Int Urol Nephrol (2014) 46:2127–2132 DOI 10.1007/s11255-014-0780-4
UROLOGY - ORIGINAL PAPER
Renal insufficiency is associated with an increased risk of papillary renal cell carcinoma histology Solomon L. Woldu • Aaron C. Weinberg • Arindam RoyChoudhury Herbert Chase • Sean D. Kalloo • James M. McKiernan • G. Joel DeCastro
•
Received: 11 March 2014 / Accepted: 24 June 2014 / Published online: 8 July 2014 Ó Springer Science+Business Media Dordrecht 2014
Abstract Purpose End-stage renal disease (ESRD) and acquired renal cystic disease associated with dialysis are known risk factors of papillary renal cell carcinoma (pRCC); however, it is not known whether renal insufficiency alone is a risk factor for pRCC. Our aim was to test whether renal insufficiency is associated with an increased preponderance of pRCC. Methods Retrospective review of institutional database to identify all patients who underwent extirpative renal surgery for renal cell carcinoma (RCC) with complete records from 1992 to 2012. We excluded those patients with preoperative ESRD as defined by GFR \ 15 mL/min/1.73 m2. The dependent variable was histologic RCC subtype. Independent variables included demographic data, comorbidities, and renal functional data. Multivariate analysis by binary logistic regression was used to determine factors that independently were associated with pRCC development. S. L. Woldu (&) A. C. Weinberg J. M. McKiernan G. J. DeCastro Department of Urology, Columbia University Medical Center, Herbert Irving Pavilion – 11th Floor 161 Ft. Washington Ave, New York, NY 10032, USA e-mail: [email protected]; [email protected] A. RoyChoudhury Department of Biostatistics, Columbia University Medical Center, New York, NY, USA H. Chase Department of Biomedical Informatics, Columbia University Medical Center, New York, NY, USA S. D. Kalloo Department of Nephrology, Columbia University Medical Center, New York, NY, USA
Results A total of 1,226 patients met inclusion criteria, of which 15 % were pRCC. There was a positive association between likelihood of pRCC histology of RCC and increasing preoperative chronic kidney disease (CKD) stage (p = 0.021). Multivariate regression analysis indicated that male gender, race, and declining renal function categorized both by GFR and CKD stage were independently associated with a higher likelihood of pRCC histology as compared to other RCC histology. Conclusions Within a large cohort of patients with a diagnosis of RCC, declining renal function was independently associated with an increased likelihood of pRCC histology. This finding and the available molecular evidence indicating protein expression similarity between pRCC and resident stem cells, which appear to be upregulated with kidney damage, suggest a possible causal relationship between renal injury and pRCC. Keywords Renal cell carcinoma Papillary Renal insufficiency Risk factor
Introduction Renal cell carcinoma (RCC) is the sixth most common cancer in men and eighth most common cancer in women in the USA [1]. Its incidence is rising, in part due to increased detection of asymptomatic renal masses related to increased use of cross-sectional imaging [2]. There are a variety of RCC histologic subtypes; clear cell RCC represents more than 70 % of cases, followed by papillary (10–15 %), and less commonly chromophobe RCC, oncocytoma, collecting duct RCC, and other rarer subtypes [3]. Each histologic subtype is believed to be the result of distinct genetic pathways and possibly from
123
2128
distinct cell lineages within the kidney. RCC is an adenocarcinoma and therefore arises from the tubular epithelium of the nephron. Much of our understanding of the etiology and distinctness of the subtypes of RCC comes from studying patients with a familial RCC predisposition syndrome. Papillary RCC (pRCC) is the second most common histologic subtype and appears to be both morphologically and genetically distinct. pRCC is characterized by trisomy of chromosomes 7 and 17 and other abnormalities on chromosomes 1, 12, 16, 20, and Y. Researchers have demonstrated mutations in the c-MET proto-oncogene on chromosome 7 in patients with hereditary pRCC, a familial predisposition syndrome [4]. However, these mutations are reported only in a small subset of pRCC overall, and the driving factor for the majority of mutations has not been elucidated [5]. There is evidence that end-stage renal disease (ESRD) and hemodialysis are risk factors for the development of acquired renal cystic disease, which in turn is strongly associated with the development of pRCC [6]. Furthermore, there are recent reports of the presence of a stem cell population located within the adult human kidney that can respond and proliferate in response to tubular injury [7]. Most provocatively, these purported stems cells share protein expression patterns with pRCC [8]. We sought to determine whether renal insufficiency alone, without ESRD or dialysis, is associated with an increased likelihood of pRCC histology at the time of nephrectomy for RCC.
Patients and methods Columbia University Medical Center has maintained an Institutional Review Board approved database of all renal surgeries performed for suspected malignancy. We reviewed this database from 1992 to 2012 to analyze all patients with pathologically confirmed RCC and known preoperative renal function. Data obtained included age, gender, race, BMI, comorbidities (HTN, DM, obesity, smoking status), and preoperative renal function as estimated by serum creatinine and by the glomerular filtration rate (GFR) calculated by the Modification of Diet and Renal Disease formula [9]. Patients with ESRD as defined as GFR \15 mL/min/1.73 m2 were excluded, as this is a known risk factor for the development of pRCC, and the aim of our study was to define the relationship between the risk of pRCC and renal function in patients without ESRD and the requirement of dialysis. Our dependent variable was whether the histology of the RCC was papillary or non-papillary. We separately analyzed the independent variable of renal function as a continuous variable (GFR),
123
Int Urol Nephrol (2014) 46:2127–2132
categorical variable (stage of CKD), and the presence or absence of CKD (as defined by a GFR \60 mL/min/ 1.73 m2). CKD staging was defined as follows: stage 1: GFR C90 mL/min/1.73 m2, stage 2: GFR 60–89 mL/min/ 1.73 m2, stage 3: 30–59 mL/min/1.73 m2, stage 4: 15–29 mL/min/1.73 m2, and stage 5 (ESRD): GFR \15 mL/min/1.73 m2 [10]. Despite collecting data on patient comorbidities associated with developing CKD, we were unable to attribute the CKD to a specific cause with our retrospective database. There was no distinction made between Type 1 and Type 2 pRCC histology, as the pathologists did not consistently report this distinction. ANOVA was performed to compare likelihood of dependent variables based on categorical independent variables. Multivariate analysis by binary logistic regression was used to determine factors that were independently associated with dependent variables. SPSS Version 21.0 was used for statistical analysis (IBM Corp, Armonk, NY). p value of \0.05 was considered significant.
Results A total of 1,226 patients were identified that met our inclusion criteria. Tumor histology was classified as clear cell RCC in 806 (65.7 %), pRCC in 184 (15.0 %), chromophobe RCC in 122 (10.0 %), oncocytoma in 112 (9.1 %), and other unspecified RCC in 2 (0.2 %) patients. CKD staging was 1 for 319 patients, 2 for 633 patients, 3 for 257 patients, and 4 for 17 patients. When stratifying histology into papillary versus nonpapillary RCC (Table 1), those patients diagnosed with a pRCC were more likely to be male (87 % male vs. 64 % female, p = 0.001) and have worse baseline kidney function, as indicated by a lower mean GFR (72.0 vs. 76.8 mL/ min/1.73 m2, p = 0.004), and increased likelihood of being defined as having CKD (29 % with CKD vs. 21 % without CKD, p = 0.013). There was a strong positive association between likelihood of papillary histology of RCC and increasing stage of CKD, p = 0.021 (Table 1; Fig. 1). Multivariate regression analysis to determine predictors of pRCC histology indicated that male gender, race, and worsening renal function were independently associated with a higher likelihood of pRCC histology as compared to other RCC histology (Table 2). Worsening renal function remained consistently an independent factor associated with increased risk of pRCC versus other RCC histology when analyzed separately: (1) as a continuous variable (GFR), (2) as a dichotomous variable defined by the presence or absence of CKD as defined as GFR \60 mL/min/ 1.73 m2, and (3) as an ordinal variable defined by specific CKD stage. An increasing stage of CKD was associated
Int Urol Nephrol (2014) 46:2127–2132
2129
with a step-wise increase in risk of having papillary RCC histology; when compared to CKD stage 1 patients, CKD stage 2 patients had an adjusted OR 1.48 (95 % CI 0.94–2.33), CKD stage 3 patients had an adjusted OR 2.20 (95 % CI 1.27–3.80), and CKD stage 4 patients had an
adjusted OR 2.95 (95 % CI 0.85–10.20) of being diagnosed with pRCC versus another histologic subtype.
Table 1 Patient characteristics
The pathogenesis of pRCC is not well elucidated. While Hereditary Papillary RCC Syndrome is believed to be the result of mutations in the c-MET oncogene on chromosome 7, this accounts for the small minority of cases of pRCC [4]. Beyond lacking clarity on the genetic alterations that lead to the development of most cases of pRCC, there is also a lack of consensus on the progenitor cell type in the tubule that leads to pRCC. pRCC has often been reported to originate from the proximal tubule, but this is controversial [11]. Yang et al. recently reported on 35 cases of pRCC, which expressed markers consistent with both proximal and distal tubule expression patterns. In addition, the pRCC also expressed markers absent from both proximal and distal tubules, and the authors concluded that pRCC was more likely to be derived from multipotent stem cells rather than a specific tubular location [12]. While most cases of RCC are thought to be sporadic, there are certain factors that appear to result in an increased risk of developing RCC. Cigarette smoking is the most consistently established risk factor and accounts for about 20 % of RCC cases. There also appears to be an association between obesity and increased risk of RCC [13]. Hypertension and its treatment with antihypertensive medications have been postulated to be associated with an increased risk. However, the interpretation of the literature on this subject is difficult, as pre-diagnostic renal tumors may lead to an increased blood pressure and imaging for hypertension can lead to the discovery of incidental renal masses [14]. Furthermore, it is difficult to separate the influences
Histology Nonpapillary RCC
Papillary RCC
p value
61.80
62.13
0.748
Female
376 (36 %)
24 (13 %)
0.001*
Male
666 (64 %)
160 (87 %)
Asian
30 (3 %)
4 (2 %)
Black
74 (7 %)
22 (12 %)
Age (mean, years) Gender
Race 0.150
Other/missing
109 (10 %)
19 (10 %)
White Hispanic
829 (80 %)
139 (76 %)
Hypertension
631 (61 %)
121 (66 %)
0.181
Obesity
143 (14 %)
21 (11 %)
0.396
Smoker
128 (12 %)
31 (17 %)
0.089
Diabetic
297 (29 %)
44 (24 %)
0.200
Preoperative GFR (mL/min/1.73 m2)
76.77
72.02
0.004*
CKD (GFR \60 mL/min/1.73 m2)
220 (21 %)
54 (29 %)
0.013*
1
285 (27 %)
34 (18 %)
0.021*
2
537 (52 %)
96 (52 %)
3
207 (20 %)
50 (27 %)
4
13 (1.2 %)
4 (2 %)
CKD stage
* Statistical significance
Discussion
Fig. 1 Breakdown of RCC histology stratified by CKD stage
123
2130
Int Urol Nephrol (2014) 46:2127–2132
Table 2 Multivariate analysis of predictors of papillary histology of RCC Odds ratio
95 % CI for odds ratio Lower
Upper
p value
Male gender
3.720
2.351
5.887
0.001*
Age
0.994
0.979
1.009
0.404
Obesity
0.835
0.490
1.422
0.506
Smoking
1.388
0.882
2.182
0.156
Hypertension
1.229
0.852
1.771
0.270
Diabetes
0.668
0.443
1.006
0.053
Race White
0.012* Reference
Black
2.271
0.689
7.484
0.178
Asian
1.099
0.332
3.633
0.877
Other or missing
0.904
0.301
2.716
CKD staging
0.857 0.028*
CKD stage 1
Reference
CKD stage 2 CKD stage 3
1.481 2.201
0.942 1.273
2.328 3.803
0.089 0.005*
CKD stage 4
2.949
0.853
10.199
0.088
Presence of CKD** (GFR \60)
1.630
1.105
2.405
0.014*
Declining GFR**
1.012
1.003
1.021
0.012*
* Statistical significance ** Both the variables of ‘‘presence of CKD’’ and worse GFR were analyzed in a separate multivariate analysis from the variable of ‘‘CKD staging,’’ but are shown here on the same table for ease of interpretation. There was no significant difference in the associations of the other variables
on RCC risk of hypertension versus its treatment. While some antihypertensive medications (e.g., diuretics) act on the renal tubules, no particular class of antihypertensive medication has consistently been associated with RCC risk, thereby pointing toward HTN, rather than its treatment, as the source of RCC risk [15]. Links between phenacetincontaining analgesics, nonsteroidal anti-inflammatory drugs, diet, alcohol consumption, occupation, and RCC have not been well established [16–18]. Robust evidence exists for an increased risk of RCC among patients undergoing renal dialysis. Stewart et al. [6] performed a retrospective review of over 800,000 patients who were receiving dialysis for ESRD stemming from different etiologies. Overall, they found an increased risk of kidney cancer (3.6 times higher) than would be expected in the general population. This increased risk was more pronounced in women and increased with longer duration of dialysis. There was no clear relationship between types of dialysis. Renal cystic disease associated with dialysis is a condition that has been implicated as a risk factor for RCC. It
123
commonly develops in patients receiving any method of dialysis, and its prevalence increases with duration of dialysis [19]. These cysts are postulated to undergo malignant degeneration, leading to an increased risk of RCC. However, there is some evidence that cystic disease may actually predate dialysis [20]. Renal transplantation, alone, confers an additional cancer risk, as the incidence of RCC appears to be increased in native kidneys in renal transplant patients [21]. These RCCs are more likely to be multifocal (40 %), bilateral (20 %), and more commonly papillary than reported in non-transplant series (32–38 %) [22]. Since patients with ESRD require dialysis or transplantation for long-term survival, it is difficult to determine whether ESRD or its treatments are behind the observed increased prevalence of RCC in these patients. However, there is basis for the claim that renal failure itself may result in increased RCC risk and that dialysis simply prolongs the exposure of the patient to the state of kidney failure, namely, the mode of dialysis does not appear to affect RCC risk, risk of RCC was still over 3 times higher than expected in patients on dialysis for just 1 year, and cystic degeneration can occur prior to starting dialysis [6, 20]. In a recent population-based study, Hofmann et al. [23] found an association between a patient’s self-reporting of a diagnosis of CKD and increased incidence of RCC diagnosis. While this study was suggestive and does correspond with our findings, it lacked power, specifics about GFR, and actual tumor histology. Weng et al. [24] performed a population-based review in Taiwan and found an increased mortality attributed to kidney cancers in nondialysis-dependent patients with CKD. While this study did not address incidence of RCC or histology specifically, it is also suggestive of an association between CKD and kidney cancer. Molecular studies on kidney regeneration give further insight on the possible link between renal injury and development of RCC. Insults to the kidney leading to acute renal failure are often the result of acute tubular necrosis. Recovery of renal function is a result of the replacement of necrotic tubular cells [25]. The origin of these new renal cells is not clearly defined, but as in the case of other organs, the kidney is purported to contain organ-specific pluripotent cells (renal stem cells). Several different groups of researchers have reported on a cell population within the normal adult kidney that expresses the stem cell marker CD133 and is capable of expansion, self-renewal, and differentiation into epithelial cells or endothelial cells for repair of damaged renal tubules [26, 27]. Following a similar thread, Schwartz et al. [8] have reported on the use of CD133 as a marker to confirm acute tubular injury and found a positive correlation with CD133 expression in renal biopsy specimens and serum creatinine levels.
Int Urol Nephrol (2014) 46:2127–2132
Interestingly, Lindgren et al. [28] noted that the protein expression pattern of the CD133? cells was similar to that of pRCC and distinct from other RCC subtypes. Schwartz et al. [8] also analyzed CD133 expression in a variety of histologic subtypes of RCC and noted significantly higher expression in pRCC. Taken together, these studies suggest a biological explanation for our finding that the likelihood of diagnosis of a pRCC (as opposed to another RCC subtype) increased with worsening renal function. As a patient’s renal function declines due to chronic disease or suffers episodic insults, the stem cell population may proliferate in order to attempt to bring about regeneration of tubules. Assuming that some small fraction of stem cells may become constitutively proliferative and undergo malignant transformation, it stands to reason that patients who have experienced kidney damage are more prone to stem cell-derived renal malignancy. As these stem cells are CD133?, a marker associated with pRCCs, it further stands to reason that those who have suffered renal injury (either acute or chronic) are more likely to develop pRCC. In the absence of a control comparison group, our data simply demonstrate an association between declining renal function and increased risk of pRCC. However, the possibility of a causative link is strengthened by these results. It is difficult to conceive that the risk of RCC overall stays the same regardless of renal function, while only the likelihood of pRCC versus another histologic subtype increases as renal function declines. Furthermore, pRCC is more likely to be multifocal and bilateral than the other subtypes [29]. This characteristic of pRCC supports the notion that its etiology may be due to aberrant regeneration and proliferation in response to kidney injury, rather than to an isolated genetic mutation as would be expected in histologic subtypes that tend to form solitary and isolated tumors. Our study is limited by its retrospective nature and its inclusion only of patients with a diagnosis of RCC who have undergone surgery. The immediate cause of renal impairment for each individual patient is also unknown. While we do have information on comorbidities known to be associated with renal damage, we do not have detailed information about acute events that may have contributed to the renal insufficiency, such as acute tubular nephritis and periods of hypotension. It is possible that the increased risk of pRCC occurs only in those with a chronic process (i.e., HTN, DM) or only in those who have suffered some acute kidney damage. Further study is warranted to better define the association between declining renal function and the risk of RCC (and specifically pRCC risk). Our data and a growing body of literature identifying kidney stem cells and linking their protein expression patterns to pRCC subtypes are
2131
provocative and may give epidemiological rationale to work by basic scientists who are attempting to tackle this important clinical problem.
Conclusions Within a large cohort of patients with a diagnosis of RCC that excluded those with ESRD, declining renal function was independently associated with an increased likelihood of pRCC histology. This finding and the available molecular evidence indicating similarity in protein expression between pRCC and resident stem cells, which appear to upregulated with kidney damage, suggest a possible causal relationship between renal injury and pRCC. Conflict of interest of interest.
The authors declare that they have no conflict
Ethical standard This retrospective human study has been approved by the Columbia University Medical Center institutional review board and has therefore been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments. All patients have given their informed consent to be part of our institutional surgical registry of genitourinary cancers.
References 1. Siegel R, Naishadham D, Jemal A (2013) Cancer statistics. CA Cancer J Clin 63:11–30 2. Jayson M, Sanders H (1998) Increased incidence of serendipitously discovered renal cell carcinoma. Urology 51:203–205 3. Rini BI, Campbell SC, Escudier B (2009) Renal cell carcinoma. Lancet 373:1119–1132 4. Schmidt L, Junker K, Nakaigawa N et al (1999) Novel mutations of the MET proto-oncogene in papillary renal carcinomas. Oncogene 18:2343–2350 5. Schmidt LS, Nickerson ML, Angeloni D et al (2004) Early onset hereditary papillary renal carcinoma: germline missense mutations in the tyrosine kinase domain of the met proto-oncogene. J Urol 172:1256–1261 6. Stewart JH, Buccianti G, Agodoa L et al (2003) Cancers of the kidney and urinary tract in patients on dialysis for end-stage renal disease: analysis of data from the United States, Europe, and Australia and New Zealand. J Am Soc Nephrol 14:197–207 7. Bussolati B, Bruno S, Grange C et al (2005) Isolation of renal progenitor cells from adult human kidney. Am J Pathol 166:545–555 8. Schwartz JDDFH JM, Wilson GD, Wolforth SC, Rooney MT, Wei L, Zhang PL (2013) CD133 Staining detects acute kidney injury and differentiates clear cell papillary renal cell carcinoma from other renal tumors. ISRN Biomark 2013, 8 9. Levey AS, Stevens LA, Schmid CH et al (2009) A new equation to estimate glomerular filtration rate. Ann Intern Med 150:604–612 10. Levey AS, Coresh J, Balk E et al (2003) National Kidney Foundation practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Ann Intern Med 139:137–147
123
2132 11. Campbell SCLBR (2012) Malignant renal tumors. In: Wein AJ (ed) Campbell-Walsh urology, vol 3, 10th edn. Elsevier, Philadelphia PA, p 4320 12. Ximing J, Yang KD, Jia L, Gersbach E, Lin F, Shen S, Teh BT (2013) Is papillary renal cell carcinoma derived from proximal convoluted tubules?. United States and Canadian Academy of Pathology Annual Meeting, Baltimore, MA 13. Renehan AG, Tyson M, Egger M et al (2008) Body-mass index and incidence of cancer: a systematic review and meta-analysis of prospective observational studies. Lancet 371:569–578 14. Hajjar I, Kotchen TA (2003) Trends in prevalence, awareness, treatment, and control of hypertension in the United States, 1988–2000. J Am Med Assoc 290:199–206 15. Fryzek JP, Poulsen AH, Johnsen SP et al (2005) A cohort study of antihypertensive treatments and risk of renal cell cancer. Br J Cancer 92:1302–1306 16. Lipworth L, Tarone RE, McLaughlin JK (2006) The epidemiology of renal cell carcinoma. J Urol 176:2353–2358 17. Ljungberg B, Campbell SC, Choi HY et al (2011) The epidemiology of renal cell carcinoma. Eur Urol 60:615–621 18. McCredie M, Pommer W, McLaughlin JK et al (1995) International renal-cell cancer study. II. Analgesics. Int J Cancer 60:345–349 19. Park JH, Kim YO, Park JH et al (2000) Comparison of acquired cystic kidney disease between hemodialysis and continuous ambulatory peritoneal dialysis. Korean J Intern Med 15:51–55 20. Liu JS, Ishikawa I, Horiguchi T (2000) Incidence of acquired renal cysts in biopsy specimens. Nephron 84:142–147
123
Int Urol Nephrol (2014) 46:2127–2132 21. Kasiske BL, Snyder JJ, Gilbertson DT et al (2004) Cancer after kidney transplantation in the United States. Am J Transplant 4:905–913 22. Klatte T, Marberger M (2011) Renal cell carcinoma of native kidneys in renal transplant patients. Curr Opin Urol 21:376–379 23. Hofmann JN, Schwartz K, Chow WH et al (2013) The association between chronic renal failure and renal cell carcinoma may differ between black and white Americans. Cancer Causes Control 24:167–174 24. Weng PH, Hung KY, Huang HL et al (2011) Cancer-specific mortality in chronic kidney disease: longitudinal follow-up of a large cohort. Clin J Am Soc Nephrol 6:1121–1128 25. Bonventre JV, Weinberg JM (2003) Recent advances in the pathophysiology of ischemic acute renal failure. J Am Soc Nephrol 14:2199–2210 26. Angelotti ML, Ronconi E, Ballerini L et al (2012) Characterization of renal progenitors committed toward tubular lineage and their regenerative potential in renal tubular injury. Stem Cells 30:1714–1725 27. Kim K, Park BH, Ihm H et al (2011) Expression of stem cell marker CD133 in fetal and adult human kidneys and pauciimmune crescentic glomerulonephritis. Histol Histopathol 26:223–232 28. Lindgren D, Bostrom AK, Nilsson K et al (2011) Isolation and characterization of progenitor-like cells from human renal proximal tubules. Am J Pathol 178:828–837 29. Reuter VE (2006) The pathology of renal epithelial neoplasms. Semin Oncol 33:534–543
UROLOGY - ORIGINAL PAPER
Renal insufficiency is associated with an increased risk of papillary renal cell carcinoma histology Solomon L. Woldu • Aaron C. Weinberg • Arindam RoyChoudhury Herbert Chase • Sean D. Kalloo • James M. McKiernan • G. Joel DeCastro
•
Received: 11 March 2014 / Accepted: 24 June 2014 / Published online: 8 July 2014 Ó Springer Science+Business Media Dordrecht 2014
Abstract Purpose End-stage renal disease (ESRD) and acquired renal cystic disease associated with dialysis are known risk factors of papillary renal cell carcinoma (pRCC); however, it is not known whether renal insufficiency alone is a risk factor for pRCC. Our aim was to test whether renal insufficiency is associated with an increased preponderance of pRCC. Methods Retrospective review of institutional database to identify all patients who underwent extirpative renal surgery for renal cell carcinoma (RCC) with complete records from 1992 to 2012. We excluded those patients with preoperative ESRD as defined by GFR \ 15 mL/min/1.73 m2. The dependent variable was histologic RCC subtype. Independent variables included demographic data, comorbidities, and renal functional data. Multivariate analysis by binary logistic regression was used to determine factors that independently were associated with pRCC development. S. L. Woldu (&) A. C. Weinberg J. M. McKiernan G. J. DeCastro Department of Urology, Columbia University Medical Center, Herbert Irving Pavilion – 11th Floor 161 Ft. Washington Ave, New York, NY 10032, USA e-mail: [email protected]; [email protected] A. RoyChoudhury Department of Biostatistics, Columbia University Medical Center, New York, NY, USA H. Chase Department of Biomedical Informatics, Columbia University Medical Center, New York, NY, USA S. D. Kalloo Department of Nephrology, Columbia University Medical Center, New York, NY, USA
Results A total of 1,226 patients met inclusion criteria, of which 15 % were pRCC. There was a positive association between likelihood of pRCC histology of RCC and increasing preoperative chronic kidney disease (CKD) stage (p = 0.021). Multivariate regression analysis indicated that male gender, race, and declining renal function categorized both by GFR and CKD stage were independently associated with a higher likelihood of pRCC histology as compared to other RCC histology. Conclusions Within a large cohort of patients with a diagnosis of RCC, declining renal function was independently associated with an increased likelihood of pRCC histology. This finding and the available molecular evidence indicating protein expression similarity between pRCC and resident stem cells, which appear to be upregulated with kidney damage, suggest a possible causal relationship between renal injury and pRCC. Keywords Renal cell carcinoma Papillary Renal insufficiency Risk factor
Introduction Renal cell carcinoma (RCC) is the sixth most common cancer in men and eighth most common cancer in women in the USA [1]. Its incidence is rising, in part due to increased detection of asymptomatic renal masses related to increased use of cross-sectional imaging [2]. There are a variety of RCC histologic subtypes; clear cell RCC represents more than 70 % of cases, followed by papillary (10–15 %), and less commonly chromophobe RCC, oncocytoma, collecting duct RCC, and other rarer subtypes [3]. Each histologic subtype is believed to be the result of distinct genetic pathways and possibly from
123
2128
distinct cell lineages within the kidney. RCC is an adenocarcinoma and therefore arises from the tubular epithelium of the nephron. Much of our understanding of the etiology and distinctness of the subtypes of RCC comes from studying patients with a familial RCC predisposition syndrome. Papillary RCC (pRCC) is the second most common histologic subtype and appears to be both morphologically and genetically distinct. pRCC is characterized by trisomy of chromosomes 7 and 17 and other abnormalities on chromosomes 1, 12, 16, 20, and Y. Researchers have demonstrated mutations in the c-MET proto-oncogene on chromosome 7 in patients with hereditary pRCC, a familial predisposition syndrome [4]. However, these mutations are reported only in a small subset of pRCC overall, and the driving factor for the majority of mutations has not been elucidated [5]. There is evidence that end-stage renal disease (ESRD) and hemodialysis are risk factors for the development of acquired renal cystic disease, which in turn is strongly associated with the development of pRCC [6]. Furthermore, there are recent reports of the presence of a stem cell population located within the adult human kidney that can respond and proliferate in response to tubular injury [7]. Most provocatively, these purported stems cells share protein expression patterns with pRCC [8]. We sought to determine whether renal insufficiency alone, without ESRD or dialysis, is associated with an increased likelihood of pRCC histology at the time of nephrectomy for RCC.
Patients and methods Columbia University Medical Center has maintained an Institutional Review Board approved database of all renal surgeries performed for suspected malignancy. We reviewed this database from 1992 to 2012 to analyze all patients with pathologically confirmed RCC and known preoperative renal function. Data obtained included age, gender, race, BMI, comorbidities (HTN, DM, obesity, smoking status), and preoperative renal function as estimated by serum creatinine and by the glomerular filtration rate (GFR) calculated by the Modification of Diet and Renal Disease formula [9]. Patients with ESRD as defined as GFR \15 mL/min/1.73 m2 were excluded, as this is a known risk factor for the development of pRCC, and the aim of our study was to define the relationship between the risk of pRCC and renal function in patients without ESRD and the requirement of dialysis. Our dependent variable was whether the histology of the RCC was papillary or non-papillary. We separately analyzed the independent variable of renal function as a continuous variable (GFR),
123
Int Urol Nephrol (2014) 46:2127–2132
categorical variable (stage of CKD), and the presence or absence of CKD (as defined by a GFR \60 mL/min/ 1.73 m2). CKD staging was defined as follows: stage 1: GFR C90 mL/min/1.73 m2, stage 2: GFR 60–89 mL/min/ 1.73 m2, stage 3: 30–59 mL/min/1.73 m2, stage 4: 15–29 mL/min/1.73 m2, and stage 5 (ESRD): GFR \15 mL/min/1.73 m2 [10]. Despite collecting data on patient comorbidities associated with developing CKD, we were unable to attribute the CKD to a specific cause with our retrospective database. There was no distinction made between Type 1 and Type 2 pRCC histology, as the pathologists did not consistently report this distinction. ANOVA was performed to compare likelihood of dependent variables based on categorical independent variables. Multivariate analysis by binary logistic regression was used to determine factors that were independently associated with dependent variables. SPSS Version 21.0 was used for statistical analysis (IBM Corp, Armonk, NY). p value of \0.05 was considered significant.
Results A total of 1,226 patients were identified that met our inclusion criteria. Tumor histology was classified as clear cell RCC in 806 (65.7 %), pRCC in 184 (15.0 %), chromophobe RCC in 122 (10.0 %), oncocytoma in 112 (9.1 %), and other unspecified RCC in 2 (0.2 %) patients. CKD staging was 1 for 319 patients, 2 for 633 patients, 3 for 257 patients, and 4 for 17 patients. When stratifying histology into papillary versus nonpapillary RCC (Table 1), those patients diagnosed with a pRCC were more likely to be male (87 % male vs. 64 % female, p = 0.001) and have worse baseline kidney function, as indicated by a lower mean GFR (72.0 vs. 76.8 mL/ min/1.73 m2, p = 0.004), and increased likelihood of being defined as having CKD (29 % with CKD vs. 21 % without CKD, p = 0.013). There was a strong positive association between likelihood of papillary histology of RCC and increasing stage of CKD, p = 0.021 (Table 1; Fig. 1). Multivariate regression analysis to determine predictors of pRCC histology indicated that male gender, race, and worsening renal function were independently associated with a higher likelihood of pRCC histology as compared to other RCC histology (Table 2). Worsening renal function remained consistently an independent factor associated with increased risk of pRCC versus other RCC histology when analyzed separately: (1) as a continuous variable (GFR), (2) as a dichotomous variable defined by the presence or absence of CKD as defined as GFR \60 mL/min/ 1.73 m2, and (3) as an ordinal variable defined by specific CKD stage. An increasing stage of CKD was associated
Int Urol Nephrol (2014) 46:2127–2132
2129
with a step-wise increase in risk of having papillary RCC histology; when compared to CKD stage 1 patients, CKD stage 2 patients had an adjusted OR 1.48 (95 % CI 0.94–2.33), CKD stage 3 patients had an adjusted OR 2.20 (95 % CI 1.27–3.80), and CKD stage 4 patients had an
adjusted OR 2.95 (95 % CI 0.85–10.20) of being diagnosed with pRCC versus another histologic subtype.
Table 1 Patient characteristics
The pathogenesis of pRCC is not well elucidated. While Hereditary Papillary RCC Syndrome is believed to be the result of mutations in the c-MET oncogene on chromosome 7, this accounts for the small minority of cases of pRCC [4]. Beyond lacking clarity on the genetic alterations that lead to the development of most cases of pRCC, there is also a lack of consensus on the progenitor cell type in the tubule that leads to pRCC. pRCC has often been reported to originate from the proximal tubule, but this is controversial [11]. Yang et al. recently reported on 35 cases of pRCC, which expressed markers consistent with both proximal and distal tubule expression patterns. In addition, the pRCC also expressed markers absent from both proximal and distal tubules, and the authors concluded that pRCC was more likely to be derived from multipotent stem cells rather than a specific tubular location [12]. While most cases of RCC are thought to be sporadic, there are certain factors that appear to result in an increased risk of developing RCC. Cigarette smoking is the most consistently established risk factor and accounts for about 20 % of RCC cases. There also appears to be an association between obesity and increased risk of RCC [13]. Hypertension and its treatment with antihypertensive medications have been postulated to be associated with an increased risk. However, the interpretation of the literature on this subject is difficult, as pre-diagnostic renal tumors may lead to an increased blood pressure and imaging for hypertension can lead to the discovery of incidental renal masses [14]. Furthermore, it is difficult to separate the influences
Histology Nonpapillary RCC
Papillary RCC
p value
61.80
62.13
0.748
Female
376 (36 %)
24 (13 %)
0.001*
Male
666 (64 %)
160 (87 %)
Asian
30 (3 %)
4 (2 %)
Black
74 (7 %)
22 (12 %)
Age (mean, years) Gender
Race 0.150
Other/missing
109 (10 %)
19 (10 %)
White Hispanic
829 (80 %)
139 (76 %)
Hypertension
631 (61 %)
121 (66 %)
0.181
Obesity
143 (14 %)
21 (11 %)
0.396
Smoker
128 (12 %)
31 (17 %)
0.089
Diabetic
297 (29 %)
44 (24 %)
0.200
Preoperative GFR (mL/min/1.73 m2)
76.77
72.02
0.004*
CKD (GFR \60 mL/min/1.73 m2)
220 (21 %)
54 (29 %)
0.013*
1
285 (27 %)
34 (18 %)
0.021*
2
537 (52 %)
96 (52 %)
3
207 (20 %)
50 (27 %)
4
13 (1.2 %)
4 (2 %)
CKD stage
* Statistical significance
Discussion
Fig. 1 Breakdown of RCC histology stratified by CKD stage
123
2130
Int Urol Nephrol (2014) 46:2127–2132
Table 2 Multivariate analysis of predictors of papillary histology of RCC Odds ratio
95 % CI for odds ratio Lower
Upper
p value
Male gender
3.720
2.351
5.887
0.001*
Age
0.994
0.979
1.009
0.404
Obesity
0.835
0.490
1.422
0.506
Smoking
1.388
0.882
2.182
0.156
Hypertension
1.229
0.852
1.771
0.270
Diabetes
0.668
0.443
1.006
0.053
Race White
0.012* Reference
Black
2.271
0.689
7.484
0.178
Asian
1.099
0.332
3.633
0.877
Other or missing
0.904
0.301
2.716
CKD staging
0.857 0.028*
CKD stage 1
Reference
CKD stage 2 CKD stage 3
1.481 2.201
0.942 1.273
2.328 3.803
0.089 0.005*
CKD stage 4
2.949
0.853
10.199
0.088
Presence of CKD** (GFR \60)
1.630
1.105
2.405
0.014*
Declining GFR**
1.012
1.003
1.021
0.012*
* Statistical significance ** Both the variables of ‘‘presence of CKD’’ and worse GFR were analyzed in a separate multivariate analysis from the variable of ‘‘CKD staging,’’ but are shown here on the same table for ease of interpretation. There was no significant difference in the associations of the other variables
on RCC risk of hypertension versus its treatment. While some antihypertensive medications (e.g., diuretics) act on the renal tubules, no particular class of antihypertensive medication has consistently been associated with RCC risk, thereby pointing toward HTN, rather than its treatment, as the source of RCC risk [15]. Links between phenacetincontaining analgesics, nonsteroidal anti-inflammatory drugs, diet, alcohol consumption, occupation, and RCC have not been well established [16–18]. Robust evidence exists for an increased risk of RCC among patients undergoing renal dialysis. Stewart et al. [6] performed a retrospective review of over 800,000 patients who were receiving dialysis for ESRD stemming from different etiologies. Overall, they found an increased risk of kidney cancer (3.6 times higher) than would be expected in the general population. This increased risk was more pronounced in women and increased with longer duration of dialysis. There was no clear relationship between types of dialysis. Renal cystic disease associated with dialysis is a condition that has been implicated as a risk factor for RCC. It
123
commonly develops in patients receiving any method of dialysis, and its prevalence increases with duration of dialysis [19]. These cysts are postulated to undergo malignant degeneration, leading to an increased risk of RCC. However, there is some evidence that cystic disease may actually predate dialysis [20]. Renal transplantation, alone, confers an additional cancer risk, as the incidence of RCC appears to be increased in native kidneys in renal transplant patients [21]. These RCCs are more likely to be multifocal (40 %), bilateral (20 %), and more commonly papillary than reported in non-transplant series (32–38 %) [22]. Since patients with ESRD require dialysis or transplantation for long-term survival, it is difficult to determine whether ESRD or its treatments are behind the observed increased prevalence of RCC in these patients. However, there is basis for the claim that renal failure itself may result in increased RCC risk and that dialysis simply prolongs the exposure of the patient to the state of kidney failure, namely, the mode of dialysis does not appear to affect RCC risk, risk of RCC was still over 3 times higher than expected in patients on dialysis for just 1 year, and cystic degeneration can occur prior to starting dialysis [6, 20]. In a recent population-based study, Hofmann et al. [23] found an association between a patient’s self-reporting of a diagnosis of CKD and increased incidence of RCC diagnosis. While this study was suggestive and does correspond with our findings, it lacked power, specifics about GFR, and actual tumor histology. Weng et al. [24] performed a population-based review in Taiwan and found an increased mortality attributed to kidney cancers in nondialysis-dependent patients with CKD. While this study did not address incidence of RCC or histology specifically, it is also suggestive of an association between CKD and kidney cancer. Molecular studies on kidney regeneration give further insight on the possible link between renal injury and development of RCC. Insults to the kidney leading to acute renal failure are often the result of acute tubular necrosis. Recovery of renal function is a result of the replacement of necrotic tubular cells [25]. The origin of these new renal cells is not clearly defined, but as in the case of other organs, the kidney is purported to contain organ-specific pluripotent cells (renal stem cells). Several different groups of researchers have reported on a cell population within the normal adult kidney that expresses the stem cell marker CD133 and is capable of expansion, self-renewal, and differentiation into epithelial cells or endothelial cells for repair of damaged renal tubules [26, 27]. Following a similar thread, Schwartz et al. [8] have reported on the use of CD133 as a marker to confirm acute tubular injury and found a positive correlation with CD133 expression in renal biopsy specimens and serum creatinine levels.
Int Urol Nephrol (2014) 46:2127–2132
Interestingly, Lindgren et al. [28] noted that the protein expression pattern of the CD133? cells was similar to that of pRCC and distinct from other RCC subtypes. Schwartz et al. [8] also analyzed CD133 expression in a variety of histologic subtypes of RCC and noted significantly higher expression in pRCC. Taken together, these studies suggest a biological explanation for our finding that the likelihood of diagnosis of a pRCC (as opposed to another RCC subtype) increased with worsening renal function. As a patient’s renal function declines due to chronic disease or suffers episodic insults, the stem cell population may proliferate in order to attempt to bring about regeneration of tubules. Assuming that some small fraction of stem cells may become constitutively proliferative and undergo malignant transformation, it stands to reason that patients who have experienced kidney damage are more prone to stem cell-derived renal malignancy. As these stem cells are CD133?, a marker associated with pRCCs, it further stands to reason that those who have suffered renal injury (either acute or chronic) are more likely to develop pRCC. In the absence of a control comparison group, our data simply demonstrate an association between declining renal function and increased risk of pRCC. However, the possibility of a causative link is strengthened by these results. It is difficult to conceive that the risk of RCC overall stays the same regardless of renal function, while only the likelihood of pRCC versus another histologic subtype increases as renal function declines. Furthermore, pRCC is more likely to be multifocal and bilateral than the other subtypes [29]. This characteristic of pRCC supports the notion that its etiology may be due to aberrant regeneration and proliferation in response to kidney injury, rather than to an isolated genetic mutation as would be expected in histologic subtypes that tend to form solitary and isolated tumors. Our study is limited by its retrospective nature and its inclusion only of patients with a diagnosis of RCC who have undergone surgery. The immediate cause of renal impairment for each individual patient is also unknown. While we do have information on comorbidities known to be associated with renal damage, we do not have detailed information about acute events that may have contributed to the renal insufficiency, such as acute tubular nephritis and periods of hypotension. It is possible that the increased risk of pRCC occurs only in those with a chronic process (i.e., HTN, DM) or only in those who have suffered some acute kidney damage. Further study is warranted to better define the association between declining renal function and the risk of RCC (and specifically pRCC risk). Our data and a growing body of literature identifying kidney stem cells and linking their protein expression patterns to pRCC subtypes are
2131
provocative and may give epidemiological rationale to work by basic scientists who are attempting to tackle this important clinical problem.
Conclusions Within a large cohort of patients with a diagnosis of RCC that excluded those with ESRD, declining renal function was independently associated with an increased likelihood of pRCC histology. This finding and the available molecular evidence indicating similarity in protein expression between pRCC and resident stem cells, which appear to upregulated with kidney damage, suggest a possible causal relationship between renal injury and pRCC. Conflict of interest of interest.
The authors declare that they have no conflict
Ethical standard This retrospective human study has been approved by the Columbia University Medical Center institutional review board and has therefore been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments. All patients have given their informed consent to be part of our institutional surgical registry of genitourinary cancers.
References 1. Siegel R, Naishadham D, Jemal A (2013) Cancer statistics. CA Cancer J Clin 63:11–30 2. Jayson M, Sanders H (1998) Increased incidence of serendipitously discovered renal cell carcinoma. Urology 51:203–205 3. Rini BI, Campbell SC, Escudier B (2009) Renal cell carcinoma. Lancet 373:1119–1132 4. Schmidt L, Junker K, Nakaigawa N et al (1999) Novel mutations of the MET proto-oncogene in papillary renal carcinomas. Oncogene 18:2343–2350 5. Schmidt LS, Nickerson ML, Angeloni D et al (2004) Early onset hereditary papillary renal carcinoma: germline missense mutations in the tyrosine kinase domain of the met proto-oncogene. J Urol 172:1256–1261 6. Stewart JH, Buccianti G, Agodoa L et al (2003) Cancers of the kidney and urinary tract in patients on dialysis for end-stage renal disease: analysis of data from the United States, Europe, and Australia and New Zealand. J Am Soc Nephrol 14:197–207 7. Bussolati B, Bruno S, Grange C et al (2005) Isolation of renal progenitor cells from adult human kidney. Am J Pathol 166:545–555 8. Schwartz JDDFH JM, Wilson GD, Wolforth SC, Rooney MT, Wei L, Zhang PL (2013) CD133 Staining detects acute kidney injury and differentiates clear cell papillary renal cell carcinoma from other renal tumors. ISRN Biomark 2013, 8 9. Levey AS, Stevens LA, Schmid CH et al (2009) A new equation to estimate glomerular filtration rate. Ann Intern Med 150:604–612 10. Levey AS, Coresh J, Balk E et al (2003) National Kidney Foundation practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Ann Intern Med 139:137–147
123
2132 11. Campbell SCLBR (2012) Malignant renal tumors. In: Wein AJ (ed) Campbell-Walsh urology, vol 3, 10th edn. Elsevier, Philadelphia PA, p 4320 12. Ximing J, Yang KD, Jia L, Gersbach E, Lin F, Shen S, Teh BT (2013) Is papillary renal cell carcinoma derived from proximal convoluted tubules?. United States and Canadian Academy of Pathology Annual Meeting, Baltimore, MA 13. Renehan AG, Tyson M, Egger M et al (2008) Body-mass index and incidence of cancer: a systematic review and meta-analysis of prospective observational studies. Lancet 371:569–578 14. Hajjar I, Kotchen TA (2003) Trends in prevalence, awareness, treatment, and control of hypertension in the United States, 1988–2000. J Am Med Assoc 290:199–206 15. Fryzek JP, Poulsen AH, Johnsen SP et al (2005) A cohort study of antihypertensive treatments and risk of renal cell cancer. Br J Cancer 92:1302–1306 16. Lipworth L, Tarone RE, McLaughlin JK (2006) The epidemiology of renal cell carcinoma. J Urol 176:2353–2358 17. Ljungberg B, Campbell SC, Choi HY et al (2011) The epidemiology of renal cell carcinoma. Eur Urol 60:615–621 18. McCredie M, Pommer W, McLaughlin JK et al (1995) International renal-cell cancer study. II. Analgesics. Int J Cancer 60:345–349 19. Park JH, Kim YO, Park JH et al (2000) Comparison of acquired cystic kidney disease between hemodialysis and continuous ambulatory peritoneal dialysis. Korean J Intern Med 15:51–55 20. Liu JS, Ishikawa I, Horiguchi T (2000) Incidence of acquired renal cysts in biopsy specimens. Nephron 84:142–147
123
Int Urol Nephrol (2014) 46:2127–2132 21. Kasiske BL, Snyder JJ, Gilbertson DT et al (2004) Cancer after kidney transplantation in the United States. Am J Transplant 4:905–913 22. Klatte T, Marberger M (2011) Renal cell carcinoma of native kidneys in renal transplant patients. Curr Opin Urol 21:376–379 23. Hofmann JN, Schwartz K, Chow WH et al (2013) The association between chronic renal failure and renal cell carcinoma may differ between black and white Americans. Cancer Causes Control 24:167–174 24. Weng PH, Hung KY, Huang HL et al (2011) Cancer-specific mortality in chronic kidney disease: longitudinal follow-up of a large cohort. Clin J Am Soc Nephrol 6:1121–1128 25. Bonventre JV, Weinberg JM (2003) Recent advances in the pathophysiology of ischemic acute renal failure. J Am Soc Nephrol 14:2199–2210 26. Angelotti ML, Ronconi E, Ballerini L et al (2012) Characterization of renal progenitors committed toward tubular lineage and their regenerative potential in renal tubular injury. Stem Cells 30:1714–1725 27. Kim K, Park BH, Ihm H et al (2011) Expression of stem cell marker CD133 in fetal and adult human kidneys and pauciimmune crescentic glomerulonephritis. Histol Histopathol 26:223–232 28. Lindgren D, Bostrom AK, Nilsson K et al (2011) Isolation and characterization of progenitor-like cells from human renal proximal tubules. Am J Pathol 178:828–837 29. Reuter VE (2006) The pathology of renal epithelial neoplasms. Semin Oncol 33:534–543
