SEARCH

SEARCH BY CITATION

Keywords:

  • renal cell carcinoma;
  • neoplasm staging;
  • kidney neoplasms;
  • adrenalectomy;
  • survival

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

BACKGROUND

The significance of adrenal invasion and tumor thrombus in renal cell carcinoma (RCC) has been debated recently. The authors evaluated the associations of direct adrenal invasion, perinephric fat invasion, and tumor thrombus level with outcome to determine whether reclassification would improve the prognostic accuracy of the current primary tumor classification.

METHODS

The authors studied 697 patients treated with nephrectomy for pT3 and pT4 RCC between 1970 and 2000. Associations with outcome were evaluated using Cox proportional hazards regression and prognostic accuracy was measured using the c index.

RESULTS

Among patients with pT3 RCC, direct adrenal invasion was significantly associated with death from RCC (risk ratio, 2.11; P = 0.004). No significant difference in survival was found between patients with pT4 RCC and pT3 tumors with direct adrenal invasion (P = 0.490). Among patients with pT3b RCC, those with level I–III tumor thrombus were significantly more likely to die of RCC compared with patients harboring level 0 tumor thrombus (risk ratio, 1.62; P < 0.001). In addition, patients with fat invasion were more likely to die of RCC compared with pT3 patients without fat invasion (risk ratio, 1.87; P < 0.001). Therefore, patients with pT3 RCC were reclassified into 4 prognostic groups, and this reclassification significantly improved prediction of death from RCC compared with the current classification (c indices of 0.61 vs. 0.55, respectively).

CONCLUSIONS

Direct adrenal invasion from RCC should be reclassified as pT4. In addition, the proposed reclassification for patients with pT3 RCC improved prognostic accuracy. Cancer 2005;. © 2005 American Cancer Society.

According to the 2002 American Joint Committee on Cancer (AJCC), the primary tumor classification for pT3 and pT4 renal cell carcinoma (RCC) incorporates the features of perinephric fat invasion, direct ipsilateral adrenal involvement, tumor thrombus, and extension beyond Gerota's fascia.1 Recently, several authors have published reviews on the significance of these pathologic features.2–4 In response to the recently submitted International Union Against Cancer (UICC) commentary regarding continuous improvement of the TNM classification,5 our study was undertaken to determine if alterations in the current primary tumor classification utilizing these features could improve its prognostic accuracy.

Direct ipsilateral adrenal gland invasion from RCC is rare, found in only 2.5% of radical nephrectomy specimens.6 Although indirect involvement of the adrenal gland by RCC is classified as pM1, direct extension into the adrenal gland is currently classified as pT3a disease along with RCC that invades the perinephric or renal sinus fat.1 Multiple centers have reported that direct adrenal gland invasion renders a poor prognosis compared with perinephric or renal sinus fat invasion alone, and have suggested that adrenal gland invasion be staged as pT3d7 or pT4.6 In addition, patients with renal vein only tumor thrombus are similarly classified as pT3b along with patients with tumor thrombus extending to the level of the diaphragm. We have previously reported data showing that patients with tumor thrombus involving the renal vein only were less likely to die of RCC than patients with a larger tumor thrombus burden.8 Moreover, no distinction is made in the pT3b classification between patients with and without fat invasion. As such, the objective of our study was to evaluate the associations of direct ipsilateral adrenal invasion, fat invasion, and tumor thrombus level with patient outcome to determine if further subclassification of pT3 RCC tumors based on these features could improve the prognostic ability of the current primary tumor classification.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

After approval from the Mayo Clinic institutional review board was obtained, we identified 2667 patients treated with radical nephrectomy or nephron-sparing surgery for unilateral, sporadic RCC between 1970 and 2000. There were 697 (26.1%) patients with pT3a, pT3b, pT3c, or pT4 tumors. A single urologic pathologist (J.C.C.) reviewed the microscopic slides from all tumor specimens without knowledge of patient outcome. Histologic subtype was classified according to the UICC, AJCC, and Heidelberg guidelines.9, 10 Invasion of the adrenal gland can occur through direct extension from the primary kidney tumor or through metastasis. Direct invasion was defined as contiguous spread of the kidney tumor through the peripheral perinephric fat into the ipsilateral adrenal gland. Metastatic invasion was characterized as a discrete nodule of RCC in the ipsilateral adrenal gland with uninvolved adjacent perinephric fat. Peripheral perinephric fat invasion was characterized by tumor extending into the fat surrounding the renal capsule and renal sinus fat invasion was characterized by tumor extending into the fat of the renal sinus. Tumors that were circumscribed and surrounded by a discrete capsule that impinged on the peripheral perinephric fat or renal sinus fat were not considered invasive into fat.

The level of tumor thrombus was classified as 0 (thrombus limited to the renal vein, detected clinically or during assessment of the pathologic specimen), I (thrombus extending ≤ 2 cm above the renal vein), II (thrombus extending > 2 cm above the renal vein, but below the confluence of the hepatic veins), III (thrombus at the level of or above the hepatic veins but below the diaphragm), or IV (thrombus extending above the diaphragm).

Associations among the pathologic features were evaluated using chi-square and Fisher exact tests. Cancer-specific survival was estimated using the Kaplan–Meier method. The duration of follow-up was calculated from the date of surgery to the date of death or last follow-up. Cause of death was determined from the death certificate or physician correspondence and deaths from causes other than RCC were censored. Associations with death from RCC were assessed using log-rank tests and Cox proportional hazards regression models. The predictive ability of the features in these models was evaluated using a c (for concordance) index proposed by Harrell et al.11 The interpretation of the c index is similar to the interpretation of the area under a receiver operating characteristic curve. A value of 1.0 indicates that the features in the model perfectly separate patients with different outcomes, whereas a value of 0.5 indicates that the features contain prognostic information equal to that obtained by chance alone. Statistical analyses were performed using the SAS software package (SAS Institute, Cary, NC) and Ps < 0.05 were considered statistically significant.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Of the 697 patients studied, 214 (30.7%) had pT3a RCC, 436 (62.6%) had pT3b RCC, 19 (2.7%) had pT3c RCC, and 28 (4.0%) had pT4 RCC. Almost all patients had clear cell RCC (n = 633 [90.8%]). The remaining patients had papillary RCC (n = 38 [5.5%]), chromophobe RCC (n = 18 [2.6%]), collecting duct RCC (n = 4 [0.6%]), or RCC not otherwise specified (n = 4 [0.6%]). Five hundred fifty patients were deceased at last follow-up, including 400 patients who died of RCC at a median of 1.4 years after nephrectomy (range, 0–25 years). The median follow-up for the 147 patients who were still alive at last follow-up was 7.6 years (range, 0–33 years).

Direct Ipsilateral Adrenal Invasion

For the purpose of evaluating the impact of direct adrenal invasion, only the 424 (60.8%) patients who underwent adrenalectomy were included in the analysis. There were 28 patients (6.6%) with tumors that directly invaded the ipsilateral adrenal gland. Among these 28 patients, 8 had pT3a disease, 14 had pT3b disease, and 6 were classified as pT4 RCC (Table 1). No patient with pT3c disease had direct adrenal invasion. Of the 424 patients who had an adrenalectomy at the time of nephrectomy, 275 were pNx/pN0 and pM0. The remaining 149 were either pNx/pN0 and pM1 (n = 93), pN1/pN2 and pM0 (n = 31), or pN1/pN2 and pM1 (n = 25). Of the 275 patients with pNx/pN0 and pM0 RCC, 9 had direct adrenal invasion including 7 patients with pT3a/pT3b RCC.

Table 1. Associations of Direct Ipsilateral Adrenal Invasion with the 2002 Primary Tumor Classification
 Patients with direct ipsilateral adrenal invasion, no. (%)
NoYes
  1. P = 0.003

2002 Classification  
pT3a114 (93.4) 8 (6.6)
pT3b253 (94.8)14 (5.2)
pT3c 16 (100.0) 0 (0.0)
pT4 13 (68.4) 6 (31.6)

A comparison of pathologic features among the patients with pT3a or pT3b tumors with and without direct adrenal invasion and all patients with pT4 tumors is shown in Table 2. Patients with pT3a or pT3b disease with direct adrenal invasion were more likely to have regional lymph node involvement (P = 0.016), distant metastases (P = 0.002), and sarcomatoid differentiation at the time of nephrectomy (P = 0.008) compared with patients with pT3a or pT3b tumors without direct adrenal invasion. There were no significant differences in the pathologic features summarized between the 22 patients with pT3a or pT3b tumors with direct adrenal invasion and the 19 patients with pT4 tumors.

Table 2. Comparison of Pathologic Features by the 2002 Primary Tumor Classification and Direct Ipsilateral Adrenal Invasion, No. (%)
FeatureWithout adrenal invasion pT3a/pT3b (%) (n = 367)With adrenal invasion pT3a/pT3b (%) (n = 22)pT4 (%) (n = 19)
  1. RCC: renal cell carcinoma.

RCC histologic subtype   
 Clear cell330 (89.9)19 (86.4)19 (100.0)
 Papillary25 (6.8)2 (9.1)0 (0.0)
 Chromophobe7 (1.9)1 (4.6)0 (0.0)
 Collecting duct2 (0.5)0 (0.0)0 (0.0)
 Not otherwise specified3 (0.8)0 (0.0)0 (0.0)
Regional lymph nodes   
 pN0 or pNx323 (88.0)15 (68.2)15 (79.0)
 pN1 or pN244 (12.0)7 (31.8)4 (21.0)
Distant metastases   
 pM0275 (74.9)10 (45.5)8 (42.1)
 pM192 (25.1)12 (54.6)11 (57.9)
Nuclear grade   
 1/295 (25.9)2 (9.0)0 (0.0)
 3/4272 (74.1)20 (90.9)19 (100.0)
Histologic tumor necrosis   
 No170 (46.3)6 (27.3)1 (5.3)
 Yes197 (56.7)16 (72.7)18 (94.7)
Sarcomatoid differentiation   
 No339 (92.4)16 (72.7)14 (73.7)
 Yes28 (7.6)6 (27.3)5 (26.3)

Cancer-specific survival for the 22 patients with pT3a or pT3b tumors with direct adrenal invasion and patients with pT3a, pT3b, or pT3c tumors without direct adrenal invasion is shown in Figure 1. For comparison, cancer-specific survival for pT4 tumors is also depicted. The estimated 5-year cancer-specific survival rates for patients with pT3a, pT3b, and pT3c tumors that did not invade the adrenal gland were 53.9%, 42.7%, and 42.7%, respectively. Conversely, the estimated 5-year cancer-specific survival rates for the 22 patients with pT3a and pT3b tumors that directly invaded the adrenal gland and patients with pT4 tumors were 20.2% and 14.0%, respectively. The cancer-specific survival for patients with pT3a or pT3b tumors that directly invaded the adrenal gland was significantly different from the cancer-specific survival for patients with pT3a (P < 0.001) or pT3b (P = 0.011) disease that did not invade the adrenal gland. Among patients with pT3a and pT3b RCC, direct adrenal invasion was significantly associated with death from RCC (risk ratio, 2.11; 95% confidence interval [CI], 1.27–3.52; P = 0.004). Even among the subset of pT3a and pT3b patients with pNx/pN0 pM0 RCC, direct adrenal invasion was significantly associated with death from RCC (risk ratio, 2.90; 95% CI, 1.27–6.61; P = 0.011; Fig. 2). Furthermore, there was not a significant difference in cancer-specific survival between the patients with pT3a and pT3b tumors that directly invaded the adrenal gland and patients with pT4 tumors, either among all the patients studied (P = 0.490) or among the subset with pNx/pN0 pM0 RCC (P = 0.754).

thumbnail image

Figure 1. Cancer-specific survival by the 2002 primary tumor classification for 424 patients who underwent adrenalectomy for pT3a (fat invasion only), pT3b, pT3c, or pT4 RCC, or pT3 renal cell carcinoma with direct ipsilateral adrenal invasion.

Download figure to PowerPoint

thumbnail image

Figure 2. Cancer-specific survival by the 2002 primary tumor classification for the 275 patients who underwent adrenalectomy for pNx/pN0 pM0 RCC with pT3a (fat invasion only), pT3b, pT3c, or pT4 tumors, or pT3 tumors with direct ipsilateral adrenal invasion.

Download figure to PowerPoint

Tumor Thrombus and Fat Invasion

For the purpose of evaluating the impact of tumor thrombus level and perinephric fat invasion, patients with pT3 disease with direct adrenal invasion were excluded from analysis. Of the remaining 675 patients, 450 were classified as pNx/pN0 pM0 at the time of nephrectomy. Table 3 summarizes the presence and level of tumor thrombus and the type of fat invasion by the 2002 tumor classifications.

Table 3. Summary of Tumor Thrombus Level and Fat Invasion by the 2002 Primary Tumor Classification, No. (%)
Features2002 Classification
pT3apT3bpT3cpT4
Tumor thrombus level    
 None206 (100)0 (0)0 (0)13 (46.4)
 00 (0)283 (67.1)0 (0)10 (35.7)
 I0 (0)52 (12.3)0 (0)3 (10.7)
 II0 (0)64 (15.2)0 (0)2 (7.1)
 III0 (0)23 (5.5)0 (0)0 (0)
 IV0 (0)0 (0)19 (100)0 (0)
Fat invasion    
 None0 (0)189 (44.8)5 (26.3)0 (0)
 Perinephric166 (80.6)121 (28.7)12 (63.2)24 (85.7)
 Renal sinus13 (6.3)47 (11.1)0 (0)0 (0)
 Both27 (13.1)65 (15.4)2 (10.5)4 (14.3)

Cancer-specific survival by the 2002 primary tumor classification is shown in Figure 3. The 5-year estimated cancer-specific survival rates after nephrectomy were 54.7%, 45.9%, 34.4%, and 18.1% for patients with pT3a, pT3b, pT3c, and pT4 RCC, respectively. The associations of the 2002 tumor classification with death from RCC, both univariately and after adjusting for regional lymph node involvement and distant metastases at the time of nephrectomy, are summarized in Table 4. The c index for the model containing the 2002 tumor classification alone was 0.548. The c index increased to 0.704 after accounting for regional lymph node involvement and distant metastases at the time of nephrectomy, but the 2002 tumor classification was no longer statistically significantly associated with outcome.

thumbnail image

Figure 3. Cancer-specific survival by the 2002 primary tumor classification for 675 patients with pT3 and pT4 renal cell carcinoma.

Download figure to PowerPoint

Table 4. Association of the 2002 Primary Tumor Classification with Death from RCC
UnivariateRisk ratio (95% CI)P value
  1. RCC: renal cell carcinoma; 95% CI: 95% confidence interval.

pT3a1.0 (reference) 
pT3b1.25 (1.00–1.57)0.055
pT3c1.77 (0.95–3.29)0.073
pT42.75 (1.72–4.39)< 0.001
 C index = 0.548 
Adjusted for N and M classification
pT3a1.0 (reference) 
pT3b1.14 (0.91–1.43)0.268
pT3c1.54 (0.83–2.88)0.171
pT41.54 (0.95–2.48)0.079
 index = 0.704 

Of the 422 patients with pT3b RCC, patients with level I–III tumor thrombus were significantly more likely to die of RCC compared with patients with level 0 tumor thrombus (risk ratio, 1.62; 95% CI, 1.25–2.11; P < 0.001) even after adjusting for regional lymph node involvement and distant metastases (risk ratio, 1.66; 95% CI, 1.27–2.16; P < 0.001). Similarly, patients with perinephric or renal sinus fat invasion were close to twice as likely to die of RCC compared with patients without fat invasion both univariately (risk ratio, 1.87; 95% CI, 1.45–2.42; P < 0.001) and after adjusting for regional lymph node involvement and distant metastases at the time of nephrectomy (risk ratio, 1.84; 95% CI, 1.42–2.38; P < 0.001). Therefore, patients with pT3 RCC were reclassified into 4 groups (new primary tumor classification noted with the superscript “N”): pT3aN, thrombus level 0 without fat invasion (n = 140); pT3bN, fat invasion only (n = 206); pT3cN, thrombus level 0 with fat invasion or thrombus level I–III without fat invasion (n = 192); and pT3dN, thrombus level I–III with fat invasion or thrombus level IV (n = 109). There were 28 patients with tumors that extended beyond Gerota's fascia into adjacent organs, which remained classified as pT4. Cancer-specific survival for these 5 groups is shown in Figure 4. The estimated 5-year cancer-specific survival rates after nephrectomy were 63.5%, 54.7%, 42.4%, 25.5%, and 18.1%, respectively, for the 5 groups.

thumbnail image

Figure 4. Cancer-specific survival by the proposed primary tumor classification for 675 patients with pT3 and pT4 RCC: pT3aN, thrombus level 0 without fat invasion; pT3bN, fat invasion only; pT3cN, thrombus level 0 with fat invasion or thrombus level I–III without fat invasion; pT3dN, thrombus level I–III with fat invasion or thrombus level IV; pT4, extension beyond Gerota's fascia.

Download figure to PowerPoint

The associations of this proposed primary tumor classification with death from RCC, both univariately and after adjusting for regional lymph node involvement and distant metastases at the time of nephrectomy, are summarized in Table 5. The c index for the model with the proposed reclassification alone was 0.610, which increased to 0.733 after adjusting for regional lymph node involvement and distant metastases at the time of nephrectomy. It is noteworthy that all components of the proposed reclassification were statistically significantly different from the reference group consisting of patients with level 0 tumor thrombus in the absence of fat invasion after adjusting for regional lymph node involvement and distant metastases.

Table 5. Association of the Proposed Primary Tumor Classification with Death from RCCa
UnivariateRisk ratio (95% CI)P value
  • RCC: renal cell carcinoma; 95%; CI: 95% confidence interval.

  • a

    pT3aN, thrombus level 0 without fat invasion; pT3bN, fat invasion only; pT3cN, thrombus level 0 with fat invasion or thrombus level I–III without fat invasion; pT3dN, thrombus level I–III with fat invasion or thrombus level IV; pT4, extension beyond Gerota's fascia.

pT3aN1.0 (reference) 
pT3bN1.27 (0.93–1.73)0.127
pT3cN1.82 (1.35–2.46)< 0.001
pT3dN2.57 (1.83–3.60)< 0.001
pT43.50 (2.14–5.74)< 0.001
 c index = 0.610 
Adjusted for N and M classification
pT3aN1.0 (reference) 
pT3bN1.41 (1.03–1.92)0.031
pT3cN1.85 (1.37–2.50)< 0.001
pT3dN2.54 (1.80–3.58)< 0.001
pT42.14 (1.30–3.52)0.003
 c index = 0.733 

The associations of the 2002 and proposed primary tumor classifications with cancer-specific survival among the subset of 450 patients with pNx/pN0 pM0 RCC are illustrated in Figures 5 and 6 and summarized in Table 6. Even among this subset, the predictive ability of the proposed classification was superior to that of the 2002 classification (c indexes of 0.635 vs. 0.561, respectively).

thumbnail image

Figure 5. Cancer-specific survival by the 2002 primary tumor classification for 450 patients with pT3–4 pNx/pN0 pM0 renal cell carcinoma.

Download figure to PowerPoint

thumbnail image

Figure 6. Cancer-specific survival by the proposed primary tumor classification for 450 patients with pT3–4 pNx/pN0 pM0 RCC: pT3aN, thrombus level 0 without fat invasion; pT3bN, fat invasion only; pT3cN, thrombus level 0 with fat invasion or thrombus level I–III without fat invasion; pT3dN, thrombus level I–III with fat invasion or thrombus level IV; pT4, extension beyond Gerota's fascia.

Download figure to PowerPoint

Table 6. Association of the 2002 and Proposed Primary Tumor Classifications with Death from RCC among Patients with pNx/pN0, pM0 RCCa
 Risk ratio (95% CI)P value
  • RCC: renal cell carcinoma; 95%; CI: 95% confidence interval.

  • a

    pT3aN, thrombus level 0 without fat invasion; pT3bN, fat invasion only; pT3cN, thrombus level 0 with fat invasion or thrombus level I–III without fat invasion; pT3dN, thrombus level I–III with fat invasion or thrombus level IV; pT4, extension beyond Gerota's fascia.

2002 Classification  
pT3a1.0 (reference) 
pT3b1.36 (0.98–1.87)0.064
pT3c2.26 (0.90–5.68)0.082
pT45.58 (2.52–12.38)< 0.001
 c index = 0.561 
Proposed classification  
 pT3aN1.0 (reference) 
 pT3bN1.28 (0.83–1.98)0.262
 pT3cN2.16 (1.42–3.27)< 0.001
 pT3dN3.03 (1.87–4.92)< 0.001
 pT47.35 (3.22–16.76)< 0.001
 c index = 0.635 

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Tumor classification systems provide important prognostic information, improve international communication, identify targets for research, and guide follow-up recommendations. The 2002 primary tumor classification for pT3 and pT4 RCC incorporates the features of fat invasion, direct ipsilateral adrenal involvement, tumor thrombus, and extension beyond Gerota's fascia.2 The significance of each of these pathologic features has been debated,2–4, 6, 8 and our study was undertaken to determine if alterations in the current classification regarding these features could improve prognostic accuracy.

Direct adrenal gland invasion is rare. Han et al.6 reported that only 2.5% of 1087 patients demonstrated direct adrenal invasion. Similarly, only 28 (1.0%) of the 2667 patients with RCC treated with nephrectomy at our institution had confirmed direct adrenal invasion. Of 214 patients with pT3a disease, only 4 (1.9%) patients had direct adrenal invasion without positive lymph nodes or metastatic disease at the time of nephrectomy.

Due to the infrequent nature of direct ipsilateral adrenal gland invasion, few centers have reported its prognostic significance. Although reports have suggested that adrenal invasion from RCC should be classified in a different category,6, 7 no change to the recent update occurred. Han et al.6 compared survival in 27 patients with pT3a disease with direct adrenal invasion with 187 patients with pT3a disease without adrenal invasion. The median survival for patients with adrenal invasion was significantly worse compared with patients without adrenal invasion (12.5 months vs. 36 months, P < 0.001), and was similar to patients with pT4 disease (11 months). However, the natural history of surgically treated RCC in the study by Han et al.6 was influenced by the administration of immunotherapy, to which there was a 15% response in patients with adrenal invasion compared with a 29% response in patients without adrenal invasion. Although this difference was not statistically significant, 76% of patients with metastatic disease and adrenal invasion received immunotherapy compared with 96% of patients without adrenal invasion and metastatic disease. The authors suggested that adrenal gland invasion be staged as pT4, but recognized that external validation was needed.

Direct adrenal invasion has a poor prognosis. The 5-year cancer-specific survival rate for patients with pT3a and pT3b disease and direct adrenal invasion was 20.2% in the current study. Conversely, patients without direct adrenal invasion had significantly higher 5-year cancer-specific survival rates: 53.9% (P < 0.001) and 42.7% (P = 0.011) for patients with pT3a and pT3b disease, respectively. Furthermore, there was not a significant difference in survival between patients with pT4 disease and patients with direct adrenal invasion (P = 0.490). The current study supports the growing belief that direct adrenal invasion is biologically more aggressive compared with perinephric or renal sinus fat invasion alone. Multiple other tumors that invade adjacent organs, including the prostate, bladder, colon, and stomach, are classified as pT4.1 In addition, primary adrenal carcinoma that invades the kidney is also classified as pT4 disease.12, 13 Our findings, in conjunction with previously published reports, help to solidify the recommendation that adrenal gland invasion from RCC should similarly be classified as pT4.

Currently, the 2002 primary tumor classification for RCC does not differentiate between patients with tumor thrombus below the diaphragm of various levels. This is supported by a recent review of 226 patients with tumor thrombus by Kim et al.2, which demonstrated that the level of tumor thrombus did not impact prognosis when other variables were considered. However, we have previously reported that patients with tumor thrombus involving the renal vein only were less likely to die of RCC compared with patients with a larger tumor thrombus burden.8 This is in agreement with the findings of Moinzadeh and Libertino3 in a review of 153 patients with venous tumor thrombus. In the current study, patients with pT3b RCC and renal vein only tumor thrombus had a better prognosis than patients with pT3b RCC and level I–III tumor thrombus.

The impact of fat invasion on prognosis for patients with RCC is well documented. Gettman et al.14 previously demonstrated that perinephric fat invasion portends worse prognosis among patients with venous tumor thrombus. The current study demonstrates improved prognostic ability by accounting for both tumor thrombus level and fat invasion among patients with pT3 RCC.

Our data indicate that both tumor thrombus level and the presence of perinephric or renal sinus fat invasion are important considerations when predicting outcome for patients with pT3 RCC, even after accounting for regional lymph node involvement and distant metastases at the time of nephrectomy. Furthermore, among patients with pT3 and pT4 RCC, the prognostic ability of the current classification is minimal with a c index of 0.548, which is only slightly better than chance alone. After accounting for the complete TNM classification, the c index increased to 0.704. Our proposed tumor classification, conversely, accounts for the importance of tumor thrombus level and fat invasion and still contains information to predict death from RCC after accounting for regional lymph node involvement and distant metastases. The c indices for the proposed classification univariately and after adjusting for regional lymph node involvement and distant metastases were 0.610 and 0.733, respectively.

Although the primary tumor classification proposed in the current study demonstrated an improvement in prognostic accuracy using our patient population, further improvements and external validation will be necessary in the future. In addition to refinements to the tumor classification, other pathologic, molecular, and genetic features should be considered in multifactorial prognostic models, several of which already have been proposed and found to be superior to the prognostic ability of the primary tumor classification alone.15–17

In summary, the 2002 primary tumor classification for RCC makes no distinction between a tumor that is limited to the renal vein with no fat invasion and one that extends above the renal vein or involves the peripheral perinephric or renal sinus fat. We propose a tumor classification that accounts for the independent effects of fat invasion and tumor thrombus level. Our findings also indicate that patients with tumors directly invading the ipsilateral adrenal gland experience similar outcomes to patients with pT4 tumors, and we believe they should be similarly staged. Our suggested revisions to the current primary tumor classification represent a significant improvement in its prognostic accuracy.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES