Urologic Oncology: Seminars and Original Investigations ] (2016) ∎∎∎–∎∎∎
Original article
Baseline tumor volume in assessing prognosis of patients with intermediate-risk synchronous metastatic renal cell carcinoma Roderick E. de Bruijn, M.D.a,*, Jasper Nijkamp, Ph.D.b, Allard Noe, M.D.c, Simon Horenblas, Ph.D., M.D.d, John B.A.G. Haanen, Ph.D., M.D.e, Warner Prevoo, M.D.f, Axel Bex, Ph.D., M.D.g a Department of Urology, Sint Lucas Andreas Hospital, Amsterdam, The Netherlands Department of Radiotherapy, The Netherlands Cancer Institute, Amsterdam, The Netherlands c Department of Urology, Sint Antonius Hospital, Nieuwegein, The Netherlands d Department of Urology, The Netherlands Cancer Institute, Amsterdam, The Netherlands e Department of Medical Oncology, The Netherlands Cancer Institute Amsterdam, The Netherlands f Department of Radiology, The Netherlands Cancer Institute, Amsterdam, The Netherlands g Department of Urology, The Netherlands Cancer Institute, Amsterdam, The Netherlands b
Received 17 July 2015; received in revised form 7 December 2015; accepted 14 December 2015
Abstract Objective: To analyze if prediction of survival for patients with synchronous metastatic renal cell cancer (mRCC) could be further refined by baseline volume of the primary tumor, the metastases, or the remaining volume after surgery; this study was performed because survival expectancies of patients with intermediate-risk mRCC vary substantially. Material and methods: The predictive value of the different volumes on overall survival (OS) was analyzed retrospectively in patients with intermediate Memorial Sloan-Kettering Cancer Center (MSKCC) risk profile and r3 International Metastatic Renal Cell Carcinoma Database Consortium (IMDC) factors, who received sunitinib in our institute. Tumor volumes were calculated on segmented computed tomography using in-house developed software. A multivariate analysis was performed including number of metastatic sites and baseline tumor burden (TB). Results: A total of 68 patients were included. Median OS for patients without cytoreductive nephrectomy (CN) was 6 months (95% CI: 3.0–8.9 mo) vs. 31 months (95% CI: 23.1–38.8 mo) for those with CN, respectively. More second-line treatment was given after CN (49% vs. 17%, P ¼ 0.125). There was no correlation between tumor volume and TB measured by Response Evaluation Criteria in Solid Tumors. Of all included clinical and volumetric parameters, remaining volume after CN, CN status and 2 vs. 3 IMDC factors were significantly correlated with OS. In the Cox regression analysis, CN was the only remaining significant parameter (P ¼ 0.003). Conclusions: None of the baseline volumetric parameters is an independent prognostic factor in patients with intermediate MSKCC risk mRCC with r3 IMDC factors receiving sunitinib. Only CN status correlated significantly with prognosis. None of the baseline volumes nor TB by Response Evaluation Criteria in Solid Tumors was associated with CN status, suggesting that extent of disease had no significant influence on the decision to perform surgery. r 2016 Elsevier Inc. All rights reserved.
Keywords: Renal cell carcinoma; Volume; Metastasis; MSKCC; Survival
1. Introduction Despite the introduction of drugs targeting vascular endothelial growth factor or the mammalian target of Corresponding author. Tel.: þ31-622-975-020; fax: þ31-703-244-002. E-mail address: [email protected] (R.E. de Bruijn). *
http://dx.doi.org/10.1016/j.urolonc.2015.12.007 1078-1439/r 2016 Elsevier Inc. All rights reserved.
rapamycin, metastatic renal cell cancer (mRCC) is still the tenth cancer-related death in developed countries [1]. Patients respond heterogeneously to targeted therapy [2], and outcomes can differ substantially between individuals with similar clinical and pathological characteristics [3]. Prognostic models like the Memorial Sloan-Kettering Cancer Center (MSKCC) and International Metastatic
2
R.E. de Bruijn et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
Renal Cell Carcinoma Database Consortium (IMDC) risk assessment have been validated to stratify the patient risk of cancer progression and overall survival (OS) in the era of targeted therapy [4]. However, patients assigned to identical risk groups still have a broad range of survival expectancy. For example, intermediate-risk patients have a median survival of 27 months [4,5], but 25% do not survive the first year and 25% live 4 years and longer [5]. In a recent IMDC retrospective study consisting of exclusively synchronous mRCC, only those patients who underwent cytoreductive nephrectomy (CN) had a prognosis comparable to the survival expectancy reported for intermediate risk in the literature [6]. Survival for patients without a CN was 10 months shorter despite sharing the same risk group. The survival difference became nonsignificant with Z4 IMDC factors present. Although next generation sequencing is underway to unravel predictive and prognostic gene signatures, no molecular biomarkers have been identified to augment the existing risk models. Several clinical parameters have been suggested to improve prediction of survival expectancy in mRCC. A recurrent and consistent feature is that the extent of disease at the time of diagnosis may determine the heterogeneous outcomes observed within identical risk groups. The definition of disease extent varies in literature, using number of metastatic sites, baseline tumor burden (TB) according to Response Evaluation Criteria in Solid Tumors (RECIST) or volume of disease [7–9]. Baseline TB according to RECIST is easy to measure on diagnostic computed tomography (CT) scans. However, the correlation between longest diameter and tumor volume (TV) is not perfect [10]. A study revealed significant variability of lung tumor measurements by RECIST on CT scans taken within 15 minutes [11]. Volume of disease, and especially the distribution of volume between the primary tumor and the metastasis have been used to develop guidelines for selecting patients for CN. Pierorazio et al. estimated that if 490% of the total TV is located in the primary tumor, patients might benefit from a CN. Barbastefano et al. [9] indicated that a threshold of 95% should be used. The objective of this study was to analyze if prediction of survival for patients with synchronous mRCC and an identical MSKCC or IMDC risk profile, who receive sunitinib could be further refined by various volumetric parameters of primary tumor and metastases. Owing to the primary tumor in situ at the time of diagnosis and the option to do a CN, synchronous mRCC is a model disease to investigate the different baseline and actual volumes in relation to outcome under targeted therapy. In addition, we compared several volumetric parameters to other prognostic factors including number of metastatic sites and baseline TB by RECIST.
2. Material and methods 2.1. Hypothesis and secondary objections The aim of this study was to analyze if prediction of survival for patients with primary mRCC of intermediaterisk groups could be further refined by the volume of primary tumor or metastatic volume. We indicted the hypothesis that TV or metastatic volume, or a ratio of these volume parameters would have a predictive value on OS. In all, 2 secondary objectives were set to further essay this hypothesis: Is there an association between TV and the fact that patients undergoing CN or not? And could OS be further predicted by residual volume after CN? 2.2. Patient population Patients with histologically proven clear-cell primary mRCC of intermediate MSKCC risk or r3 IMDC risk factors treated at our institute between January 2006 and December 2012 were included in this retrospective singleinstitution descriptive series. Patients were derived from a prospectively maintained institutional tumor registry with review board approval. MSKCC criteria were evaluated based on the 5 risk factors: low Karnofsky performance status (o80), high lactate dehydrogenase level (41.5 times the upper limit of normal), low serum hemoglobin, high corrected serum calcium (410 mg/dl), and time from initial diagnosis to sunitinib treatment of less than 1 year [12]. To analyze IMDC risk, neutrophile and platelet counts were additionally collected. All patients were treated with orally administrated sunitinib at a dose of 50 mg daily, consisting of 4 weeks on treatment followed by a 2-week rest period in cycles of 6 weeks. Sunitinib was reduced (to 37.5 mg and then to 25 mg) depending on the type and severity of adverse events. If patients had symptoms of progressive disease during the rest period, continuous dosing of sunitinib at 37.5 mg/d was offered. Patients were only included if pretreatment CT scans of the abdomen and thorax were available to evaluate the volume of primary tumor and metastases. Pretreatment scans were only chosen for analysis if they were taken within 2 months before the start of treatment. For each patient, sex, age, OS, CN status, Karnofsky score, and number of metastatic sites were registered. 2.3. TV measurements The primary tumor and the metastasis were manually segmented on pretreatment diagnostic CT scans of the thorax and the abdomen. For the manual segmentation, in-house developed software (Radiotherapy department) was used, in which the tumor and metastasis were segmented on the transversal slices. The segmentations were
R.E. de Bruijn et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
3
2.4. Statistical analysis Pretreatment clinical and volumetric parameters were first correlated to OS to identify interesting parameters for the prediction of long-term survival before treatment with sunitinib. For the correlation, a nonparametric Spearman's rho correlation was used. The following volumetric parameters were evaluated as predictors: baseline TB, TBactual, baseline TVprimary, TVmets, TVtotal, and TVactual. Also, number of metastasis, number of metastatic sites, IMDC 2 vs. 3 factors, and CN status were evaluated. The parameters that significantly correlated were input for a Cox regression analysis to find significant predictive factors in a multivariable setting. The remaining significant multivariable parameters were used for patient selection, and OS between patient groups was calculated using the Kaplan-Meier method. The SPSS software was used for statistical analysis (SPSS for Windows, version 20). The level of significance for all comparisons was chosen at P r 0.05. 3. Results 3.1. General patient factors Fig. 1. Reconstruction of renal tumour and lymphnode metastases after manual segmentation. The segmentations were subsequently triangulated into volumetric meshes of which the volume could be determined. (Color version of figure is available online.)
subsequently triangulated into volumetric meshes of which the volume could be determined (Fig. 1). The diagnostic CT scans had an in-plane resolution of 1 by 1 mm and a slice distance of either 3 or 5 mm. For the primary TV calculation, no difference was made between necrotic and solid parts. Only metastases with a diameter of more than 1 cm were segmented, as the additional value of smaller metastasis to the total TV is limited. All segmentations were done by one observer (Rd.B.), and visually evaluated by another observer (J.N.). For each patient, the primary TV (TVprimary), total metastatic volume (TVmets), and total TV (TVtotal) were determined. To correct for a CN performed, the actual TV (TVactual) was calculated by subtracting TVprimary from TVtotal. In case no CN was performed, TVactual was equivalent to TVtotal. The tumor segmentations were also used to automatically calculate the maximal diameter per tumor/metastasis. For each patient, the baseline TB was defined according to RECIST 1.1 as the sum of the maximal diameters of each target lesion (in cm), with a maximum of 10 lesions [7]. Also TB was corrected for the cases where a CN was performed (TBactual). The tumour and metastatic volumes were also used to derive the fractional percentage of tumor volume (FPTV), which would be removed if a CN would be performed (Pierorazio) (TVprimary/TVtotal*100).
From 146 patients with synchronous mRCC treated since 2006, 68 patients were identified with intermediate MSKCC prognosis and r3 factors of IMDC risk, systemic therapy with sunitinib, and complete pretherapeutic CT of the abdomen and chest (Table 1). Median follow-up of these patients was 49 months. In all, 2 patients who were planned on having sunitinib as first-line treatment did not receive this treatment. Overall, 1 patient refused treatment, and the other could not be started because of rapid progression. These patients are kept in the analysis. In total, 24 patients received second-line treatment. Totally, 17 patients were treated with everolimus and 5 patients received pazopanib. In 2 patients, sunitinib was second-line treatment. 3.2. Predictive parameters To illustrate the lack of correlation between TB and TV, patient data are presented in a scatter plot (Figs. 2 and 3). Of all included, clinical and volumetric parameters, TBactual, IMDC 2 vs. 3 factors, CN, and TVactual had a significant correlation with survival, with correlation coefficients of 0.379 (P = 0.001), 0.250 (P = 0.033), 0.524 (P o 0.001), and 0.413 (P o 0.001), respectively. In the Cox regression analysis, CN (P = 0.003) was the only remaining significant parameter. The median OS for patients without CN was estimated at 6 months (95% CI: 3.0–8.9 mo); whereas for patients with CN, the estimated OS was 31 months (95% CI: 23.1– 38.8 mo) (Fig. 4). It is noteworthy that although not significant (P ¼ 0.125), 19 (49%) of the 39 patients who underwent CN also received second-line treatment, whereas
4
R.E. de Bruijn et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
patients surviving 42 years vs. o2 years after CN. However, the difference does not reach statistical significance (P ¼ 0.069).
Table 1 Clinical characteristics of included patients Age, y Median Range
59 33–81
Sex Male Female
51 17
Karnofsky o80% 480%
6 62
Number of metastatic sites 1 2 3 Z4
17 28 18 5
Number of metastasis Median Average Range
6 7.8 1–38
Pretreatment nephrectomy No Yes
29 39
Sunitinib first-line treatment No Yes
4 64
Second-line treatment No Yes
46 22
4. Discussion
only 5 (17%) of the 29 other patients received second-line treatment. 3.3. Pretherapeutic TV and CN status To evaluate whether patients were selected for CN, the volumetric and clinical input parameters were also compared between patients with and without a CN using a 2-sided independent t test. There was no association of TVmets with CN status (Table 2). Although TVmets was higher in patients without CN, the difference was not statistically significant (P ¼ 0.149). In addition, there was no association of the ratio TVmets-TVprimary with CN status. There was no association between larger primary tumors with higher TVmets.
In this study, primary tumour and metastatic volume was measured by precise segmentation and reconstruction by triangulation to analyze if prediction of survival for patients with synchronous mRCC and an intermediate MSKCC and r3-factor IMDC risk profile, who receive systemic therapy with sunitinib could be further refined by pretherapeutic volumetry. Combining prognostic models with extent of disease has been investigated previously. For example, the International Kidney Cancer Working Group added number of metastatic sites to MSKCC factors in their prognostic model [8]. In the literature, several definitions of extent of disease were used to investigate an association with survival but it remains unresolved if TV is equivalent to TB by RECIST. Iacovelli et al. [7] showed that baseline TB, defined as the sum of the maximum diameter of the 10 largest metastases, can be used to predict OS. A 46-patient retrospective study found a relation of renal cancer tumor load with outcome by associating a 495% of TB removed by CN with progression-free survival (PFS) [9]. In this study, baseline TB was measured by RECIST, defined as the sum of the longest diameter of all measurable lesions (41.0 cm). A similar study had been performed previously in the cytokine era (Pierorrazio). In their study, an FPTV cutoff of 90% was selected for stratification, showing that patients who underwent a CN with an FPTV 490% did better than others (11.6 vs. 2.9 months disease specific survival). Our results shown in Fig. 5 indicate that an actual selection of a cutoff point cannot easily be done. Also note that OS in the current study after CN, independent of further patient selection, is much better than the results of
3.4. Percentage of TV removed by CN An average of 23% of the total TV was located in the metastases. In only 27 and 20 of the 68 patients, the FPTV was 490% (Barbastefano et al. [9]) and 495%, respectively. Additionally, in only 39 patients, TVprimary was actually removed by CN. In Fig. 5, the overall survival of these 39 patients is shown in respect to the FPTV. No clear cutoff values can be derived from this figure. Further, differences in remaining TVmets suggest identification of
Fig. 2. Relation between tumour burden as determined by RECIST using the maximum diameter of maximally 10 lesions vs. the actual tumour volume determined on the same lesions. These data are collected before the start of sunitinib treatment, corrected for cytoreductive nephrectomy if applicable.
R.E. de Bruijn et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
5
Fig. 3. Scatter plot of the overall survival in relation to the actual tumour volume (left) stratified for cytoredutive nephrectomy. On the right, the same figure for tumour burden.
Pierorazio et al. Capitanio et al. [13] showed that patients undergoing CN with high metastastatic burden by RECIST do worse than patients with low metastatic burden. We chose to analyze only patients with intermediate risk profile because a wide range of OS has been reported for this group, which may benefit from refined prognosis assessment. Only CN status was significantly associated with improved survival in this retrospective analysis. Additionally, none of the pretherapeutic volumetric parameters nor TB by RECIST was associated with CN status. This suggests that extent of the disease had no influence on the decision to perform CN. We found no association with baseline TVtotal (including TVprimary and TVmets) on survival nor did we find that larger primary tumors were associated with higher TVmets. This finding is in line with a genetic analysis of intratumor heterogeneity, which clearly demonstrated that the potential
Fig. 4. Kaplan-Meier curve showing overall survival difference between patient treated with sunitinib after a cytoreductive nephrectomy is performed or not. The difference in median overall survival was significant (P o 0.001). (Color version of figure is available online.)
to metastasize develops randomly and not as a function of tumor growth over time in primary renal tumors [14]. Although it seems conceivable that larger primary tumors would have had more time to develop metastatic disease, it has been demonstrated that spatial evolution and subclones in tumors are driving mRCC rather than growth over time alone [3]. However, TVprimary can effectively be reduced by CN and the prognostic significance of TVmets alone could be evaluated in patients who had a CN. A previous report observed an association of percentage of TB removed with PFS but not with OS [9]. This report was a small retrospective study, TB was measured by RECIST longest diameter and only patients with 495% TB removed had a significant improvement of PFS. In addition, a 95% tumour reduction by CN may not be achievable for most patients with primary mRCC of intermediate risk. In our series, CN would have resulted in removal of 495% of the TV in only 29%. In only 39 patients, TVprimary was removed by CN. An explanation could be that patients with 495% tumour removal by CN often have oligometastatic disease and a long interval between diagnosis and disease progression requiring targeted systemic therapy. These patients are often favorable risk by MSKCC or IMDC factors and were unlikely to be included in our analysis of intermediaterisk patients receiving sunitinib. The observed difference in remaining TVmets suggests identification of long-term survivors, but was not statistically significant (P ¼ 0.069). Unexpectedly, we found no association of TVmets nor of the ratio TVmets-TVprimary with CN status, suggesting that neither metastatic volume nor its ratio to TVprimary was a factor in the decision for or against CN. This is contrary to our assumption that intermediate-risk patients with low metastatic volume would have been offered CN more often than patients with high metastatic volume or small primary tumour. However, the decision to remove the primary tumour substracted the substantial volume from the baseline TV. In the Cox regression analysis, only CN status was therefore significantly associated with survival in patients with intermediate-risk synchronous mRCC, who were
R.E. de Bruijn et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
6
Table 2 Comparison of the volumetric and clinical parameters between patients with and without a CN CN ¼ No (n ¼ 29)
CN ¼ Yes (n ¼ 39)
P value
TVprimary (median cc) TVmets (median cc) TVtotal (median cc)
285 (iqr ¼ 448) 67 (iqr ¼ 147) 416 (iqr ¼ 496)
313 (iqr ¼ 469) 60 (iqr ¼ 107) 351 (iqr ¼ 478)
0.977a 0.149a 0.674a
Karnofsky o80% 480%
3 26
2 37
0.956b
IMDC 3 Factors 2 Factors
13 16
9 30
0.463b
Number of metastatic sites (median) Number of metastasis (median) Second-line treatment
2 5 (iqr ¼ 7) 5
2 6 (iqr ¼ 5) 19
0.638a 0.452a 0.125b
iqr ¼ interquartile range. a 2-Sided independent t test. b Chi-square test.
treated with sunitinib. This is in line with previous trials in the cytokine era where the benefit of CN was proven [15,16]. However, taking the other parameters that were part of the regression analysis and which were each individually associated with a survival benefit into account (TBactual, IMDC 2 vs. 3 factors, and TVactual ) it is likely that CN status summarizes a favorable prognostic subgroup rather than being proof of a beneficial effect of debulking nephrectomy in the era of targeted therapy. Presently, only retrospective data are available that suggest that appropriately selected patients with mRCC benefit from integration of CN with systemic targeted therapy [6,17–19]. Therefore, the results of the ongoing randomized controlled CARMENA trial that compares CN followed by sunitinib to sunitinib alone would have to be awaited to definitely settle this question.
Actual TV correlated better with OS than actual TB by RECIST. However, none of the volumetric parameters was associated with survival in the final Cox regression analysis. In this particular group of patients with primary mRCC treated with sunitinib, the time consuming effort of segmentation volumetry did not add a benefit. This study has several limitations. The data were collected retrospectively and are biased by unaccounted factors. In addition, by limiting the analysis to MSKCC intermediate risk, the cohort was relatively small. Despite an identical MSKCC risk group, patients differed statistically significant between 2 vs. 3 IMDC factors in favor of CN with only 2 factors.
5. Conclusion Although in patients with intermediate-risk MSKCC mRCC receiving sunitinib, several volumetric and clinical parameters were significantly associated with survival, only CN status remained as independent prognostic factor in the Cox regression analysis. None of the baseline volume parameters was associated with CN suggesting that extent of disease had no significant influence on the decision to perform surgery. It is likely that CN status summarizes a favorable prognostic subgroup rather than being proof of a beneficial effect of debulking nephrectomy in the era of targeted therapy. References
Fig. 5. The overall survival of patients in whom the primary tumour was removed by CN is shown in respect to the FPTV. No clear cutoff values can be derived from this figure.
[1] Ferlay J, Steliarova-Foucher E, Lortet-Tieulent J, Rosso S, Coebergh JW, Comber H, et al. Cancer incidence and mortality patterns in Europe: estimates for 40 countries in 2012. Eur J Cancer 2013;49: 1374–403. [2] Albiges L, Choueiri T, Escudier B, Galsky M, George D, Hofmann F, et al. A systematic review of sequencing and combinations of
R.E. de Bruijn et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
systemic therapy in metastatic renal cancer. Eur Urol 2015;67:100– 10. Gerlinger M, Horswell S, Larkin J, Rowan AJ, Salm MP, Varela I, et al. Genomic architecture and evolution of clear cell renal cell carcinomas defined by multiregion sequencing. Nat Genet 2014;46: 225–33. Heng DY, Xie W, Regan MM, Harshman LC, Bjarnason GA, Vaishampayan UN, et al. External validation and comparison with other models of the International Metastatic Renal-Cell Carcinoma Database Consortium prognostic model: a population-based study. Lancet Oncol 2013;14:141–8. Harshman LC, Xie W, Bjarnason GA, Knox JJ, Mackenzie M, Wood L, et al. Conditional survival of patients with metastatic renal-cell carcinoma treated with VEGF-targeted therapy: a population-based study. Lancet Oncol 2012;13:150–5. Heng DY, Wells JC, Rini BI, Beuselinck B, Lee JL, Knox JJ, et al. Cytoreductive nephrectomy in patients with synchronous metastases from renal cell carcinoma: results from the international metastatic renal cell carcinoma database consortium. Eur Urol 2014;66:704–10. Iacovelli R, Lanoy E, Albiges L, Escudier B. Tumour burden is an independent prognostic factor in metastatic renal cell carcinoma. BJU Int 2012;110:1747–53. Manola J, Royston P, Elson P, McCormack JB, Mazumdar M, Negrier S, et al. Prognostic model for survival in patients with metastatic renal cell carcinoma: results from the International Kidney Cancer Working Group. Clin Cancer Res 2011;17:5443–50. Barbastefano J, Garcia JA, Elson P, Wood LS, Lane BR, Dreicer R, et al. Association of percentage of tumour burden removed with debulking nephrectomy and progression-free survival in patients with metastatic renal cell carcinoma treated with vascular endothelial growth factor-targeted therapy. BJU Int 2010;106:1266–9. van der Veldt AA, Meijerink MR, van den Eertwegh AJ, Bex A, de Gast GC, Haanen J, et al. Sunitinib for treatment of advanced renal cell cancer: primary tumor response. Clin Cancer Res 2008;14: 2431–6.
7
[11] Oxnard GR, Zhao B, Sima CS, Ginsberg MS, James LP, Lefkowitz RA, et al. Variability of lung tumor measurements on repeat computed tomography scans taken within 15 minutes. J Clin Oncol 2011;29:3114–9. [12] Motzer RJ, Bacik J, Murphy BA, Russo P, Mazumdar M. Interferonalfa as a comparative treatment for clinical trials of new therapies against advanced renal cell carcinoma. J Clin Oncol 2002;20:289–96. [13] Capitanio U, Abdollah F, Matloob R, Salonia A, Suardi N, Briganti A, et al. Effect of number and location of distant metastases on renal cell carcinoma mortality in candidates for cytoreductive nephrectomy: implications for multimodal therapy. Int J Urol 2013;20:572–9. [14] Gerlinger M, Rowan AJ, Horswell S, Larkin J, Endesfelder D, Gronroos E, et al. Intratumor heterogeneity and branched evolution revealed by multiregion sequencing. N Engl J Med 2012;366: 883–92. [15] Flanigan RC, Salmon SE, Blumenstein BA, Bearman SI, Roy V, McGrath PC, et al. Nephrectomy followed by interferon alfa-2b compared with interferon alfa-2b alone for metastatic renal-cell cancer. N Engl J Med 2001;345:1655–9. [16] Mickisch GH, Garin A, Van Poppel H, de Prijck L, Sylvester R. Radical nephrectomy plus interferon-alfa-based immunotherapy compared with interferon alfa alone in metastatic renal-cell carcinoma: a randomised trial. Lancet 2001;358:966–70. [17] Choueiri TK, Xie W, Kollmannsberger C, North S, Knox JJ, Lampard JG, et al. The impact of cytoreductive nephrectomy on survival of patients with metastatic renal cell carcinoma receiving vascular endothelial growth factor targeted therapy. J Urol 2011;185:60–6. [18] Aben KK, Heskamp S, Janssen-Heijnen ML, Koldewijn EL, van Herpen CM, Kiemeney LA, et al. Better survival in patients with metastasised kidney cancer after nephrectomy: a population-based study in the Netherlands. Eur J Cancer 2011;47:2023–32. [19] Abern MR, Scosyrev E, Tsivian M, Messing EM, Polascik TJ, Dudek AZ. Survival of patients undergoing cytoreductive surgery for metastatic renal cell carcinoma in the targeted-therapy era. Anticancer Res 2014;34:2405–11.
Original article
Baseline tumor volume in assessing prognosis of patients with intermediate-risk synchronous metastatic renal cell carcinoma Roderick E. de Bruijn, M.D.a,*, Jasper Nijkamp, Ph.D.b, Allard Noe, M.D.c, Simon Horenblas, Ph.D., M.D.d, John B.A.G. Haanen, Ph.D., M.D.e, Warner Prevoo, M.D.f, Axel Bex, Ph.D., M.D.g a Department of Urology, Sint Lucas Andreas Hospital, Amsterdam, The Netherlands Department of Radiotherapy, The Netherlands Cancer Institute, Amsterdam, The Netherlands c Department of Urology, Sint Antonius Hospital, Nieuwegein, The Netherlands d Department of Urology, The Netherlands Cancer Institute, Amsterdam, The Netherlands e Department of Medical Oncology, The Netherlands Cancer Institute Amsterdam, The Netherlands f Department of Radiology, The Netherlands Cancer Institute, Amsterdam, The Netherlands g Department of Urology, The Netherlands Cancer Institute, Amsterdam, The Netherlands b
Received 17 July 2015; received in revised form 7 December 2015; accepted 14 December 2015
Abstract Objective: To analyze if prediction of survival for patients with synchronous metastatic renal cell cancer (mRCC) could be further refined by baseline volume of the primary tumor, the metastases, or the remaining volume after surgery; this study was performed because survival expectancies of patients with intermediate-risk mRCC vary substantially. Material and methods: The predictive value of the different volumes on overall survival (OS) was analyzed retrospectively in patients with intermediate Memorial Sloan-Kettering Cancer Center (MSKCC) risk profile and r3 International Metastatic Renal Cell Carcinoma Database Consortium (IMDC) factors, who received sunitinib in our institute. Tumor volumes were calculated on segmented computed tomography using in-house developed software. A multivariate analysis was performed including number of metastatic sites and baseline tumor burden (TB). Results: A total of 68 patients were included. Median OS for patients without cytoreductive nephrectomy (CN) was 6 months (95% CI: 3.0–8.9 mo) vs. 31 months (95% CI: 23.1–38.8 mo) for those with CN, respectively. More second-line treatment was given after CN (49% vs. 17%, P ¼ 0.125). There was no correlation between tumor volume and TB measured by Response Evaluation Criteria in Solid Tumors. Of all included clinical and volumetric parameters, remaining volume after CN, CN status and 2 vs. 3 IMDC factors were significantly correlated with OS. In the Cox regression analysis, CN was the only remaining significant parameter (P ¼ 0.003). Conclusions: None of the baseline volumetric parameters is an independent prognostic factor in patients with intermediate MSKCC risk mRCC with r3 IMDC factors receiving sunitinib. Only CN status correlated significantly with prognosis. None of the baseline volumes nor TB by Response Evaluation Criteria in Solid Tumors was associated with CN status, suggesting that extent of disease had no significant influence on the decision to perform surgery. r 2016 Elsevier Inc. All rights reserved.
Keywords: Renal cell carcinoma; Volume; Metastasis; MSKCC; Survival
1. Introduction Despite the introduction of drugs targeting vascular endothelial growth factor or the mammalian target of Corresponding author. Tel.: þ31-622-975-020; fax: þ31-703-244-002. E-mail address: [email protected] (R.E. de Bruijn). *
http://dx.doi.org/10.1016/j.urolonc.2015.12.007 1078-1439/r 2016 Elsevier Inc. All rights reserved.
rapamycin, metastatic renal cell cancer (mRCC) is still the tenth cancer-related death in developed countries [1]. Patients respond heterogeneously to targeted therapy [2], and outcomes can differ substantially between individuals with similar clinical and pathological characteristics [3]. Prognostic models like the Memorial Sloan-Kettering Cancer Center (MSKCC) and International Metastatic
2
R.E. de Bruijn et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
Renal Cell Carcinoma Database Consortium (IMDC) risk assessment have been validated to stratify the patient risk of cancer progression and overall survival (OS) in the era of targeted therapy [4]. However, patients assigned to identical risk groups still have a broad range of survival expectancy. For example, intermediate-risk patients have a median survival of 27 months [4,5], but 25% do not survive the first year and 25% live 4 years and longer [5]. In a recent IMDC retrospective study consisting of exclusively synchronous mRCC, only those patients who underwent cytoreductive nephrectomy (CN) had a prognosis comparable to the survival expectancy reported for intermediate risk in the literature [6]. Survival for patients without a CN was 10 months shorter despite sharing the same risk group. The survival difference became nonsignificant with Z4 IMDC factors present. Although next generation sequencing is underway to unravel predictive and prognostic gene signatures, no molecular biomarkers have been identified to augment the existing risk models. Several clinical parameters have been suggested to improve prediction of survival expectancy in mRCC. A recurrent and consistent feature is that the extent of disease at the time of diagnosis may determine the heterogeneous outcomes observed within identical risk groups. The definition of disease extent varies in literature, using number of metastatic sites, baseline tumor burden (TB) according to Response Evaluation Criteria in Solid Tumors (RECIST) or volume of disease [7–9]. Baseline TB according to RECIST is easy to measure on diagnostic computed tomography (CT) scans. However, the correlation between longest diameter and tumor volume (TV) is not perfect [10]. A study revealed significant variability of lung tumor measurements by RECIST on CT scans taken within 15 minutes [11]. Volume of disease, and especially the distribution of volume between the primary tumor and the metastasis have been used to develop guidelines for selecting patients for CN. Pierorazio et al. estimated that if 490% of the total TV is located in the primary tumor, patients might benefit from a CN. Barbastefano et al. [9] indicated that a threshold of 95% should be used. The objective of this study was to analyze if prediction of survival for patients with synchronous mRCC and an identical MSKCC or IMDC risk profile, who receive sunitinib could be further refined by various volumetric parameters of primary tumor and metastases. Owing to the primary tumor in situ at the time of diagnosis and the option to do a CN, synchronous mRCC is a model disease to investigate the different baseline and actual volumes in relation to outcome under targeted therapy. In addition, we compared several volumetric parameters to other prognostic factors including number of metastatic sites and baseline TB by RECIST.
2. Material and methods 2.1. Hypothesis and secondary objections The aim of this study was to analyze if prediction of survival for patients with primary mRCC of intermediaterisk groups could be further refined by the volume of primary tumor or metastatic volume. We indicted the hypothesis that TV or metastatic volume, or a ratio of these volume parameters would have a predictive value on OS. In all, 2 secondary objectives were set to further essay this hypothesis: Is there an association between TV and the fact that patients undergoing CN or not? And could OS be further predicted by residual volume after CN? 2.2. Patient population Patients with histologically proven clear-cell primary mRCC of intermediate MSKCC risk or r3 IMDC risk factors treated at our institute between January 2006 and December 2012 were included in this retrospective singleinstitution descriptive series. Patients were derived from a prospectively maintained institutional tumor registry with review board approval. MSKCC criteria were evaluated based on the 5 risk factors: low Karnofsky performance status (o80), high lactate dehydrogenase level (41.5 times the upper limit of normal), low serum hemoglobin, high corrected serum calcium (410 mg/dl), and time from initial diagnosis to sunitinib treatment of less than 1 year [12]. To analyze IMDC risk, neutrophile and platelet counts were additionally collected. All patients were treated with orally administrated sunitinib at a dose of 50 mg daily, consisting of 4 weeks on treatment followed by a 2-week rest period in cycles of 6 weeks. Sunitinib was reduced (to 37.5 mg and then to 25 mg) depending on the type and severity of adverse events. If patients had symptoms of progressive disease during the rest period, continuous dosing of sunitinib at 37.5 mg/d was offered. Patients were only included if pretreatment CT scans of the abdomen and thorax were available to evaluate the volume of primary tumor and metastases. Pretreatment scans were only chosen for analysis if they were taken within 2 months before the start of treatment. For each patient, sex, age, OS, CN status, Karnofsky score, and number of metastatic sites were registered. 2.3. TV measurements The primary tumor and the metastasis were manually segmented on pretreatment diagnostic CT scans of the thorax and the abdomen. For the manual segmentation, in-house developed software (Radiotherapy department) was used, in which the tumor and metastasis were segmented on the transversal slices. The segmentations were
R.E. de Bruijn et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
3
2.4. Statistical analysis Pretreatment clinical and volumetric parameters were first correlated to OS to identify interesting parameters for the prediction of long-term survival before treatment with sunitinib. For the correlation, a nonparametric Spearman's rho correlation was used. The following volumetric parameters were evaluated as predictors: baseline TB, TBactual, baseline TVprimary, TVmets, TVtotal, and TVactual. Also, number of metastasis, number of metastatic sites, IMDC 2 vs. 3 factors, and CN status were evaluated. The parameters that significantly correlated were input for a Cox regression analysis to find significant predictive factors in a multivariable setting. The remaining significant multivariable parameters were used for patient selection, and OS between patient groups was calculated using the Kaplan-Meier method. The SPSS software was used for statistical analysis (SPSS for Windows, version 20). The level of significance for all comparisons was chosen at P r 0.05. 3. Results 3.1. General patient factors Fig. 1. Reconstruction of renal tumour and lymphnode metastases after manual segmentation. The segmentations were subsequently triangulated into volumetric meshes of which the volume could be determined. (Color version of figure is available online.)
subsequently triangulated into volumetric meshes of which the volume could be determined (Fig. 1). The diagnostic CT scans had an in-plane resolution of 1 by 1 mm and a slice distance of either 3 or 5 mm. For the primary TV calculation, no difference was made between necrotic and solid parts. Only metastases with a diameter of more than 1 cm were segmented, as the additional value of smaller metastasis to the total TV is limited. All segmentations were done by one observer (Rd.B.), and visually evaluated by another observer (J.N.). For each patient, the primary TV (TVprimary), total metastatic volume (TVmets), and total TV (TVtotal) were determined. To correct for a CN performed, the actual TV (TVactual) was calculated by subtracting TVprimary from TVtotal. In case no CN was performed, TVactual was equivalent to TVtotal. The tumor segmentations were also used to automatically calculate the maximal diameter per tumor/metastasis. For each patient, the baseline TB was defined according to RECIST 1.1 as the sum of the maximal diameters of each target lesion (in cm), with a maximum of 10 lesions [7]. Also TB was corrected for the cases where a CN was performed (TBactual). The tumour and metastatic volumes were also used to derive the fractional percentage of tumor volume (FPTV), which would be removed if a CN would be performed (Pierorazio) (TVprimary/TVtotal*100).
From 146 patients with synchronous mRCC treated since 2006, 68 patients were identified with intermediate MSKCC prognosis and r3 factors of IMDC risk, systemic therapy with sunitinib, and complete pretherapeutic CT of the abdomen and chest (Table 1). Median follow-up of these patients was 49 months. In all, 2 patients who were planned on having sunitinib as first-line treatment did not receive this treatment. Overall, 1 patient refused treatment, and the other could not be started because of rapid progression. These patients are kept in the analysis. In total, 24 patients received second-line treatment. Totally, 17 patients were treated with everolimus and 5 patients received pazopanib. In 2 patients, sunitinib was second-line treatment. 3.2. Predictive parameters To illustrate the lack of correlation between TB and TV, patient data are presented in a scatter plot (Figs. 2 and 3). Of all included, clinical and volumetric parameters, TBactual, IMDC 2 vs. 3 factors, CN, and TVactual had a significant correlation with survival, with correlation coefficients of 0.379 (P = 0.001), 0.250 (P = 0.033), 0.524 (P o 0.001), and 0.413 (P o 0.001), respectively. In the Cox regression analysis, CN (P = 0.003) was the only remaining significant parameter. The median OS for patients without CN was estimated at 6 months (95% CI: 3.0–8.9 mo); whereas for patients with CN, the estimated OS was 31 months (95% CI: 23.1– 38.8 mo) (Fig. 4). It is noteworthy that although not significant (P ¼ 0.125), 19 (49%) of the 39 patients who underwent CN also received second-line treatment, whereas
4
R.E. de Bruijn et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
patients surviving 42 years vs. o2 years after CN. However, the difference does not reach statistical significance (P ¼ 0.069).
Table 1 Clinical characteristics of included patients Age, y Median Range
59 33–81
Sex Male Female
51 17
Karnofsky o80% 480%
6 62
Number of metastatic sites 1 2 3 Z4
17 28 18 5
Number of metastasis Median Average Range
6 7.8 1–38
Pretreatment nephrectomy No Yes
29 39
Sunitinib first-line treatment No Yes
4 64
Second-line treatment No Yes
46 22
4. Discussion
only 5 (17%) of the 29 other patients received second-line treatment. 3.3. Pretherapeutic TV and CN status To evaluate whether patients were selected for CN, the volumetric and clinical input parameters were also compared between patients with and without a CN using a 2-sided independent t test. There was no association of TVmets with CN status (Table 2). Although TVmets was higher in patients without CN, the difference was not statistically significant (P ¼ 0.149). In addition, there was no association of the ratio TVmets-TVprimary with CN status. There was no association between larger primary tumors with higher TVmets.
In this study, primary tumour and metastatic volume was measured by precise segmentation and reconstruction by triangulation to analyze if prediction of survival for patients with synchronous mRCC and an intermediate MSKCC and r3-factor IMDC risk profile, who receive systemic therapy with sunitinib could be further refined by pretherapeutic volumetry. Combining prognostic models with extent of disease has been investigated previously. For example, the International Kidney Cancer Working Group added number of metastatic sites to MSKCC factors in their prognostic model [8]. In the literature, several definitions of extent of disease were used to investigate an association with survival but it remains unresolved if TV is equivalent to TB by RECIST. Iacovelli et al. [7] showed that baseline TB, defined as the sum of the maximum diameter of the 10 largest metastases, can be used to predict OS. A 46-patient retrospective study found a relation of renal cancer tumor load with outcome by associating a 495% of TB removed by CN with progression-free survival (PFS) [9]. In this study, baseline TB was measured by RECIST, defined as the sum of the longest diameter of all measurable lesions (41.0 cm). A similar study had been performed previously in the cytokine era (Pierorrazio). In their study, an FPTV cutoff of 90% was selected for stratification, showing that patients who underwent a CN with an FPTV 490% did better than others (11.6 vs. 2.9 months disease specific survival). Our results shown in Fig. 5 indicate that an actual selection of a cutoff point cannot easily be done. Also note that OS in the current study after CN, independent of further patient selection, is much better than the results of
3.4. Percentage of TV removed by CN An average of 23% of the total TV was located in the metastases. In only 27 and 20 of the 68 patients, the FPTV was 490% (Barbastefano et al. [9]) and 495%, respectively. Additionally, in only 39 patients, TVprimary was actually removed by CN. In Fig. 5, the overall survival of these 39 patients is shown in respect to the FPTV. No clear cutoff values can be derived from this figure. Further, differences in remaining TVmets suggest identification of
Fig. 2. Relation between tumour burden as determined by RECIST using the maximum diameter of maximally 10 lesions vs. the actual tumour volume determined on the same lesions. These data are collected before the start of sunitinib treatment, corrected for cytoreductive nephrectomy if applicable.
R.E. de Bruijn et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
5
Fig. 3. Scatter plot of the overall survival in relation to the actual tumour volume (left) stratified for cytoredutive nephrectomy. On the right, the same figure for tumour burden.
Pierorazio et al. Capitanio et al. [13] showed that patients undergoing CN with high metastastatic burden by RECIST do worse than patients with low metastatic burden. We chose to analyze only patients with intermediate risk profile because a wide range of OS has been reported for this group, which may benefit from refined prognosis assessment. Only CN status was significantly associated with improved survival in this retrospective analysis. Additionally, none of the pretherapeutic volumetric parameters nor TB by RECIST was associated with CN status. This suggests that extent of the disease had no influence on the decision to perform CN. We found no association with baseline TVtotal (including TVprimary and TVmets) on survival nor did we find that larger primary tumors were associated with higher TVmets. This finding is in line with a genetic analysis of intratumor heterogeneity, which clearly demonstrated that the potential
Fig. 4. Kaplan-Meier curve showing overall survival difference between patient treated with sunitinib after a cytoreductive nephrectomy is performed or not. The difference in median overall survival was significant (P o 0.001). (Color version of figure is available online.)
to metastasize develops randomly and not as a function of tumor growth over time in primary renal tumors [14]. Although it seems conceivable that larger primary tumors would have had more time to develop metastatic disease, it has been demonstrated that spatial evolution and subclones in tumors are driving mRCC rather than growth over time alone [3]. However, TVprimary can effectively be reduced by CN and the prognostic significance of TVmets alone could be evaluated in patients who had a CN. A previous report observed an association of percentage of TB removed with PFS but not with OS [9]. This report was a small retrospective study, TB was measured by RECIST longest diameter and only patients with 495% TB removed had a significant improvement of PFS. In addition, a 95% tumour reduction by CN may not be achievable for most patients with primary mRCC of intermediate risk. In our series, CN would have resulted in removal of 495% of the TV in only 29%. In only 39 patients, TVprimary was removed by CN. An explanation could be that patients with 495% tumour removal by CN often have oligometastatic disease and a long interval between diagnosis and disease progression requiring targeted systemic therapy. These patients are often favorable risk by MSKCC or IMDC factors and were unlikely to be included in our analysis of intermediaterisk patients receiving sunitinib. The observed difference in remaining TVmets suggests identification of long-term survivors, but was not statistically significant (P ¼ 0.069). Unexpectedly, we found no association of TVmets nor of the ratio TVmets-TVprimary with CN status, suggesting that neither metastatic volume nor its ratio to TVprimary was a factor in the decision for or against CN. This is contrary to our assumption that intermediate-risk patients with low metastatic volume would have been offered CN more often than patients with high metastatic volume or small primary tumour. However, the decision to remove the primary tumour substracted the substantial volume from the baseline TV. In the Cox regression analysis, only CN status was therefore significantly associated with survival in patients with intermediate-risk synchronous mRCC, who were
R.E. de Bruijn et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
6
Table 2 Comparison of the volumetric and clinical parameters between patients with and without a CN CN ¼ No (n ¼ 29)
CN ¼ Yes (n ¼ 39)
P value
TVprimary (median cc) TVmets (median cc) TVtotal (median cc)
285 (iqr ¼ 448) 67 (iqr ¼ 147) 416 (iqr ¼ 496)
313 (iqr ¼ 469) 60 (iqr ¼ 107) 351 (iqr ¼ 478)
0.977a 0.149a 0.674a
Karnofsky o80% 480%
3 26
2 37
0.956b
IMDC 3 Factors 2 Factors
13 16
9 30
0.463b
Number of metastatic sites (median) Number of metastasis (median) Second-line treatment
2 5 (iqr ¼ 7) 5
2 6 (iqr ¼ 5) 19
0.638a 0.452a 0.125b
iqr ¼ interquartile range. a 2-Sided independent t test. b Chi-square test.
treated with sunitinib. This is in line with previous trials in the cytokine era where the benefit of CN was proven [15,16]. However, taking the other parameters that were part of the regression analysis and which were each individually associated with a survival benefit into account (TBactual, IMDC 2 vs. 3 factors, and TVactual ) it is likely that CN status summarizes a favorable prognostic subgroup rather than being proof of a beneficial effect of debulking nephrectomy in the era of targeted therapy. Presently, only retrospective data are available that suggest that appropriately selected patients with mRCC benefit from integration of CN with systemic targeted therapy [6,17–19]. Therefore, the results of the ongoing randomized controlled CARMENA trial that compares CN followed by sunitinib to sunitinib alone would have to be awaited to definitely settle this question.
Actual TV correlated better with OS than actual TB by RECIST. However, none of the volumetric parameters was associated with survival in the final Cox regression analysis. In this particular group of patients with primary mRCC treated with sunitinib, the time consuming effort of segmentation volumetry did not add a benefit. This study has several limitations. The data were collected retrospectively and are biased by unaccounted factors. In addition, by limiting the analysis to MSKCC intermediate risk, the cohort was relatively small. Despite an identical MSKCC risk group, patients differed statistically significant between 2 vs. 3 IMDC factors in favor of CN with only 2 factors.
5. Conclusion Although in patients with intermediate-risk MSKCC mRCC receiving sunitinib, several volumetric and clinical parameters were significantly associated with survival, only CN status remained as independent prognostic factor in the Cox regression analysis. None of the baseline volume parameters was associated with CN suggesting that extent of disease had no significant influence on the decision to perform surgery. It is likely that CN status summarizes a favorable prognostic subgroup rather than being proof of a beneficial effect of debulking nephrectomy in the era of targeted therapy. References
Fig. 5. The overall survival of patients in whom the primary tumour was removed by CN is shown in respect to the FPTV. No clear cutoff values can be derived from this figure.
[1] Ferlay J, Steliarova-Foucher E, Lortet-Tieulent J, Rosso S, Coebergh JW, Comber H, et al. Cancer incidence and mortality patterns in Europe: estimates for 40 countries in 2012. Eur J Cancer 2013;49: 1374–403. [2] Albiges L, Choueiri T, Escudier B, Galsky M, George D, Hofmann F, et al. A systematic review of sequencing and combinations of
R.E. de Bruijn et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
systemic therapy in metastatic renal cancer. Eur Urol 2015;67:100– 10. Gerlinger M, Horswell S, Larkin J, Rowan AJ, Salm MP, Varela I, et al. Genomic architecture and evolution of clear cell renal cell carcinomas defined by multiregion sequencing. Nat Genet 2014;46: 225–33. Heng DY, Xie W, Regan MM, Harshman LC, Bjarnason GA, Vaishampayan UN, et al. External validation and comparison with other models of the International Metastatic Renal-Cell Carcinoma Database Consortium prognostic model: a population-based study. Lancet Oncol 2013;14:141–8. Harshman LC, Xie W, Bjarnason GA, Knox JJ, Mackenzie M, Wood L, et al. Conditional survival of patients with metastatic renal-cell carcinoma treated with VEGF-targeted therapy: a population-based study. Lancet Oncol 2012;13:150–5. Heng DY, Wells JC, Rini BI, Beuselinck B, Lee JL, Knox JJ, et al. Cytoreductive nephrectomy in patients with synchronous metastases from renal cell carcinoma: results from the international metastatic renal cell carcinoma database consortium. Eur Urol 2014;66:704–10. Iacovelli R, Lanoy E, Albiges L, Escudier B. Tumour burden is an independent prognostic factor in metastatic renal cell carcinoma. BJU Int 2012;110:1747–53. Manola J, Royston P, Elson P, McCormack JB, Mazumdar M, Negrier S, et al. Prognostic model for survival in patients with metastatic renal cell carcinoma: results from the International Kidney Cancer Working Group. Clin Cancer Res 2011;17:5443–50. Barbastefano J, Garcia JA, Elson P, Wood LS, Lane BR, Dreicer R, et al. Association of percentage of tumour burden removed with debulking nephrectomy and progression-free survival in patients with metastatic renal cell carcinoma treated with vascular endothelial growth factor-targeted therapy. BJU Int 2010;106:1266–9. van der Veldt AA, Meijerink MR, van den Eertwegh AJ, Bex A, de Gast GC, Haanen J, et al. Sunitinib for treatment of advanced renal cell cancer: primary tumor response. Clin Cancer Res 2008;14: 2431–6.
7
[11] Oxnard GR, Zhao B, Sima CS, Ginsberg MS, James LP, Lefkowitz RA, et al. Variability of lung tumor measurements on repeat computed tomography scans taken within 15 minutes. J Clin Oncol 2011;29:3114–9. [12] Motzer RJ, Bacik J, Murphy BA, Russo P, Mazumdar M. Interferonalfa as a comparative treatment for clinical trials of new therapies against advanced renal cell carcinoma. J Clin Oncol 2002;20:289–96. [13] Capitanio U, Abdollah F, Matloob R, Salonia A, Suardi N, Briganti A, et al. Effect of number and location of distant metastases on renal cell carcinoma mortality in candidates for cytoreductive nephrectomy: implications for multimodal therapy. Int J Urol 2013;20:572–9. [14] Gerlinger M, Rowan AJ, Horswell S, Larkin J, Endesfelder D, Gronroos E, et al. Intratumor heterogeneity and branched evolution revealed by multiregion sequencing. N Engl J Med 2012;366: 883–92. [15] Flanigan RC, Salmon SE, Blumenstein BA, Bearman SI, Roy V, McGrath PC, et al. Nephrectomy followed by interferon alfa-2b compared with interferon alfa-2b alone for metastatic renal-cell cancer. N Engl J Med 2001;345:1655–9. [16] Mickisch GH, Garin A, Van Poppel H, de Prijck L, Sylvester R. Radical nephrectomy plus interferon-alfa-based immunotherapy compared with interferon alfa alone in metastatic renal-cell carcinoma: a randomised trial. Lancet 2001;358:966–70. [17] Choueiri TK, Xie W, Kollmannsberger C, North S, Knox JJ, Lampard JG, et al. The impact of cytoreductive nephrectomy on survival of patients with metastatic renal cell carcinoma receiving vascular endothelial growth factor targeted therapy. J Urol 2011;185:60–6. [18] Aben KK, Heskamp S, Janssen-Heijnen ML, Koldewijn EL, van Herpen CM, Kiemeney LA, et al. Better survival in patients with metastasised kidney cancer after nephrectomy: a population-based study in the Netherlands. Eur J Cancer 2011;47:2023–32. [19] Abern MR, Scosyrev E, Tsivian M, Messing EM, Polascik TJ, Dudek AZ. Survival of patients undergoing cytoreductive surgery for metastatic renal cell carcinoma in the targeted-therapy era. Anticancer Res 2014;34:2405–11.
