The Evolving Presentation of Renal Carcinoma in the United States: Trends From the Surveillance, Epidemiology, and End Results Program

The Evolving Presentation of Renal Carcinoma in the United States: Trends From the Surveillance, Epidemiology, and End Results Program Mike M. Nguyen,* Inderbir S. Gill and Lars M. Ellison From the Department of Urology, University of California at Davis, Sacramento California, and Section of Laparoscopic and Robotic Surgery, Glickman Urological Institute, Cleveland Clinic Foundation (ISG), Cleveland, Ohio

Purpose: The incidence of renal cancer is increasing, while cases series suggest that tumor size is decreasing. This has important implications for treatment planning. We evaluated national trends in renal cancer size and observed survival in patients diagnosed in the 3 periods 1988 to 1992, 1993 to 1997 and 1998 to 2002. Materials and Methods: From the Surveillance, Epidemiology, and End Results database we identified 29,053 patients diagnosed with primary renal cancer. Patients were stratified into size categories and 5-year time cohorts. Size distribution was compared across cohorts. Kaplan-Meier survival curves and Cox proportional hazards modeling were used to examine trends in overall and stage specific survival. Results: From 1988 through 2002 renal tumor size decreased from 66.8 to 58.6 mm, while the age adjusted incidence of renal cancer increased from 8.6 to 11.2 cases per 100,000 individuals. Kaplan-Meier analysis showed steadily deteriorating survival with increased cancer size above 4 cm with a median survival of 105 months for 4 to 7 cm vs 46 months for more than 7 cm. Cox modeling demonstrated significantly improved survival in patients diagnosed in the latter cohorts. With adjustment for size the latter cohorts remained significantly improved compared to the earliest cohort, although the 1998 to 2002 cohort was no longer significantly different than the 1993 to 1997 cohort. Conclusions: Nationally renal tumor size at presentation has steadily and consistently decreased. Patients more recently diagnosed had improved survival, which could be attributable to decreased tumor size in the latter cohorts. Patients more recently diagnosed also demonstrated a relative survival advantage independent of size compared to the earliest patients studied. Key Words: kidney, kidney neoplasms, SEER program, mortality, diagnosis

dvances in medical technology have changed the presentation of renal cancer in the last decade. Widespread use of imaging has contributed to an increased incidence of renal cancer through earlier detection but it appears that other undefined factors are also leading to a real increase in incidence.1 In contrast, tumor size at presentation appears to be decreasing based on findings in single institution case series.2 A shift to smaller tumors has significant implications as patients and their physicians become forced to make clinical decisions about the management of low stage disease. Decreasing size at presentation may also affect aggregate survival rates because tumor size has been shown to strongly predict survival.3 We examined national trends in renal tumor size at presentation, and the resulting impact on overall and stage specific survival. Additionally, we questioned whether survival improvements within tumor size strata have occurred. To answer these questions we analyzed 3 time cohorts of

A

Submitted for publication October 13, 2005. * Correspondence: Department of Urology, University of California at Davis, 4860 Y St., Suite 3500, Sacramento, California 95817 (telephone: 916-734-2893; FAX: 916-734-8094; e-mail: mike.m.nguyen@ gmail.com).

0022-5347/06/1766-2397/0 THE JOURNAL OF UROLOGY® Copyright © 2006 by AMERICAN UROLOGICAL ASSOCIATION

patients, including 1988 to 1992, 1993 to 1997 and 1988 to 2002, from the National Cancer Institute SEER program. METHODS The SEER database is a national prospective cancer registry maintained at the National Cancer Institute. This database includes information from population based cancer registries from various areas of the country, including Connecticut, Iowa, New Mexico, Utah and Hawaii, and the metropolitan areas of Detroit, San Francisco, Seattle-Puget Sound and Atlanta. Together these regions currently include approximately 26% of the population in the United States. The recently released 1973 to 2002 SEER data set was used.4 The disease site code chosen was kidney and renal pelvis. Histological criteria were used to exclude transitional cell cancer, neuroblastoma and Wilms tumor. Cases with no size recorded were excluded. The 14 cases with sizes that were recorded as greater than 40 cm were also excluded to remove clinical and statistical outliers. A total of 29,053 patients were included in the analysis. The primary dependent variables were tumor size and survival. The independent variables were diagnosis year, patient age, sex and histology. Analysis was performed using the STATA® 8 statistical software package. Average sizes at diagnosis and age adjusted incidence rates were compared by year. Kaplan-Meier

2397

Vol. 176, 2397-2400, December 2006 Printed in U.S.A. DOI:10.1016/j.juro.2006.07.144

2398

EVOLVING PRESENTATION OF RENAL CANCER IN UNITED STATES

survival analysis and Cox proportional hazards modeling were performed for survival grouped into the 5-year cohorts (1988 to 1992, 1993 to 1997 and 1998 to 2002), and for tumor size stratified into 1, 2, 3, 4 and 7 cm cutoffs. Four and 7 cm cutoffs were chosen to correspond to the 2002 TNM T1a and T1b subgroups. Sizes less than 4 cm were chosen to evaluate potential survival advantages below 4 cm. The age adjusted incidence was determined from the SEER incidence data set5 using SEER*stat version 6.1.4 software (www.seer. cancer.gov/seerstat). The standard population used was the population of the United States, as recorded in the 2000 census. RESULTS Mean study population age was 63.5 years with males comprising 62% of cases (table 1). Average overall tumor size was 62.4 mm. Patient age and sex did not differ among 5-year cohorts. Mean tumor size decreased sequentially among the 3 cohorts, that is 65.1, 63.4 and 59.9 mm, respectively. Conventional renal cell cancer accounted for 89.7% of histological subtypes in the total cohort. Additional subtypes were papillary in 2.3% of cases, chromophobe in 0.1% and medullary tumors in less than 0.1%. Less common subtypes and malignancies not otherwise specified comprised the remaining 7.9%. On an annual basis there was a steady decrease in overall average renal tumor size at presentation from 66.8 mm in 1988 to 58.6 mm in 2002 (fig. 1). As defined by size category, there was an increase in the percent of smaller tumors (4 cm or less) presenting in each successive 5-year cohort, that is 30%, 34% and 39%, respectively. There was a corresponding decrease in larger tumors of more than 7 cm during the study period, that is 34%, 33% and 30%, respectively. For intermediate (4.1 to 7 cm tumors) the percent of cases also decreased, that is 36%, 33% and 31%, respectively (table 1). During the same period there was an increase in the age adjusted incidence of renal cell cancer from 8.6 to 11.2/ 100,000 individuals (fig. 2). Kaplan-Meier analysis showed steadily deteriorating survival with increased renal cancer size above 4 cm (log rank test p ⫽ 0.000, fig. 3). Tumors less than 4 cm showed uniform survival, as evidenced by nonsignificant HR differences (data not shown). Overall patients diagnosed in the 1993 to 1997 and 1998 to 2002 cohorts had improved survival relative to the 1988 to 1992 cohort on Cox proportional hazards testing (HR 0.928 and 0.882, respectively). When the latter 2 cohorts were compared with each other, the 1998 to 2002 cohort retained a survival advantage over the 1993 to 1997 cohort (HR 0.950, table 2 and fig. 4). When adjusted for size, these HRs were decreased but they remained significant for the second and third cohorts compared to the earliest cohort.

FIG. 1. Average renal cancer size at presentation by diagnosis year. Dotted lines indicate SD. Mean 1988 vs 2002 size unpaired t test p value.

However, the survival difference between the second and third cohorts no longer remained significant when adjusted for size (table 3). DISCUSSION Renal cancer size steadily decreased from 66.8 to 58.6 mm nationally in the last 14 years for which data are available. Consistent with this a higher percent of cases presented with smaller tumors in the latter cohorts, that is 30%, 34% and 39% for tumors 4 cm or less. As cross-sectional imaging rates increase, we expect that size at diagnosis will continue to decrease before stabilizing at a new lower baseline. Size at presentation was a strong predictor of survival with smaller size leading to improved survival. However, a true threshold occurred at 4 cm, below which further decreases in size did not result in continued improvements in survival. These findings are consistent with changes made in the TNM renal cancer classification in 2002, when T1 tumors were divided into the subgroups T1a—less than 4 cm and T1b— 4 cm or greater to less than 7 cm.6 Other investigators have also found significant differences in outcome between these 2 subgroups in single institution series.7,8 Although it was modest, the decrease in size from 66.8 to 58.6 mm during the study period appears to have contributed to a survival advantage in patients diagnosed within

TABLE 1. Baseline demographics by 5-year cohort Characteristic No. pts Av age % Male No./cm size cohort (%): 4 or Less 4.1–7 Greater than 7 Av size (mm)

1988–1992

1993–1997

7,755 63.5 62

9,304 63.4 63

2,324 (30) 2,767 (36) 2,664 (34) 65.1

3,182 (34) 3,056 (33) 3,066 (33) 63.4

1998–2002 11,994 63.1 63 4,650 (39) 3,703 (31) 3,641 (30) 59.9

Totals 29,053 63.3 63 10,156 (35) 9,526 (33) 9,371 (32) 62.4

EVOLVING PRESENTATION OF RENAL CANCER IN UNITED STATES

2399

TABLE 2. Cox proportional hazards of 5-year cohorts compared to 1988 to 1992 and 1993 to 1997 cohorts unadjusted and adjusted for size Baseline Cohort (comparison)

HR

p Value

95% CI

0.001 0.000

0.890–0.969 0.840–0.927

1.077 0.001 0.950 0.040 Adjusted for size

1.032–1.124 0.905–0.998

0.943 0.928

0.007 0.003

0.903–0.984 0.883–0.975

1.061 0.985

0.007 0.532

1.017–1.107 0.938–1.034

Unadjusted 1988–1992: 1993–1997 1998–2002 1993–1997: 1988–1992 1998–2002

FIG. 2. Incidence of renal cancer per 100,000 individuals with patient age adjusted to United States 2000 standard population. Dotted lines indicate 95% CI.

the more recent 5-year cohorts of 1993 to 1997 and 1998 to 2002 compared to patients diagnosed in 1988 to 1992. When adjusted for size, survival advantages between the cohorts decreased. However, they remained significant when the earliest cohort was compared to the second or third cohorts but not for comparisons between the second and third cohorts, suggesting that other factors lead to improved survival in more recently diagnosed patients. However, these factors appear to have stabilized in the last 10 years observed in this study (table 3). While we were not able to further explore these factors in this study, possibilities may include improved surgical treatment or diagnostic followup in patients in the latter cohorts. Analysis of a more detailed data set may allow the identification of these factors in a future study. Smaller tumor size has important clinical implications for patients and clinicians to consider when deciding how to treat these lower stage cancers. As tumors decrease in size at presentation, nephron sparing treatments become increasingly applicable, in addition to traditional radical nephrectomy. These treatments include partial nephrectomy using open or minimally invasive approaches and energy ablative therapies, such as cryotherapy, radio frequency ablation and high intensity focused ultrasound.9 Initially re-

FIG. 3. Kaplan-Meier survival stratified by tumor size. Log rank test p value.

1988–1992: 1993–1997 1998–2002 1993–1997: 1988–1992 1998–2002

0.928 0.882

served for patients with solitary kidney or impaired renal function, nephron sparing surgery has been shown also to have benefits in patients with a normal contralateral kidney and normal preoperative renal function. Nephron sparing surgery was shown to result in a lower risk of renal failure in these patients10 and quality of life measures have been found to be higher in patients who have more renal parenchyma remaining after surgery for localized renal cancer independent of postoperative renal function.11 These potential benefits in renal function and quality of life are in addition to the improved survival experienced by patients with smaller tumors.

CONCLUSIONS Renal cancers have decreased in size at presentation on an annual basis. Decreases in size have contributed to improved aggregate outcomes in patients diagnosed with renal cancer. Patients more recently diagnosed also demonstrated a relative survival advantage independent of size compared to the earliest patients studied. However, this size independent survival advantage was lost during the last 10 years compared in the study.

FIG. 4. Kaplan-Meier survival stratified by 5-year cohorts. Y axis represents logarithmic scale. Log rank test p value.

2400

EVOLVING PRESENTATION OF RENAL CANCER IN UNITED STATES

Abbreviations and Acronyms SEER ⫽ Surveillance, Epidemiology, and End Results REFERENCES 1.

2.

3.

4.

5.

6.

7.

8.

Chow, W. H., Devesa, S. S., Warren, J. L. and Fraumeni, J. F., Jr.: Rising incidence of renal cell cancer in the United States. JAMA, 281: 1628, 1999 Patard, J. J., Tazi, H., Bensalah, K., Rodriguez, A., Vincendeau, S., Rioux-Leclercq, N. et al: The changing evolution of renal tumours: a single center experience over a two-decade period. Eur Urol, 45: 490, 2004 Giuliani, L., Giberti, C., Martorana, G. and Rovida, S.: Radical extensive surgery for renal cell carcinoma: long-term results and prognostic factors. J Urol, 143: 168, 1990 Surveillance, Epidemiology, and End Results (SEER) Program Populations (1973–2002), National Cancer Institute, Department of Cancer Control and Population Sciences, Surveillance Research Program, Cancer Statistics Branch, released April 2005. Available at www.seer.cancer.gov/ popdata. Accessed June 10, 2005 Surveillance, Epidemiology, and End Results (SEER) Program SEER*Stat Database: Incidence—SEER 9 Regs Public-Use, Nov 2004 Sub (1973–2002), National Cancer Institute, Department of Cancer Control and Population Sciences, Surveillance Research Program, Cancer Statistics Branch, released April 2005 based on the November 2004 submission. Available at www.seer.cancer.gov. Accessed June 10, 2005 Sobin, L. H. and Wittekind, C.: TNM Classification of Malignant Tumors, 6th ed. New York: Wiley-Liss, pp. 193–195, 2003 Salama, M. E., Guru, K., Stricker, H., Peterson, E., Peabody, J., Menon, M. et al: pT1 substaging in renal cell carcinoma: validation of the 2002 TNM staging modification of malignant renal epithelial tumors. J Urol, 173: 1492, 2005 Ficarra, V., Novara, G., Galfano, A., Novella, G., Schiavone, D. and Artibani, W.: Application of TNM, 2002 version, in

localized renal cell carcinoma: is it able to predict different cancer-specific survival probability? Urology, 63: 1050, 2004 9. Gill, I. S.: Minimally invasive nephron-sparing surgery. Urol Clin North Am, 30: 551, 2003 10. McKiernan, J., Simmons, R., Katz, J. and Russo, P.: Natural history of chronic renal insufficiency after partial and radical nephrectomy. Urology, 59: 816, 2002 11. Clark, P. E., Schover, L. R., Uzzo, R. G., Hafez, K. S., Rybicki, L. A. and Novick, A. C.: Quality of life and psychological adaptation after surgical treatment for localized renal cell carcinoma: impact of the amount of remaining renal tissue. Urology, 57: 252, 2001

EDITORIAL COMMENT These authors emphasize what most urological oncologists know, that is that size matters, especially in kidney cancer. While the widespread adoption of abdominal imaging since the 1980s (reference 1 in article) has led to an increase in the incidence of kidney cancer, previous analyses of SEER data suggested that these incidental renal masses have spanned the gamut from large to small1 with no attendant improvement in survival. The current study provides new insights by showing that mean tumor size has in fact decreased somewhat and survival from kidney cancer has improved, albeit modestly. Left unanswered is the central question of how we can further improve survival from kidney cancer beyond making the diagnosis earlier. Mark S. Litwin Department of Urology and Health Services University of California-Los Angeles Los Angeles, California 1.

Hock, L. M., Lynch, J. and Balaji, K. C.: Increasing incidence of all stages of kidney cancer in the last 2 decades in the United States: an analysis of surveillance, epidemiology and end results program data. J Urol, 167: 57, 2002