Evaluation of renicapsular involvement in Stages I and II renal cell carcinoma from the morphological and prognostic point of view

Urologic Oncology: Seminars and Original Investigations 28 (2010) 274 –279

Original article

Evaluation of renicapsular involvement in Stages I and II renal cell carcinoma from the morphological and prognostic point of view Matthias May, M.D.a,1, Sabine Brookman-Amissah, M.D.a,1, Jan Roigas, M.D.b, Christian Peter Gilfrich, M.D.a, Sandra Pflanz, M.D.c, Bernd Hoschke, M.D.c, Sven Gunia, M.D.d,* b

a Department of Urology, St. Elisabeth Hospital, Straubing, Germany Department of Urology, Vivantes Hospital Am Urban, Berlin, Germany c Department of Urology, Carl-Thiem Hospital, Cottbus, Germany d Institute of Pathology, HELIOS Hospital, Bad Saarow, Germany

Received 16 June 2008; received in revised form 14 July 2008; accepted 14 July 2008

Abstract Background: Prognostic factors are essential for predicting postsurgical outcome in renal cell cancer (RCC). This study aimed to evaluate the prognostic impact of renicapsular involvement (RCI; invasion without penetration) in Stage I (pT1N0M0) and Stage II (pT2N0M0) RCC and to histomorphologically compare the structure of fibrous tumoral capsule with the pattern of RCI, the differentiation of which might by challenging in localized RCCs spreading near the renicapsule. Materials and Methods: We retrospectively investigated a cohort of 635 study group patients (396 men and 239 women; mean age: 60.9 years; range: 18 – 84 years) in terms of histomorphology and clinical outcome after surgery (nephrectomy or elective nephron-sparing surgery) at Stages I and II RCC (pT1-2N0M0). In 489 patients who were still alive at the end of the study, median follow-up was 80 months (mean 86.1 months). Disease-free survival (DFS) was calculated using the Kaplan Meier method. Univariate and multivariate Cox proportional hazards regression models were fit to determine possible associations between various parameters and survival. Another 55 control group patients (38 men and 17 women) aged between 44 and 75 years (mean age 61.4 years) with pT3a RCC were analyzed for statistical comparison (mean and median follow-up of the survivors were 85.7 and 84 months). Results: The 5-year DFS rate for patients with and without RCI was determined to be 76.9% and 86.3%, respectively (P ⬍ 0.01). Patients with histopathologically confirmed RCI appear to have the same adverse prognostic outcome as patients with RCC invading perinephric tissue (pT3aN0M0; P ⫽ 0.493). Histopathologically, fibrous tumoral capsule and RCI conventionally show a different morphology, making their separation straightforward. Conclusions: RCI reflects adverse prognostic outcome in surgically treated Stages I and II RCC. It can be determined by the pathologist without additional expense in time and cost. Hence, clinical pathologists should render a clear statement concerning RCI when reporting on small localized RCC specimens in order to provide additional prognostic information in individual cases and to facilitate selection of appropriate patients to be included in further standardized prospective studies, which are required to confirm the prognostic impact of RCI in Stages I and II RCC. © 2010 Elsevier Inc. All rights reserved. Keywords: Localized renal cell carcinoma; Renicapsular involvement; Fibrous tumoral capsule; Histopathology; Prognostic outcome

1. Introduction Renal cell carcinoma (RCC) is a frequent disease. For 2008, 54,390 estimated new cases of RCC are expected * Corresponding author. Tel.: ⫹49 33631 72312; fax: ⫹49 33631 73010. E-mail address: [email protected] (S. Gunia). 1 M.M. and S.B.-A. contributed equally to this work. 1078-1439/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.urolonc.2008.07.008

for both sexes in the USA alone [1]. Owing to the widespread use of abdominal imaging, small localized tumors (stage pT1-2N0M0) are diagnosed with increasing frequency nowadays [2,3]. Nevertheless, despite appropriate surgical management of RCC, some patients experience unexpected tumor progress or relapse during postsurgical follow-up [4]. Therefore, prognostic parameters are essential in terms of reliably predicting individual postsurgical outcome.

M. May et al. / Urologic Oncology: Seminars and Original Investigations 28 (2010) 274 –279

Currently, tumor stage and grade represent established prognostic tools to be applied for individual risk assessment in patients diagnosed with RCC [5]. However, predicting postsurgical outcome in localized Stages I and II RCC remains to be challenging for urologists and clinical pathologists because not all localized RCCs demonstrate the same biologic behavior and postsurgical clinical course [3]. Hence, additional reliable parameters are needed in order to make straightforward prognostic evaluation in individual cases, and they have been the target of several investigations in the past [6 – 8]. Since the prognostic impact of renicapsular involvement (RCI; invasion without penetration) seems to have attracted limited attention in the past, little is known about its prognostic implication in localized RCC [9,10]. Especially, appropriate long-term postsurgical follow-up data and standardized prospective studies are currently not available. We present a comprehensive study surveying the so far longest mean postsurgical follow-up of 86.1 months in a cohort including a total of 635 patients, aiming to evaluate the so far neglected long-term prognostic implication of RCI in Stages I and II RCC. From the histomorphological point of view, the focus was placed on the structural comparison between RCI and the fibrous tumor capsule frequently encountered in clear cell and papillary RCC. Differentiation of these connective tissue structures might occasionally be challenging when dealing with localized RCC spreading in the renal periphery in close vicinity to the renicapsule [11].

2. Materials and methods 2.1. Subjects The study group comprised a total of 635 patients (396 men and 239 women) 18 to 84 years of age (mean age: 60.9 years; median age: 62.2 years) with histopathologically confirmed Stages I and II RCC (pT1-2N0M0). All patients underwent either radical nephrectomy (n ⫽ 523) or elective nephron-sparing surgery (n ⫽ 112) at the Department of Urology, Carl-Thiem Hospital Cottbus, Germany, between January 1992 and December 2006. Clinical aftercare was performed at a urological outpatient setting of the latter hospital. During the first 2 years after surgery, aftercare included physical examination every 3 months. Ultrasound examination as well as CT scan of the thorax and the abdomen was performed every 6 months. From the third postsurgical year, physical and ultrasound examination were carried out at 6-month intervals, and a CT scan of the thorax and the abdomen was performed once a year. Additional examinations (i.e., radioisotope bone scanning, MRI, and cranial CT scan) were carried out individually according to clinical symptoms or in order to complete diagnostic procedures in patients coming to clinical attention with tumor progress.

275

For statistical comparison, an additional 55 control patients aged between 44 and 75 years (38 men and 17 women; mean age 61.4 years; median age 61.6 years) with stage pT3N0M0 RCC were also included in this study. All control group individuals showed RCC with invasion of perinephric fat but without adrenal involvement. The mean postsurgical follow-up was 85.7 months (median 84 months). 2.2. Study design As suggested by the 2007 European Association of Urology (EAU) guidelines, all patients underwent computed tomography (CT) scan or, alternatively, magnetic resonance imaging (MRI) of the abdomen prior to surgery. Preoperative diagnostic procedures comprised X-ray or CT scan of the thorax. Patients presenting with neurological symptoms additionally underwent cranial CT scan. Those with bone pain at the time of diagnosis additionally underwent radioisotope bone scanning. Pertinent demographics including age, sex, ASA score, and body mass index (BMI) were recorded (Table 1). Body mass index (BMI) was categorized based on the WHO definitions (normal: 25 kg/m2 or less, overweight-obese: ⬎25 kg/m2). The following pathological data were recorded in all cases: histopathological type, tumor size, pathologic staging, pattern of growth (compact, tubulo-cystic, and tubulopapillary, respectively), and 3-tiered nuclear grading according to Thoenes et al. [12]. Histopathological staging was determined according to the 2002 TNM classification system [13], and histopathological types were classified according to the Heidelberg classification of renal cell tumors [14]. Using 5 ␮m thick wax-embedded hematoxylin-eosin (H & E) stained tissue sections, all slices were read by at least one experienced clinical histopathologists at the Institute of Pathology of the Carl-Thiem Klinikum Cottbus (S.G., P.S., O.K.) at the time of initial pathology report. If present, microvascular invasion (MVI), renicapsular involvement (RCI), and collecting system invasion (CSI) were consistently rendered by the pathologist. Notably, MVI was defined as the unequivocal presence of tumor cells within the lumen of small to medium-sized endothelialized vessels. The presence of endothelial lining was immunohistochemically confirmed in all cases using antibodies directed against CD31 and CD34. No attempt was made to differentiate between lymphatic and vascular vessels. RCI was strictly defined as the histopathologically confirmed presence of tumor tissue within the renicapsule with characteristic splaying of the renicapsular fibers (Fig. 1). In order to avoid possible sampling errors during surgical pathology dissection, the portions of the tumor macroscopically bordering the renicapsule were completely embedded for subsequent histopathological assessment. Any possible tumor spread beyond the renicapsule with invasion of perinephric tissue was histopathologically excluded in all study

276

M. May et al. / Urologic Oncology: Seminars and Original Investigations 28 (2010) 274 –279

Table 1 Pertinent data of the study group (n ⫽ 635) Variable Age (years) Median (mean) Range Sex Male Female Tumor stage pT1aN0M0 (Stage Ia) pT1bN0M0 (Stage Ib) pT2N0M0 (Stage II) Nuclear grading (Thoenes, 3-tiered) Grade 1 Grade 2 Grade 3 Microvascular invasion (MVI) Absent Present Collecting system invasion (CSI) Absent Present Renicapsular involvement (RCI) Absent Present Tumor size (in cm) Median (mean) Range Histological type (Heidelberg classification) Clear cell Papillary Chromophobe Pleomorph-sarcomatoid Pattern of growth Compact Tubolopapillary Tubolocystic Body mass index (kg/m2) ⱕ25 ⬎25 ASA score 1–2 3–4

n (%) 62.2 (60.9) 18–84 396 (62.4%) 239 (37.6%) 287 (45.2%) 258 (40.6%) 90 (14.2%) 131 (20.6%) 458 (72.1%) 46 (7.2%) 631 (99.4%) 4 (0.6%) 595 (93.7%) 40 (6.3%) 489 (77.0%) 146 (23.0%) 4.5 (4.8) 0.8–17.0

was still alive at the end of the study (median and mean follow-up 80 and 86.1 months, respectively). In case of metastatic spread during postsurgical followup, different treatment options, such as surgery, systemic immunotherapy, radiotherapy, or supportive care, were chosen individually according to the pattern of metastatic spread and the general health condition of the patients concerned. Single local recurrences (LR) were surgically resected without exception. The primary endpoint of this investigation was reflected by the disease-free survival (DFS; defined as the period of time between surgery and tumor progression). In all patients with tumor progression, type, localization, and therapeutic management were recorded. For statistical analysis, survival curves were estimated by means of the product limit-method of Kaplan-Meier and compared by means of the log-rank test. Univariate and multivariate Cox proportional hazards models were fit to determine the parameters that were significantly associated with survival. The relationship between clinical outcome and the different parameters studied were statistically reflected by means of Hazard ratios (HR) with the corresponding 95% confidence intervals (95% CI) and P values. Statistical analyses were performed using SPSS, version 16.0 (Chicago, IL), with any P value ⱕ 0.05 considered to be statistically significant.

496 (78.1%) 96 (15.1%) 33 (5.2%) 10 (1.6%)

3. Results

526 (82.8%) 74 (11.7%) 35 (5.5%)

3.1. Histomorphologic comparison of fibrous tumor capsule and renicapsular involvement

201 (31.7%) 434 (68.3%)

The presence of an investing fibrous tumor capsule was histopathologically confirmed in all cases of clear cell and

387 (60.9%) 248 (39.1%)

group specimens by means of several serial sections. Histomorphologically, emphasis was placed on the structural comparison between RCI and the fibrous tumor capsule investing the tumor tissue. Penetration of the urothelial lined calyxes, infundibulum, and/or renal pelvis by the tumor was strictly required to establish the diagnosis of CSI, whereas simple compression or displacement of the urothelial lining by the tumor was not sufficient to qualify as CSI. All study group patients (n ⫽ 635) were re-evaluated after a median follow-up of 70 months postsurgically (mean: 78.7 months). Amongst these, 72 patients died from RCC (median and mean follow-up 40.5 and 51.8 months, respectively), and another 74 patients died due to other causes. Thus, a cohort of 489 study group patients (77%)

Fig. 1. Papillary RCC with renicapsular involvement. Note the characteristic splaying of renicapsular fibers with small tumor cell clusters being easily discernable within virtually empty spaces. Hematoxylin-eosin (H & E), ⫻4 objective.

M. May et al. / Urologic Oncology: Seminars and Original Investigations 28 (2010) 274 –279

Fig. 2. Clear cell RCC with fibrous tumoral capsule. Note haphazardly arranged elongated archiform fibrous septae extending through the tumor tissue. Van Gieson, ⫻2 objective.

papillary RCC investigated. Nearly all cases showed at least focally detectable haphazardly arranged elongated archiform fibrous septae continuously extending from the tumor capsule through the adjacent tumor tissue (Fig. 2). This morphology occasionally appeared to be somewhat reminiscent of the pseudolobular structural change encountered in liver cirrhosis, especially in areas of RCC bordering its surrounding tumor capsule. In contrast, renicapsular involvement seemed to be characterized by splaying of renicapsular fibers with small tumor cell clusters being easily discernable within virtually empty spaces (Fig. 1). 3.2. Renicapsular involvement and its prognostic implications Among the 635 study group specimens, 146 (23%) tumors showed RCI (Table 1). The mean tumor diameter in localized RCC with RCI was significantly increased in comparison with equally staged tumors without RCI (5.5. vs. 4.6 cm; P ⬍ 0.01). Concerning the entire study group, the mean 5-year and 10-year DFS rates were found to be 84.1% and 77.6% (standard error: 0.016% and 0.021%, respectively). The 5-year DFS rate for patients with and without RCI was determined to be 76.9% and 86.3%, respectively (P ⬍ 0.01; Fig. 3). Concerning Stage I tumors, the 5-year DFS rate with and without RCI was 80.2% and 86.9% (P ⫽ 0.046), and 64.5% and 84.4% in Stage II tumors (P ⬍ 0.01), respectively. In addition to RCI, the following parameters were found to significantly influence DFS in univariate Kaplan-Meier analyses: collecting system invasion (CSI; P ⫽ 0.024), histopathological growth pattern (P ⫽ 0.03), Thoenes nuclear grading (P ⫽ 0.017), tumor size (P ⬍ 0.01), and tumor stage (P ⬍ 0.01; Table 1). All study parameters presented in Table 1 were found to significantly influence DFS in uni-

277

Fig. 3. Kaplan-Meier survival estimates for DFS as function of renicapsular involvement in patients with RCC Stages I and II (pT1-2N0M0; n ⫽ 635).

variate analysis. Multivariate analysis was used to identify independent prognostic factors in terms of DFS. However, simultaneous analysis of tumor size and pT stage appeared not to be meaningful owing to the high intercorrelation between these two parameters (P ⬎ 0.9; P ⬍ 0.01). Therefore, the continuous parameter tumor size and the categorical variables RCI, CSI, histopathological growth pattern, and Thoenes nuclear grading were examined by means of multivariate Cox proportional hazard model. In this model, RCI and tumor size were found to be the only independent prognostic factors in terms of DFS (Table 2) with patients with histopathologically confirmed RCI determined to be at 1.63-fold increased risk of tumor progression compared with equally staged tumors without RCI (P ⫽ 0.016). Patients with Stages I and II (pT1-2N0M0) RCC with histopathologically confirmed RCI failed to differ significantly in terms of DFS in comparison with patients with pT3aN0M0 RCC (P ⫽ 0.493; Fig. 4). At the end of this study, 32 of the control patients were still alive. Thirteen control patients died due to RCC with the mean tumor progression recorded to have occurred after 44.3 months (median tumor progression after 52 months). Ten control patients died due to other causes.

Table 2 Cox proportional hazard regression model for DFS in Stages I and II RCC Parameter

HR

95% CI

P value

RCI (absent vs. present) CSI (absent vs. present) Nuclear grading (Thoenes, 3-tiered) Pattern of growth (three types) Tumor size (in cm)

1.63 1.41 1.34 1.62 1.13

1.10–2.41 0.78–2.56 0.90–2.01 0.99–2.63 1.05–1.21

P P P P P

⫽ ⫽ ⫽ ⫽ ⬍

0.016 0.258 0.151 0.052 0.001

CI ⫽ confidence interval; CSI ⫽ collecting system invasion; DFS ⫽ disease-free survival; HR ⫽ hazard ratio; RCI ⫽ renicapsular involvement.

278

M. May et al. / Urologic Oncology: Seminars and Original Investigations 28 (2010) 274 –279

Fig. 4. Kaplan-Meier survival estimates for DFS comparing patients with RCC Stages I and II (pT1-2N0M0) and renicapsular involvement (n ⫽ 146) vs. patients with RCC stage pT3aN0M0 (n ⫽ 55).

4. Discussion The frequent diagnosis of localized RCC due to the advent of modern imaging modalities accounts for increasing demand of prognostic parameters predictive of postsurgical outcome in early-stage tumors (pT1-2N0M0). Concerning the latter, limited studies are currently available addressing the issue of the prognostic impact of RCI in localized RCC [9,10]. Jeong et al. reported RCI to be associated with adverse prognostic outcome in pT2 but not in pT1 RCC [10]. These results might be attributable to the relatively low number of pT1 tumors with RCI investigated in this study (n ⫽ 78) and stand out in stark contrast with the data presented by Klatte et al., who suggested RCI to be an adverse prognostic factor in Stages I and II RCC [9]. However, median follow-up was limited to only 49 months in the latter study, which represents an important drawback because late metastasis even after several years postsurgically is not uncommon in RCC [4]. Surveying the so far longest postsurgical mean follow-up of 86.1 months in the currently largest cohort of patients investigated, our data confirm the results published by Klatte et al. [9] by indicating that patients with Stages I and II RCC with histopathologically proven RCI appear to have the same adverse prognostic outcome in terms of DFS as patients diagnosed with RCC invading the perinephric tissue (pT3aN0M0). Given that tumor size represents an important prognostic factor [5,13], one explanation might be that localized RCC with RCI showed significantly increased mean diameters compared with equally staged tumors without RCI in our study. Indeed, according to our data, tumor size might play a larger prognostic role than generally acknowledged since RCI appeared only to significantly stratify patients with pT2 disease (P ⬍ 0.01) while just achieving significance (P ⫽ 0.05) for pT1 tumors. Our data strongly support the notion that the issue of RCI in localized RCC is not only crucial to be reported by the pathologist, but indeed might give rise to suggest a revision

of the stage pT3a. Including patients with localized RCC with histopathologically confirmed RCI in this category appears to improve prognostic validity of the TNM classification system [9]. However, appropriate standardized prospective studies are clearly needed to back this up. Since tumor size is the only prognostic indicator applied in the 2002 TNM classification system for localized RCC [13], many clinical pathologists do not render a concise statement concerning RCI when reporting on RCC specimens in daily practice. This trend unfortunately not only results in an avoidable lack of prognostic information in individual cases but also dramatically limits standardized prospective multi-center studies because selection of appropriate patients to be included in further prospective studies is chiefly based on concise histopathology reports. Conventional histopathological evaluation of RCI is usually straightforward in the setting of routine procedures without additional expense in time and cost, but differentiation between the connective tissue of RCI and the fibrous tumor capsule might occasionally be challenging in localized tumors spreading in close vicinity to the renicapsule. We observed haphazardly arranged elongated archiform fibrous septae continuously extending from the tumor capsule through the adjacent tumor tissue to occur at least focally in all cases of clear cell and papillary RCCs investigated (Fig. 2). In contrast, RCI conventionally showed a strikingly different morphology, characterized by splaying of renicapsular fibers with small tumor cell clusters being easily discernable within virtually empty spaces (Fig. 1). These different patterns might aid in this differential diagnostic scenario. It merits emphasis that in daily pathological practice, addressing the presence or absence of RCI is easy and straightforward without needing any special stains or extra slides to be obtained. We are aware that the retrospective nature of our study as well as the restriction to one single institution from which the material investigated was retrieved might be viewed as potential limitations. However, since the institution involved does not represent a reference center, we feel our data to be comparable with the general urological community because previous selection of the material investigated was avoided. Nevertheless, our data strongly hint at the demand for further standardized prospective multi-center studies, which are clearly needed to confirm the effect of RCI in patients with localized RCC.

5. Conclusions RCI reflects adverse prognostic outcome in surgically treated Stages I and II RCCs (pT1-2N0M0) and abrogates the prognostic difference between localized and stage pT3aN0M0 RCC in terms of DFS. Since conventional pathological assessment of RCI is straightforward and does not require additional expense in time and cost, clinical pathologists should render a concise statement concerning

M. May et al. / Urologic Oncology: Seminars and Original Investigations 28 (2010) 274 –279

RCI when reporting on localized RCC specimens. This would provide not only additional prognostic information in individual cases but also enable selection of patients to be included in further standardized prospective studies, which are clearly needed to confirm the prognostic impact of RCI in localized RCC.

Acknowledgments The authors gratefully acknowledge Inna Spivak, Ph.D., Department of Applied Mathematics, Faculty of Mathematics, Natural Sciences and Computer Science, Brandenburg University of Technology Cottbus (BTU), Germany, for her excellent assistance in the statistical evaluation of the data. They also gratefully acknowledge Peter Stosiek, M.D. and Olaf Kaufmann, M.D., for their contribution in the histopathological assessment of the material investigated, and for allowing the authors to include the data analyzed in this study.

References [1] Jemal A, Siegel R, Ward E, et al. Cancer statistics, 2008. CA Cancer J Clin 2008;58:71–96. [2] Jayson M, Sanders H. Increased incidence of serendipitously discovered renal cell carcinoma. Urology 1998;51:203–5.

279

[3] Pantuck AJ, Zisman A, Belldegrun AS. The changing natural history of renal cell carcinoma. J Urol 2001;166:1611–23. [4] Wallen EM, Pruthi RS, Joyce GF, et al. Urologic Diseases in America Project. Kidney Cancer J Urol 2007;177:2006 –18. [5] Lam JS, Shvarts O, Leppert JT, et al. Renal cell carcinoma 2005: New frontiers in staging, prognostication, and targeted molecular therapy. J Urol 2005;173:1853– 62. [6] Horiguchi A, Ito K, Sumitomo M, et al. Intratumoral lymphatics and lymphatic invasion are associated with tumor aggressiveness and poor prognosis in renal cell carcinoma. Urology 2008;71:928 –32. [7] Mancini V, Battaglia M, Ditonno P, et al. Current insights in renal cell cancer pathology. Urol Oncol 2008;26:225–38. [8] Amin MB, Amin MB, Tamboli P, et al. Prognostic impact of histologic subtyping of adult renal epithelial neoplasms: an experience of 405 cases. Am J Surg Pathol 2002;26:281–91. [9] Klatte T, Chung J, Leppert JT, et al. Prognostic relevance of capsular involvement and collecting system invasion in Stages I and II renal cell carcinoma. BJU Int 2007;99:821– 4. [10] Jeong IG, Jeong CW, Hong SK, et al. Prognostic implication of capsular invasion without perinephric fat infiltration in localized renal cell carcinoma. Urology 2006;67:709 –12. [11] Thomas DH, Verghese A, Kynaston HG, et al. Analysis of the prognostic implications of different tumor margin types in renal cell carcinoma. Histopathology 2003;43:374 – 80. [12] Thoenes W, Störkel S, Rumpelt HJ. Histopathology and classification of renal cell tumors (adenomas, oncocytomas, and carcinomas). The basic cytological and histopathological elements and their use for diagnostics. Pathol Res Pract 1986;181:125– 43. [13] American Joint Committee on Cancer. Kidney. AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002. p 323–5. [14] Kovacs G, Akhtar M, Beckwith BJ, et al. The Heidelberg classification of renal cell tumors. J Pathol 1997;183:131–3.