Original Study

Characteristics of Long-Term andShort-Term Survivors of Metastatic RenalCell Carcinoma Treated With TargetedTherapies: Results From the International

mRCC Database ConsortiumAndré P. Fay,1 Wanling L. Xie,1 Jae-Lyn Lee,2 Lauren C. Harshman,1,3

Georg A. Bjarnason,4 Jennifer J. Knox,5 Scott Ernst,6 Lori Wood,7

Ulka N. Vaishamayan,8 Takeshi Yuasa,9 Min-Han Tan,10 Sun-Young Rha,11

Frede Donskov,12 Neeraj Agarwal,13 Christian K. Kollmannsberger,14

Scott A. North,15 Brian I. Rini,16 Toni K. Choueiri,1 Daniel Y.C. Heng17

AbstractPatients with mRCC may have variable clinical courses when treated with targeted therapy. The two extremesof the survival spectrum need to be characterized. Analyzing data from a large database (International Met-astatic Renal Cell Carcinoma Database Consortium - IMDC) we found that baseline prognostic criteria andabsence of PD after first- and second-line targeted therapy may discriminate long-term survival.Background: Targeted therapies improve survival in metastatic renal cell carcinoma (mRCC). However, survivalpatterns can be divergent, and patients at the 2 extremes of the survival spectrum need to be characterized. Patientsand Methods: Data from 2161 patients included in the International mRCC Database Consortium (IMDC) wereanalyzed. We identified patients on the basis of their duration of survival. Long-term survival (LTS) was defined asoverall survival (OS) of � 4 years, and short-term survival (STS) was defined as OS of � 6 months from the start oftargeted therapy. Baseline characteristics, including demographic, clinicopathologic, and laboratory data, werecompared between LTS and STS. Treatment response by the RECIST criteria was summarized for the 2 survivalgroups. Results: A total of 152 patients experienced LTS and 218 experienced STS. Adverse clinical and laboratoryprognostic factors previously described in the IMDC prognostic model were significantly more frequent in theSTS group (P < .0001). In the LTS group, 138 patients (91%) had nonprogressive disease (non-PD) as best responseto first-line targeted therapy, and 56 (60%) of 94 patients who received second-line therapy had non-PD. In theSTS group, only 51 patients (23%) had non-PD on first-line therapy. None of 21 the patients who received second-line therapy had non-PD as best response. In LTS, the median duration of therapy was 23.6 months (range 0.4 to

The last 2 authors contributed equally to this article, and both should be consideredsenior author.

1Dana Farber Cancer Institute, Brigham and Women’s Hospital, Harvard MedicalSchool, Boston, MA2University of Ulsan College of Medicine, Asan, South Korea3Stanford Cancer Institute, Stanford, CA4Sunnybrook Odette Cancer Center, Toronto, Ontario, Canada5Princess Margaret Hospital, Toronto, Canada6London Health Sciences Center, London, ON, Canada7Queen Elizabeth II Health Sciences Center, Halifax, NS, Canada8Karmanos Cancer Center, Detroit, MI9Cancer Institute Hospital, Japanese Foundation for Cancer Research, Tokyo, Japan10National Cancer Center, Singapore

11Yonsei Cancer Center, Yonsei University College of Medicine, Seoul, South Korea12Aarhus University Hospital, Aarhus, Denmark13University of Utah Huntsman Cancer Institute, Salt Lake City, UT14British Columbia Cancer Agency, Vancouver, Canada15Cross Cancer Institute, Edmonton, AB, Canada16Cleveland Clinic Taussig Cancer Institute, Cleveland, OH17Tom Baker Cancer Center and University of Calgary, Calgary, Canada

Submitted: Jun 13, 2014; Revised: Sep 19, 2014; Accepted: Sep 25, 2014

Address for correspondence: Daniel Y. C. Heng, University of Calgary; Tom BakerCancer Center 1331 29th Street NW, Calgary, AB T2N 4N2, CanadaFax: (403) 283-1651; e-mail contact: daniel.heng@albertahealthservices.ca

1558-7673/$ - see frontmatter ª 2014 Elsevier Inc. All rights reserved.http://dx.doi.org/10.1016/j.clgc.2014.09.003 Clinical Genitourinary Cancer Month 2014 - 1

Long- and Short-Term Survivors in mRCC

2 - Cli

81.8þ months) for first-line therapy and 11.5 months (range 0.6 to 45.7 months) for second-line therapy, compared to2.0 and 0.8 months for the STS group, respectively. Conclusion: Baseline prognostic criteria and absence of PD afterfirst and second-line targeted therapy may characterize long-term survival.

Clinical Genitourinary Cancer, Vol. -, No. -, --- ª 2014 Elsevier Inc. All rights reserved.Keywords: Long-term survival, Overall survival, Primary refractory disease, Prognostic factors, Renal cell carcinoma,

Targeted therapies

Figure 1 Study Design

IntroductionThe incidence and mortality of renal cell carcinoma (RCC) ap-

pears to be rising in the United States, with approximately 65,000new cases of RCC diagnosed every year.1,2 Complete surgicalresection remains the only curative treatment for localized disease.3

However, about 30% of patients will develop metastatic diseaseafter nephrectomy, which is generally associated with poor long-term survival.4

In the last decade, major advances have contributed to theunderstanding of the biology underlying RCC. The knowledgethat this disease is driven by dysregulation of angiogenesis pro-vided a rationale for the development of new targeted agents.5

Drugs targeting vascular endothelial growth factor (VEGF) ormammalian target of rapamycin (mTOR) produce objectiveresponse or disease stabilization in the majority of patients,resulting in a median survival approaching 30 months with met-astatic RCC (mRCC).6-10 However, variable outcomes in terms ofduration of survival and degree of response to targeted therapieshave been reported.

Several clinical and laboratory factors have been identified asimportant prognostic factors and have been integrated into prog-nostic models to estimate survival in the metastatic setting.11,12

Patients are stratified into poor-, intermediate-, and favorable-riskcategories.11 Recently, conditional survival defined as the proba-bility of surviving an additional period of time after the patient hasalready survived a specific time period was shown to provide rele-vant individual prognostic information.13

However, patients with similar survival estimates at baseline mayhave very divergent clinical outcomes.14 For this reason, charac-terization of patients at the 2 extremes of the survival spectrum mayprovide insights into the biology of RCC. We analyzed data fromour large international multi-institutional database, the Interna-tional Metastatic Renal Cell Carcinoma Database Consortium(IMDC), to provide information about treatment response andclinical features of patients with overall survival (OS) of greater than4 years or less than 6 months.

Patients and MethodsThe IMDC database includes 20 centers from North America

(Canada, United States), Europe (Denmark, Greece), and Asia(Singapore, Japan, South Korea). As of January 2012, the databasecontained data of 2161 patients with mRCC who had receivedtargeted therapy. Patients who survived at least 4 years and patientswho survived 6 months or less over the same period (2004-2007)were identified. Long-term survival (LTS) was defined as an OSof � 4 years after initiating therapy, and short-term survival (STS)was defined as an OS of � 6 months after initiating therapy

nical Genitourinary Cancer Month 2014

(Figure 1). Survival cutoffs were defined arbitrarily based on medianOS reported in clinical trials with targeted agents.3 Baseline patientcharacteristics, including demographic, clinicopathologic, and lab-oratory data, as well as outcomes data were collected from medicalchart reviews and electronic records, as previously outlined.11

Laboratory values were standardized against institutional upperlimit of normal (ULN) and lower limit of normal (LLN) valueswhen appropriate. Response to targeted therapy was assessed byResponse Evaluation Criteria in Solid Tumors (RECIST v.1.0).15

Uniform data collection templates were used to ensure consistentdata. Most of the patients had been treated with standardly availableagents, but a subset may have been participants in clinical trials.All patients were collected in consecutive series to avoid selectionbias. All IMDC centers obtained local institutional review boardapprovals before data acquisition.

The primary objective of this study was to evaluate patient andtumor characteristics as well as tumor response to the targetedtherapy according to the 2 divergent survival outcomes (LTS vs.STS). OS was defined as the time period between targeted therapyinitiation and the date of death, or censored on the day of the lastfollow-up visit. Duration of targeted therapy was defined as the timeperiod between treatment from targeted therapy initiation to pro-gression, drug cessation, death, or censored at the last follow-upvisit. Radiographic criteria with RECIST v.1.0 grouped patientsinto 4 categories according to their best response during treatment:complete response (CR), partial response (PR), stable disease (SD),and progressive disease (PD). Non-PD was defined as CR, PR, orSD and clinically implies the absence of primary refractory disease totargeted therapy. Patient and tumor characteristics outlined inTable 1 were compared between LTS and STS by Fisher’s exact testfor categorical variables. The percentage of patients who experienced

Table 1 Patient and Disease Characteristics at Initiationof Targeted Therapy

Characteristic

Short-TermSurvival

(£6 months,n [ 218)

Long-TermSurvival(‡4 years,n [ 152) P

Age NS

<60 years 97 (44%) 69 (45%)

�60 years 121 (56%) 83 (55%)

Gender NS

Female 61 (28%) 30 (20%)

Male 157 (72%) 122 (80%)

No. of Metastases .019

1 36 (17%) 41 (27%)

>1 181 (83%) 111 (73%)

Brain Metastasis NS

No 196 (90%) 143 (94%)

Yes 21 (10%) 9 (6%)

Histology NS

Clear cell histology 171 (85%) 127 (90%)

Noneclear-cell histology 31 (15%) 14 (10%)

Sarcomatoid Pathology .0001

No 163 (84%) 129 (97%)

Yes 31 (16%) 4 (3%)

Previous Nephrectomy <.0001

No 76 (35%) 8 (5%)

Yes 142 (65%) 144 (95%)

Previous Immunotherapy NS

No 138 (63%) 96 (63%)

Yes 80 (37%) 56 (37%)

LDH >1.5 ULN .03

Yes 50 (23%) 20 (13%)

No 168 (77%) 132 (87%)

IMDC Prognostic Risk Factor

KPS <.0001

�80 99 (47%) 136 (93%)

<80 111 (53%) 11 (7%)

Diagnosis to TKI Therapy<1 Year

<.0001

No 85 (39%) 110 (72%)

Yes 131 (61%) 42 (28%)

Low Hemoglobin Level (<LLN) <.0001

No 41 (19%) 96 (65%)

Yes 170 (81%) 52 (35%)

Hypercalcemia <.0001

No 155 (75%) 142 (97%)

Yes 53 (25%) 4 (3%)

Neutrophilia (>ULN) <.0001

No 137 (66%) 142 (98%)

Yes 70 (34%) 3 (2%)

Thrombophilia (>ULN) <.0001

No 123 (58%) 142 (96%)

Yes 87 (42%) 6 (4%)

Table 1 Continued

Characteristic

Short-TermSurvival

(£6 months,n [ 218)

Long-TermSurvival(‡4 years,n [ 152) P

IMDC Risk Group <.0001

Favorable, 0 RF 5 (2%) 64 (44%)

Intermediate, 1 RF 21 (10%) 55 (38%)

Intermediate, 2 RFs 54 (25%) 20 (14%)

Poor, 3-6 RFs 130 (62%) 5 (4%)

Abbreviations: IMDC ¼ International mRCC Database Consortium; KPS ¼ Karnofsky perfor-mance status; LDH ¼ lactate dehydrogenase; LLN ¼ lower limit of normal; NS ¼ notsignificant; RF ¼ risk factor; TKI ¼ tyrosine kinase inhibitor; ULN ¼ upper limit of normal.

André P. Fay et al

non-PD was summarized by the 2 survival extremes as well as by theIMDC prognostic risk groups. Duration of therapy was estimatedby the Kaplan-Meier methodology.

All statistical computations were performed by SAS v.9.2 (SASInstitute, Cary, NC, USA), and a P value (2-sided) of < .05 wasconsidered statistically significant.

ResultsPatient and Tumor Characteristics

Patient characteristics at initiation of targeted therapy are out-lined in Table 1. The study cohorts included a total of 370 patientsout of 2161 patients registered in the IMDC database. There were152 patients who survived at least 4 years (LTS) and 218 patientswho survived < 6 months (STS) since therapy initiation. All first-line targeted therapies were administered between 2004 and2007. The majority of patients received a first-line VEGF targetedtherapy of sunitinib (72%), sorafenib (22%), and bevacizumab(5%) in the STS group; the same distribution was 55%, 34%, and11% in the LTS group. Thirty-seven percent of patients in bothgroups had been treated with previous immunotherapy.

IMDC Prognostic Risk Criteria in LTS and STS GroupsKarnofsky performance status of < 80% was observed in 53% of

patients in the STS group and only 7% in the LTS group. STS weremore likely to have anemia (81% vs. 35%) and time from diagnosisto targeted therapy of < 1 year (61% vs. 28%). Presence of otherIMDC laboratory risk factors, including neutrophilia, thrombocy-tosis, and hypercalcemia, were extremely low (less than 5%) in theLTS group. Overall, the presence of the IMDC adverse prognosticrisk factors was significantly associated with STS (P < .0001,Table 1).

By the IMDC prognostic risk group criteria, the majority ofpatients (82%) in the LTS group had favorable risk (44%) or in-termediate risk with 1 risk factor (38%). By contrast, most of pa-tients (87%) in the STS group had poor (3 to 6 risk factors, 62%) orintermediate risk with 2 risk factors (25%).

Response to Targeted Therapy in LTS and STS GroupsIn the LTS group, the majority of patients (91%, 95% confi-

dence interval 85-95) had non-PD as best response to first-linetargeted therapy, including CR (4%), PR (34%), and SD (53%).

Clinical Genitourinary Cancer Month 2014 - 3

Table 2 Nonprogressive Disease by Prognostic Group

Characteristic

Short-Term Survival (£6 months, n [ 218) Long-Term Survival (‡48 months, n [ 152)

First-Line Therapy Second-Line Therapy First-Line Therapy Second-Line Therapy

All 51/218 (23%) 0/21 138/152 (91%) 56/94 (60%)

IMDC Risk Groupa

Favorable 3/5 (60%) 0/1 59/64 (92%) 21/43 (49%)

Intermediate 23/75 (31%) 0/9 68/75 (91%) 32/45 (71%)

Poor 25/130 (19%) 0/11 5/5 (100%) 2/3 (67%)

aExcludes 16 patients without prognostic data.

Long- and Short-Term Survivors in mRCC

4 - Cli

In addition, 59 (92%) of 64 patients who had favorable prognosticrisk score, 68 (91%) of 75 patients with intermediate risk score, andall the 5 patients in the poor prognostic risk score category had non-PD as best response to first-line targeted therapy.

Ninety-four (62%) of 152 LTS received second-line sequentialtherapy. In this setting, 21 (49%) of 43 patients with favorableprognostic risk score, 32 (71%) of 45 patients with intermediate riskscore, and 2 (67%) of 3 with poor risk score had non-PD as bestresponse (Table 2). Overall, 56 (60%) of all 94 patients had non-PD on second-line therapy.

In the STS group, PD was the best response to first-line targetedtherapy in 93 patients (43%). Only 42 patients (19%) had stabledisease, and in 74 (34%) of 218 patients, the best response was notavailable in the database (Table 1). Three (60%) of 5 patients whohad favorable prognostic risk scores, 23 (31%) of 75 patients whohad intermediate risk scores, and 25 (19%) of 130 who had poorprognostic risk scores had non-PD as best response to first-linetargeted therapy. Overall, 51 patients (23%) had non-PD on first-line therapy.

Only 21 of 218 patients in the STS group received second-linetargeted therapy; all had PD as best response, independent of riskscore category (Table 2).

Table 3 Response and Duration of First- and Second-Line Therapy

Characteristic

First-Line Therapy

n (%)Median Duration ofTherapy (months)

Short-Term Survival

CR þ PR 9 (4%) 3.7

SD 42 (19%) 2.8

PD 93 (43%) 1.6

Unknown 74 (34%) 1.7

All 218 2.0

Long-Term Survival

CR þ PR 57 (38%) 35.9

SD 81 (53%) 17.0

PD 5 (3%) 2.9

Unknown 9 (6%) 28.5

All 152 23.6

Abbreviations: CR ¼ complete response; IMDC ¼ International mRCC Database Consortium; PD ¼

nical Genitourinary Cancer Month 2014

The median duration of receipt of first-line therapy was 23.6months (range 0.4 to 81.8þ months) for LTS (Table 3). In addi-tion, for patients who received second-line targeted therapy, themedian treatment duration was 11.5 months (range 0.6 to 45.7months). In the STS group, the median duration of first-line andsecond-line therapy was 2.0 (range 0.2 to 5.9 months) and 0.8months (range 0.2 to 4.1 months), respectively. The duration oftherapy according to response to first-line therapy is outlined inTable 3.

In the LTS group, 57 patients (38%) who had CR or PR as bestresponse to first-line targeted therapy had a median duration oftherapy of 35.9 months; 32 of these 57 patients also receivedsecond-line therapy for a median of 5.0 months. The medianduration of therapy for the 5 patients (3%) who had PD whilereceiving first-line therapy was 2.9 months. All these patientsreceived second-line treatment, and if they had non-PD as bestresponse, the median duration of therapy was 20.1 months.

Among STS, 9 patients (4%) who had CR or PR to first-linetreatment achieved a median duration of receipt of therapy of 3.7months. In patients who had PD (n ¼ 93, 43%) as best response,the median duration of treatment was 1.6 months. As previouslyreported, response data from 74 patients (34%) were not available

Second-Line TherapyNon-PD While

Receiving Second-LineTherapy, n (%)n (%)

Median Duration ofTherapy (months)

— — —

4 — —

14 — —

3 — —

21 0.8 0

32 5.0 13 (41%)

53 12.7 36 (68%)

5 20.1 4 (80%)

4 7.6 3 (75%)

94 11.5 56 (60%)

progressive disease; PR ¼ partial response; SD ¼ stable disease.

André P. Fay et al

for analysis; the median duration of treatment in this subgroup was1.7 months.

DiscussionData from large randomized clinical trials and prognostic

models have provided survival estimates for patients with meta-static RCC.11,12 The Memorial Sloan-Kettering Cancer Center(MSKCC) system was developed based on patients previouslytreated with interferon a. The study identified 5 adverse prog-nostic factors associated with different clinical outcome: (1)an interval from diagnosis to treatment < 1 year; (2) Karnofskyperformance status < 80%; (3) serum lactate dehydrogenase> 1.5 times the ULN; (4) corrected serum calcium greater thanULN; and (5) serum hemoglobin less than the LLN.12 In the eraof targeted therapies, the IMDC has validated 4 of these 5 factorsand incorporated 2 new markers of poor prognosis: neutrophiland platelet counts greater than ULN.11

As expected, our data demonstrate that the presence of adverseclinical and laboratory prognostic factors previously described bythe IMDC were significantly more frequent in the STS group.These findings confirm the importance of the baseline prognosticcriteria to aid in the discrimination between long- and short-termsurvival. Intermediate risk score is defined as 1 or 2 adverse prog-nostic factors.11 Interestingly, 54 (72%) of 75 patients in the STSgroup who had intermediate risk prognostic scores had 2 risk fac-tors. On the other hand, the majority of subjects with LTS in thesame category (n ¼ 112; 74%) had only 1 risk factor. Thesefindings suggest that the presence of 1 additional adverse prognosticfeature may be related to distinct clinical outcomes.

As previously reported, our analysis showed that tumor histologywith sarcomatoid features, more than 1 site of metastasis, andabsence of nephrectomy were significantly more prevalent in STS.16

In addition, metastatic RCC with noneclear-cell histologies hasbeen associated with poor prognosis.17,18 In this series, although asmall number of patients presented this pathologic covariate, thepresence of this feature was not significantly associated with dura-tion of survival (Table 1).

A retrospective analysis of 1059 patients with mRCC treated withsunitinib on clinical trials demonstrated that patients who hadobjective response to this agent achieved significantly longer survival.The median OS was 48.8 months for those with disease thatresponded to therapy and 14.5 months for those whose disease didnot respond to therapy (P < .001). The survival duration was notassociated with time to tumor response, and those whose diseaseresponded to therapy also had significantly better performancestatus and more favorable MSKCC prognostic factors.19 It isimportant to note that this analysis was not designed to compareLTS and STS. Another study reported that tumor reduction within60 days of treatment initiation was associated with 74% decreasedrisk of death and early response was considered an independentpredictor of better OS (hazard ratio 0.26; P ¼ .031).20 In addition,Grünwald and colleagues21 presented at the 2013 European CancerCongress the results from a study that explored the impact of tumorshrinkage in the clinical outcome of 2749 patients with mRCC whowere treated as part of clinical trials. In this analysis, the tumorshrinkage was an independent predictor of OS. In our study, 57patients (38%) in the LTS group had CR or PR to first-line therapy

compared with 9 patients (4%) in the STS group. In addition, 81patients (53%) in the first group had stable disease compared with42 patients (19%) in the other. Over 90% of patients in the LTSachieved disease control (non-PD), independent of the prognosticrisk category. Interestingly, although rare, even patients with poorprognostic risk scores experienced LTS if their disease responded totargeted therapies. Alternatively, response to therapy does not ensurelong-term survival, as 9 subjects in the STS group with objectiveresponses had a median duration of therapy of only 3.7 months.

To our knowledge, this is the largest series characterizing patientsin the 2 extremes of survival spectrum. Among the 2161 patientsregistered in the IMDC database, we identified 152 patients (7%)who have survived more than 4 years. This subgroup of patientsexperienced higher response rates, longer treatment durations, andexposure to second-line agents compared to patients in the STSgroup, although we recognize that patients in the LTS group hadmore favorable baseline prognostic risk factors. A better character-ization of these 2 extreme phenotypes of survival may help inplanning clinical decisions. Predictions are clinically useful forthe following: (1) selecting therapy (eg, define best subsequenttherapy or good candidates for new therapeutic strategies, such asimmunotherapy); (2) prognostication; and (3) planning end-of-lifeclinical care. However, they should be used carefully; patientswith similar survival estimates may present different clinical out-comes. In addition, known and emerging pathologic and geneticbiomarkers in these 2 groups of patients might be informative.

Although we have evaluated a large database, our study has severallimitations. There is potential selection bias in any retrospectiveanalysis. Despite standard data collection templates and inclusion ofconsecutive patients from each center in the database, data weremissing for several patients, especially in the STS group. The shorttime of receipt of therapy for patients missing response data (1.7months) is possibly due to the fact that these patients had rapid clinicalprogression and were taken off therapy, sometimes without detailedimaging. Furthermore, patients included in this analysis were treatedwith different targeted agents (although only VEGF and mTOR in-hibitors); how it could distinctly impact clinical outcome needs toaddressed in future studies. Unfortunately, tumor specimens andblood samples were not available, and molecular characterization ofthese 2 groups was not an objective of this work. However, the role oftumor biology in the survival duration should be addressed in futurestudies. Finally, our analysis was descriptive in nature, and no formalcomparisonwasmade for treatment outcomes, given that we included2 extremes of survival spectrum by the study design. The strengths ofour study include the large number of patients and the adjustment forthe validated IMDC prognostic risk criteria.

In summary, baseline prognostic criteria used in the targetedtherapy era may predict 2 extreme survival patterns. Our analysissupports response to targeted therapies and the absence of PD asimportant elements related to long-term survival.

Clinical Practice Points

� Patients with metastatic RCC have variable clinical courses interms of survival and response to targeted therapy.

� Prognostic models have been widely used to predict patientoutcomes. However, patients at the two extremes of the survivalspectrum need to be characterized.

Clinical Genitourinary Cancer Month 2014 - 5

Long- and Short-Term Survivors in mRCC

6 - Cli

� In this analysis of patients from a large international database(IMDC), baseline prognostic criteria were able to discriminatebetween long- and short- term survivors.

� In addition, patients with poor prognostic factors may be able toachieve long-term survival if they could respond to targetedtherapy.

AcknowledgmentsThis work was conducted as part of the IMDC database. We

thank the DF/HCC Kidney Cancer SPORE, Trust Family, and theMichael Brigham Funds for Kidney Cancer Research (T.K.C.). Thiswork was supported in part by DF/HCC Kidney Cancer SPOREP50 CA101942-01.

DisclosureL.C.H.: Advisory boards at Aveo, Pfizer, Bristol-Myers Squibb;

Past research funding: BMS, Novartis. M.H.T. has filed for patentson molecular predictors of survival, drug response, and toxicity inpatients with RCC. U.N.V.: Consultant speaker and researchsupport from Novartis and speaker for Pfizer. J.J.K.: Researchsupport: Pfizer and Novartis. G.A.B.: Speaker for Pfizer, GSK, andNovartis; grant support from Pfizer. T.K.C.: Consultancy: Pfizer,Novartis; Advisory board: Pfizer, Novartis, Aveo, GlaxoSmithKline,Exelixis; Research: Pfizer; No Speakers bureau. C.K.: Advisory rolesat Pfizer, Novartis, and GlaxoSmithKline; Honoraria and researchfunding from Pfizer, Novartis, and GlaxoSmithKline. B.I.R.:Advisory role at Pfizer, GlaxoSmithKline, Aveo, Bayer, Onyx, andhas received research funding from GlaxoSmithKline and Pfizer.F.D.: Research grants from Novartis and GSK. L.W.: Research:Pfizer, Novartis, GSK; advisory boards, but no financial compen-sation, from Pfizer and Novartis. D.Y.C.H.: Consultancy: Bayer,Novartis, and Pfizer. All remaining authors have declared no con-flicts of interest.

References1. Sun M, Thuret R, Abdollah F, et al. Age-adjusted incidence, mortality, and sur-

vival rates of stage-specific renal cell carcinoma in North America: a trend analysis.Eur Urol 2011; 59:135-41.

nical Genitourinary Cancer Month 2014

2. SiegelR,Ma J,ZouZ, JemalA.Cancer statistics, 2014.CACancer JClin2014;64:9-29.3. Courtney KD, Choueiri TK. Optimizing recent advances in metastatic renal cell

carcinoma. Curr Oncol Rep 2009; 11:218-26.4. Janzen NK, Kim HL, Figlin RA, Belldegrun AS. Surveillance after radical or partial

nephrectomy for localized renal cell carcinoma and management of recurrentdisease. Urol Clin North Am 2003; 30:843-52.

5. Choueiri TK, Fay AP, Gagnon R, et al. The role of aberrant VHL/HIF pathwayelements in predicting clinical outcome to pazopanib therapy in patients withmetastatic clear-cell renal cell carcinoma. Clin Cancer Res 2013; 19:5218-26.

6. Escudier B, Bellmunt J, Negrier S, et al. Phase III trial of bevacizumab plusinterferon alfa-2a in patients with metastatic renal cell carcinoma (AVOREN): finalanalysis of overall survival. J Clin Oncol 2010; 28:2144-50.

7. Motzer RJ, Hutson TE, Tomczak P, et al. Overall survival and updated resultsfor sunitinib compared with interferon alfa in patients with metastatic renal cellcarcinoma. J Clin Oncol 2009; 27:3584-90.

8. Rini BI, Halabi S, Rosenberg JE, et al. Phase III trial of bevacizumab plus inter-feron alfa versus interferon alfa monotherapy in patients with metastatic renal cellcarcinoma: final results of CALGB 90206. J Clin Oncol 2010; 28:2137-43.

9. Sternberg CN, Davis ID, Mardiak J, et al. Pazopanib in locally advanced ormetastatic renal cell carcinoma: results of a randomized phase III trial. J Clin Oncol2010; 28:1061-8.

10. Motzer RJ, Hutson TE, Cella D, et al. Pazopanib versus sunitinib in metastaticrenal-cell carcinoma. N Engl J Med 2013; 369:722-31.

11. Heng DY, Xie W, Regan MM, et al. Prognostic factors for overall survival inpatients with metastatic renal cell carcinoma treated with vascular endothelialgrowth factor-targeted agents: results from a large, multicenter study. J Clin Oncol2009; 27:5794-9.

12. Motzer RJ, Bacik J, Murphy BA, Russo P, Mazumdar M. Interferon-alfa as acomparative treatment for clinical trials of new therapies against advanced renal cellcarcinoma. J Clin Oncol 2002b; 20:289-96.

13. Harshman LC, Xie W, Bjarnason GA, et al. Conditional survival of patients withmetastatic renal-cell carcinoma treated with VEGF-targeted therapy: a population-based study. Lancet Oncol 2012; 13:927-35.

14. Leibovich BC, Cheville JC, Lohse CM, et al. A scoring algorithm to predictsurvival for patients with metastatic clear cell renal cell carcinoma: a stratificationtool for prospective clinical trials. J Urol 2005; 174:1759-63.

15. Eisenhauer EA, Therasse P, Bogaerts J, et al. New response evaluation criteria insolid tumours: revised RECIST guideline (version 1.1). Eur J Cancer 2009; 45:228-47.

16. Golshayan AR, George S, Heng DY, et al. Metastatic sarcomatoid renal cell car-cinoma treated with vascular endothelial growth factor-targeted therapy. J ClinOncol 2009; 27:235-41.

17. Choueiri TK, Plantade A, Elson P, et al. Efficacy of sunitinib and sorafenib inmetastatic papillary and chromophobe renal cell carcinoma. J Clin Oncol 2008; 26:127-31.

18. Motzer RJ, Bacik J, Mariani T, Russo P, Mazumdar M, Reuter V. Treatmentoutcome and survival associated with metastatic renal cell carcinoma of non-clear-cell histology. J Clin Oncol 2002a; 20:2376-81.

19. Molina AM, Lin X, Korytowsky B, et al. Sunitinib objective response in metastaticrenal cell carcinoma: analysis of 1059 patients treated on clinical trials. Eur JCancer. In press.

20. Abel EJ, Culp SH, Tannir NM, Tamboli P, Matin SF, Wood CG. Early primarytumor size reduction is an independent predictor of improved overall survival inmetastatic renal cell carcinoma patients treated with sunitinib. Eur Urol 2011; 60:1273-9.

21. V. Grünwald, D. Kalanovic, R. McKay, J. Perkins, R. Simantov, T.K. Choueiri,Tumor response is an independent prognostic factor in patients (pts) treated formetastatic renal cell carcinoma (mRCC). Paper presented at: 2013 EuropeanCancer Congress.