- Email: [email protected]
Quantitative tumour necrosis is an independent predictor of overall survival in clear cell renal cell carcinoma
Pathology (January 2015) 47(1), pp. 34–37
ANATOMICAL PATHOLOGY
Quantitative tumour necrosis is an independent predictor of overall survival in clear cell renal cell carcinoma ANDREW A. RENSHAW1
AND
JOHN C. CHEVILLE2
1Department of Pathology, Baptist Hospital of Miami, Miami, FL, and 2Professor of Laboratory Medicine and
Pathology, Department of Pathology, Mayo Clinic, Rochester, MN, United States
Summary Previous studies have reached conflicting results regarding whether tumour necrosis is a predictor of survival in clear cell renal cell carcinoma. In addition, studies quantifying the extent of necrosis are limited. The aim of this study was to determine if quantifying tumour necrosis could improve its predictive value for survival in clear cell renal cell carcinoma. We reviewed the clinical pathological information contained in The Cancer Genome Atlas for clear cell renal cell carcinoma and correlated it with overall survival using a Cox proportional hazard model. Necrosis was quantified on a single frozen section slide taken at the time of tissue harvesting for molecular studies. For all tumours, the presence of tumour necrosis was a significant predictor of overall survival ( p < 0.001) on univariate analysis. When quantitated, >10% necrosis was associated with survival, but 10% necrosis was not. On multivariate analysis, age ( p ¼ 0.004), T3b stage ( p ¼ 0.02), M1 stage ( p < 0.001), necrosis >30% ( p < 0.001), and elevated serum calcium ( p ¼ 0.003) remained significant. For clinical stage 1–2 (T1-T2N0M0) tumours, necrosis >20% was significant on univariate analysis ( p 0.005), and remained so on multivariate analysis ( p < 0.001). We conclude that quantitating the extent of tumour necrosis adds prognostic information in clear cell renal cell carcinomas, including organ confined tumours. Key words: Clear cell, necrosis, prognosis, quantitative, renal cell carcinoma, survival. Received 23 June, revised 7 August, accepted 13 August 2014
INTRODUCTION Previous studies have shown conflicting results regarding whether coagulative tumour necrosis is a predictor of survival in renal cell carcinoma. Most studies have agreed that the presence of tumour necrosis is associated with worse cause specific survival1–6 and overall survival7–11 in renal cell carcinomas of clear cell renal cell and chromophobe type,1 including in organ confined (T1-T2N0M0) clear cell tumours.4,8 Nevertheless, some studies have failed to show that tumour necrosis is an independent predictor of prognosis when considered in the context of other prognostic features.12 In addition, previous studies examining the extent of necrosis are limited. To our knowledge only two previous studies attempted to stratify the extent of necrosis.6,8 One study suggested that tumour necrosis of greater than 50% of the entire tumour was associated with worse disease specific Print ISSN 0031-3025/Online ISSN 1465-3931 DOI: 10.1097/PAT.0000000000000193
#
survival but not overall survival, and the other suggested that quantification was not of value in predicting overall survival. However, there were significant limitations in the way in which the extent of necrosis was quantified in these studies. In particular, the extent of necrosis was estimated for the entire tumour and the extent of necrosis was stratified into a few very large groups. Stratification into smaller groups, especially in those tumours with lesser degrees of necrosis, was not assessed. We recently noticed that in a previously published study addressing the molecular characteristics of clear cell renal cell carcinoma from the Tumour Cancer Genome Atlas (TCGA),13 tumour necrosis was assessed as part of a quality assessment measure for molecular studies. In this study, tumour necrosis was estimated from examination of a single frozen section slide of tissue that was taken for molecular studies. We wondered if this assessment might shed light on the significance of quantifying necrosis in clear cell renal cell carcinoma.
METHODS All data were taken from the supplementary tables of the Cancer Genome Atlas Research Network publication13 (Data File S2, clinical dataset) which also contain additional methodological information. In brief and as described in the methods of that paper and paraphrased here, tumours were flash-frozen and shipped to a centralised processing centre (Biospecimen Core Resource, BCR) for additional pathological review and nucleic acids extraction. Biospecimens were collected from newly diagnosed patients with renal clear cell carcinoma undergoing surgical resection who had received no prior treatment for their disease, including chemotherapy or radiotherapy. All cases were collected, regardless of surgical stage or histological grade. Cases were staged according to the American Joint Committee on Cancer (AJCC) staging system. Cases were graded using the Fuhrman grading system, which was the standard at the time the study was performed. Each frozen tumour specimen submitted to the BCR weighed at least 30 mg and was typically under 200 mg. Specimens were shipped overnight from one of 13 tissue source sites using a cryoport that maintained an average temperature of less than –1808C. Tumour necrosis was assessed microscopically from a single frozen section of this biorepository specimen. Complete clinical data elements were compiled for all specimens included and reflected current data as of 13 April 2012. Clinical/demographic data included: sample code, primary site (kidney for all specimens), gender, age at diagnosis, race, ethnicity, and year of tumour collection. Samples of questionable authenticity as clear cell RCC tumours due to unexpected molecular analysis results were evaluated by secondary pathological review. Tumour information recorded complete pathological information regarding the tumour. All specimens included in this analysis were coded as kidney clear cell renal carcinoma. The table records: laterality (right/left), Fuhrman nuclear grade, maximum tumour dimension (cm), T stage, lymph node involvement (based on pathological staging), and M stage (intended to be indicative of a review of clinical evidence for metastatic disease, but was often provided from available pathological information only, so should be interpreted with caution).
2014 Royal College of Pathologists of Australasia
Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.
NECROSIS IN CLEAR CELL RCC
A compiled tumour stage using standard AJCC staging criteria using the Tumour Node Metastasis universal schema was reported. Pathological data came from board certified pathologists in the originating institution. Additional laboratory data including lactate dehydrogenase, calcium, haemoglobin, white blood cell count, and platelet count were also included. Clinical status of patients at the point of enrolment, and as available at last follow-up was recorded. Sites were asked to indicate if patients following surgical resection were tumour free, or with tumour. We also recorded vital status (living/deceased) at the time of enrolment. Follow-up data were requested for subjects out to a minimum of 2 years from the time of sample collection. The patient tumour status (tumour free/with tumour) was again recorded, along with vital status (living/deceased) from the most recent followup data form completion at the time of data collection. Time to recurrence was recorded as the number of days to a new tumour event. We also recorded the days to last contact at the point of enrolment or most recent follow up. Finally, the days from diagnosis (sample collection) to death were recorded at both enrolment and in the most recent follow-up forms. These data provided the information to explore survival-based outcomes and median follow-up for patients included in this study. Cause of death, from cancer or other causes, was not recorded. Survival analysis was performed using a Cox proportional hazard ratio (HR). A threshold of p < 0.05 was considered significant.
RESULTS A total of 446 cases were included and there were 145 deaths; 257 organ confined (T1-T2N0M0) cases were included, and included 45 deaths. Tumour necrosis was present in 181 (40.6%) tumours, and 79 (30.7%) organ confined tumours, and ranged from 1–30%. Necrosis of greater than 5% was present in 34% of tumours. For all cases, age, type of surgery, size, T stage, M stage, presence of necrosis, elevated calcium, elevated WBC, low haemoglobin, and elevated platelet count were all significantly associated with overall survival on univariate analysis (Table 1). When quantitated, necrosis >10% was associated with survival, but necrosis 10% was not (Table 2). On multivariate analysis, age ( p ¼ 0.004), T3b stage ( p ¼ 0.02), M1 stage ( p < 0.001), necrosis >10% ( p 0.04), necrosis >30% ( p < 0.001), and elevated serum calcium ( p ¼ 0.003) remained significant (Table 3). Kaplan–Meier curves for tumours with >10% necrosis versus 10% necrosis are shown in Fig. 1. For clinical stage 1–2 (T1-T2N0M0) tumours, necrosis >20% was significant on univariate analysis ( p 0.005), and remained so on multivariate analysis ( p < 0.001) (Tables 4 and 5). Kaplan–Meier curves for tumours with >20% necrosis versus 20% necrosis are shown in Fig. 2.
DISCUSSION In this paper we have shown that tumour necrosis is a significant independent prognostic factor for clear cell renal cell carcinoma both in general and for organ confined disease as well as when evaluated with both traditional pathological factors and additional clinical laboratory data. In addition, we have shown that stratifying necrosis into different quantities creates subgroups with significantly different overall survival. Overall, the cases included in this series are similar to other studies, as has been discussed in prior publications.13 One notable difference is that in this series Fuhrman grading was not a significant prognostic factor. At the time of this study, Fuhrman grading was the standard for grading renal cell carcinoma. While there are several possible reasons for this, one factor may be that the grades were assigned by multiple board certified pathologists rather than a single pathologist. It
35
Table 1 Univariate analysis for all renal cell carcinomas for overall survival (n ¼ 446) Factor
p value
95% CI
Age Sex Partial nephrectomy (vs total nephrectomy) Grade Tumour size T2 stage (vs T1) T3a T3b T3c T4 N1 stage (vs N0) M1 stage (vs M0) Necrosis present Elevated LDH Elevated calcium (vs normal) Low calcium (vs normal) Elevated WBC (vs normal) Low WBC (vs normal) Elevated Hgb (vs normal) Low Hgb (vs normal) Elevated platelets (vs normal) Low platelets (vs normal)
<0.001 0.43 <0.001 0.99 <0.001 0.06 <0.001 <0.001 0.004 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 0.01 0.004 0.09 0.003 <0.001 <0.001 0.01
1.02–1.05 0.62–1.22 0.18–0.62 NA 1.09–1.19 0.98–3.15 2.32–5.24 2.24–6.16 1.92–33.49 3.95–25.64 1.81–7.94 3.15–6.20 1.06–1.08 3.56–3.99 1.98–8.87 0.42–0.89 0.38–0.83 0.87–4.59 1.79–19.00 1.31–2.83 2.10–5.16 1.16–3.11
CI, confidence interval; Hgb, haemoglobin; LDH, lactate dehydrogenase; NA, not applicable; WBC, white blood cell count.
has previously been shown that Fuhrman grading is not completely reproducible,2,4,14 and the results in this study may reflect that fact. In addition, the cases in this study may represent a selection bias for larger higher stage tumours (which may be more easily harvested for molecular studies). As others have also shown, in this series an elevated serum calcium was a significant negative prognostic factor for renal cell carcinoma in general.15–19 However, its significance in organ confined disease is less clear. Other authors in much larger series have shown that it is a significant prognostic factor in this setting as well, and for this factor our current series may be underpowered. Similar issues may also apply to the other laboratory values reviewed in this study.20–26 Although several laboratory values including serum calcium have been examined in series that assessed necrosis, a direct comparison between the two has not been previously made. Specifically, in one study the significance of laboratory values after stratifying for the SSIGN score (Stage, Size, Grade, and Necrosis), which includes tumour necrosis, was performed.15 To our knowledge the current study is the first study to directly compare the relative importance of these laboratory values against tumour necrosis as an independent value.
Table 2 Effect of tumour necrosis quantitation on overall survival for all renal cell carcinomas compared with no necrosis (n ¼ 446) Extent of necrosis
p value
95% CI
1–5% 6–10% 11–15% 16–20% 21–25% 26–30%
0.33 0.09 0.01 0.03 0.04 <0.001
0.75–2.36 0.90–4.07 1.21–5.45 1.14–11.95 1.09–18.66 4.43–9.65
CI, confidence interval.
Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.
36
Pathology (2015), 47(1), January
RENSHAW and CHEVILLE
Table 3 Multivariate analysis for overall survival on all renal cell carcinomas (n ¼ 446)
Table 4 Effect of tumour necrosis quantitation on overall survival for T1-T2N0M0 renal cell carcinomas compared with no necrosis (n ¼ 257)
Factor
p value
95% CI
Extent of necrosis
p value
95% CI
Age Partial nephrectomy (vs total nephrectomy) Tumour size T2 (vs T1) T3a T3b T3c T4 N1 stage (vs N0) M1 stage (vs M0) Necrosis 1–5% (vs none) Necrosis 6–10% Necrosis 11–15% Necrosis 16–20% Necrosis 21–25% Necrosis 25% Elevated LDH Elevated calcium Elevated WBC Low Hgb Elevated platelets
0.004 0.71 0.49 0.44 0.31 0.02 0.49 0.53 0.28 <0.001 0.76 0.14 0.06 0.10 0.14 <0.001 0.45 0.003 0.16 0.99 0.11
1.01–1.05 0.54–2.50 0.95–1.11 0.63–2.91 0.73–2.67 1.16–4.30 0.36–8.25 0.39–6.05 0.64–4.55 2.16–6.16 0.59–2.06 0.81–4.35 0.96–5.11 0.81–10.26 0.69–13.56 2.40–6.84 0.14–2.38 1.64–11.81 0.85–2.59 0.63–1.60 0.90–2.63
1–5% 6–10% 11–15% 16–20% 21–25% 26–30%
0.45 0.34 0.34 0.99 0.005 <0.001
0.54–3.99 0.47–8.96 0.47–8.87 NA 2.41–146.38 2.80–11.04
CI, confidence interval; Hgb, haemoglobin; LDH, lactate dehydrogenase; WBC, white blood cell count.
% survived
Table 5 Multivariate analysis for overall survival on T1–T2N0M0 renal cell carcinomas (n ¼ 257) Factor
p value
95% CI
Age Partial nephrectomy (vs total nephrectomy) Tumour size Necrosis 1–5% (vs none) Necrosis 6–10% Necrosis 11–15% Necrosis 16–20% Necrosis 21–25% Necrosis 25% Elevated LDH Elevated calcium Elevated WBC Low Hgb Elevated platelets
<0.001 0.21 0.85 0.21 0.17 0.76 0.99 <0.001 <0.001 0.84 NA 0.66 0.86 0.22
1.02–1.09 0.69–5.29 0.85–1.22 0.68–5.71 0.60–17.64 0.25–6.78 NA 6.03–811.84 2.19–18.30 0.07–22.65 NA 0.44–3.59 0.47–2.46 0.57–10.33
CI, confidence interval; Hgb, haemoglobin; LDH, lactate dehydrogenase; NA, not applicable; WBC, white blood cell count.
studies stratified cases by either quartiles or halves and included cases with over 75% necrosis. In the current study we stratified cases by 5% intervals and had no cases with necrosis of more than 30%. It appears likely that the differing results we obtained with necrosis quantification may be related to these different methods of stratification. In conclusion, we have shown that tumour necrosis is a significant independent prognostic factor for clear cell renal cell carcinoma both overall and in organ confined disease when evaluated against both traditional pathological factors and additional clinical laboratory data. In addition, we have shown for the first time that stratifying necrosis into different quantities creates subgroups with significantly different overall survival. Further evaluation of tumour necrosis quantitation in clear cell renal cell carcinoma appears warranted.
0.0 0.2 0.4 0.6 0.8 1.0
It is important to note that the method used to evaluate tumour necrosis in this study was very different than that used in all other studies. Other studies examined all of the routine histological slides and assessed necrosis for the tumour as a whole. In this study, tumour necrosis was assessed on a single frozen section slide of tissue submitted for molecular studies. As a result, this single slide likely represented tumour that grossly appeared well preserved. The relative importance of this selection criteria and the use of frozen sections rather than formalin fixed sections to make this evaluation is not entirely clear. In addition, the histological criteria that were used to determine necrosis in this study were not detailed. Nevertheless, the selection of a single area of the tumour for evaluation of pathological features rather than the entire tumour is not a new idea. Fuhrman grading of renal tumours and Nottingham grading of breast tumours are performed on selected regions of the tumour rather than the tumour as a whole. Our results suggest that evaluation of the extent of tumour necrosis may also be of value when performed on a single section of representative tumour selected from well preserved areas rather than assessed for the tumour as a whole. Finally, the method of quantitation of the necrosis in this study was different than that of previous studies. Previous
CI, confidence interval; NA, not applicable.
Necrosis <10% Necrosis >10%
0
500
1000
1500
2000
2500
3000
3500
Days Fig. 1 Overall survival for all clear cell renal cell carcinomas.
Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.
37
0.6 0.4
Necrosis<20% Necrosis>20%
0.0
0.2
% survived
0.8
1.0
NECROSIS IN CLEAR CELL RCC
0
500
1000
1500
2000
2500
3000
3500
Days Fig. 2 Overall survival for T1-T2N0M0 clear cell renal cell carcinomas.
Conflicts of interest and sources of funding: The authors state that there are no conflicts of interest to disclose. Address for correspondence: Dr Andrew Renshaw, Department of Pathology, Baptist Hospital of Miami, 8900 N. Kendall Dr, Miami, FL 33176, USA. E-mail: [email protected]
References 1. Cheville JC, Lohse CM, Zincke H, Weaver AL, Blute ML. Comparisons of outcome and prognostic features among histologic subtypes of renal cell carcinoma. Am J Surg Pathol 2003; 27: 612–24. 2. Ficarra V, Martignoni G, Lohse C, et al. External validation of the Mayo Clinic Stage, Size, Grade and Necrosis (SSIGN) score to predict cancer specific survival using a European series of conventional renal cell carcinoma. J Urol 2006; 175: 1235–9. 3. Delahunt B, McKenney JK, Lohse CM, et al. A novel grading system for clear cell renal cell carcinoma incorporating tumor necrosis. Am J Surg Pathol 2013; 37: 311–22. 4. Cheville JC, Blute ML, Zincke H, Lohse CM, Weaver AL. Stage pT1 conventional (clear cell) renal cell carcinmoa: pathological features associated with cancer specific survival. J Urol 2001; 166: 453–6. 5. Sengupta S, Lohse CM, Leibovich BC, et al. Histologic coagulative tumor necrosis as a prognostic indicator of renal cell carcinoma aggressiveness. Cancer 2005; 104: 511–20. 6. Katz MD, Serrano MF, Grubb RL 3rd, et al. Percent microscopic tumor necrosis and survival after curative surgery for renal cell carcinoma. J Urol 2010; 183: 909–14. 7. Chang YH, Chuang CK, Pang ST, et al. Prognostic value of TNM stage and tumor necrosis for renal cell carcinoma. Kaohsiung J Med Sci 2011; 27: 59–63. 8. Lam JS, Shvarts O, Said JW, et al. Clinicopathologic and molecular correlations of necrosis in the primary tumor of patients with renal cell carcinoma. Cancer 2005; 103: 2517–25. 9. Moch H, Gasser T, Amin MB, Torhorst J, Sauter G, Mihatsch MJ. Prognostic utility of the recently recommended histologic classification and revised TNM staging system of renal cell carcinoma: a Swiss experience with 588 tumors. Cancer 2000; 89: 604–14. 10. Breau RH, Cheville JC, Lohse CM, Kwon ED, Blute ML. Re: Presence of tumor necrosis is not a significant predictor of survival in clear cell renal cell carcinoma: higher prognostic accuracy of extent based rather than presence/absence classification. T. Klatte, J. W. Said, M. de Martino, J. Larochelle, B. Shuch, J. Y. Rao, G. V. Thomas, F. F. Kabbinavar, A. S. Belldegrun and A. J. Pantuck. J Urol 2009; 181: 1558–1564. J Urol 2009; 182: 2979–80; author reply 80–1. 11. Pichler M, Hutterer GC, Chromecki TF, et al. Histologic tumor necrosis is an independent prognostic indicator for clear cell and papillary renal cell carcinoma. Am J Clin Pathol 2012; 137: 283–9.
12. Klatte T, Said JW, de Martino M, et al. Presence of tumor necrosis is not a significant predictor of survival in clear cell renal cell carcinoma: higher prognostic accuracy of extent based rather than presence/absence classification. J Urol 2009; 181: 1558–64. 13. Cancer Genome Atlas Research Network. Comprehensive molecular characterization of clear cell renal cell carcinoma. Nature 2013; 499: 43–9. 14. Sika-Paotonu D, Bethwaite PB, McCredie MR, Jordan WT, Delahunt B. Nucleolar grade but not Fuhrman grade is applicable to papillary renal cell carcinoma. Am J Surg Pathol 2006; 30: 1091–6. 15. Magera JS Jr, Leibovich BC, Lohse CM, et al. Association of abnormal preoperative laboratory values with survival after radical nephrectomy for clinically confined clear cell renal cell carcinoma. Urology 2008; 71: 278– 82. 16. Chasan SA, Pothel LR, Huben RP. Management and prognostic significance of hypercalcemia in renal cell carcinoma. Urology 1989; 33: 167–70. 17. Fahn HJ, Lee YH, Chen MT, Huang JK, Chen KK, Chang LS. The incidence and prognostic significance of humoral hypercalcemia in renal cell carcinoma. J Urol 1991; 145: 248–50. 18. Motzer RJ, Escudier B, Bukowski R, et al. Prognostic factors for survival in 1059 patients treated with sunitinib for metastatic renal cell carcinoma. Br J Cancer 2013; 108: 2470–7. 19. Yoon HJ, Paeng JC, Kwak C, et al. Prognostic implication of extrarenal metabolic tumor burden in advanced renal cell carcinoma treated with targeted therapy after nephrectomy. Ann Nucl Med 2013; 27: 748–55. 20. Lee SE, Byun SS, Han JH, Han BK, Hong SK. Prognostic significance of common preoperative laboratory variables in clear cell renal cell carcinoma. BJU Int 2006; 98: 1228–32. 21. Heng DY, Xie W, Regan MM, et al. Prognostic factors for overall survival in patients with metastatic renal cell carcinoma treated with vascular endothelial growth factor-targeted agents: results from a large, multicenter study. J Clin Oncol 2009; 27: 5794–9. 22. Lehmann J, Retz M, Nurnberg N, et al. The superior prognostic value of humoral factors compared with molecular proliferation markers in renal cell carcinoma. Cancer 2004; 101: 1552–62. 23. Kim HL, Belldegrun AS, Freitas DG, et al. Paraneoplastic signs and symptoms of renal cell carcinoma: implications for prognosis. J Urol 2003; 170: 1742–6. 24. Azuma T, Matayoshi Y, Nagase Y, Oshi M. Neutrophil number after interferon-alfa treatment is an independent predictive marker of overall survival in metastatic renal cell carcinoma. Clin Genitourin Cancer 2012; 10: 180–4. 25. Du J, Zheng JH, Chen XS, et al. High preoperative plasma fibrinogen is an independent predictor of distant metastasis and poor prognosis in renal cell carcinoma. Int J Clin Oncol 2013; 18: 517–23. 26. Fox P, Hudson M, Brown C, et al. Markers of systemic inflammation predict survival in patients with advanced renal cell cancer. Br J Cancer 2013; 109: 147–53.
Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.
ANATOMICAL PATHOLOGY
Quantitative tumour necrosis is an independent predictor of overall survival in clear cell renal cell carcinoma ANDREW A. RENSHAW1
AND
JOHN C. CHEVILLE2
1Department of Pathology, Baptist Hospital of Miami, Miami, FL, and 2Professor of Laboratory Medicine and
Pathology, Department of Pathology, Mayo Clinic, Rochester, MN, United States
Summary Previous studies have reached conflicting results regarding whether tumour necrosis is a predictor of survival in clear cell renal cell carcinoma. In addition, studies quantifying the extent of necrosis are limited. The aim of this study was to determine if quantifying tumour necrosis could improve its predictive value for survival in clear cell renal cell carcinoma. We reviewed the clinical pathological information contained in The Cancer Genome Atlas for clear cell renal cell carcinoma and correlated it with overall survival using a Cox proportional hazard model. Necrosis was quantified on a single frozen section slide taken at the time of tissue harvesting for molecular studies. For all tumours, the presence of tumour necrosis was a significant predictor of overall survival ( p < 0.001) on univariate analysis. When quantitated, >10% necrosis was associated with survival, but 10% necrosis was not. On multivariate analysis, age ( p ¼ 0.004), T3b stage ( p ¼ 0.02), M1 stage ( p < 0.001), necrosis >30% ( p < 0.001), and elevated serum calcium ( p ¼ 0.003) remained significant. For clinical stage 1–2 (T1-T2N0M0) tumours, necrosis >20% was significant on univariate analysis ( p 0.005), and remained so on multivariate analysis ( p < 0.001). We conclude that quantitating the extent of tumour necrosis adds prognostic information in clear cell renal cell carcinomas, including organ confined tumours. Key words: Clear cell, necrosis, prognosis, quantitative, renal cell carcinoma, survival. Received 23 June, revised 7 August, accepted 13 August 2014
INTRODUCTION Previous studies have shown conflicting results regarding whether coagulative tumour necrosis is a predictor of survival in renal cell carcinoma. Most studies have agreed that the presence of tumour necrosis is associated with worse cause specific survival1–6 and overall survival7–11 in renal cell carcinomas of clear cell renal cell and chromophobe type,1 including in organ confined (T1-T2N0M0) clear cell tumours.4,8 Nevertheless, some studies have failed to show that tumour necrosis is an independent predictor of prognosis when considered in the context of other prognostic features.12 In addition, previous studies examining the extent of necrosis are limited. To our knowledge only two previous studies attempted to stratify the extent of necrosis.6,8 One study suggested that tumour necrosis of greater than 50% of the entire tumour was associated with worse disease specific Print ISSN 0031-3025/Online ISSN 1465-3931 DOI: 10.1097/PAT.0000000000000193
#
survival but not overall survival, and the other suggested that quantification was not of value in predicting overall survival. However, there were significant limitations in the way in which the extent of necrosis was quantified in these studies. In particular, the extent of necrosis was estimated for the entire tumour and the extent of necrosis was stratified into a few very large groups. Stratification into smaller groups, especially in those tumours with lesser degrees of necrosis, was not assessed. We recently noticed that in a previously published study addressing the molecular characteristics of clear cell renal cell carcinoma from the Tumour Cancer Genome Atlas (TCGA),13 tumour necrosis was assessed as part of a quality assessment measure for molecular studies. In this study, tumour necrosis was estimated from examination of a single frozen section slide of tissue that was taken for molecular studies. We wondered if this assessment might shed light on the significance of quantifying necrosis in clear cell renal cell carcinoma.
METHODS All data were taken from the supplementary tables of the Cancer Genome Atlas Research Network publication13 (Data File S2, clinical dataset) which also contain additional methodological information. In brief and as described in the methods of that paper and paraphrased here, tumours were flash-frozen and shipped to a centralised processing centre (Biospecimen Core Resource, BCR) for additional pathological review and nucleic acids extraction. Biospecimens were collected from newly diagnosed patients with renal clear cell carcinoma undergoing surgical resection who had received no prior treatment for their disease, including chemotherapy or radiotherapy. All cases were collected, regardless of surgical stage or histological grade. Cases were staged according to the American Joint Committee on Cancer (AJCC) staging system. Cases were graded using the Fuhrman grading system, which was the standard at the time the study was performed. Each frozen tumour specimen submitted to the BCR weighed at least 30 mg and was typically under 200 mg. Specimens were shipped overnight from one of 13 tissue source sites using a cryoport that maintained an average temperature of less than –1808C. Tumour necrosis was assessed microscopically from a single frozen section of this biorepository specimen. Complete clinical data elements were compiled for all specimens included and reflected current data as of 13 April 2012. Clinical/demographic data included: sample code, primary site (kidney for all specimens), gender, age at diagnosis, race, ethnicity, and year of tumour collection. Samples of questionable authenticity as clear cell RCC tumours due to unexpected molecular analysis results were evaluated by secondary pathological review. Tumour information recorded complete pathological information regarding the tumour. All specimens included in this analysis were coded as kidney clear cell renal carcinoma. The table records: laterality (right/left), Fuhrman nuclear grade, maximum tumour dimension (cm), T stage, lymph node involvement (based on pathological staging), and M stage (intended to be indicative of a review of clinical evidence for metastatic disease, but was often provided from available pathological information only, so should be interpreted with caution).
2014 Royal College of Pathologists of Australasia
Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.
NECROSIS IN CLEAR CELL RCC
A compiled tumour stage using standard AJCC staging criteria using the Tumour Node Metastasis universal schema was reported. Pathological data came from board certified pathologists in the originating institution. Additional laboratory data including lactate dehydrogenase, calcium, haemoglobin, white blood cell count, and platelet count were also included. Clinical status of patients at the point of enrolment, and as available at last follow-up was recorded. Sites were asked to indicate if patients following surgical resection were tumour free, or with tumour. We also recorded vital status (living/deceased) at the time of enrolment. Follow-up data were requested for subjects out to a minimum of 2 years from the time of sample collection. The patient tumour status (tumour free/with tumour) was again recorded, along with vital status (living/deceased) from the most recent followup data form completion at the time of data collection. Time to recurrence was recorded as the number of days to a new tumour event. We also recorded the days to last contact at the point of enrolment or most recent follow up. Finally, the days from diagnosis (sample collection) to death were recorded at both enrolment and in the most recent follow-up forms. These data provided the information to explore survival-based outcomes and median follow-up for patients included in this study. Cause of death, from cancer or other causes, was not recorded. Survival analysis was performed using a Cox proportional hazard ratio (HR). A threshold of p < 0.05 was considered significant.
RESULTS A total of 446 cases were included and there were 145 deaths; 257 organ confined (T1-T2N0M0) cases were included, and included 45 deaths. Tumour necrosis was present in 181 (40.6%) tumours, and 79 (30.7%) organ confined tumours, and ranged from 1–30%. Necrosis of greater than 5% was present in 34% of tumours. For all cases, age, type of surgery, size, T stage, M stage, presence of necrosis, elevated calcium, elevated WBC, low haemoglobin, and elevated platelet count were all significantly associated with overall survival on univariate analysis (Table 1). When quantitated, necrosis >10% was associated with survival, but necrosis 10% was not (Table 2). On multivariate analysis, age ( p ¼ 0.004), T3b stage ( p ¼ 0.02), M1 stage ( p < 0.001), necrosis >10% ( p 0.04), necrosis >30% ( p < 0.001), and elevated serum calcium ( p ¼ 0.003) remained significant (Table 3). Kaplan–Meier curves for tumours with >10% necrosis versus 10% necrosis are shown in Fig. 1. For clinical stage 1–2 (T1-T2N0M0) tumours, necrosis >20% was significant on univariate analysis ( p 0.005), and remained so on multivariate analysis ( p < 0.001) (Tables 4 and 5). Kaplan–Meier curves for tumours with >20% necrosis versus 20% necrosis are shown in Fig. 2.
DISCUSSION In this paper we have shown that tumour necrosis is a significant independent prognostic factor for clear cell renal cell carcinoma both in general and for organ confined disease as well as when evaluated with both traditional pathological factors and additional clinical laboratory data. In addition, we have shown that stratifying necrosis into different quantities creates subgroups with significantly different overall survival. Overall, the cases included in this series are similar to other studies, as has been discussed in prior publications.13 One notable difference is that in this series Fuhrman grading was not a significant prognostic factor. At the time of this study, Fuhrman grading was the standard for grading renal cell carcinoma. While there are several possible reasons for this, one factor may be that the grades were assigned by multiple board certified pathologists rather than a single pathologist. It
35
Table 1 Univariate analysis for all renal cell carcinomas for overall survival (n ¼ 446) Factor
p value
95% CI
Age Sex Partial nephrectomy (vs total nephrectomy) Grade Tumour size T2 stage (vs T1) T3a T3b T3c T4 N1 stage (vs N0) M1 stage (vs M0) Necrosis present Elevated LDH Elevated calcium (vs normal) Low calcium (vs normal) Elevated WBC (vs normal) Low WBC (vs normal) Elevated Hgb (vs normal) Low Hgb (vs normal) Elevated platelets (vs normal) Low platelets (vs normal)
<0.001 0.43 <0.001 0.99 <0.001 0.06 <0.001 <0.001 0.004 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 0.01 0.004 0.09 0.003 <0.001 <0.001 0.01
1.02–1.05 0.62–1.22 0.18–0.62 NA 1.09–1.19 0.98–3.15 2.32–5.24 2.24–6.16 1.92–33.49 3.95–25.64 1.81–7.94 3.15–6.20 1.06–1.08 3.56–3.99 1.98–8.87 0.42–0.89 0.38–0.83 0.87–4.59 1.79–19.00 1.31–2.83 2.10–5.16 1.16–3.11
CI, confidence interval; Hgb, haemoglobin; LDH, lactate dehydrogenase; NA, not applicable; WBC, white blood cell count.
has previously been shown that Fuhrman grading is not completely reproducible,2,4,14 and the results in this study may reflect that fact. In addition, the cases in this study may represent a selection bias for larger higher stage tumours (which may be more easily harvested for molecular studies). As others have also shown, in this series an elevated serum calcium was a significant negative prognostic factor for renal cell carcinoma in general.15–19 However, its significance in organ confined disease is less clear. Other authors in much larger series have shown that it is a significant prognostic factor in this setting as well, and for this factor our current series may be underpowered. Similar issues may also apply to the other laboratory values reviewed in this study.20–26 Although several laboratory values including serum calcium have been examined in series that assessed necrosis, a direct comparison between the two has not been previously made. Specifically, in one study the significance of laboratory values after stratifying for the SSIGN score (Stage, Size, Grade, and Necrosis), which includes tumour necrosis, was performed.15 To our knowledge the current study is the first study to directly compare the relative importance of these laboratory values against tumour necrosis as an independent value.
Table 2 Effect of tumour necrosis quantitation on overall survival for all renal cell carcinomas compared with no necrosis (n ¼ 446) Extent of necrosis
p value
95% CI
1–5% 6–10% 11–15% 16–20% 21–25% 26–30%
0.33 0.09 0.01 0.03 0.04 <0.001
0.75–2.36 0.90–4.07 1.21–5.45 1.14–11.95 1.09–18.66 4.43–9.65
CI, confidence interval.
Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.
36
Pathology (2015), 47(1), January
RENSHAW and CHEVILLE
Table 3 Multivariate analysis for overall survival on all renal cell carcinomas (n ¼ 446)
Table 4 Effect of tumour necrosis quantitation on overall survival for T1-T2N0M0 renal cell carcinomas compared with no necrosis (n ¼ 257)
Factor
p value
95% CI
Extent of necrosis
p value
95% CI
Age Partial nephrectomy (vs total nephrectomy) Tumour size T2 (vs T1) T3a T3b T3c T4 N1 stage (vs N0) M1 stage (vs M0) Necrosis 1–5% (vs none) Necrosis 6–10% Necrosis 11–15% Necrosis 16–20% Necrosis 21–25% Necrosis 25% Elevated LDH Elevated calcium Elevated WBC Low Hgb Elevated platelets
0.004 0.71 0.49 0.44 0.31 0.02 0.49 0.53 0.28 <0.001 0.76 0.14 0.06 0.10 0.14 <0.001 0.45 0.003 0.16 0.99 0.11
1.01–1.05 0.54–2.50 0.95–1.11 0.63–2.91 0.73–2.67 1.16–4.30 0.36–8.25 0.39–6.05 0.64–4.55 2.16–6.16 0.59–2.06 0.81–4.35 0.96–5.11 0.81–10.26 0.69–13.56 2.40–6.84 0.14–2.38 1.64–11.81 0.85–2.59 0.63–1.60 0.90–2.63
1–5% 6–10% 11–15% 16–20% 21–25% 26–30%
0.45 0.34 0.34 0.99 0.005 <0.001
0.54–3.99 0.47–8.96 0.47–8.87 NA 2.41–146.38 2.80–11.04
CI, confidence interval; Hgb, haemoglobin; LDH, lactate dehydrogenase; WBC, white blood cell count.
% survived
Table 5 Multivariate analysis for overall survival on T1–T2N0M0 renal cell carcinomas (n ¼ 257) Factor
p value
95% CI
Age Partial nephrectomy (vs total nephrectomy) Tumour size Necrosis 1–5% (vs none) Necrosis 6–10% Necrosis 11–15% Necrosis 16–20% Necrosis 21–25% Necrosis 25% Elevated LDH Elevated calcium Elevated WBC Low Hgb Elevated platelets
<0.001 0.21 0.85 0.21 0.17 0.76 0.99 <0.001 <0.001 0.84 NA 0.66 0.86 0.22
1.02–1.09 0.69–5.29 0.85–1.22 0.68–5.71 0.60–17.64 0.25–6.78 NA 6.03–811.84 2.19–18.30 0.07–22.65 NA 0.44–3.59 0.47–2.46 0.57–10.33
CI, confidence interval; Hgb, haemoglobin; LDH, lactate dehydrogenase; NA, not applicable; WBC, white blood cell count.
studies stratified cases by either quartiles or halves and included cases with over 75% necrosis. In the current study we stratified cases by 5% intervals and had no cases with necrosis of more than 30%. It appears likely that the differing results we obtained with necrosis quantification may be related to these different methods of stratification. In conclusion, we have shown that tumour necrosis is a significant independent prognostic factor for clear cell renal cell carcinoma both overall and in organ confined disease when evaluated against both traditional pathological factors and additional clinical laboratory data. In addition, we have shown for the first time that stratifying necrosis into different quantities creates subgroups with significantly different overall survival. Further evaluation of tumour necrosis quantitation in clear cell renal cell carcinoma appears warranted.
0.0 0.2 0.4 0.6 0.8 1.0
It is important to note that the method used to evaluate tumour necrosis in this study was very different than that used in all other studies. Other studies examined all of the routine histological slides and assessed necrosis for the tumour as a whole. In this study, tumour necrosis was assessed on a single frozen section slide of tissue submitted for molecular studies. As a result, this single slide likely represented tumour that grossly appeared well preserved. The relative importance of this selection criteria and the use of frozen sections rather than formalin fixed sections to make this evaluation is not entirely clear. In addition, the histological criteria that were used to determine necrosis in this study were not detailed. Nevertheless, the selection of a single area of the tumour for evaluation of pathological features rather than the entire tumour is not a new idea. Fuhrman grading of renal tumours and Nottingham grading of breast tumours are performed on selected regions of the tumour rather than the tumour as a whole. Our results suggest that evaluation of the extent of tumour necrosis may also be of value when performed on a single section of representative tumour selected from well preserved areas rather than assessed for the tumour as a whole. Finally, the method of quantitation of the necrosis in this study was different than that of previous studies. Previous
CI, confidence interval; NA, not applicable.
Necrosis <10% Necrosis >10%
0
500
1000
1500
2000
2500
3000
3500
Days Fig. 1 Overall survival for all clear cell renal cell carcinomas.
Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.
37
0.6 0.4
Necrosis<20% Necrosis>20%
0.0
0.2
% survived
0.8
1.0
NECROSIS IN CLEAR CELL RCC
0
500
1000
1500
2000
2500
3000
3500
Days Fig. 2 Overall survival for T1-T2N0M0 clear cell renal cell carcinomas.
Conflicts of interest and sources of funding: The authors state that there are no conflicts of interest to disclose. Address for correspondence: Dr Andrew Renshaw, Department of Pathology, Baptist Hospital of Miami, 8900 N. Kendall Dr, Miami, FL 33176, USA. E-mail: [email protected]
References 1. Cheville JC, Lohse CM, Zincke H, Weaver AL, Blute ML. Comparisons of outcome and prognostic features among histologic subtypes of renal cell carcinoma. Am J Surg Pathol 2003; 27: 612–24. 2. Ficarra V, Martignoni G, Lohse C, et al. External validation of the Mayo Clinic Stage, Size, Grade and Necrosis (SSIGN) score to predict cancer specific survival using a European series of conventional renal cell carcinoma. J Urol 2006; 175: 1235–9. 3. Delahunt B, McKenney JK, Lohse CM, et al. A novel grading system for clear cell renal cell carcinoma incorporating tumor necrosis. Am J Surg Pathol 2013; 37: 311–22. 4. Cheville JC, Blute ML, Zincke H, Lohse CM, Weaver AL. Stage pT1 conventional (clear cell) renal cell carcinmoa: pathological features associated with cancer specific survival. J Urol 2001; 166: 453–6. 5. Sengupta S, Lohse CM, Leibovich BC, et al. Histologic coagulative tumor necrosis as a prognostic indicator of renal cell carcinoma aggressiveness. Cancer 2005; 104: 511–20. 6. Katz MD, Serrano MF, Grubb RL 3rd, et al. Percent microscopic tumor necrosis and survival after curative surgery for renal cell carcinoma. J Urol 2010; 183: 909–14. 7. Chang YH, Chuang CK, Pang ST, et al. Prognostic value of TNM stage and tumor necrosis for renal cell carcinoma. Kaohsiung J Med Sci 2011; 27: 59–63. 8. Lam JS, Shvarts O, Said JW, et al. Clinicopathologic and molecular correlations of necrosis in the primary tumor of patients with renal cell carcinoma. Cancer 2005; 103: 2517–25. 9. Moch H, Gasser T, Amin MB, Torhorst J, Sauter G, Mihatsch MJ. Prognostic utility of the recently recommended histologic classification and revised TNM staging system of renal cell carcinoma: a Swiss experience with 588 tumors. Cancer 2000; 89: 604–14. 10. Breau RH, Cheville JC, Lohse CM, Kwon ED, Blute ML. Re: Presence of tumor necrosis is not a significant predictor of survival in clear cell renal cell carcinoma: higher prognostic accuracy of extent based rather than presence/absence classification. T. Klatte, J. W. Said, M. de Martino, J. Larochelle, B. Shuch, J. Y. Rao, G. V. Thomas, F. F. Kabbinavar, A. S. Belldegrun and A. J. Pantuck. J Urol 2009; 181: 1558–1564. J Urol 2009; 182: 2979–80; author reply 80–1. 11. Pichler M, Hutterer GC, Chromecki TF, et al. Histologic tumor necrosis is an independent prognostic indicator for clear cell and papillary renal cell carcinoma. Am J Clin Pathol 2012; 137: 283–9.
12. Klatte T, Said JW, de Martino M, et al. Presence of tumor necrosis is not a significant predictor of survival in clear cell renal cell carcinoma: higher prognostic accuracy of extent based rather than presence/absence classification. J Urol 2009; 181: 1558–64. 13. Cancer Genome Atlas Research Network. Comprehensive molecular characterization of clear cell renal cell carcinoma. Nature 2013; 499: 43–9. 14. Sika-Paotonu D, Bethwaite PB, McCredie MR, Jordan WT, Delahunt B. Nucleolar grade but not Fuhrman grade is applicable to papillary renal cell carcinoma. Am J Surg Pathol 2006; 30: 1091–6. 15. Magera JS Jr, Leibovich BC, Lohse CM, et al. Association of abnormal preoperative laboratory values with survival after radical nephrectomy for clinically confined clear cell renal cell carcinoma. Urology 2008; 71: 278– 82. 16. Chasan SA, Pothel LR, Huben RP. Management and prognostic significance of hypercalcemia in renal cell carcinoma. Urology 1989; 33: 167–70. 17. Fahn HJ, Lee YH, Chen MT, Huang JK, Chen KK, Chang LS. The incidence and prognostic significance of humoral hypercalcemia in renal cell carcinoma. J Urol 1991; 145: 248–50. 18. Motzer RJ, Escudier B, Bukowski R, et al. Prognostic factors for survival in 1059 patients treated with sunitinib for metastatic renal cell carcinoma. Br J Cancer 2013; 108: 2470–7. 19. Yoon HJ, Paeng JC, Kwak C, et al. Prognostic implication of extrarenal metabolic tumor burden in advanced renal cell carcinoma treated with targeted therapy after nephrectomy. Ann Nucl Med 2013; 27: 748–55. 20. Lee SE, Byun SS, Han JH, Han BK, Hong SK. Prognostic significance of common preoperative laboratory variables in clear cell renal cell carcinoma. BJU Int 2006; 98: 1228–32. 21. Heng DY, Xie W, Regan MM, et al. Prognostic factors for overall survival in patients with metastatic renal cell carcinoma treated with vascular endothelial growth factor-targeted agents: results from a large, multicenter study. J Clin Oncol 2009; 27: 5794–9. 22. Lehmann J, Retz M, Nurnberg N, et al. The superior prognostic value of humoral factors compared with molecular proliferation markers in renal cell carcinoma. Cancer 2004; 101: 1552–62. 23. Kim HL, Belldegrun AS, Freitas DG, et al. Paraneoplastic signs and symptoms of renal cell carcinoma: implications for prognosis. J Urol 2003; 170: 1742–6. 24. Azuma T, Matayoshi Y, Nagase Y, Oshi M. Neutrophil number after interferon-alfa treatment is an independent predictive marker of overall survival in metastatic renal cell carcinoma. Clin Genitourin Cancer 2012; 10: 180–4. 25. Du J, Zheng JH, Chen XS, et al. High preoperative plasma fibrinogen is an independent predictor of distant metastasis and poor prognosis in renal cell carcinoma. Int J Clin Oncol 2013; 18: 517–23. 26. Fox P, Hudson M, Brown C, et al. Markers of systemic inflammation predict survival in patients with advanced renal cell cancer. Br J Cancer 2013; 109: 147–53.
Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.