- Email: [email protected]
Tumor Size Does Not Predict Risk of Metastatic Disease or Prognosis of Small Renal Cell Carcinomas
Tumor Size Does Not Predict Risk of Metastatic Disease or Prognosis of Small Renal Cell Carcinomas Tobias Klatte, Jean-Jacques Patard, Michela de Martino, Karim Bensalah, Gregory Verhoest, Alexandre de la Taille, Clément-Claude Abbou, Ernst Peter Allhoff, Giuseppe Carrieri, Stephen B. Riggs, Fairooz F. Kabbinavar, Arie S. Belldegrun and Allan J. Pantuck* From the Departments of Urology (TK, MM, SBR, ASB, AJP) and Medicine (FFK), David Geffen School of Medicine, University of California-Los Angeles, Los Angeles, California, Service d’Urologie, Centre Hospitalier Universitaire Pontchaillou, Rennes (JJP, KB, GV), Service d’Urologie, Centre Hospitalier Universitaire Henri Mondor, Créteil (ADT, CCA), France, Urologische Universitätsklinik, Universität Magdeburg, Germany (EPA), and Dipartimento di Scienze Chirurgiche, Divisione di Urologia e Centro Trapianti, Università di Foggia, Italy (GC)
Purpose: We characterized the clinicopathological features and the prognosis of small solid renal tumors defined as tumors 4 cm or smaller. Materials and Methods: We identified 1,208 patients who were treated with nephrectomy at 5 international academic centers for small solid renal tumors. Clinicopathological parameters and outcome data were collected for each patient and analyzed. Results: Of the tumors 88% were renal cell carcinoma and 12% were benign. Of those with renal cell carcinoma 995 (93%) were localized (N0M0) and 72 (7%) presented with metastatic disease. Tumor size did not predict synchronous metastatic disease. The incidence of metastatic disease in the tumor size ranges 0.1 to 1.0, 1.1 to 2.0, 2.1 to 3.0 and 3.1 to 4.0 cm was 7%, 6%, 5% and 8%, respectively (p ⫽ 0.322). Survival rates were excellent. The majority of patients who died of renal cell carcinoma (54%) presented with synchronous metastatic disease, but 3% of patients with localized disease also died of renal cell carcinoma. In patients with localized disease there was a 7% chance of recurrence post nephrectomy at 5 years. Progression-free survival (28 months) was better than for patients with metastatic disease having a primary tumor greater than 4 cm (8 months). Tumor size was not retained as an independent prognostic factor of survival in multivariate analyses. The University of California Integrated Staging System and the Karakiewicz nomogram were the best predictors of cancer specific survival for all renal cell carcinoma stages (c-index 0.87). Conclusions: More than 85% of small solid renal tumors are renal cell carcinoma. The majority of localized small renal tumors can be cured with existing surgical approaches. However, there is a small but not insignificant risk of synchronous and metachronous metastatic disease and cancer associated death. Patients considering experimental therapies such as ablation and surveillance should be aware of this. Tumor size alone is not sufficient to distinguish renal cell carcinoma with benign behavior from aggressive small renal cell carcinoma. Survival of patients with small metastatic renal cell carcinoma is better then expected. The biology of these unique tumors should be further studied. Key Words: kidney neoplasms, prognosis, nomograms
enal cell carcinoma is the third most common urological malignancy and accounts for approximately 3% of all cancers in adults. It has been estimated for 2007 that about 51,000 people will be diagnosed with RCC in the United States, and approximately 13,000 will die of the disease.1 The incidence of RCC has steadily increased during the last 3 decades, mainly due to the widespread use of routine abdominal imaging and the detection of incidental, small solid renal tumors (4 cm or smaller) thereby.2 Fre-
R
Submitted for publication August 28, 2007. Study received institutional review board approval. * Correspondence and requests for reprints: Department of Urology, David Geffen School of Medicine at UCLA, 10833 Le Conte Ave., Room B7-298A CHS, Los Angeles, California 90025-1738 (telephone: 310-206-2436; FAX: 310-206-408; e-mail: apantuck@mednet. ucla.edu).
See Editorial on page 1657. For another article on a related topic see pages 2010 and 2057.
0022-5347/08/1795-1719/0 THE JOURNAL OF UROLOGY® Copyright © 2008 by AMERICAN UROLOGICAL ASSOCIATION
quently these tumors are benign or RCC with favorable pathological characteristics and prognosis.3 However, it is known that a subset of small renal tumors exhibit aggressive properties including synchronous or metachronous metastatic spread. Patients with synchronous metastatic disease will be treated with surgery followed by targeted or immune based therapy. However, the prognosis of patients with small metastatic RCC is yet to be specifically determined. For patients who are treated for localized disease (N0M0), prediction of prognosis is critical to provide tailored surveillance and patient allocation into adjuvant clinical trials.4 In addition, a better understanding of the biology of small, localized RCC would allow easier patient
Editor’s Note: This article is the second of 5 published in this issue for which category 1 CME credits can be earned. Instructions for obtaining credits are given with the questions on pages 2070 and 2071.
1719
Vol. 179, 1719-1726, May 2008 Printed in U.S.A. DOI:10.1016/j.juro.2008.01.018
1720
SMALL SOLID RENAL TUMORS
counseling for new nonsurgical approaches such as radio frequency ablation, cryotherapy and active surveillance which have been evolved to considerable alternative options for a subset of patients, although surgical extirpation remains the standard of care.5 We report on 1,208 patients who presented with a small solid renal tumor at 5 international institutions. Because all patients underwent surgical extirpation, this series allows the definition of the actual incidence of renal tumor types and pathological staging. Analyses were performed 1) to define the incidence of benign tumors, localized RCC, metastatic RCC, clinical and pathological characteristics, and prognosis, 2) to characterize the malignant potential of small renal tumors, defined as the probability of metachronous metastatic disease after undergoing surgery for localized disease, and 3) to evaluate the performance of prognostic models to predict survival of small RCC. PATIENTS AND METHODS Patient Population We studied retrospectively 1,208 patients, who were diagnosed with unilateral, sporadic, small solid renal tumors from 1989 to 2006 at 5 international academic urological centers in Créteil, France; Rennes, France; Magdeburg, Germany; Foggia, Italy; and Los Angeles, California. A small
solid renal tumor was defined as a solid renal tumor 4 cm or smaller. The study was approved by the institutional review board at each institution. Patient charts were reviewed and clinicopathological data recorded. The data that was abstracted included age, gender, symptoms,6 ECOG PS, T, N and M stage according to 2002 definitions, Fuhrman grade, and histological type. Pathological examination was performed by 1 anatomical pathologist at each institution. The type of renal tumor was defined according to WHO criteria. For practical purposes, we distinguished between benign and malignant renal tumors. Benign tumors were further stratified into oncocytoma, angiomyolipoma, atypical cyst and other. Malignant renal tumors were all RCC in this series and were divided into clear cell, papillary, chromophobe, collecting duct and unclassified subtype. All patients underwent open surgical extirpation. A radical nephrectomy was performed in 834 patients (69%), while 374 (31%) underwent nephron-sparing surgery. Surgical margins were negative in all patients. No adjuvant therapy was administered postoperatively. Patients who presented with metastatic disease underwent immunotherapy following nephrectomy. Postoperative surveillance was performed according to guidelines established at each institution. Median followup was 3.1 years (range 0.1 to 18.0).
TABLE 1. Patient and tumor characteristics
No. pts Mean pt age (SD) No. gender (%): Female Male No. ECOG PS (%): 0 1 or Greater No. presentation (%): Incidental Local Systemic Mean tumor size (SD) No. cm tumor (%): 0.1–1.0 1.1–2.0 2.1–3.0 3.1–4.0 No. benign tumor type (%): Oncocytoma Angiomyolipoma Complex cyst Other No. RCC type (%): Clear cell Papillary Chromophobe Collecting duct Unclassified No. T stage (%): T1a T3a T3b No. N/M stage (%): Nx or N0/M0 N⫹/M0 N⫹/M1 N0/M1 No. Fuhrman grade (%): G1 G2 G3–G4
Benign Tumor
Localized RCC
Metastatic RCC
141 60.3 (14.7)
995 61.3 (11.9)
72 60.6 (11.3)
Totals 61.2
(12.3)
65 76
(46) (54)
380 615
(38) (62)
21 51
(29) (71)
466 742
(39) (61)
127 14
(90) (10)
806 189
(81) (19)
33 39
(46) (54)
966 242
(80) (20)
118 (84) 21 (15) 2 (1) 2.7 (0.8) 5 32 65 39
(3) (23) (46) (28)
75 33 16 17
(53) (23) (11) (12)
782 (79) 201 (20) 12 (1) 2.9 (0.8)
31 (43) 34 (47) 7 (10) 3.1 (0.9)
931 256 21 2.9
(77) (21) (2) (0.9)
26 177 372 420
2 12 20 38
33 221 457 497
(3) (18) (38) (41)
(3) (18) (37) (42) —
(3) (17) (28) (53) —
—
—
802 145 48 0 0
(81) (15) (5) (0) (0)
62 7 1 1 1
(86) (10) (1) (1) (1)
864 152 49 1 1
(81) (14) (5) (less than 1) (less than 1)
—
895 83 17
(90) (8) (2)
44 23 5
(61) (32) (7)
939 106 22
(88) (10) (2)
995 0 0 0
(100) (0) (0) (0)
0 8 8 56
(0) (11) (11) (78)
995 8 8 56
(93) (less than 1) (less than 1) (5)
281 554 160
(28) (56) (16)
7 36 29
(10) (50) (40)
288 590 189
(27) (55) (18)
—
—
SMALL SOLID RENAL TUMORS Statistical Analysis The statistical software R v2.4 (www.r-project.org) was used for all analyses. The chi-square and Student t tests were used to compare categorical and continuous data, respectively. The relationship between tumor size and histology (benign tumor vs RCC) as well as metastatic status (localized vs metastatic RCC) was further analyzed with logistic regression models. End points of this study included overall survival, cancer specific survival, recurrence-free survival and progressionfree survival, which were all calculated from the date of surgery to the date of death, death from RCC, disease recurrence in localized disease and disease progression in metastatic disease, respectively. The Kaplan-Meier method was used to estimate the survivor functions, which were compared with the log rank test. Confidence intervals for survival data were determined by the Greenwood method. Multivariate Cox proportional hazard models were performed to detect variables independently associated with survival. Prognostic assessment was performed according to the 3 prognostic models of the UISS,7 the Kattan nomogram8 and the Karakiewicz nomogram.9 The predictive accuracy of these models was measured by the concordance index (cindex). The c-index corresponds to the area under the receiver operating characteristics (ROC) curve adapted for survival data and ranges from 0.5 (no predictive accuracy) to 1.0 (perfect predictive accuracy). RESULTS Of the 1,208 consecutive surgical patients 742 were men (61%) and 466 were women (39%). Mean age at diagnosis was 61 years (median 62, range 17 to 89). An ECOG PS of 0
1721
was assigned to 80% of the patients. A total of 931 patients (77%) were completely asymptomatic at the time of diagnosis and had incidentally discovered tumors, 256 (21%) presented with local symptoms, and 21 (2%) had tumors associated with systemic symptoms. A total of 1,067 patients (88%) had RCC, of whom 995 (93%) presented without evidence of metastatic disease and 72 (7%) had lymph node and/or distant metastases. Distant metastatic disease was confirmed by biopsy in 56% of the patients. The most frequently assigned histological subtypes were clear cell (81% of RCCs, 72% of small renal tumors), papillary (14%, 13%), and chromophobe (5%, 4%). There were 141 patients (12%) who had benign tumors, most frequently oncocytoma (53% of benign tumors, 6% of small renal tumors), angiomyolipoma (23%, 3%), complex cysts (11%, 1%), and other entities (12%, 1%). Patient and tumor characteristics are summarized in table 1.
Clinicopathological Features Benign renal tumor vs RCC. Tumor size was statistically significant but clinically insignificantly smaller in benign tumors than RCC (mean ⫾ SD, 2.7 ⫾ 0.8 cm vs 2.9 ⫾ 0.9 cm, p ⫽ 0.004). The frequency of RCC in the tumor size ranges 0.1 to 1.0, 1.1 to 2.0 and 2.1 to 3.0 cm was similar (85%, 86% and 86%, p ⫽ 0.987) and increased significantly in tumors measuring 3.1 to 4.0 cm (92%, p ⫽ 0.001). Patients with benign tumors had significantly better ECOG PS (ECOG PS 0, 90% vs 79%, p ⫽ 0.001) and were more likely to present with incidental tumors (84% vs 76%, p ⫽ 0.047). Also, benign tumors were more likely to be diagnosed in women (46% vs 38%), although this difference did not quite reach statistical
FIG. 1. Kaplan-Meier survival estimates for overall survival and 95% CI. Patients with benign renal tumors had significantly better survival than patients with localized RCC (p ⫽ 0.008). Numbers of patients at risk are indicated.
1722
SMALL SOLID RENAL TUMORS
significance (p ⫽ 0.051). No difference was detected in age (60.3 ⫾ 14.7 years vs 61.3 ⫾ 11.9 years, p ⫽ 0.436). The same clinicopathological features were compared between patients with benign tumors and localized RCC (995). Patients with benign tumors had smaller tumor sizes (2.7 ⫾ 0.8 cm vs 2.9 ⫾ 0.8 cm, p ⫽ 0.003), corresponding to a 34% increase in risk of RCC by 1 cm increase in tumor size (risk ratio 1.34, 95% CI 1.10 –1.65). An ECOG PS of 0 was more frequent in benign tumors (90% vs 81%, p ⫽ 0.009), while no difference was detected in the frequency of incidental tumors (84% vs 79%, p ⫽ 0.163) or age (p ⫽ 0.415). Again, the percentage of women was higher in patients with benign lesions than in RCC (46% vs 38% p ⫽ 0.072). Localized (N0M0) vs metastatic RCC. A total of 995 patients with RCC (93%) had no evidence of metastatic disease at the time of diagnosis, while 72 (7%) had metastatic disease including 56 with N0M1, 8 N⫹M0 and 8 N⫹M1 disease. Factors associated with metastatic disease in patients with RCC were an ECOG PS 1 or greater (metastatic vs localized, 54% vs 19%, p ⬍0.001), presentation with symptoms (57% vs 21%, p ⬍0.001), higher T stage (incidence of metastatic disease, pT1a 5%, pT3a 22%, pT3b 23%, p ⬍0.001), higher grades (p ⬍0.001) and histological type (p ⬍0.001). The incidence of clear cell RCC was higher in patients with metastatic disease (86% vs 81%), while the frequency of papillary RCC (10% vs 15%) and chromophobe RCC (1% vs 5%) was lower. Tumor size did not differ between patients with localized and metastatic disease (2.9 ⫾ 0.8 cm vs 3.1 ⫾ 0.9 cm, p ⫽ 0.096, risk ratio 1.29 [0.96-1.73]). The incidence of metastatic disease in the tumor size ranges 0.1 to 1.0, 1.1 to 2.0, 2.1 to 3.0 and 3.1 to 4.0 cm was 7%, 6%, 5% and 8%, respectively (p ⫽ 0.322). Tumor size in patients with metastatic disease ranged from 0.6 cm to 4.0 cm. Prognosis At the time of analysis 189 deaths (16%) had occurred including 61 (32%) from RCC. Of the 61 deaths from RCC 33 (54%) occurred in patients with synchronous metastatic disease representing 46% of these patients, while only 28 patients with N0M0 disease died following recurrent RCC. These 28 patients accounted for only 3% of the patients who underwent surgery for localized RCC. Overall survival. Five and 10-year overall survival rates (95% CI) for the entire cohort were 83% (80 to 85) and 64% (57 to 69), respectively (fig. 1). Overall survival was significantly better for patients with benign tumors than for patients with localized RCC (5-year 92% [85 to 96] vs 85% [81 to 87], 10-year 85% [71 to 92] vs 63% [55 to 69], p ⫽ 0.008, fig. 1). This survival difference was also present when comparing survival of patients with benign tumors to those with pT1aN0M0 RCC (p ⫽ 0.0277). Median overall survival time for metastatic RCC was 36 months (range 20 to 52), and the 1 and 5-year overall survival rates were 74% (62 to 83) and 36% (23 to 49) (fig. 1). Factors independently associated with overall survival included age, gender, ECOG PS and type of histology (table 2). Cancer specific survival. For the entire RCC cohort the 5 and 10-year survival probabilities were 92% (90 to 94) and 87% (82 to 91) (fig. 2). Patients with localized RCC had 5 and
TABLE 2. Multivariate analysis for overall survival, cancer specific survival and recurrence-free survival Covariate Age Male ECOG PS 1 or greater Symptomatic presentation Tumor size Histology: Localized RCC Benign tumor Metastatic RCC ECOG PS 1 or greater Symptomatic presentation T stage: T1a T3a T3b N stage 1 or greater M stage 1 or greater Fuhrman grade: G1 G2 G3–4 Tumor size Histological subtype
HR
95% CI
p Value
Overall survival 1.03 1.57 1.91 1.19 1.14
1.02–1.05 1.15–2.15 1.36–2.68 0.85–1.66 0.97–1.37
⬍0.001 0.005 ⬍0.001 0.320 0.098
0.27–0.89 2.89–6.36
0.020 ⬍0.001
Ca specific survival 2.26 1.26–4.09 1.53 0.83–2.81
0.006 0.172
1 0.49 4.29
1 2.78 5.08 1.98 10.5
1.48–5.25 2.13–12.1 0.81–4.87 5.94–18.7
0.002 ⬍0.001 0.136 ⬍0.001
1 3.64 5.10 1.22 0.98
1.09–12.2 1.44–18.0 0.85–1.76 0.57–1.69
0.036 0.011 0.282 0.942
Recurrence-free survival ECOG PS 1 or greater 2.00 1.06–3.75 Symptomatic presentation 0.87 0.45–1.69 T stage: T1a 1 T3a 6.01 3.27–11.1 T3b 8.23 3.24–20.9 Fuhrman grade: G1 1 G2 3.19 1.23–8.24 G3–4 3.36 1.14–9.90 Tumor size 1.18 0.81–1.71 Histological subtype:* Clear cell 1 Nonclear cell 0.27 0.06–1.12
0.032 0.681 ⬍0.001 ⬍0.001 0.017 0.028 0.395 0.072
* Due to the small number of events, papillary and chromophobe RCC were classified as nonclear cell subtype for this analysis.
10-year cancer specific survival rates of 96% (94 to 98) and 91% (85 to 94). For patients who presented with metastatic disease, median survival time was 45 months (range 26 to 64) with corresponding 1 and 5-year survival rates of 76% (64 to 85) and 42% (28 to 56). Multivariate Cox regression analysis revealed that ECOG PS, T stage, presence of distant metastases and Fuhrman grade, but not tumor size, were independent prognostic factors of cancer specific survival (table 2). Recurrence-free and progression-free survival. A total of 52 patients (5%) had RCC recurrence after nephrectomy for localized RCC (fig. 3). Characteristics of these patients are shown in table 3. The 5 and 10-year recurrence-free survival rates were 93% (90 to 95) and 84% (77 to 89). Stratified by T stage the 5-year recurrence-free rates for pT1aN0M0, pT3aN0M0 and pT3bN0M0 were 96% (93 to 97), 69% (52 to 81) and 58% (26 to 80), respectively (fig. 3, A). Multivariate analysis showed that ECOG PS, T stage and Fuhrman grade, but not tumor size independently impacted recurrence-free survival (table 2). In patients with metastatic disease median progressionfree survival was 28 months (range 6 to 50). The 1 and 5-year progression-free survival rates were 64% (51 to 74) and 39% (26 to 53) (fig. 3, B). This progression-free survival
SMALL SOLID RENAL TUMORS
1723
FIG. 2. Kaplan-Meier survival estimates for cancer specific survival, and 95% CI for patients with localized (N0M0) and metastatic (N⫹M0 or M1) RCC. Numbers of patients at risk are indicated.
time was significantly longer than for patients with metastatic disease and a primary tumor more than 4 cm in size (median progression free survival 8 months, p ⬍0.0001). Prognostic models for small RCC. We compared the accuracy of 3 prognostic models to predict cancer specific survival and recurrence-free survival. The results are reported in table 4. In terms of cancer specific survival the UISS and the Karakiewicz nomogram reached a c-index of approximately 0.87. The Kattan nomogram was the best predictor of cancer specific survival (c-index 0.778) and recurrence-free survival (c-index 0.755) in patients with localized disease. DISCUSSION The use of routine abdominal imaging modalities has led to increasing detection of solid renal tumors 4 cm or smaller. Therefore, urologists are increasingly confronted with their management and followup. We studied 1,208 small renal masses and found that 88% were RCC and 12% were benign tumors. As others3 we demonstrated a relationship between tumor size and the risk of localized RCC, which increased by 34% by each cm increase in tumor size. Of the patients with RCC, 995 (93%) were localized, but 72 (7%) presented with metastatic disease. We could not show a significant relationship between tumor size and the risk of synchronous metastatic disease and could not identify a tumor size cutoff, under which the risk of concomitant metastatic disease is negligible. Similarly, in a study of 10,420 cases from the Surveillance, Epidemiology, and End Results database, no significant correlation was observed.10 In this study the risk of synchronous metastatic disease in the tumor size ranges less than 1, 1 to 2, 2 to 3 and 3 to 4 cm was 6.3%, 4.7%, 5.2% and 7.2%, respectively. In contrast,
Kunkle et al11 showed in a cohort of 360 patients that the incidence of metastatic disease increases by 22% for each 1 cm increase in tumor size, but patients with all tumor sizes and not only those with tumors 4 cm or smaller were included in their analysis. In patients with RCC N0M0 disease there was a 7% chance of metachronous metastatic disease at 5 years and a 16% chance at 10 years after nephrectomy. However, the majority of localized small RCCs were cured by extirpative surgery. Minimally invasive therapies such as radio frequency ablation and cryotherapy are alternative approaches for small renal masses. Although early oncological results are promising,12,13 long-term data are required. Finally, active surveillance may be offered to patients who are not suitable candidates for surgery or minimally invasive therapies, or are not willing to accept the risks of these approaches. In a recent meta-analysis the mean growth rate of enhancing renal masses was 0.28 cm per year and the progression rate to metastatic disease was 1%.14 However, the growth rate of small renal tumors under active surveillance is variable and patient selection criteria need to be improved. Taken together, our data should serve as a caution. Patients considering these still experimental therapies should be aware of the small but not insignificant risk of metastatic disease and cancer associated death. In this regard our data clearly showed that tumor size alone does not sufficiently distinguish RCC with benign from RCC with aggressive behavior or independently predict recurrencefree and cancer specific survival, and that there is no safe cutoff under which the risk of metastases is negligible. Outcome data were generally favorable and in agreement with previous reports.7,8,15 Importantly the majority of patients who died of RCC (54%) presented with concomitant
1724
SMALL SOLID RENAL TUMORS
FIG. 3. Kaplan-Meier survival estimates for (A) recurrence-free survival (N0M0 RCC) and (B) progression-free survival (metastatic disease) and 95% confidence intervals. Numbers of patients at risk are indicated.
synchronous metastatic disease and only 3% of those presenting with localized disease died of RCC. Survival was even better than expected for patients with metastatic disease. Median overall, cancer specific and progression-free survival for this cohort was 36, 45 and 28 months, respectively, which was higher than for the population treated for metastatic RCC and primary tumors larger than 4 cm. The impact of T stage on survival of patients with metastatic disease is not well established and is usually not used for prognostic assessment,7 but our data shows that the size of the primary tumor may indeed have a prognostic impact.
This survival advantage may reflect a more indolent biology of small metastatic RCCs. The molecular pathways in these tumors should therefore be more fully characterized in future studies. Prediction of prognosis remains critical for tailoring surveillance and patient allocation into clinical trials. Traditionally the TNM staging system has been used for predicting patient outcome. However, the different stages comprise heterogeneous populations. Several institutions have recently developed prognostic models that may improve the predictive accuracy of TNM. The UISS combines ECOG PS,
SMALL SOLID RENAL TUMORS T, N and M stage, and Fuhrman grade to predict prognosis of all RCC stages.7 The UISS has been validated in a multicenter study and achieved an average c-index of 0.809 for localized and 0.651 for metastatic disease.16 Another prognostic model is the stage, size, grade, necrosis (SSIGN) for unilateral clear cell RCC, which reached a c-index of 0.849 in the original analysis.17 Kattan et al8 constructed a nomogram to predict recurrence-free survival in patients with clinically localized RCC combining clinical presentation, histological subtype, tumor size and T stage. The calculated c-index in this report was 0.74. An update of the Kattan nomogram was proposed by Sorbellini et al,18 and included tumor size, T stage, clinical presentation, grade, necrosis and vascular invasion (c-index 0.82). Karakiewicz et al9 recently published a nomogram containing T, N and M stage, grade, size and symptoms to predict survival of all RCC stages. The predictive accuracy after 5 years was 0.867. Our data showed that this nomogram and the UISS perform well in predicting survival of all RCC stages. For patients with localized disease, the Kattan nomogram may be better used to tailor postoperative approaches. However, it is expected that several molecular analyses will have a greater role in the future in predicting prognosis and in evaluating the biology of small renal tumors.19 Several limitations of our study need to be acknowledged and addressed. Based on the retrospective nature, the drawn conclusions have to be treated with reservation and need to be confirmed in prospective settings. We did not perform a pathological review by a single pathologist, although the pathological slides were examined by 1 pathologist at each institution. However, it is unlikely that our results were profoundly altered through this approach, because we performed an examination of basic pathological data, which are examined in a standardized way worldwide. Another consideration is that there was no standardization in preoperative imaging or on postoperative surveillance. However, there is currently no clear practice standard that has been established to guide such endeavors. Finally, no standardization existed regarding the performance of a lymph node dissection and, if performed, to what extent.
TABLE 3. Characteristics of the 52 patients with recurrence after surgery for localized (N0M0) RCC No. ECOG PS (%): 0 1 or Greater No. presentation (%): Incidental Local Systemic Mean tumor size (SD) No. cm tumor or (%): 0.1–1.0 1.1–2.0 2.1–3.0 3.1–4.0 No. RCC type (%): Clear cell Papillary No. T stage (%): T1a T3a T3b No. Fuhrman grade (%): G1 G2 G3–G4
36 16
TABLE 4. Comparison of the accuracy of prognostic models to predict survival of small RCC Localized RCC (N0M0) All Stages
Prognostic Model
C-index Ca Specific Survival (SD)
C-index Ca Specific Survival (SD)
C-index RecurrenceFree Survival (SD)
UISS7 Kattan et al8 Karakiewicz et al9
0.871 (0.029) Not available* 0.869 (0.030)
0.705 (0.029) 0.778 (0.047) 0.724 (0.030)
0.706 (0.020) 0.755 (0.020) 0.704 (0.026)
The Kattan nomogram approached the highest predictive accuracy for cancer specific and recurrence-free survival of localized disease (c-indices 0.778 and 0.755, respectively), while the UISS and the Karakiewicz nomogram showed high predictive accuracy for all RCC stages. * The Kattan nomogram predicts prognosis for patients with localized RCC.
Therefore, the actual incidence of patients with pN⫹ disease could be underestimated in the current study. However, recent data suggest that combination of pNx and pN0 cases for outcome modeling in RCC is appropriate.20 CONCLUSIONS Of patients presenting with small solid renal tumors more than 85% have RCC and 7% have synchronous metastatic disease. In those who present with localized disease there is a 7% chance of metachronous metastatic disease at 5 years. The majority of patients who die of RCC present with concomitant synchronous metastatic disease, but 3% of those presenting with localized disease will also die of RCC. The majority of localized small renal tumors can be cured with existing surgical approaches. However, patients considering experimental therapies such as ablation and surveillance should be aware of the small but not insignificant risk of synchronous and metachronous metastatic disease and cancer associated death. Unfortunately, tumor size alone is not sufficient to distinguish RCC with benign behavior from aggressive small RCC. Progression-free survival for patients with metastatic small RCC is better than expected. The UISS, the Karakiewicz nomogram and the Kattan nomogram are all reliable predictors of survival for patients with small RCC.
(69) (31)
38 (73) 12 (23) 2 (4) 3.2 (0.7)
1725
Abbreviations and Acronyms ECOG PS ⫽ Eastern Cooperative Oncology Group performance status RCC ⫽ renal cell carcinoma UISS ⫽ University of California Integrated Staging System
0 5 17 30
(0) (10) (33) (58)
50 2
(96) (4)
27 19 6
(52) (37) (12)
2.
5 35 12
(10) (67) (23)
3.
REFERENCES 1.
Jemal A, Siegel R, Ward E, Murray T, Xu J and Thun MJ: Cancer statistics, 2007. CA Cancer J Clin 2007; 57: 43. Chow WH, Devesa SS, Warren JL and Fraumeni JF: Rising incidence of renal cell cancer in the United States. JAMA 1999; 281: 1628. Frank I, Blute ML, Cheville JC, Lohse CM, Weaver AL and Zincke H: Solid renal tumors: an analysis of pathological features related to tumor size. J Urol 2003; 170: 2217.
1726 4.
5. 6.
7.
8.
9.
10.
11.
12.
13.
SMALL SOLID RENAL TUMORS
Lam JS, Leppert JT, Belldegrun AS and Figlin RA: Adjuvant therapy of renal cell carcinoma: patient selection and therapeutic options. BJU Int 2005; 96: 483. Cambio AJ and Evans CP: Management approaches to small renal tumours. BJU Int 2006; 97: 456. Patard JJ, Leray E, Rodriguez A, Rioux-Leclercq N, Guille F and Lobel B: Correlation between symptom graduation, tumor characteristics and survival in renal cell carcinoma. Eur Urol 2003; 44: 226. Zisman A, Pantuck AJ, Wieder J, Chao DH, Dorey F, Said JW 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 2002; 20: 4559. Kattan MW, Reuter V, Motzer RJ, Katz J and Russo P: A postoperative prognostic nomogram for renal cell carcinoma. J Urol 2001; 166: 63. Karakiewicz PI, Briganti A, Chun FK, Trinh QD, Perrotte P, Ficarra V et al: Multi-institutional validation of a new renal cancer-specific survival nomogram. J Clin Oncol 2007; 25: 1316. Nguyen MM, Stein RJ, Hafron JM, Anon M, Columbo J Jr, Gianduzzo TR et al: Tthe metastatic potential of small renal cancers is higher than previously believed. J Urol 2007; 77: 501. Kunkle DA, Crispen PL, Li T and Uzzo RG: Tumor size predicts synchronous metastatic renal cell carcinoma: implications for surveillance of small renal masses. J Urol 2007; 177: 1692. Gill IS, Remer EM, Hasan WA, Strzempkowski B, Spaliviero M, Steinberg AP et al: Renal cryoablation: outcome at 3 years. J Urol 2005; 173: 1903. Arzola J, Baughman SM, Hernandez J and Bishoff JT: Computed tomography-guided, resistance-based, percutaneous
14.
15.
16.
17.
18.
19.
20.
radiofrequency ablation of renal malignancies under conscious sedation at two years of follow-up. Urology 2006; 68: 983. Chawla SN, Crispen PL, Hanlon AL, Greenberg RE, Chen DY and Uzzo RG: The natural history of observed enhancing renal masses: meta-analysis and review of the world literature. J Urol 2006; 175: 425. Patard JJ, Shvarts O, Lam JS, Pantuck AJ, Kim HL, Ficarra V et al: Safety and efficacy of partial nephrectomy for all T1 tumors based on an international multicenter experience. J Urol 2004; 171: 2181. Patard JJ, Kim HL, Lam JS, Dorey FJ, Pantuck AJ, Zisman A et al: Use of the University of California Los Angeles integrated staging system to predict survival in renal cell carcinoma: an international multicenter study. J Clin Oncol 2004; 22: 3316. Frank I, Blute ML, Cheville JC, Lohse CM, Weaver AL and Zincke H: 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 2002; 168: 2395. Sorbellini M, Kattan MW, Snyder ME, Reuter V, Motzer R, Goetzl M et al: A postoperative prognostic nomogram predicting recurrence for patients with conventional clear cell renal cell carcinoma. J Urol 2005; 173: 48. Shvarts O, Seligson D, Lam J, Shi T, Horvath S, Figlin R et al: p53 is an independent predictor of tumor recurrence and progression after nephrectomy in patients with localized renal cell carcinoma. J Urol 2005; 173: 725. Ward JF, Blute ML, Cheville JC, Lohse CM, Weaver AL and Zincke H: The influence of pNx/pN0 grouping in a multivariate setting for outcome modeling in patients with clear cell renal cell carcinoma. J Urol 2002; 168: 56.
Purpose: We characterized the clinicopathological features and the prognosis of small solid renal tumors defined as tumors 4 cm or smaller. Materials and Methods: We identified 1,208 patients who were treated with nephrectomy at 5 international academic centers for small solid renal tumors. Clinicopathological parameters and outcome data were collected for each patient and analyzed. Results: Of the tumors 88% were renal cell carcinoma and 12% were benign. Of those with renal cell carcinoma 995 (93%) were localized (N0M0) and 72 (7%) presented with metastatic disease. Tumor size did not predict synchronous metastatic disease. The incidence of metastatic disease in the tumor size ranges 0.1 to 1.0, 1.1 to 2.0, 2.1 to 3.0 and 3.1 to 4.0 cm was 7%, 6%, 5% and 8%, respectively (p ⫽ 0.322). Survival rates were excellent. The majority of patients who died of renal cell carcinoma (54%) presented with synchronous metastatic disease, but 3% of patients with localized disease also died of renal cell carcinoma. In patients with localized disease there was a 7% chance of recurrence post nephrectomy at 5 years. Progression-free survival (28 months) was better than for patients with metastatic disease having a primary tumor greater than 4 cm (8 months). Tumor size was not retained as an independent prognostic factor of survival in multivariate analyses. The University of California Integrated Staging System and the Karakiewicz nomogram were the best predictors of cancer specific survival for all renal cell carcinoma stages (c-index 0.87). Conclusions: More than 85% of small solid renal tumors are renal cell carcinoma. The majority of localized small renal tumors can be cured with existing surgical approaches. However, there is a small but not insignificant risk of synchronous and metachronous metastatic disease and cancer associated death. Patients considering experimental therapies such as ablation and surveillance should be aware of this. Tumor size alone is not sufficient to distinguish renal cell carcinoma with benign behavior from aggressive small renal cell carcinoma. Survival of patients with small metastatic renal cell carcinoma is better then expected. The biology of these unique tumors should be further studied. Key Words: kidney neoplasms, prognosis, nomograms
enal cell carcinoma is the third most common urological malignancy and accounts for approximately 3% of all cancers in adults. It has been estimated for 2007 that about 51,000 people will be diagnosed with RCC in the United States, and approximately 13,000 will die of the disease.1 The incidence of RCC has steadily increased during the last 3 decades, mainly due to the widespread use of routine abdominal imaging and the detection of incidental, small solid renal tumors (4 cm or smaller) thereby.2 Fre-
R
Submitted for publication August 28, 2007. Study received institutional review board approval. * Correspondence and requests for reprints: Department of Urology, David Geffen School of Medicine at UCLA, 10833 Le Conte Ave., Room B7-298A CHS, Los Angeles, California 90025-1738 (telephone: 310-206-2436; FAX: 310-206-408; e-mail: apantuck@mednet. ucla.edu).
See Editorial on page 1657. For another article on a related topic see pages 2010 and 2057.
0022-5347/08/1795-1719/0 THE JOURNAL OF UROLOGY® Copyright © 2008 by AMERICAN UROLOGICAL ASSOCIATION
quently these tumors are benign or RCC with favorable pathological characteristics and prognosis.3 However, it is known that a subset of small renal tumors exhibit aggressive properties including synchronous or metachronous metastatic spread. Patients with synchronous metastatic disease will be treated with surgery followed by targeted or immune based therapy. However, the prognosis of patients with small metastatic RCC is yet to be specifically determined. For patients who are treated for localized disease (N0M0), prediction of prognosis is critical to provide tailored surveillance and patient allocation into adjuvant clinical trials.4 In addition, a better understanding of the biology of small, localized RCC would allow easier patient
Editor’s Note: This article is the second of 5 published in this issue for which category 1 CME credits can be earned. Instructions for obtaining credits are given with the questions on pages 2070 and 2071.
1719
Vol. 179, 1719-1726, May 2008 Printed in U.S.A. DOI:10.1016/j.juro.2008.01.018
1720
SMALL SOLID RENAL TUMORS
counseling for new nonsurgical approaches such as radio frequency ablation, cryotherapy and active surveillance which have been evolved to considerable alternative options for a subset of patients, although surgical extirpation remains the standard of care.5 We report on 1,208 patients who presented with a small solid renal tumor at 5 international institutions. Because all patients underwent surgical extirpation, this series allows the definition of the actual incidence of renal tumor types and pathological staging. Analyses were performed 1) to define the incidence of benign tumors, localized RCC, metastatic RCC, clinical and pathological characteristics, and prognosis, 2) to characterize the malignant potential of small renal tumors, defined as the probability of metachronous metastatic disease after undergoing surgery for localized disease, and 3) to evaluate the performance of prognostic models to predict survival of small RCC. PATIENTS AND METHODS Patient Population We studied retrospectively 1,208 patients, who were diagnosed with unilateral, sporadic, small solid renal tumors from 1989 to 2006 at 5 international academic urological centers in Créteil, France; Rennes, France; Magdeburg, Germany; Foggia, Italy; and Los Angeles, California. A small
solid renal tumor was defined as a solid renal tumor 4 cm or smaller. The study was approved by the institutional review board at each institution. Patient charts were reviewed and clinicopathological data recorded. The data that was abstracted included age, gender, symptoms,6 ECOG PS, T, N and M stage according to 2002 definitions, Fuhrman grade, and histological type. Pathological examination was performed by 1 anatomical pathologist at each institution. The type of renal tumor was defined according to WHO criteria. For practical purposes, we distinguished between benign and malignant renal tumors. Benign tumors were further stratified into oncocytoma, angiomyolipoma, atypical cyst and other. Malignant renal tumors were all RCC in this series and were divided into clear cell, papillary, chromophobe, collecting duct and unclassified subtype. All patients underwent open surgical extirpation. A radical nephrectomy was performed in 834 patients (69%), while 374 (31%) underwent nephron-sparing surgery. Surgical margins were negative in all patients. No adjuvant therapy was administered postoperatively. Patients who presented with metastatic disease underwent immunotherapy following nephrectomy. Postoperative surveillance was performed according to guidelines established at each institution. Median followup was 3.1 years (range 0.1 to 18.0).
TABLE 1. Patient and tumor characteristics
No. pts Mean pt age (SD) No. gender (%): Female Male No. ECOG PS (%): 0 1 or Greater No. presentation (%): Incidental Local Systemic Mean tumor size (SD) No. cm tumor (%): 0.1–1.0 1.1–2.0 2.1–3.0 3.1–4.0 No. benign tumor type (%): Oncocytoma Angiomyolipoma Complex cyst Other No. RCC type (%): Clear cell Papillary Chromophobe Collecting duct Unclassified No. T stage (%): T1a T3a T3b No. N/M stage (%): Nx or N0/M0 N⫹/M0 N⫹/M1 N0/M1 No. Fuhrman grade (%): G1 G2 G3–G4
Benign Tumor
Localized RCC
Metastatic RCC
141 60.3 (14.7)
995 61.3 (11.9)
72 60.6 (11.3)
Totals 61.2
(12.3)
65 76
(46) (54)
380 615
(38) (62)
21 51
(29) (71)
466 742
(39) (61)
127 14
(90) (10)
806 189
(81) (19)
33 39
(46) (54)
966 242
(80) (20)
118 (84) 21 (15) 2 (1) 2.7 (0.8) 5 32 65 39
(3) (23) (46) (28)
75 33 16 17
(53) (23) (11) (12)
782 (79) 201 (20) 12 (1) 2.9 (0.8)
31 (43) 34 (47) 7 (10) 3.1 (0.9)
931 256 21 2.9
(77) (21) (2) (0.9)
26 177 372 420
2 12 20 38
33 221 457 497
(3) (18) (38) (41)
(3) (18) (37) (42) —
(3) (17) (28) (53) —
—
—
802 145 48 0 0
(81) (15) (5) (0) (0)
62 7 1 1 1
(86) (10) (1) (1) (1)
864 152 49 1 1
(81) (14) (5) (less than 1) (less than 1)
—
895 83 17
(90) (8) (2)
44 23 5
(61) (32) (7)
939 106 22
(88) (10) (2)
995 0 0 0
(100) (0) (0) (0)
0 8 8 56
(0) (11) (11) (78)
995 8 8 56
(93) (less than 1) (less than 1) (5)
281 554 160
(28) (56) (16)
7 36 29
(10) (50) (40)
288 590 189
(27) (55) (18)
—
—
SMALL SOLID RENAL TUMORS Statistical Analysis The statistical software R v2.4 (www.r-project.org) was used for all analyses. The chi-square and Student t tests were used to compare categorical and continuous data, respectively. The relationship between tumor size and histology (benign tumor vs RCC) as well as metastatic status (localized vs metastatic RCC) was further analyzed with logistic regression models. End points of this study included overall survival, cancer specific survival, recurrence-free survival and progressionfree survival, which were all calculated from the date of surgery to the date of death, death from RCC, disease recurrence in localized disease and disease progression in metastatic disease, respectively. The Kaplan-Meier method was used to estimate the survivor functions, which were compared with the log rank test. Confidence intervals for survival data were determined by the Greenwood method. Multivariate Cox proportional hazard models were performed to detect variables independently associated with survival. Prognostic assessment was performed according to the 3 prognostic models of the UISS,7 the Kattan nomogram8 and the Karakiewicz nomogram.9 The predictive accuracy of these models was measured by the concordance index (cindex). The c-index corresponds to the area under the receiver operating characteristics (ROC) curve adapted for survival data and ranges from 0.5 (no predictive accuracy) to 1.0 (perfect predictive accuracy). RESULTS Of the 1,208 consecutive surgical patients 742 were men (61%) and 466 were women (39%). Mean age at diagnosis was 61 years (median 62, range 17 to 89). An ECOG PS of 0
1721
was assigned to 80% of the patients. A total of 931 patients (77%) were completely asymptomatic at the time of diagnosis and had incidentally discovered tumors, 256 (21%) presented with local symptoms, and 21 (2%) had tumors associated with systemic symptoms. A total of 1,067 patients (88%) had RCC, of whom 995 (93%) presented without evidence of metastatic disease and 72 (7%) had lymph node and/or distant metastases. Distant metastatic disease was confirmed by biopsy in 56% of the patients. The most frequently assigned histological subtypes were clear cell (81% of RCCs, 72% of small renal tumors), papillary (14%, 13%), and chromophobe (5%, 4%). There were 141 patients (12%) who had benign tumors, most frequently oncocytoma (53% of benign tumors, 6% of small renal tumors), angiomyolipoma (23%, 3%), complex cysts (11%, 1%), and other entities (12%, 1%). Patient and tumor characteristics are summarized in table 1.
Clinicopathological Features Benign renal tumor vs RCC. Tumor size was statistically significant but clinically insignificantly smaller in benign tumors than RCC (mean ⫾ SD, 2.7 ⫾ 0.8 cm vs 2.9 ⫾ 0.9 cm, p ⫽ 0.004). The frequency of RCC in the tumor size ranges 0.1 to 1.0, 1.1 to 2.0 and 2.1 to 3.0 cm was similar (85%, 86% and 86%, p ⫽ 0.987) and increased significantly in tumors measuring 3.1 to 4.0 cm (92%, p ⫽ 0.001). Patients with benign tumors had significantly better ECOG PS (ECOG PS 0, 90% vs 79%, p ⫽ 0.001) and were more likely to present with incidental tumors (84% vs 76%, p ⫽ 0.047). Also, benign tumors were more likely to be diagnosed in women (46% vs 38%), although this difference did not quite reach statistical
FIG. 1. Kaplan-Meier survival estimates for overall survival and 95% CI. Patients with benign renal tumors had significantly better survival than patients with localized RCC (p ⫽ 0.008). Numbers of patients at risk are indicated.
1722
SMALL SOLID RENAL TUMORS
significance (p ⫽ 0.051). No difference was detected in age (60.3 ⫾ 14.7 years vs 61.3 ⫾ 11.9 years, p ⫽ 0.436). The same clinicopathological features were compared between patients with benign tumors and localized RCC (995). Patients with benign tumors had smaller tumor sizes (2.7 ⫾ 0.8 cm vs 2.9 ⫾ 0.8 cm, p ⫽ 0.003), corresponding to a 34% increase in risk of RCC by 1 cm increase in tumor size (risk ratio 1.34, 95% CI 1.10 –1.65). An ECOG PS of 0 was more frequent in benign tumors (90% vs 81%, p ⫽ 0.009), while no difference was detected in the frequency of incidental tumors (84% vs 79%, p ⫽ 0.163) or age (p ⫽ 0.415). Again, the percentage of women was higher in patients with benign lesions than in RCC (46% vs 38% p ⫽ 0.072). Localized (N0M0) vs metastatic RCC. A total of 995 patients with RCC (93%) had no evidence of metastatic disease at the time of diagnosis, while 72 (7%) had metastatic disease including 56 with N0M1, 8 N⫹M0 and 8 N⫹M1 disease. Factors associated with metastatic disease in patients with RCC were an ECOG PS 1 or greater (metastatic vs localized, 54% vs 19%, p ⬍0.001), presentation with symptoms (57% vs 21%, p ⬍0.001), higher T stage (incidence of metastatic disease, pT1a 5%, pT3a 22%, pT3b 23%, p ⬍0.001), higher grades (p ⬍0.001) and histological type (p ⬍0.001). The incidence of clear cell RCC was higher in patients with metastatic disease (86% vs 81%), while the frequency of papillary RCC (10% vs 15%) and chromophobe RCC (1% vs 5%) was lower. Tumor size did not differ between patients with localized and metastatic disease (2.9 ⫾ 0.8 cm vs 3.1 ⫾ 0.9 cm, p ⫽ 0.096, risk ratio 1.29 [0.96-1.73]). The incidence of metastatic disease in the tumor size ranges 0.1 to 1.0, 1.1 to 2.0, 2.1 to 3.0 and 3.1 to 4.0 cm was 7%, 6%, 5% and 8%, respectively (p ⫽ 0.322). Tumor size in patients with metastatic disease ranged from 0.6 cm to 4.0 cm. Prognosis At the time of analysis 189 deaths (16%) had occurred including 61 (32%) from RCC. Of the 61 deaths from RCC 33 (54%) occurred in patients with synchronous metastatic disease representing 46% of these patients, while only 28 patients with N0M0 disease died following recurrent RCC. These 28 patients accounted for only 3% of the patients who underwent surgery for localized RCC. Overall survival. Five and 10-year overall survival rates (95% CI) for the entire cohort were 83% (80 to 85) and 64% (57 to 69), respectively (fig. 1). Overall survival was significantly better for patients with benign tumors than for patients with localized RCC (5-year 92% [85 to 96] vs 85% [81 to 87], 10-year 85% [71 to 92] vs 63% [55 to 69], p ⫽ 0.008, fig. 1). This survival difference was also present when comparing survival of patients with benign tumors to those with pT1aN0M0 RCC (p ⫽ 0.0277). Median overall survival time for metastatic RCC was 36 months (range 20 to 52), and the 1 and 5-year overall survival rates were 74% (62 to 83) and 36% (23 to 49) (fig. 1). Factors independently associated with overall survival included age, gender, ECOG PS and type of histology (table 2). Cancer specific survival. For the entire RCC cohort the 5 and 10-year survival probabilities were 92% (90 to 94) and 87% (82 to 91) (fig. 2). Patients with localized RCC had 5 and
TABLE 2. Multivariate analysis for overall survival, cancer specific survival and recurrence-free survival Covariate Age Male ECOG PS 1 or greater Symptomatic presentation Tumor size Histology: Localized RCC Benign tumor Metastatic RCC ECOG PS 1 or greater Symptomatic presentation T stage: T1a T3a T3b N stage 1 or greater M stage 1 or greater Fuhrman grade: G1 G2 G3–4 Tumor size Histological subtype
HR
95% CI
p Value
Overall survival 1.03 1.57 1.91 1.19 1.14
1.02–1.05 1.15–2.15 1.36–2.68 0.85–1.66 0.97–1.37
⬍0.001 0.005 ⬍0.001 0.320 0.098
0.27–0.89 2.89–6.36
0.020 ⬍0.001
Ca specific survival 2.26 1.26–4.09 1.53 0.83–2.81
0.006 0.172
1 0.49 4.29
1 2.78 5.08 1.98 10.5
1.48–5.25 2.13–12.1 0.81–4.87 5.94–18.7
0.002 ⬍0.001 0.136 ⬍0.001
1 3.64 5.10 1.22 0.98
1.09–12.2 1.44–18.0 0.85–1.76 0.57–1.69
0.036 0.011 0.282 0.942
Recurrence-free survival ECOG PS 1 or greater 2.00 1.06–3.75 Symptomatic presentation 0.87 0.45–1.69 T stage: T1a 1 T3a 6.01 3.27–11.1 T3b 8.23 3.24–20.9 Fuhrman grade: G1 1 G2 3.19 1.23–8.24 G3–4 3.36 1.14–9.90 Tumor size 1.18 0.81–1.71 Histological subtype:* Clear cell 1 Nonclear cell 0.27 0.06–1.12
0.032 0.681 ⬍0.001 ⬍0.001 0.017 0.028 0.395 0.072
* Due to the small number of events, papillary and chromophobe RCC were classified as nonclear cell subtype for this analysis.
10-year cancer specific survival rates of 96% (94 to 98) and 91% (85 to 94). For patients who presented with metastatic disease, median survival time was 45 months (range 26 to 64) with corresponding 1 and 5-year survival rates of 76% (64 to 85) and 42% (28 to 56). Multivariate Cox regression analysis revealed that ECOG PS, T stage, presence of distant metastases and Fuhrman grade, but not tumor size, were independent prognostic factors of cancer specific survival (table 2). Recurrence-free and progression-free survival. A total of 52 patients (5%) had RCC recurrence after nephrectomy for localized RCC (fig. 3). Characteristics of these patients are shown in table 3. The 5 and 10-year recurrence-free survival rates were 93% (90 to 95) and 84% (77 to 89). Stratified by T stage the 5-year recurrence-free rates for pT1aN0M0, pT3aN0M0 and pT3bN0M0 were 96% (93 to 97), 69% (52 to 81) and 58% (26 to 80), respectively (fig. 3, A). Multivariate analysis showed that ECOG PS, T stage and Fuhrman grade, but not tumor size independently impacted recurrence-free survival (table 2). In patients with metastatic disease median progressionfree survival was 28 months (range 6 to 50). The 1 and 5-year progression-free survival rates were 64% (51 to 74) and 39% (26 to 53) (fig. 3, B). This progression-free survival
SMALL SOLID RENAL TUMORS
1723
FIG. 2. Kaplan-Meier survival estimates for cancer specific survival, and 95% CI for patients with localized (N0M0) and metastatic (N⫹M0 or M1) RCC. Numbers of patients at risk are indicated.
time was significantly longer than for patients with metastatic disease and a primary tumor more than 4 cm in size (median progression free survival 8 months, p ⬍0.0001). Prognostic models for small RCC. We compared the accuracy of 3 prognostic models to predict cancer specific survival and recurrence-free survival. The results are reported in table 4. In terms of cancer specific survival the UISS and the Karakiewicz nomogram reached a c-index of approximately 0.87. The Kattan nomogram was the best predictor of cancer specific survival (c-index 0.778) and recurrence-free survival (c-index 0.755) in patients with localized disease. DISCUSSION The use of routine abdominal imaging modalities has led to increasing detection of solid renal tumors 4 cm or smaller. Therefore, urologists are increasingly confronted with their management and followup. We studied 1,208 small renal masses and found that 88% were RCC and 12% were benign tumors. As others3 we demonstrated a relationship between tumor size and the risk of localized RCC, which increased by 34% by each cm increase in tumor size. Of the patients with RCC, 995 (93%) were localized, but 72 (7%) presented with metastatic disease. We could not show a significant relationship between tumor size and the risk of synchronous metastatic disease and could not identify a tumor size cutoff, under which the risk of concomitant metastatic disease is negligible. Similarly, in a study of 10,420 cases from the Surveillance, Epidemiology, and End Results database, no significant correlation was observed.10 In this study the risk of synchronous metastatic disease in the tumor size ranges less than 1, 1 to 2, 2 to 3 and 3 to 4 cm was 6.3%, 4.7%, 5.2% and 7.2%, respectively. In contrast,
Kunkle et al11 showed in a cohort of 360 patients that the incidence of metastatic disease increases by 22% for each 1 cm increase in tumor size, but patients with all tumor sizes and not only those with tumors 4 cm or smaller were included in their analysis. In patients with RCC N0M0 disease there was a 7% chance of metachronous metastatic disease at 5 years and a 16% chance at 10 years after nephrectomy. However, the majority of localized small RCCs were cured by extirpative surgery. Minimally invasive therapies such as radio frequency ablation and cryotherapy are alternative approaches for small renal masses. Although early oncological results are promising,12,13 long-term data are required. Finally, active surveillance may be offered to patients who are not suitable candidates for surgery or minimally invasive therapies, or are not willing to accept the risks of these approaches. In a recent meta-analysis the mean growth rate of enhancing renal masses was 0.28 cm per year and the progression rate to metastatic disease was 1%.14 However, the growth rate of small renal tumors under active surveillance is variable and patient selection criteria need to be improved. Taken together, our data should serve as a caution. Patients considering these still experimental therapies should be aware of the small but not insignificant risk of metastatic disease and cancer associated death. In this regard our data clearly showed that tumor size alone does not sufficiently distinguish RCC with benign from RCC with aggressive behavior or independently predict recurrencefree and cancer specific survival, and that there is no safe cutoff under which the risk of metastases is negligible. Outcome data were generally favorable and in agreement with previous reports.7,8,15 Importantly the majority of patients who died of RCC (54%) presented with concomitant
1724
SMALL SOLID RENAL TUMORS
FIG. 3. Kaplan-Meier survival estimates for (A) recurrence-free survival (N0M0 RCC) and (B) progression-free survival (metastatic disease) and 95% confidence intervals. Numbers of patients at risk are indicated.
synchronous metastatic disease and only 3% of those presenting with localized disease died of RCC. Survival was even better than expected for patients with metastatic disease. Median overall, cancer specific and progression-free survival for this cohort was 36, 45 and 28 months, respectively, which was higher than for the population treated for metastatic RCC and primary tumors larger than 4 cm. The impact of T stage on survival of patients with metastatic disease is not well established and is usually not used for prognostic assessment,7 but our data shows that the size of the primary tumor may indeed have a prognostic impact.
This survival advantage may reflect a more indolent biology of small metastatic RCCs. The molecular pathways in these tumors should therefore be more fully characterized in future studies. Prediction of prognosis remains critical for tailoring surveillance and patient allocation into clinical trials. Traditionally the TNM staging system has been used for predicting patient outcome. However, the different stages comprise heterogeneous populations. Several institutions have recently developed prognostic models that may improve the predictive accuracy of TNM. The UISS combines ECOG PS,
SMALL SOLID RENAL TUMORS T, N and M stage, and Fuhrman grade to predict prognosis of all RCC stages.7 The UISS has been validated in a multicenter study and achieved an average c-index of 0.809 for localized and 0.651 for metastatic disease.16 Another prognostic model is the stage, size, grade, necrosis (SSIGN) for unilateral clear cell RCC, which reached a c-index of 0.849 in the original analysis.17 Kattan et al8 constructed a nomogram to predict recurrence-free survival in patients with clinically localized RCC combining clinical presentation, histological subtype, tumor size and T stage. The calculated c-index in this report was 0.74. An update of the Kattan nomogram was proposed by Sorbellini et al,18 and included tumor size, T stage, clinical presentation, grade, necrosis and vascular invasion (c-index 0.82). Karakiewicz et al9 recently published a nomogram containing T, N and M stage, grade, size and symptoms to predict survival of all RCC stages. The predictive accuracy after 5 years was 0.867. Our data showed that this nomogram and the UISS perform well in predicting survival of all RCC stages. For patients with localized disease, the Kattan nomogram may be better used to tailor postoperative approaches. However, it is expected that several molecular analyses will have a greater role in the future in predicting prognosis and in evaluating the biology of small renal tumors.19 Several limitations of our study need to be acknowledged and addressed. Based on the retrospective nature, the drawn conclusions have to be treated with reservation and need to be confirmed in prospective settings. We did not perform a pathological review by a single pathologist, although the pathological slides were examined by 1 pathologist at each institution. However, it is unlikely that our results were profoundly altered through this approach, because we performed an examination of basic pathological data, which are examined in a standardized way worldwide. Another consideration is that there was no standardization in preoperative imaging or on postoperative surveillance. However, there is currently no clear practice standard that has been established to guide such endeavors. Finally, no standardization existed regarding the performance of a lymph node dissection and, if performed, to what extent.
TABLE 3. Characteristics of the 52 patients with recurrence after surgery for localized (N0M0) RCC No. ECOG PS (%): 0 1 or Greater No. presentation (%): Incidental Local Systemic Mean tumor size (SD) No. cm tumor or (%): 0.1–1.0 1.1–2.0 2.1–3.0 3.1–4.0 No. RCC type (%): Clear cell Papillary No. T stage (%): T1a T3a T3b No. Fuhrman grade (%): G1 G2 G3–G4
36 16
TABLE 4. Comparison of the accuracy of prognostic models to predict survival of small RCC Localized RCC (N0M0) All Stages
Prognostic Model
C-index Ca Specific Survival (SD)
C-index Ca Specific Survival (SD)
C-index RecurrenceFree Survival (SD)
UISS7 Kattan et al8 Karakiewicz et al9
0.871 (0.029) Not available* 0.869 (0.030)
0.705 (0.029) 0.778 (0.047) 0.724 (0.030)
0.706 (0.020) 0.755 (0.020) 0.704 (0.026)
The Kattan nomogram approached the highest predictive accuracy for cancer specific and recurrence-free survival of localized disease (c-indices 0.778 and 0.755, respectively), while the UISS and the Karakiewicz nomogram showed high predictive accuracy for all RCC stages. * The Kattan nomogram predicts prognosis for patients with localized RCC.
Therefore, the actual incidence of patients with pN⫹ disease could be underestimated in the current study. However, recent data suggest that combination of pNx and pN0 cases for outcome modeling in RCC is appropriate.20 CONCLUSIONS Of patients presenting with small solid renal tumors more than 85% have RCC and 7% have synchronous metastatic disease. In those who present with localized disease there is a 7% chance of metachronous metastatic disease at 5 years. The majority of patients who die of RCC present with concomitant synchronous metastatic disease, but 3% of those presenting with localized disease will also die of RCC. The majority of localized small renal tumors can be cured with existing surgical approaches. However, patients considering experimental therapies such as ablation and surveillance should be aware of the small but not insignificant risk of synchronous and metachronous metastatic disease and cancer associated death. Unfortunately, tumor size alone is not sufficient to distinguish RCC with benign behavior from aggressive small RCC. Progression-free survival for patients with metastatic small RCC is better than expected. The UISS, the Karakiewicz nomogram and the Kattan nomogram are all reliable predictors of survival for patients with small RCC.
(69) (31)
38 (73) 12 (23) 2 (4) 3.2 (0.7)
1725
Abbreviations and Acronyms ECOG PS ⫽ Eastern Cooperative Oncology Group performance status RCC ⫽ renal cell carcinoma UISS ⫽ University of California Integrated Staging System
0 5 17 30
(0) (10) (33) (58)
50 2
(96) (4)
27 19 6
(52) (37) (12)
2.
5 35 12
(10) (67) (23)
3.
REFERENCES 1.
Jemal A, Siegel R, Ward E, Murray T, Xu J and Thun MJ: Cancer statistics, 2007. CA Cancer J Clin 2007; 57: 43. Chow WH, Devesa SS, Warren JL and Fraumeni JF: Rising incidence of renal cell cancer in the United States. JAMA 1999; 281: 1628. Frank I, Blute ML, Cheville JC, Lohse CM, Weaver AL and Zincke H: Solid renal tumors: an analysis of pathological features related to tumor size. J Urol 2003; 170: 2217.
1726 4.
5. 6.
7.
8.
9.
10.
11.
12.
13.
SMALL SOLID RENAL TUMORS
Lam JS, Leppert JT, Belldegrun AS and Figlin RA: Adjuvant therapy of renal cell carcinoma: patient selection and therapeutic options. BJU Int 2005; 96: 483. Cambio AJ and Evans CP: Management approaches to small renal tumours. BJU Int 2006; 97: 456. Patard JJ, Leray E, Rodriguez A, Rioux-Leclercq N, Guille F and Lobel B: Correlation between symptom graduation, tumor characteristics and survival in renal cell carcinoma. Eur Urol 2003; 44: 226. Zisman A, Pantuck AJ, Wieder J, Chao DH, Dorey F, Said JW 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 2002; 20: 4559. Kattan MW, Reuter V, Motzer RJ, Katz J and Russo P: A postoperative prognostic nomogram for renal cell carcinoma. J Urol 2001; 166: 63. Karakiewicz PI, Briganti A, Chun FK, Trinh QD, Perrotte P, Ficarra V et al: Multi-institutional validation of a new renal cancer-specific survival nomogram. J Clin Oncol 2007; 25: 1316. Nguyen MM, Stein RJ, Hafron JM, Anon M, Columbo J Jr, Gianduzzo TR et al: Tthe metastatic potential of small renal cancers is higher than previously believed. J Urol 2007; 77: 501. Kunkle DA, Crispen PL, Li T and Uzzo RG: Tumor size predicts synchronous metastatic renal cell carcinoma: implications for surveillance of small renal masses. J Urol 2007; 177: 1692. Gill IS, Remer EM, Hasan WA, Strzempkowski B, Spaliviero M, Steinberg AP et al: Renal cryoablation: outcome at 3 years. J Urol 2005; 173: 1903. Arzola J, Baughman SM, Hernandez J and Bishoff JT: Computed tomography-guided, resistance-based, percutaneous
14.
15.
16.
17.
18.
19.
20.
radiofrequency ablation of renal malignancies under conscious sedation at two years of follow-up. Urology 2006; 68: 983. Chawla SN, Crispen PL, Hanlon AL, Greenberg RE, Chen DY and Uzzo RG: The natural history of observed enhancing renal masses: meta-analysis and review of the world literature. J Urol 2006; 175: 425. Patard JJ, Shvarts O, Lam JS, Pantuck AJ, Kim HL, Ficarra V et al: Safety and efficacy of partial nephrectomy for all T1 tumors based on an international multicenter experience. J Urol 2004; 171: 2181. Patard JJ, Kim HL, Lam JS, Dorey FJ, Pantuck AJ, Zisman A et al: Use of the University of California Los Angeles integrated staging system to predict survival in renal cell carcinoma: an international multicenter study. J Clin Oncol 2004; 22: 3316. Frank I, Blute ML, Cheville JC, Lohse CM, Weaver AL and Zincke H: 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 2002; 168: 2395. Sorbellini M, Kattan MW, Snyder ME, Reuter V, Motzer R, Goetzl M et al: A postoperative prognostic nomogram predicting recurrence for patients with conventional clear cell renal cell carcinoma. J Urol 2005; 173: 48. Shvarts O, Seligson D, Lam J, Shi T, Horvath S, Figlin R et al: p53 is an independent predictor of tumor recurrence and progression after nephrectomy in patients with localized renal cell carcinoma. J Urol 2005; 173: 725. Ward JF, Blute ML, Cheville JC, Lohse CM, Weaver AL and Zincke H: The influence of pNx/pN0 grouping in a multivariate setting for outcome modeling in patients with clear cell renal cell carcinoma. J Urol 2002; 168: 56.