates based on UISS are similar across institutions, particularly for LR and IR groups. The frequency of stage IV RCC ranged from 9.5% to 40.8% among the different centers. Similarly, the frequency of Fuhrman IV tumors ranged from 1.4% to 18%, and the frequency of ECOG PS 1 or more ranged from 6.7% to 57%. Despite these differences, the UISS risk classification stratifies patients as expected, and the overall UISS survival curves from the different institutions are reasonably similar. For patients with localized RCC, the differences in survival between institutions are most pronounced for patients in the HR group. This disparity is likely explained by differences in referral pattern, differences in accuracy of preoperative staging, and biases introduced by the subjective nature of tumor grading17 and assignment of ECOG PS. For example, at the University of California Los Angeles, all patients are staged with a chest CT, and patients with metastatic RCC are also staged with a brain magnetic resonance imaging scan and bone scan. At other centers, a chest x-ray is used, and bone scans and head CTs are reserved for symptomatic patients. Therefore, it is possible that when more sensitive methods for preoperative staging are used, patients with low-volume metastases are more likely to be excluded from the HR localized group, resulting in a more homogeneous HR patient population. Similarly, our interpretation of the wide range of hazard ratios that exists among centers is that the UISS stratification, although effective in prognostic stratification, is dependent on the differences in patient populations, accuracy of preoperative staging modalities, or differences in therapeutic choices. In the present study, the UISS was more accurate for localized RCC than for metastatic RCC. c statistics and hazard ratios for the UISS were higher for localized RCC than for metastatic RCC. In fact, in two of eight centers, the system did not discriminate well for metastatic patients. This failure may reflect the limitations of staging and the interobserver variability in grading and assigning ECOG PS in patients with extensive disease. Furthermore, for the vast majority of patients with metastatic RCC, the prognosis is uniformly poor. Therefore, stratifying this group based on survival may not be possible or as useful. Some of the variability in survival between institutions may also result from differences in treatment. For example, at the University of California Los Angeles, the standard treatment for metastatic RCC during the last 10 years has been a cytoreductive nephrectomy followed by high-dose IL-2 therapy. However, the treatment of choice at the three French centers is an outpatient regimen of IFN and IL-2, while the other centers in the study preferred outpatient regimens with either IFN or IL-2. Furthermore, there is evidence to suggest that not all immunotherapy regimens are equally effective. For example, a recent randomized study from the National Cancer Institute reported that high-dose IL-2 produced greater biologic activity than low-dose IL-2, as evidenced by a higher response rate.18 Limitations of the study include the lack of central pathology review to concerns regarding variability in tumor grading between institutions. Ideally, all tumors should be graded by a single set of pathologists. However, logistically it would be impossible to re-review more than 4,000 RCC cases from eight different centers. In addition, our current approach better estimates how the UISS works in the "real world." Another limitation of using databases from multiple institutions is that additional predictive variables cannot be evaluated for possible inclusion in UISS. This limitation is a rationale for establishing an international consensus on what variables should be collected in RCC data sets. This study better defines the general applicability of the UISS for predicting survival in patients with RCC. The UISS is clearly an accurate predictor of survival for patients with localized RCC. The variability between centers in survival for patients with localized RCC categorized as LR and IR is small. The greater variability for patients with localized RCC categorized as HR may be partly explained by differences in the methods used for preoperative staging. The UISS is also useful for predicting survival for patients with metastatic RCC. However, given the differences in treatment strategies for metastatic RCC at various institutions, the UISS may not stratify patient survival based on the defined risk groups.
The authors indicated no potential conflicts of interest.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
1. Pantuck AJ, Zisman A, Belldegrun A: The changing natural history of renal cell carcinoma. J Urol 166:1611-1623, 2001[CrossRef][Medline] 2. Folkman J: Role of angiogenesis in tumor growth and metastasis. Semin Oncol 29:15-18, 2002[Medline] 3. Guinan P, Sobin LH, Algaba F, et al: TNM staging of renal cell carcinoma: Workgroup No. 3—Union International Contre le Cancer (UICC) and the American Joint Committee on Cancer (AJCC). Cancer 80:992-993, 1997[CrossRef][Medline] 4. Fuhrman SA, Lasky LC, Limas C: Prognostic significance of morphologic parameters in renal cell carcinoma. Am J Surg Pathol 6:655-663, 1982[Medline] 5. Kattan MW, Reuter V, Motzer RJ, et al: A postoperative prognostic nomogram for renal cell carcinoma. J Urol 166:63-67, 2001[CrossRef][Medline] 6. Frank I, Blute ML, Cheville JC, et al: An outcome prediction model for patients with clear cell renal cell carcinoma treated with radical nephrectomy based on tumor stage, size, grade and necrosis: The SSIGN score. J Urol 168:2395-2400, 2002[CrossRef][Medline]
7. Zisman A, Pantuck AJ, Dorey F, et al: Improved prognostication of renal cell carcinoma using an integrated staging system. J Clin Oncol 19:1649-1657, 2001
8. Zisman A, Pantuck AJ, Wieder J, et al: Risk group assessment and clinical outcome algorithm to predict the natural history of patients with surgically resected renal cell carcinoma. J Clin Oncol 20:4559-4566, 2002 9. Kovacs G, Akhtar M, Beckwith BJ, et al: The Heidelberg classification of renal cell tumours. J Pathol 183:131-133, 1997[CrossRef][Medline] 10. Oken MM, Creech RH, Tormey DC, et al: Toxicity and response criteria of the Eastern Cooperative Oncology Group. Am J Clin Oncol 5:649-655, 1982[Medline] 11. Harrell FE Jr, Lee KL, Mark DB: Multivariable prognostic models: Issues in developing models, evaluating assumptions and adequacy, and measuring and reducing errors. Stat Med 15:361-387, 1996[CrossRef][Medline]
12. Elson PJ, Witte RS, Trump DL: Prognostic factors for survival in patients with recurrent or metastatic renal cell carcinoma. Cancer Res 48:7310-7313, 1988
13. Motzer RJ, Mazumdar M, Bacik J, et al: Survival and prognostic stratification of 670 patients with advanced renal cell carcinoma. J Clin Oncol 17:2530-2540, 1999
14. Motzer RJ, Bacik J, Murphy BA, et al: Interferon-alfa as a comparative treatment for clinical trials of new therapies against advanced renal cell carcinoma. J Clin Oncol 20:289-296, 2002 15. Abou-Jawde RM, Mekhail T, Bou Merhi GF: Prognostic factors (PF) for survival in previously untreated metastatic renal cell cancer (RCC): A comprehensive evaluation and validation of established risk groups. Proc Am Soc Clin Oncol 22:385, 2003(abstr 1545) 16. Han KR, Bleumer I, Pantuck AJ, et al: Validation of an integrated staging system toward improved prognostication of patients with localized renal cell carcinoma in an international population. J Urol 170:2221-2224, 2003[CrossRef][Medline] 17. Al-Aynati M, Chen V, Salama S, et al: Interobserver and intraobserver variability using the Fuhrman grading system for renal cell carcinoma. Arch Pathol Lab Med 127:593-596, 2003[Medline]
18. Yang JC, Sherry RM, Steinberg SM, et al: Randomized study of high-dose and low-dose interleukin-2 in patients with metastatic renal cancer. J Clin Oncol 21:3127-3132, 2003 Submitted September 22, 2003; accepted May 20, 2004.
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2004 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
