Does Renal Capsular Invasion Have AnyPrognostic Value in Localized Renal Cell Carcinoma?Evren Süer,* Gül Ergün, Sümer Baltacı and Yasar BedükFrom the Department of Urology, Faculty of Medicine, University of Ankara and Department of Statistics, Faculty of Science, Universityof Hacettepe (GE), Ankara, Turkey

Purpose: We evaluated the impact of renal capsular invasion without perirenal fat invasion in the outcome of patients withlocalized renal cell cancer treated surgically.Materials and Methods: We retrospectively reviewed the records of 249 consecutive patients with localized renal cell cancer(pT1-pT2N0M0) who underwent partial or radical nephrectomy between 1995 and 2007. Pathological staging was doneaccording to the 2002 TNM staging system. Association of clinical and pathological parameters with capsular invasion wasdetermined using the chi-square test. Kaplan-Meier estimations of disease specific survival were generated according tocapsular invasion and the log rank test was used to compare survival according to the variable.Results: Renal capsular invasion was detected in 79 of the 249 patients (31.7%). Of the patients 19 of 96 (19.8%) with pT1a,26 of 87 (29.9%) with pT1b and 34 of 66 (51.1%) with pT2 had renal capsular invasion. Tumor diameter and high grade wereassociated with renal capsular invasion (p �0.05). Mean followup was 40.7 months. Five-year disease specific survival inpatients with and without capsular invasion was 92.7% and 90.4%, respectively (p �0.05). In pT1a, pT1b and pT2 casesdifferences in Kaplan-Meier estimations according to renal capsular invasion were statistically insignificant (p �0.05). Onmultivariate analysis renal capsular invasion was not detected as an independent prognostic factor for disease specificsurvival (HR 0.6582, p � 0.5373).Conclusions: Tumor diameter and high grade were associated with renal capsular invasion in stage pT1–2 renal cell cancer.However, capsular invasion did not have any independent impact on patient survival.

Key Words: kidney; carcinoma, renal cell; mortality; neoplasm invasiveness; neoplasm staging

A s a result of the advance and spreading use of imag-ing, an increase in the localized RCC rate has becomeevident. Most of these tumors are incidental and low

stage (T1 and T2). However, a considerable number of pa-tients with low stage RCC have recurrence that leads todeath as a result of metastasis.1 It is known that RCC is anextremely heterogeneous disease, altering with histopatho-logical properties even in low stage RCC. The TNM stagingsystem is the most studied and most accurate prognosticpredictor available for RCC.2,3 Fuhrman grade, tumor size(which is also related to TNM stage) and microvascularinvasion are the other prognostic factors.4–6 Especially inlow stage RCC cases predicting relapse is more complicated.At this stage identifying prognostic factors divides patientsinto different surveillance protocols and, therefore, interven-tion against tumor relapse is more rapid.

Regardless of tumor size, renal tumors invading perire-nal fat tissue are categorized as pT3a in the TNM stagingsystem. Although there is a significant number of patientswith tumor penetrating into the renal capsule without peri-renal fat tissue invasion, the prognostic value of capsular

Submitted for publication December 8, 2007.* Correspondence: Kelebek Sokak 4/5, GOP Ankara, Turkey,

0090 (telephone: �312 4450016; FAX: 0090 �312 4450017; e-mail:evrenos97@yahoo.com).For other articles on a related topic see pages 338, 343

and 352.

0022-5347/08/1801-0068/0THE JOURNAL OF UROLOGY®

Copyright © 2008 by AMERICAN UROLOGICAL ASSOCIATION

68

invasion is still unclear. Only 2 publications describe therole of capsular invasion in RCC.7,8 The collective findings ofthese studies show that capsular invasion is a poor prognos-tic factor in low stage RCC. In this manner we evaluated ourrenal cell cancer data and analyzed the prognostic role ofcapsular invasion in low stage RCC.

PATIENTS AND METHODS

A total of 249 patients with low stage renal cell cancer(pT1–T2) treated with open partial or radical nephrectomybetween January 1995 and June 2007 were assessed in thisstudy. All patients were evaluated preoperatively with com-puterized tomography and ultrasonography. After surgeryall tumor specimens were examined for tumor size, Fuhr-man grade, histological cell subtype and capsular invasion.Mean postoperative followup was 40.7 months (range 1 to145). Capsular invasion was defined as tumor involvementin the capsule without perirenal fat tissue infiltration. The2002 TNM classification was used for pathological tumorstaging.

The followup protocol included abdominal and chest im-aging twice in year 1. After year 1 chest and abdominalimaging was performed annually.

The chi-square test was applied to evaluate the associa-tion between categorical variables. The t test was used tocompare all mean values of continuous variables. Kaplan-

Meier estimates of disease specific survival were obtained

Vol. 180, 68-71, July 2008Printed in U.S.A.

DOI:10.1016/j.juro.2008.03.060

PROGNOSTIC VALUE OF RENAL CAPSULAR INVASION IN RENAL CELL CANCER 69

according to capsular invasion and differences were exam-ined by the log rank test. To determine independent prog-nostic factors multivariate survival analysis was performedby a Cox regression model with respect to potential influ-encing factors, including capsular invasion, patient age, andlesion size and grade. Statistical significance was consideredat p �0.05. All statistical analyses were performed usingSPSS® software.

RESULTS

Of 249 patients 79 (31.7%) had capsular invasion. Mean ageof patients with and without capsular invasion was 56.7 and57.7 years, respectively. Table 1 shows assessed correlationsbetween capsular invasion and other pathological features.The incidence of capsular invasion in the pathologicalgroups was 19.8%, 29.9% and 51.5% for pT1A, pT1B andpT2, respectively (p � 0.0000). Similarly in high grade(grade 3–4) RCC cases capsular invasion incidence was higherthan in lower grade cases (45.8% vs 27.4%, p � 0.008). Con-sidering histological cell subtype, capsular invasion washigher in the clear cell group, although statistically thisdifference was not significant.

In patients with stage pT1A disease with capsular inva-sion mean � SD pathological tumor size was 3.3 � 0.54 cmand in patients without capsular invasion it was 3.41 � 0.59cm (p � 0.0838). The mean size of stage pT1B tumors inpatients with and without capsular invasion was 5.758 �1.261 and 5.608 � 0.906 cm, respectively (p � 0.534). Meantumor size was 10.11 � 3.14 cm in stage pT2 cases withcapsular invasion and 10.77 � 2.85 cm in cases withoutcapsular invasion (p � 0.370). Consequently in patients withand without capsular invasion compared for every pT stagetumor size was not statistically different in each pT stage.

At the time of analysis tumor had recurred in 2 of 31partial nephrectomy cases (6.45%) and in 16 of 218 radicalnephrectomy cases (7.3%). Five-year disease specific sur-vival in patients with and without capsular invasion was92.68% and 90.4%, respectively. Results were similar for allpathological stages (figs. 1 to 3). For pT1A, pT1B and pT2

TABLE 1. Patient characteristics

Variable

No. CapsularInvasion (%)

TotalNo. (%)

pValueWith Without

Age:Younger than 60 41 (29.3) 99 (70.7) 140 (100) 0.34860 or Older 38 (34.9) 71 (65.1) 109 (100)

Sex:M 44 (28.4) 111 (71.6) 155 (100) 0.146F 35 (37.2) 59 (62.8) 94 (100)

Side:Lt 34 (28.6) 85 (71.4) 119 (100) 0.306Rt 45 (34.6) 85 (65.4) 130 (100)

Stage:pT1AN0M0 19 (19.8) 77 (80.2) 96 (100) 0.0000*pT1BN0M0 26 (29.9) 61 (70.1) 87 (100)pT2N0M0 34 (51.5) 32 (48.5) 66 (100)

Grade:Low (1–2) 52 (27.4) 138 (72.6) 190 (100) 0.008*High (3–4) 27 (45.8) 32 (54.2) 59 (100)

Histological type:Clear cell 67 (35.3) 123 (64.7) 190 0.130Papillary 9 (21.4) 33 (78.6) 42Chromophobe 3 (20.0) 12 (80.0) 15

* Statistically significant (p �0.05).

these parameters were 87.5% and 88.17%, 100% and 96.5%,and 90.03% and 87.45%, respectively, in patients with andwithout capsular invasion. Multivariate analysis was usedto identify independent prognostic factors for disease specificsurvival in all patients. In this model capsular invasion wasnot an independent prognostic factor of disease specific sur-vival (table 2).

DISCUSSION

Morbidity and mortality are the possible outcomes of renalcell cancer. Of the patients 40% die of cancer progressionand metastasis. This makes RCC the most deadly canceramong urological malignancies. Therefore, standardizingand classifying prognostic factors would optimize patienttreatment after surgery. Pathological prognostic factors are

Time (Months)

140120100806040200

Sur

viva

l Pro

babi

lity

1,0

,9

,8

,7

CAP(-)

CAP (+)

FIG. 1. Disease specific survival probability stratified by capsularinvasion in patients with stage pT1A (log rank p � 0.835). CAP(�),capsule negative. CAP(�), capsule positive.

Time (Months)

160140120100806040200

Sur

viva

l Pro

babi

lity

1,00

,90

,80

,70

CAP (-)

CAP (+)

FIG. 2. Disease specific survival probability stratified by capsular

invasion in patients with stage pT1B (log rank p � 0.363). CAP(�),capsule negative. CAP(�), capsule positive.

PROGNOSTIC VALUE OF RENAL CAPSULAR INVASION IN RENAL CELL CANCER70

the most prominent, investigated and important prognosticfactors for RCC.

All staging systems, including those proposed in 1958 byFlocks and Kadesky,9 the Robson staging system10 and theTNM staging system, that are currently in use for RCCprovide valuable prognostic information. The main limita-tions of the Robson staging system is that it fails to distin-guish between patients with venous and lymphatic involve-ment. Using the TNM staging system and separating localand locally advanced RCC improves the predictive power ofstaging systems.

Tumor size is one of the most important adjusting factorsin the TNM staging system. Gulliani11 and Guinan12 et alreported a strong correlation between tumor size and prog-nosis in RCC. Kontak and Campbell noted that tumor size is

Time (Months)

160140120100806040200

Sur

viva

l Pro

babi

lity

1,00

,90

,80

,70

CAP (-)

CAP (+)

FIG. 3. Disease specific survival probability stratified by capsularinvasion in patients with stage pT2 (log rank p � 0.770). CAP(�),capsule negative. CAP(�), capsule positive.

TABLE 2. Multivariate analysis of disease-free survival

Covariate HR 95% CI p Value

Pt age 0.9761 0.9312–1.0231 0.3129Tumor size 1.0101 0.8737–1.1678 0.8923Capsular invasion 0.6582 0.1743–2.4853 0.5373Grade 2.1925 0.6540–7.3508 0.2034

TABLE 3. Current

Present Series

No. pts 249Followup (mos) Mean 40.7No. capsular invasion (%): 79 (31.7)

Stage pT1 45 (24.5)Stage pT2 34 (51.5)Low grade (1–2) 52 (27.4)High grade (3–4) 27 (45.8)

% 5-Yr survival:Pos capsular invasion, pT1a Disease specific 87.5Pos capsular invasion, pTb Disease specific 100Pos capsular invasion, pT2 Disease specific 90Neg capsular invasion, pT1a Disease specific 88.2Neg capsular invasion, pT1b Disease specific 96.5

Neg capsular invasion, pT2 Disease specific 87.5

one of the most considerable prognostic predictive factor forRCC.13

Perirenal fat tissue invasion is a pathological feature thatis the fundamental base of the TNM staging system. Basedon several studies we know that patients with perirenal fattissue invasion have worse 5-year survival than those withlow stage RCC.14,15

Capsular invasion is another pathological feature that isnot used in tumor staging systems. When reviewing ourdata, we found 79 patients with capsular invasion withoutperirenal fat tissue invasion. This number represented ap-proximately a third of the patients with low stage RCC.

Only 2 publications have considered renal capsular inva-sion for low stage RCC. Jeong et al analyzed the prognosticvalue of capsular invasion separately in pT1 and pT2 con-ventional RCC cases.7 They reported that capsular invasionis a poor prognostic factor for pT2 RCC. Related to theseoutcomes, renal capsule involvement was associated withhigher T stage. Klatte et al assessed 519 patients, more thanin the previous study.8 Capsular invasion was more frequentin pT2 stage tumors and it was noted to be a negativeprognostic factor in low stage RCC. Unlike the previousstudy, an association between higher grades and capsularinvasion was noted. Interestingly no survival differenceswere detected between patients with RCC who had capsularinvasion (pT1/2N0M0) and those with perirenal fat tissueinvasion (pT3AN0M0).

In harmony with these publications, we observed a rela-tion between capsular invasion for higher Fuhrman gradeand pathological stage tumors. We evaluated all pathologi-cal stages independently according to capsular invasion for5-year disease specific survival. No statistical difference wasnoted. Table 3 shows similarities and differences among the3 studies.

It is difficult to provide an adequate explanation for thedifferent findings in these 3 studies. Klatte et al reportedthat capsular involvement was a prognostic factor, possiblybecause it was associated with larger and biologically ag-gressive tumors.8 Jeong et al reported that capsular inva-sion was a significant prognostic factor only in pT2 cases,although in that series the mean size of pT2 tumors with andwithout capsular invasion was similar.7 In our study meantumor size was not significantly different in patients withand without capsular invasion at each pT stage. Accordingto the database presented it is suggested that capsular in-vasion is not an indicator of aggressive biological behavior.It could merely reflect the fact that a tumor had a peripheral

previous reports

Jeong et al7 Klatte et al8

288 519Mean 61 Median 49108 (37.5) 112 (21.6)78 (33.9)30 (51.7)65 (36.5)43 (39.1)

Disease specific 92.6 Recurrence-free 75.6

Disease specific 73.8Disease specific 95.6 Recurrence-free 86.9

and

Disease specific 90.5

PROGNOSTIC VALUE OF RENAL CAPSULAR INVASION IN RENAL CELL CANCER 71

origin and its growth commenced from a location closer tothe renal capsule.

In contrast to these studies, we did not detect capsularinvasion as an unfavorable prognostic factor in low stageRCC. Since this is a retrospective study with a relatively lowpatient number, this study has some limitations. Neverthe-less, despite inadequate data on the prognostic value ofcapsular invasion, to our knowledge our study is the firstthat does not demonstrate any relationship between capsu-lar invasion and poor prognosis.

CONCLUSIONS

Although capsular invasion is a frequent pathological fea-ture in patients with low stage RCC, it is not included inTNM staging systems. According to our study capsular in-vasion does not seem to be a prognostic factor in low stageRCC. For more convincing data and clarification in thisregard prospective, randomized studies with a high numberof patients are required.

Abbreviations and Acronyms

RCC � renal cell cancer

REFERENCES

1. Leibovich BC, Pantuck AJ, Bui MH, Ryu Han K, Zisman Aet al: Current staging in renal cell carcinoma. Urol ClinNorth Am 2003; 30: 481.

2. Frank I, Blutte ML, Leibovich BC, Cheville JC, Lohse CM andZincke H: Independent validation of the 2002 AmericanJoint Committee on cancer primary tumor classification forrenal cell carcinoma using a large, single institution cohort.J Urol 2005; 173: 1889.

3. Ficarra V, Martignoni G, Lohse C, Novara G, Pea M, Cavalleri

S et al: External validation of the Mayo Clinic Stage, Size,

Grade and Necrosis (SSIGN) score to predict cancer specificsurvival using a European series of conventional renal cellcarcinoma. J Urol 2006; 175: 1235.

4. Dall’Oglio MF, Arap MA, Antunes AA, Cury J, Leite KR andSrougi M: Impact of clinicopathological parameters inpatients treated for renal cell carcinoma. J Urol 2007;177: 1687.

5. Knight DA and Stadler WM: Prognostic factors in localisedrenal cell cancer. BJU Int 2007; 99: 1212.

6. Park WH and Eisen T: Prognostic factors in renal cell cancer.BJU Int 2007; 99: 1277.

7. Jeong IG, Jeong CW, Hong SK, Kwak C, Lee E and Lee SE:Prognostic implication of capsular invasion without perine-phritic fat infiltration in localized renal cell carcinoma.Urology 2006; 67: 709.

8. Klatte T, Chung J, Leppert JT, Lam JS, Pantuck AJ, Figlin RAet al: Prognostic relevance of capsular involvement andcollecting system invasion in stage I and II renal cell car-cinoma. BJU Int 2007; 99: 821.

9. Flocks RH and Kadesky MC: Malignant neoplasms of the kid-ney: an analysis of 353 patients followed five years of more.J Urol 1958; 79: 196.

10. Robson CJ, Churchill BM and Andersen W: The results ofradical nephrectomy for renal cell carcinoma. J Urol 1969;101: 297.

11. Guilliani L, Giberti C, Martorana G and Rovida S: Radicalextensive surgery for renal cell carcinoma: long-term re-sults and prognostic factors. J Urol 1990; 143: 468.

12. Guinan PD, Vogelzang NJ, Frengen AM, Chimel JS, SylvesterJL, Sener SF et al: Renal cell carcinoma: tumor size, stageand survival. J Urol 1995; 153: 901.

13. Kontak JA and Campbell SC: Prognostic factors in renal cellcarcinoma. Urol Clin North Am 2003; 30: 467.

14. Patard JJ, Leray E, Cindolo L, Ficarra V, Rodriguez A, De LaTaille A et al: Multi-institutional validation of a symptombased classification for renal cell carcinoma. J Urol 2004;172: 858.

15. Lam JS, Klatte T, Patard JJ, Goel RH, Guillè F, Lobel B et al:Prognostic relevance of tumour size in T3a renal cell carci-

noma: a multicentre experience. Eur Urol 2007; 52: 155.