UROLOGICAL ONCOLOGY Characteristics and Prognostic Value of Papillary Histologic Subtype in Non- metastatic Renal Cell Carcinoma in Korea: A Multicenter Study Won Ki Lee,1 Sang Eun Lee,2 Sung Kyu Hong,2 Chang Wook Jeong,2 Yong Hyun Park,2 Seok Ho Kang,3 Yong-June Kim,4 Sung-Hoo Hong,5 Won Suk Choi,6 Seok-Soo Byun2* Purpose: To analyze the characteristics of nonmetastatic papillary renal cell carcinomas (RCC) and the prognostic value of RCC histologic subtyping, based on a large multicenter experience in Korea. Materials and Methods: A total of 2,905 patients with nonmetastatic RCC (TxN0M0) at the time of surgery were retrospectively enrolled from five institutions between 1999 and 2011 in Korea. Among these, patients with clear cell subtype (n = 2,488, 85.6%) and papillary subtype (n = 192, 6.6%) were included in our study. Results: Patients with papillary subtype did not differ significantly from those with clear cell subtype on the following parameters: age (P = .694), gender (P = .511), body mass index (P = .136), patient performance status (P = .419), symptoms at presentation (P = .419), tumor size (P = .778) and pathologic stage (P = .367). However, high Fuhrman’s grades were more common in papillary subtypes compared with clear cell subtypes (P = .001). The 5-year recurrence-free survival rates in patients with clear cell subtype and papillary subtype were 84.9% and 86.7%, respectively (P = .167). The 5-year cancer-specific survival rates in patients with clear cell subtype and papillary subtype were 92.0% and 93.1%, respectively (P = .931). Histologic subtype was not an independent prognostic factor of recurrence-free and cancer-specific survival (P = .107 and P = .998, respectively). Conclusion: Our study suggests that the characteristics and prognosis of papillary subtype might be comparable to those of clear cell subtype in non-metastatic RCC, especially in Asia. Keywords: kidney neoplasms; carcinoma; renal cell; papillary; pathology; follow-up studies; retrospective studies; prognosis. INTRODUCTION It is well known that renal cell carcinoma (RCC) is a heterogeneous and complex disease, and its natu-ral history is greatly variable.(1) Prediction of disease progression is, therefore, critical in optimal clinical deci- sion making with and counseling of patients. A variety of findings have been considered as prognostic factors for RCC. However, only a few prognostic factors including tumor-node-metastasis (TNM) stage, Fuhrman’s grade and tumor size are almost undisputed prognostics factor for RCC, especially non-metastatic RCC.(2) The histologic subtypes of RCCs, according to the 2004 World Health Organization (WHO) classification, in- clude clear cell, papillary, chromophobe, collecting duct and unclassified.(3) Histologic subtype has been tradition- ally considered as a prognostic factor for RCC, based on molecular and genetic studies which have shown that RCCs have genetic and pathologic differences among the histologic subtypes.(4) However, the tumor characteristics and prognostic value of the histologic subtypes have not been verified, and remain debated.(2) It is also remarkable that almost all studies involving the prognostic value of histologic subtypes have focused on the Western popula- tion. Differences in characteristics of cancer among dif- ferent races have well been established in other urologic cancers, such as prostate cancer.(5) Also, because many previous studies have reported the existence of racial/ ethnic disparities in incidence and survival of RCC,(6-8) the prognostic value of histologic subtypes in Asians may differ from that in Westerners. In the current study, we analyzed the characteristics and prognostic values of papillary RCC (pRCC) compared with clear cell RCC (cRCC) in nonmetastatic RCC, based on a large, multicenter experience in Korea. MATERIALS AND METHODS Study Subjects Our study was a clinical case series with retrospective de- sign based on a large, multicenter experience in Korea. A total of 2,905 patients with nonmetastatic RCC (Tx- N0M0), which was determined by imaging modalities at the time of surgery, were enrolled from five institutions 1Department of Urology, College of Medicine, Hallym University, Chun Cheon, Korea. 2Department of Urology, College of Medicine, Seoul National University, Seoul, Korea. 3Department of Urology, College of Medicine, Korea University, Seoul, Korea. 4Department of Urology, College of Medicine, Chungbuk National University, Cheongju, Korea. 5Department of Urology, College of Medicine, Catholic University, Seoul, Korea. 6Department of Urology, Choi Won Suk Urology Clinic, Yongin, Korea. * Correspondence: Department of Urology, Seoul National University Bundang Hospital, 166, Gumi-ro, Bundang-gu, Seongnam, Kyunggi-do, 463-707, Korea. Tel: +82 31787 7342; Fax: +82 31787 4057. E-mail: ssbyun@snubh.org. Received August 2013 & Accepted March 2014 Urological Oncology 1884 between 1999 and 2011. All patients had undergone sur- gery with curative intent. Patients with hereditary syn- drome were excluded. Of the 2,905 patients identified, the 2,680 who were diagnosed with cRCC or pRCC, which are the 2 major histologic variants, were final- ly included in our study. One hundred thirteen patients underwent lymph node dissection, because of suspicious malignant lymph node lesions during the operation. How- ever, all patients had pathologically N0. Thirty patients had positive surgical margins, and any of them did not receive any adjuvant treatments. After receiving approv- al from the relevant institutional review board (approval No. B-1202-145-102), patients’ clinical and pathologic data were reviewed. Variables The following variables were noted in all patients: fol- low-up duration including recurrence-free survival (RFS) and cancer-specific survival (CSS), age, gender, body mass index (BMI), Eastern Cooperative Oncology Group performance status (ECOG PS), symptoms at presenta- tion, tumor size and laterality, type and method of sur- gery, pathologic stage, Fuhrman’s grade and histologic subtype. The follow-up consisted of a history, physical examina- tion, comprehensive metabolic panel, abdominal comput- ed tomography and chest radiography performed every 3 months for 6 months, every 6 months for 3 years and yearly after surgery for most of the patients. Bone or brain scan was performed only when clinically indicated. RFS and CSS were determined from the date of surgery to the date of recurrence and cancer-specific death, respectively, and identified using the imaging studies. Tumor size was determined from the pathologic specimen by recording the greatest diameter. Pathologic stage was determined according to the 2002 TNM classification system. Tumor grade was determined according to the Fuhrman’s nucle- ar grade. Histologic subtyping was performed according to the 2004 WHO classification. To assess pathologic fea- tures, urological pathologists reviewed all specimens at Variables pRCC cRCC P Value Number of subjects 192 2488 ----- Follow up duration, mean ± SD, months 38.0 ± 26.8 37.8 ± 29.7 .950 Age, mean ± SD, years 56.4 ± 13.5 56.1 ± 12.4 .694 Gender, no (%) Male 142 (74.0) 1785 (71.7) Female 50 (26.0) 703 (28.3) .511 BMI, mean ± SD, kg/m2 24.2 ± 3.4 24.5 ± 3.3 .136 ECOG PS ≥ 1, no (%) 114 (59.4) 1515 (60.9) .678 Symptoms at presentation, no (%) .419 No symptoms 156 (81.3) 2074 (83.4) Hematuria 16 (8.3) 235 (9.4) Flank pain 15 (7.8) 135 (5.4) Others 5 (2.6) 44 (1.8) Tumor size Continuous, mean ± SD, cm 4.3 ± 3.1 4.2 ± 2.9 .778 Category, no (%) .225 < 4 112 (58.3) 1410 (56.7) 4-7 43 (22.4) 682 (27.4) > 7 37 (19.3) 396 (15.9) Side, no (%) Left 105 (54.7) 1270 (51.0) Right 86 (44.8) 1207 (48.5) Bilateral 1 (0.5) 11 (0.4) Type of surgery, no (%) .01 Radical nephrectomy 97 (50.5) 95 (49.5) Partial nephrectomy 1492 (60.0) 996 (40.0) Method of surgery, no (%) 0.067 Laparoscopic 127 (66.1) 1799 (72.3) Open 65 (33.9) 689 (27.7) T stage (TNM 2002), no (%) .367 pT1 151 (78.6) 1994 (80.1) pT2 20 (10.4) 203 (8.2) pT3 18 (9.4) 273 (11.0) pT4 3 (1.6) 18 (0.7) Fuhrman’s grade, no (%) .001 G1 7 (3.6) 166 (6.7) G2 86 (44.8) 1385 (55.7) G3 91 (47.4) 840 (33.8) G4 8 (4.2) 97 (3.9) Table 1. Association of different clinical and pathological variables with histologic subtype in non-metastatic renal cell carcinoma. Abbreviations: pRCC, papillary renal cell carcinoma; cRCC, clear cell renal cell carcinoma; SD, standard deviation; no, number of subjects; BMI, body mass index; ECOG PS, Eastern Cooperative Oncology Group performance status. Nonmetastatic Papillary RCC in Korea-Lee et al Vol 11. No 05 Sept-Oct 2014 1885 each institution. Statistical Analysis To compare the relationship between clinical and patho- logic characteristics of patients with cRCC and pRCC, independent t-test and Pearson’s chi-square or Fisher’s exact test were used in the comparison of continuous and categorical variables, respectively. The RFS and CSS rates were calculated using the Kaplan-Meier method, and the log-rank test was used to examine the difference in survival rate between two groups. The prognostic val- ues of variables for survival were evaluated using Cox proportional hazards models. All tests were two-sided, and P < .05 was considered to be statistically significant. Statistical Package for the Social Sciences, version 17.0 (SPSS, Chicago, IL, USA) was used for all statistical as- sessments. RESULTS Histologic subtypes of all patients included the follow- ing; 2,488 (85.6%) with cRCC, 192 (6.6%) with pRCC, 158 (5.4%) with chromophobe RCC, 14 (0.5%) with col- lecting duct RCC and 53 (1.8%) with unclassified RCC. Mean age at surgery for all patients was 55.9 ± 12.6 years, and there were 2,040 males (70.2%) and 865 females (29.8%). Mean tumor size was 4.3 ± 3.0 cm, and the pathologic stage was T1a in 1,659 (57.1%), T1b in 628 (21.6%), T2 in 256 (8.8%), T3a in 306 (10.5%), T3b in 26 (0.9%), T3c in 5 (0.2%) and T4 in 25 patients (0.9%). Fuhrman’s grade was grade 1 in 183 (6.3%), grade 2 in 1,548 (53.3%), grade 3 in 1.039 (35.8%) and grade 4 in 135 patients (4.6%). Table 1 shows the association of different clinical and pathological variables between patients with cRCC and pRCC. Patients with pRCC did not differ significant- ly from those with cRCC for most variables. However, high Fuhrman’s grades were more common in pRCCs than in cRCCs (P = .001). During the follow-up period, among 2,488 patients with cRCC, 258 (10.4%) had re- currence and among 192 patients with pRCC, 14 (7.3%) had recurrence. The 5-year RFS rates were 84.9 ± 1.0% in patients with cRCC and 86.7 ± 4.0% in patients with pRCC, and there was no significant difference between the two groups (P = .167, Figure 1). When patients were stratified according to pathologic stage, the 5-year RFS rates for cRCC and pRCC with T1 were 91.4 ± 0.9% and 94.1 ± 3.2% (P = .095), with T2 were 73.6 ± 4.1% and 87.7 ± 8.2% (P = .321), and with T3-4 were 47.5 ± 4.3% and 23.3 ± 19.3% (P = .553), respectively. When patients were stratified according to Fuhrman’s grade, the 5-year RFS rates for cRCC and pRCC with grade 1 were 94.1 ± 2.8% and 100% (P = .674), with grade 2 were 92.9 ± 1.0% and 88.3 ± 6.3% (P = .810) and with grades 3-4 were 70.6 ± 2.2% and 84.6 ± 5.3% (P = .020), respective- ly. A significant RFS difference was only present among the two groups for Fuhrman’s grade 3-4. During the follow-up period, 129 patients (5.2%) with cRCCs and 10 patients (5.2%) with pRCC died of can- cer-specific causes. The 5-year CSS rates were 92.0 ± 0.8% in patients with cRCC and 93.1 ± 2.6% in patients with pRCC, and there was no significant difference be- tween two groups (P = .931, Figure 2). When patients were stratified according to pathologic stage, the 5-year CSS rates for cRCC and pRCC with T1 were 96.8 ± 0.6% and 97.8 ± 2.2% (P = .209), with T2 were 88.6 ± 2.9% and 84.6 ± 10.0% (P = .764) and with T3-4 were 62.5 ± 4.3% and 60.5 ± 16.0% (P = .100), respectively. When patients were stratified according to Fuhrman’s grade, the 5-year CSS rates for cRCC and pRCC with grade 1 were 100% and 100% (P = .850), with grade 2 were 96.5 ± 0.7% and 100% (P = .608), and with grades 3-4 were 83.2 ± 1.9% and 87.5 ± 4.8% (P = .357), respectively. No significant CSS difference remained among the two groups stratified according to pathologic stage or Fuhrman’s grade. Independent prognostic factors of RFS in multivariate analysis were age, BMI, symptoms at presentation, tumor size, pathologic stage and Fuhrman’s grade. However, the histologic subtype was not an independent prognostic factor of RFS in univariate and multivariate analysis (P = .170 and 0.107; hazard ratio (HR) = 0.686 and 0.642; 95% confidence interval (CI) = 0.401-1.175 and 0.374- Univariate Analysis Multivariate Analysis Variables HR 95% CI P value HR 95% CI P value Age 1.019 1.009-1.030 < .001 1.013 1.002-1.023 .020 Gender: Female vs. Male 0.935 0.715-1.222 . 621 0.976 0.742-1.282 . 859 BMI (kg/m2) 0.932 0.896-0.969 < .001 0.941 0.903-0.980 .004 ECOG PS: ≥ 1 vs. 0 1.291 1.004-1.661 .047 0.816 0.624-1.067 .137 Symptoms at presentation 4.774 3.763-6.057 < .001 2.027 1.531-2.682 < .001 Tumor size 1.300 1.268-1.334 < .001 1.172 1.128-1.218 < .001 Side: Bilateral vs. Unilateral 1.584 0.394-6.369 .517 1.911 0.471-7.755 .365 T stage (2002 TNM) < .001 < .001 T2 vs. T1 3.606 2.551-5.096 < .001 0.883 0.582-1.340 .558 T3 and 4 vs. T1 9.084 6.995-11.798 < .001 2.571 1.798-3.676 < .001 Fuhrman’s grade < .001 < .001 G2 vs. G1 1.948 0.788-4.813 .149 1.630 0.658-4.040 .291 G3 and 4 vs. G1 7.997 3.287-19.453 < .001 4.106 1.666-10.120 .002 Histology: pRCC vs. cRCC 0.686 0.401-1.175 .170 0.642 0.374-1.101 .107 Table 2. Univariate and multivariate analyses predicting the probability of recurrence in non-metastatic renal cell carcinoma. Abbreviations: HR, hazard ratio; CI, confidence interval; BMI, body mass index; ECOG PS, Eastern Cooperative Oncology Group performance status; pRCC, papillary renal cell carcinoma; cRCC, clear cell renal cell carcinoma. Nonmetastatic Papillary RCC in Korea-Lee et al Urological Oncology 1886 1.101, respectively (Table 2). Similar to those of RFS, independent prognostic factors of CSS in multivariate analysis were age, BMI, symptoms at presentation, tumor size, pathologic stage and Fuhrman’s grade. The histo- logic subtype did not remain an independent prognostic factor of CSS in univariate or multivariate analysis (P = .931 and .998; HR = 1.029 and 1.001; 95% CI = 0.540- 1.958 and 0.522-1.917, respectively (Table 3). DISCUSSION It is well known that histologic subtypes of RCC show differences in genetic and morphologic parameters.(4) Nevertheless, the characteristics and prognostic value of these histologic subtypes remain controversial.(2) Deng and colleagues(2) recently analyzed the data from large cohort studies including more than 20,000 patients. They Figure 1. Kaplan-Meier curve for recurrence-free survival for patients with non-metastatic renal cell carcinoma according to histologic subtype. Abbreviations: cRCC, clear cell renal cell carcinoma; pRCC, papillary renal cell carcinoma. Figure 2. Kaplan-Meier curve for cancer-specific survival for patients with non-metastatic renal cell carcinoma according to histologic subtype. Abbreviations: cRCC, clear cell renal cell carcinoma; pRCC, papillary renal cell carcinoma. Nonmetastatic Papillary RCC in Korea-Lee et al Vol 11. No 05 Sept-Oct 2014 1887 reported that the independent prognostic value of the his- tologic subtype is not yet widely accepted. Therefore, far more studies are needed to prove the prognostic value of histologic subtype. Another important point is that almost all studies in this field have focused on Western populations. Differences in characteristics of cancer among different races have well been established in other urologic cancers, such as prostate cancer.(5) Racial disparities in RCC have also been reported in many studies. Using the national Sur- veillance, Epidemiology, and End Results (SEER) data- base, Vaishampayan and colleagues(6) reported that young blacks with localized renal cancer appeared to have had a greater rise in incidence and a poorer outcome than whites of the same age and pathologic stage. Tripathi and col- leagues(9) also reported that race was a significant predic- tor of overall survival in metastatic RCC. Recently, Staf- ford and colleagues(7) published a large population-based study comparing racial/ethnic groups using 39,434 cases of RCC. They concluded that higher incidence rates and lower survival rates were identified among blacks when compared to their counterparts, whereas Asian/Pacific Is- landers showed the opposite trend. RCC subtypes have also been reported to differ by race. Recently, Sankin and colleagues(8) reported that pRCC had a much higher oc- currence among Blacks compared to non-Blacks. More recently, Purdue and colleagues(10) analyzed data from two large case-control studies of RCC, and observed a significant difference across RCC subtypes with respect to their distribution by race. Although the issue of race has not fully been established yet in RCC, the characteris- tics and prognostic value of histologic subtypes in Asians may differ from those in Westerners. Viewed in this light, our study is remarkable, because it is one of the largest population-based studies, especially in Asia. pRCC generally accounts for approximately 10% of all RCCs, and is historically associated with smaller tumor size and presentation at an earlier stage and grade when compared with cRCC.(2,15,16) In our study, the distribu- tions of cRCC and pRCC were 2,488 (85.6%) and 192 (6.6%) patients, respectively. The proportion of pRCC was slightly lower than that reported in previous studies performed on Western populations,(2) but it was similar to those performed on the Asian population.(11-14) In two studies of Japanese patients, the incidence of pRCC was 5.4 and 5.6%, respectively.(11,12) In a study of Chinese patients, the incidence of pRCC was 4.1%.(13) In a large multicenter study analyzing 2,981 Korean patients, the incidence of pRCC was also 5.6%.(14) Previous results in- cluding ours suggest that the proportion of pRCC in Asia might be lower than that in the West. From the viewpoint of tumor characteristics, opposite of our expectations, our results showed that there were no significant differ- ences in the distribution of age, gender, tumor size, and stage between pRCC and cRCC. On the contrary, high Fuhrman’s grades were even more common in pRCC. These trends are also similar to those in studies which were performed on the Asian population.(17-19) RCC is a heterogeneous and complex disease, and has genetic and molecular differences among the histologic subtypes.(1,4) Therefore, our results might reflect the characteristics of pRCC not in the West but in Asia. Our results show that independent prognostic factors of RFS and CSS are age, BMI, symptoms at presentation, tu- mor size, stage and Fuhrman’s grade. Of these factors, the most powerful predictors are stage and Fuhrman’s grade. These results are similar to those reported in previous studies.(2,5) Stage represents the major prognostic factor used routinely in localized RCC.(2) The role of the Fuhr- man’s grade for pRCC is not widely accepted. However, recent studies have shown the prognostic value of Fuhr- man’s grade in pRCC. Klatte and colleagues(20) reported that Fuhrman’s grade should be the standard grading sys- tem for pRCC. Zucchi and colleagues(21) also suggested that the use of Fuhrman’s grade had prognostic relevance for pRCC. Our results also support the suggestion that Fuhrman’s grade has the prognostic value for pRCC. The independent prognostic value of the histologic sub- types has not yet been verified. Although most previous studies have shown the prognostic value of histologic subtypes by univariate analysis, only a few studies have shown prognostic significance on multivariate analysis. (2) Teloken and colleagues(22) in a study of 1,863 patients with localized RCCs, showed that histologic subtype re- mained significantly associated with metastasis and CSS on multivariate analysis. However, the difference be- tween pRCC and cRCC did not reach statistical signifi- cance. Recently, in a study of 3,062 patients, Leibovich and colleagues(23) reported that histologic subtype was an independent predictor of progression to distant metastasis and CSS. However, because these authors only analyzed the prognostic difference between cRCC and non-cRCC, the prognostic value of pRCC compared with cRCC is unclear. All of these large single center series have not proved that pRCC has a more favorable outcome than cRCC, when adjusted for covariates. Results from multi- center and international studies have also been similar to those from single center series. Patard and colleagues(15) analyzed 4,063 patients across the United States and Eu- rope. They concluded that histologic subtype was not an independent prognostic factor of RCCs. In another mul- ticenter study which included 2,530 patients in Europe, Karakiewicz and colleagues(24) reported that histologic subtype was not associated with outcome on multivariate analysis. Recently, Keegan and colleagues(25) published a very large multicenter study analyzing 17,605 patients using the SEER database. They reported that the effects of histologic subtype were decreased substantially af- ter accounting for covariates. Particularly, the prognos- tic value of pRCC was not significantly different from that of cRCC. Similar to previous studies, our study has shown that the RFS and CSS of pRCC are not significant- ly different compared with those of cRCC. Both groups also have comparable RFS and CSS when stratified by pathologic stage and Fuhrman’s grade except for grade 3-4. Based on univariate and multivariate analyses, his- tologic subtype is likewise not an independent prognostic factor. More recently, Steffens and colleagues(26) pub- lished a multicenter study in Germany. They reported that non-metastatic pRCC had a better prognosis compared with non-metastatic cRCC. However, it should be known that age, gender, stage and grade were only considered as a covariate. Two distinct subtypes of pRCC were introduced in 1997, and are generally known to be associated with different clinical outcomes.(27) We could not assess the subtypes Nonmetastatic Papillary RCC in Korea-Lee et al Urological Oncology 1888 of pRCC in our retrospective study. pRCC is heteroge- neous,(28,29) and can show features characteristic of type 1 and 2.(30) Furthermore, many recent studies have not identified subtype as an independent prognostic factor. (30) Therefore, the prognostic value of subtype of pRCC is yet controversial. Nevertheless, it is a weak point that subtype of pRCC was not assessed in our study. Patients with lymph node and/or distant metastasis at the time of surgery were excluded, because natural history and treat- ments, such as targeted therapy of metastatic RCCs may be different from those of non-metastatic RCCs. Never- theless, to our knowledge, our study is the largest Asian population-based study of the value of histologic subtype. Our results are noteworthy, because racial disparities in RCC have been reported in many studies,(6-10) and almost all previous studies have focused on the Western popu- lation. There are a few limitations to our study. First, our data are retrospective in nature. Second, our study lacks a central- ized pathologic review. Although most of the multicenter studies including ours have lacked centralized pathologic review, it is likely that this limitation has resulted in mis- classifications or misdiagnoses. However, the proportion of pRCC in our study was similar to that reported previ- ous studies performed on the Asian population, in which urological pathologists reviewed all specimens at each institution. Third, we could not assess potential prognos- tic factors, such as molecular markers, sarcomatoid fea- tures and tumor necrosis in all patients. These factors may better allow the identification of patients at high risk and affect the clinical outcomes. However, our study includes the most widely accepted independent prognostic fac- tors of non-metastatic RCC, including stage, Fuhrman’s grade, and tumor size,(2) and had one of the largest scale studies, especially in Asia. CONCLUSION Our results show that non-metastatic pRCC is not sig- nificantly different from non-metastatic cRCC in various clinical and pathologic parameters. RFS and CSS of non- metastatic pRCC are not significantly different compared with those of non-metastatic cRCCs, even when stratified by pathologic stage and Fuhrman’s grade. In addition, histologic subtype is not an independent prognostic fac- tor of non-metastatic RCC. Our study suggests that tumor characteristics and prognosis of pRCCs might be compa- rable to that of cRCCs, especially in Asia. CONFLICT OF INTEREST None declared. REFERENCES 1. Sun M, Shariat SF, Karakiewicz PI. Factors af- fecting outcome in renal cell carcinoma. Curr Opin Urol. 2010;20:355-60. 2. Deng FM, Melamed J. Histologic variants of renal cell carcinoma: Does tumor type influence outcome? Urol Clin N Am. 2012;39:119-32. 3. Kovacs G, Akhtar M, Beckwith BJ, et al. The Heidelberg classification of renal cell tumours. J Pathol. 1997;183:131-3. 4. Cohen HT, McGovern FJ. Renal-cell carcino- ma. N Engl J Med. 2005;353:2477-90. 5. Klein EA, Platz EA, Thompson IM. Epidemiol- ogy, etiology, and prevention of prostate cancer. In: Campbell-walsh urology, 9th ed. Elsevier Saunders, St. Louis, 2007. 6. Vaishampayan UN, Do H, Hussain M, et al. Racial disparity in incidence patterns and out- come of kidney cancer. Urology. 2003;62:1012- 7. 7. Stafford HS, Saltzstein SL, Shimasaki S, Sand- ers C, Downs TM, Sadler GR. Racial/ethnic and gender disparities in renal cell carcinoma inci- dence and survival. J Urol. 2008;179:1704-8. 8. Sankin A, Cohen J, Wang H, Macchia RJ, Ka ranikolas N. Rate of renal cell carcinoma sub- types in different races. Int Braz J Urol. 2011;37:29-32. 9. Tripathi RT, Heilbrun LK, Jain V, Vaishampay- an UN. Racial disparity in outcomes of a clinical trial population with metastatic renal cell carci- noma. Urology. 2006;68:296-301. 10. Purdue MP, Moore LE, Merino MJ, et al. An investigation of risk factors for renal cell carci- noma by histologic subtype in two case-control studies. Int J Cancer. 2013;132:2640-7. 11. Yamashita S, Ioritani N, Oikawa K, Aizawa M, Endoh M, Arai Y. Morphological subtyping of papillary renal cell carcinoma: clinicopathologi- al characteristics and prognosis. Int J Urol. 2007;14:679-83. 12. Onishi T, Ohishi Y, Goto H, Suzuki M, Miyaza wa Y. Papillary renal cell carcinoma: clinico- pathological characteristics and evaluation of prognosis in 42 patients. BJU Int. 1999;83:937- 43. 13. Chen J, Shi B, Zhang D, Jiang X, Xu Z. The clinical characteristics of renal cell carcinoma in female patients. Int J Urol. 2009;16:554-7. 14. Lee WK, Byun SS, Kim HH, et al. Character- istics and prognosis of chromophobe non-meta- static renal cell carcinoma: a multicenter study. Int J Urol. 2010;17:898-904. 15. Patard JJ, Leray E, Rioux-Leclercq N, et al. Prognostic value of histologic subtypes in renal cell carcinoma: A multicenter experience. J Clin Oncol. 2005;23:2763–71. 16. Vikram R, Ng CS, Tamboli P, et al. Papillary re- nal cell carcinoma: radiologic-pathologic cor- relation and spectrum of disease. Radiograph- ics. 2009;29:741-54. 17. Tabibi A, Parvin M, Abdi H, Bashtar R, Zamani N, Abadpour B. Correlation between size of re- nal cell carcinoma and its grade, stage, and his- tological subtype. Urol J. 2007;4:10-3. 18. Kim H, Cho NH, Kim DS, et al. Renal cell car- cinoma in South Korea: A multicenter study. Hum Pathol. 2004;35:1556-63. 19. Süer E, Baltaci S, Burgu B, Aydoğdu Ö, Göğüş Ç. Significance of tumor size in renal cell can- cer with perinephric fat infiltration: is TNM staging system adequate for predicting progno- sis? Urol J. 2013;10:774-9. 20. Klatte T, Anterasian C, Said JW, et al. Fuhrman Nonmetastatic Papillary RCC in Korea-Lee et al Vol 11. No 05 Sept-Oct 2014 1889 grade provides higher prognostic accuracy than nucleolar grade for papillary renal cell carcino- ma. J Urol. 2010;183:2143-7. 21. Zucchi A, Novara G, Costantini E, et al. Prognostic factors in a large multi-institution al series of papillary renal cell carcinoma. BJU Int. 2012;109:1140-6. 22. Teloken PE, Thompson RH, Tickoo SK, et al. Prognostic impact of histological subtype on surgically treated localized renal cell carcino- ma. J Urol. 2009;182:2132-6. 23. Leibovich BC, Lohse CM, Crispen PL, et al. Histological subtype is an independent predictor of outcome for patients with renal cell carcino- ma. J Urol. 2010;183:1309-16. 24. Karakiewicz PI, Briganti A, Chun FK, et al. Multi-institutional validation of a new renal cancer-specific survival nomogram. J Clin On- col. 2007;25:1316-22. 25. Keegan KA, Schupp CW, Chamie K, Hellenthal NJ, Evans CP, Koppie TM. Histopathology of surgically treated renal cell carcinoma: survival differences by subtype and stage. J Urol. 2012;188:391-7. 26. Steffens S, Janssen M, Roos FC, et al. Incidence and long-term prognosis of papillary compared to clear cell renal cell carcinoma-a multicentre study. Eur J Cancer. 2012;48:2347-52. 27. Delahunt B, Eble JN. Papillary renal cell carci- noma: a clinicopathologic and immuno histochemical study of 105 tumors. Mod Pathol. 1997;10:537-544. 28. Arora P, Rao S, Khurana N, Ramteke VK. Re- nal replacement lipomatosis with coexistent papillary renal cell carcinoma, renal tubulopap- illary adenomatosis, and xanthogranulomatous pyelonephritis: an extremely rare association and possible pathogenetic correlation. Urol J. 2013;10:906-8. 29. Atilgan D, Uluocak N, Erdemir F, Parlaktas BS, Koseoglu RD, Boztepe O. Multicystic renal cell carcinoma: a rare kidney tumor in children. Urol J. 2013;10:811-4. 30. Sukov WR, Lohse CM, Leibovich BC, Thomp son RH, Cheville JC. Clinical and pathological features associated with prognosis in patients with papillary renal cell carcinoma. J Urol. 2012;187:54-9. Nonmetastatic Papillary RCC in Korea-Lee et al Urological Oncology 1890