Clinicopathological features and prognosis of synchronous bilateral renal cell carcinoma: an international multicentre experience

Authors


Allan J. Pantuck, Department of Urology, David Geffen School of Medicine at UCLA, 10833 Le Conte Avenue, Room B7–298 A CHS, Los Angeles, CA 90025–1738, USA.
e-mail: apantuck@mednet.ucla.edu

Abstract

An interesting group of papers in this section is headed by two papers on synchronous bilateral renal tumours, one from an international group of authors and one from Germany. The large series of patients are examined carefully by both groups, and the findings should be useful for all who are interested in this area.

Authors from the UK examine the role of lactate dehydrogenase as a prognostic factor for testicular cancer. They found that it had limited sensitivity, specificity and positive predictive value for detecting relapse, with false-positive elevations being common.

OBJECTIVE

To present a multicentre experience and the largest cohort to date of nonmetastatic (N0M0) synchronous bilateral renal cell carcinoma (RCC), as because it is rare the single-institutional experience is limited.

PATIENTS AND METHODS

We retrospectively studied 10 337 patients from 12 urological centres to identify patients with N0M0 synchronous bilateral RCC; the clinicopathological features and cancer-specific survival were compared to a cohort treated for N0M0 unilateral RCC.

RESULTS

In all, 153 patients had synchronous bilateral solid renal tumours, of whom 135 (88%) had synchronous bilateral RCC, 118 with nonmetastatic disease; 91% had nonfamilial bilateral RCC. Bilateral clear cell RCC was the major histological subtype (76%), and papillary RCC was the next most frequent (19%). Multifocality was found in 54% of bilateral RCCs. Compared with unilateral RCC, patients did not differ in Eastern Cooperative Oncology Group performance status (ECOG PS) and T classification, but bilateral RCCs were more frequently multifocal (54% vs 16%, P < 0.001) and of the papillary subtype (19% vs 12%), and less frequently clear cell RCC (76% vs 83%, P = 0.005). For the outcome, patients with nonmetastatic synchronous bilateral RCC and unilateral RCC had a similar prognosis (P = 0.63); multifocality did not affect survival (P = 0.60). Multivariate analysis identified ECOG PS, T classification, and Fuhrman grade, but not laterality, as independent prognostic factors for cancer-specific survival.

CONCLUSIONS

Patients with N0M0 synchronous bilateral RCC and N0M0 unilateral RCC have a similar prognosis. The frequency of a familial history for RCC (von Hippel-Lindau disease or familial RCC) was significantly greater in bilateral synchronous than in unilateral RCC. The significant pathological findings in synchronous bilateral RCC are papillary subtype and multifocality.

Abbreviations
VHL

von Hippel-Lindau

NSS

nephron-sparing surgery

ECOG PS

Eastern Cooperative Oncology Group performance status.

INTRODUCTION

The number of patients developing kidney cancer has steadily increased in recent decades. For 2006, ≈ 39 000 new cases of kidney cancer and 13 000 kidney cancer-related deaths are estimated for the USA [1]. RCC represents the most common kidney cancer and bilateral RCC accounts for 3–5% of these cases [2,3].

Bilateral RCC can occur synchronously or metachronously; both can appear in hereditary and sporadic forms. Hereditary forms imply patients with a positive family history for RCC and patients with von Hippel-Lindau (VHL) disease, the latter being a syndrome with multiple visceral tumours, including RCC, which is the major cause of death in these patients [3–6]. In addition, patients with a hereditary form of kidney cancer often develop RCC when younger.

Bilateral RCC is often distinguished by multifocality [7]; in a recently reported autopsy study, 88% of the multifocal tumours occurred in synchronous bilateral RCC [8]. In addition, there is a higher proportion of papillary RCC in bilateral than unilateral RCC [7], which is considered to be more likely multifocal than other subtypes [9]. Furthermore, many patients with synchronous bilateral RCC are treated with imperative nephron-sparing surgery (NSS) to avoid the need for dialysis [10]. Thus, multifocality could impair complete surgical resection with NSS more than in unilateral RCC, in which multifocality is not as common and does not influence patient survival [9]. Taken together, the surgical treatment of synchronous bilateral RCC might represent a challenge for both long-term tumour control and renal function. Recent studies on patients with synchronous bilateral RCC indicate that NSS provides low long-term haemodialysis rates and a favourable prognosis [2,11], which is comparable to patients with unilateral RCC [7].

Even so, synchronous bilateral RCC is rare and therefore institutional experiences are limited. There are only three notably larger series reporting on 71, 33 and 29 patients, respectively [2,3,7,11]. With these considerations, we examined a larger patient population treated at 12 international academic centres, focusing on pathological data and prognosis, and compared our findings with features of patients treated for unilateral RCC.

PATIENTS AND METHODS

We retrospectively studied 10 337 patients from the institutional databases of the 12 academic centres Magdeburg, Jena (Germany), Rennes, Creteil, Saint-Etienne, Necker Paris, Toulouse, Bordeaux, Rouen, Lyon (France), Benevento (Italy) and Los Angeles, CA, USA, who were treated for a solid renal tumour from 1974 to 2006. In all, 321 patients (3%) with a bilateral solid renal tumour were identified. A synchronous bilateral solid renal tumour was defined as a solid renal tumour diagnosed in the second kidney within 6 months after being diagnosed in the first [2,12]. With this definition, 153 patients (48%) had synchronous bilateral solid renal tumours.

Of this group, 13 patients had bilateral benign tumours including bilateral oncocytoma in eight, bilateral angiomyolipoma in four, and oncocytoma with contralateral metanephric adenoma in one. In addition, five patients had RCC with a contralateral benign tumour, i.e. clear cell RCC with angiomyolipoma in two, clear cell RCC with oncocytoma in one, clear cell RCC with metanephric adenoma in one, and chromophobe RCC with oncocytoma in one. Due to uncertainty about whether distant metastases and concomitant synchronous bilateral RCC reflect extensive metastatic disease, and therefore do not represent two primary RCCs, we excluded 17 synchronous M1 patients (13%) with bilateral RCC from further analysis. Of patients with unilateral tumours available for chart review, 4596 were treated with nephrectomy, with the final diagnosis being nonmetastatic RCC. These 4596 patients and the 118 with nonmetastatic synchronous bilateral RCC form the basis of this comparative study.

Clinical data including age, Eastern Cooperative Oncology Group performance status (ECOG PS), family history for RCC and the presence of VHL disease were collected for the patients. Data on a positive family history, defined as a first-degree relative with RCC (familial RCC), or VHL disease, were available in 70 patients with synchronous bilateral RCC and 759 with unilateral RCC. Pathological data included T classification, Fuhrman grade, multifocality and histological subtype. Tumours were staged in the kidney with the highest tumour stage. Multifocality was defined as the existence of at least two RCCs in the same kidney, and was diagnosed through intraoperative observation and histological examination.

Student’s t-test and the chi-square test or Fisher’s exact test were used to compare continuous and categorical data, respectively. The point of interest was cancer-specific survival. The Kaplan–Meier method was used to generate the survival functions, which were compared using the log-rank test. Because only 15 patients died from synchronous bilateral RCC, a separate multivariate Cox regression analysis for these patients was not meaningful. Thus, we used multivariate Cox regression analysis for all patients, testing the potential prognostic factors ECOG PS, T classification, Fuhrman grade, and the dichotomised variable laterality (unilateral vs bilateral synchronous). A significance level of 0.05 was used for all statistical tests.

RESULTS

Nonmetastatic synchronous bilateral RCC was found in 118 patients, 92 men (78%) and 26 women (22%) with a mean age of 60 years. Most patients had a nonfamilial RCC (91%), while familial RCC was diagnosed in 6% and VHL disease in 3%. The ECOG PS was 0 in 75%, 1 in 22% and 2 in 3% of the patients. The most frequent T classification and Fuhrman grade was pT1 (62%) and grade 2 (60%), respectively. The observed histological types were bilateral clear cell RCC (in 86, 73%), bilateral papillary RCC (19, 16%), clear cell RCC with contralateral papillary RCC (seven, 6%), bilateral chromophobe RCC (five, 4%), and chromophobe RCC with contralateral papillary RCC (one, 1%). Thus, notably, tumours of papillary histology were involved in 27/118 (23%) cases. Multifocal tumours were found in 54% of the specimens, and was more frequent in papillary than in clear cell RCC (66% vs 53%), but this difference was not statistically significant (P = 0.18). The patient and tumour characteristics are summarized in Table 1.

Table 1. 
Patient and tumour characteristics for patients with nonmetastatic synchronous bilateral and unilateral RCC
VariableSynchronous bilateralUnilateralP
  • *

    Data on histological subtype and multifocality are shown for both anatomical sites together.

Mean (sd) age, years 60.0 (11.1)  60.5 (12.5)0.67
N (%):
ECOG PS  0.65
 0 52 (75)3106 (73) 
 ≥1 17 (25) 1152 (27) 
Familial history  0.04
 Familial RCC  4 (6)  22 (3) 
 VHL disease  2 (3)   3 (<1) 
T classification  0.51
 pT1 73 (62)2638 (57) 
 pT2 18 (15) 673 (15) 
 pT3 27 (23)1240 (27) 
 pT4  0  45 (1) 
Fuhrman grade  0.003
 G1 14 (12) 1187 (26) 
 G2 71 (60)2108 (46) 
 G3 29 (25) 1135 (25) 
 G4  4 (3) 166 (4) 
Histological subtype*  0.005
 Clear cell179 (76)3819 (83) 
 Papillary 46 (19) 531 (12) 
 Chromophobe  11 (5)  211 (5) 
 Unclassified  0  32 (<1) 
 Collecting duct  0   3 (<1) 
Multifocality*  <0.001
 Yes 97 (54)  119 (16) 
 No 82 (46) 640 (84) 

Treatment consisted of bilateral surgical procedures in 117 patients, i.e. bilateral partial nephrectomy in 62 (53%), unilateral radical nephrectomy and contralateral partial nephrectomy in 47 (40%) and bilateral radical nephrectomy in eight (7%). Fifty- four patients (46%) had a single procedure, whereas 63 (54%) had staged surgery. The remaining patient was diagnosed at autopsy with synchronous bilateral RCC.

A comparison of the clinicopathological features of bilateral synchronous with unilateral RCC is also shown in Table 1; there was no difference in age (P = 0.67), ECOG PS (P = 0.65) and T classification (P = 0.51), while patients with synchronous bilateral RCC more often had grade 2 and less often grade 1 than those with unilateral RCC. The frequency of familial history (VHL disease or familial RCC) was greater in the bilateral synchronous cases (9% vs 3%, P = 0.04) and the frequency of papillary RCC was higher (P = 0.005). Synchronous bilateral RCC was more often multifocal than in those with unilateral RCC (54% vs 16%, P < 0.001). Multifocality was present more often in bilateral synchronous than in unilateral clear cell RCC (53% vs 16%, < 0.001); the result was similar for just the papillary subtype (66% vs 24%, P < 0.001). Interestingly, the incidence of patients with synchronous bilateral RCC presenting with metastasis was significantly lower than in the unilateral population (13% vs 27%, P < 0.001).

The median (range) follow-up was 3.7  (0.1–24.9) years; in the bilateral and unilateral groups, there were 15 (13%) and 614 (13%) deaths from RCC, respectively. There was no difference when comparing cancer-specific survival among the two groups (P = 0.63) (Fig. 1a). The mean (sem) 5- and 10-year survival rates were 87 (4)% and 78 (8)% for patients with nonmetastatic bilateral synchronous RCC, and 85 (1)% and 75 (1)% for nonmetastatic unilateral RCC, respectively. The presence of multifocality did not affect survival (P = 0.60, Fig. 1b). In multivariate Cox regression analysis, ECOG PS, T classification, and Fuhrman grade, but not laterality, were independent prognostic factors for survival (Table 2).

Figure 1.

Kaplan–Meier survival estimates for: a patients with nonmetastatic synchronous bilateral and unilateral RCC (P = 0.63); and b, comparing patients with nonmetastatic synchronous bilateral RCC and at least one multifocal primary tumour to those without (P = 0.60).

Table 2.  Multivariate Cox regression analysis. ECOG PS, T classification, and Fuhrman grade were independent prognostic factors of survival
VariableCovariate hazard ratio (95% CI)P
ECOG PS1.8 (1.5–2.1)<0.001
T classification2.1 (1.9–2.4)<0.001
Fuhrman grade1.6 (1.4–1.8)<0.001
Laterality1.3 (0.7–2.5)0.437

DISCUSSION

We present a survey of nonmetastatic synchronous bilateral RCC from 12 international institutions, representing a large, worldwide experience. The main finding is that nonmetastatic synchronous bilateral RCC and nonmetastatic unilateral RCC share an equivalent prognosis. The significant pathological findings were a higher frequency of papillary RCC and multifocal tumours, but multifocality did not influence survival. Finally, to our knowledge, a synchronous papillary RCC and contralateral chromophobe RCC has not been described previously.

In the present study the incidence of bilateral RCC was 3%, and in accord with incidences reported by others [2,3]. Nearly half of the patients with bilateral tumours had synchronous bilateral RCC, while almost as many had metachronous bilateral tumours. Hereditary causes have been shown to be risk factors for synchronous bilateral RCC, and although most of the present patients had nonfamilial RCC, the proportion of those with VHL disease and/or a positive family history for RCC in the absence of VHL disease was higher than in the unilateral patients. This finding is supported by a study of Grimaldi et al.[3], who reported their experience with bilateral RCC; 88% had nonfamilial RCC. Although the proportion of patients with a heredity form of bilateral RCC is low, they are at greater risk of synchronous bilateral RCC. The age at presentation did not differ, which is in accordance with a recent study [7].

There were no significant differences between nonmetastatic synchronous bilateral RCC and nonmetastatic unilateral RCC for cancer-specific survival. As a result of the equal distribution of the prognostic factors ECOG PS, T classification and high-grade tumours (G3/4), the survival rates did not differ. As in the present study, Blute et al.[7] were unable to show a difference in cancer-specific survival between synchronous bilateral RCC and unilateral RCC. They reported on 44 patients with sporadic synchronous bilateral RCC and cited a 5-year survival rate of ≈ 80% for both groups.

Bilateral RCC has a prominent pathological feature, i.e. multifocality, which was more frequent in patients with synchronous bilateral than unilateral RCC. The high incidence of multifocality in bilateral RCC was previously reported by Blute et al.[7]. In that study, 28% of the clear cell RCCs and 33% of the papillary RCCs were multifocal; as in the present study, this difference was not statistically significant. In addition, the rates were significantly higher for synchronous bilateral than unilateral RCC. As further support, Wunderlich et al.[8] found, in their autopsy study, that 88% of the multifocal tumours were in synchronous bilateral RCC. Despite this frequency, experience with the surgical management of unilateral RCC clearly showed that multifocality does not affect survival time [9,13]. Thus our hypotheses were that multifocality is more important in bilateral than in unilateral RCC, and therefore might affect survival. With the present data, we confirmed that multifocality is a pathological landmark in synchronous bilateral RCC, but it does not affect survival.

The management of synchronous bilateral RCC presents a challenge, i.e. the effort to combine complete tumour resection and maintain renal function. It is commonly accepted to use NSS to the fullest possible extent. With this goal, a single or a staged procedure can be used. Blute et al.[7,11] used a single procedure in >70% of their patients, and showed peri-operative complication rates that were comparable to those from surgery for unilateral RCC. In addition, the long-term haemodialysis rates after bilateral surgery for synchronous bilateral RCC were 10%[11].

The question of whether bilateral RCC reflects metastatic disease remains unclear. The differentiation of bilateral primary RCC from unilateral RCC with contralateral renal metastasis can be difficult when the tumour histology is the same. According to histological guidelines proposed by Hyman et al.[14], the tumours should preferably occur simultaneously, there should be only one tumour in each kidney, each tumour should show at least partial encapsulation, and the basic histology should be different in each kidney. These guidelines are not meaningful in most cases because only half of patients with bilateral tumours have synchronous disease, multifocality is very frequent, and in most cases both tumours have the same histological type. However, genetic analysis of the tumour might be helpful to differentiate metastatic disease from a primary tumour [15–17]. The low incidence of patients presenting with concomitant metastasis, the low incidence of cancer-specific deaths, and the equivalent survival to unilateral cases are arguments against the idea that bilateral synchronous RCC represents metastatic disease. Of course, a multifocal tumour could reflect diffuse metastases, but multifocality by itself did not influence survival. We conclude that nonmetastatic synchronous bilateral RCC does not reflect a primary RCC with contralateral renal metastases.

The present study has some limitations; conclusions drawn from a retrospective study are always limited, and all participating academic centres are large tertiary referral centres. Thus, the patient composition could differ from that in other hospitals. Finally, the incidence of multifocality might be under-reported, because microscopic multifocality might be undetected during NSS. Despite this possible under-reporting, we think that it is evident that multifocality is the pathological landmark in nonmetastatic synchronous bilateral RCC.

In conclusion, patients with nonmetastatic synchronous bilateral RCC and nonmetastatic unilateral RCC have a similar prognosis. The frequency of familial history for RCC (VHL disease or familial RCC) is significantly greater in bilateral synchronous than in unilateral RCC. The significant pathological findings in synchronous bilateral RCC are papillary subtype and multifocality.

CONFLICT OF INTEREST

None declared.

Ancillary