Objective To evaluate gender as a prognostic factor in patients with renal cell carcinoma (RCC), using a retrospective review of patients with RCC stratified according to gender and analysing factors affecting prognosis.
Patients and methods From January 1957 to December 1995, 768 patients with pathologically defined RCC (all of whom underwent nephrectomy) were classified as having clear cell carcinoma in 662 (follow-up in 648), papillary RCC in 43 (follow-up in 42), chromophobe cell carcinoma in 36 (follow-up in 34) and cyst-associated RCC in 27 (all followed up) according to the criteria proposed by the World Health Organization. The survival rates were compared between men and women, calculated and stratified according to the subtype of RCC.
Results There tended to be a more favourable prognosis in women than in men but the difference was not quite significant ( P =0.061). Of those with clear cell carcinoma, women had a more favourable prognosis than men and the difference in survival was significant ( P =0.012). No other subtype of RCC was associated with a significant difference in survival between the sexes. There was a smaller proportion of patients with stage IV and a larger proportion with stage I disease in women than in men ( P <0.05). Of stage I patients, women had a more favourable prognosis than men ( P <0.011). Women had better survival after recurrence than had men, the difference being significant ( P =0.007).
Conclusion The prognosis is significantly better in women than men with clear cell carcinoma. The factors that contribute to a favourable prognosis in women are the greater proportion of lower stage disease and better survival after recurrence.
The most important clinicopathological factors affecting the prognosis of patients with RCC are the stage of disease and the grade of malignancy [1–4]. Preoperative treatment methods and postoperative therapy and surveillance schedules are decided from these two prognostic factors. Previous experience suggests that some women with recurrent RCC after surgery survive for many years and that the disease in these long-surviving women differed from that in men. Starting with this simple conjecture the survival of men and women with RCC was compared, and the factors that contribute to a favourable and unfavourable outcome in both sexes were clarified.
Patients and methods
Between January 1957 and December 1995, 768 patients with pathologically defined RCC (all of whom had undergone nephrectomy) were treated at four hospitals affiliated to the authors' institution. In these patients the RCC was classified as clear cell carcinoma (conventional) in 662 (86%, no follow-up in 14), papillary RCC in 43 (5.6%, no follow-up in one), chromophobe cell carcinoma in 36 (4.7%, no follow-up in two) and cyst-associated RCC in 27 (3.5%, all followed up) according to the criteria proposed by the WHO . Patients with collecting-duct carcinoma of the kidney, granular cell carcinoma or spindle cell carcinoma were excluded because there were so few (<1%). Furthermore, patients with RCC and acquired cystic disease of the kidney from dialysis, with von Hippel–Lindau disease, tuberous sclerosis or polycystic disease were also excluded.
The stage of the disease was determined according to Robson et al. from the pathological findings, and the nuclear grade according to Fuhrman et al.. The recurrence rate for patients with lymph node negative (N0) and metastasis negative (M0) was determined using the pathologically defined tumour stage (pT stage) classification proposed by Guinan et al.; pT1 was subdivided into pT1a (tumour 4 cm, limited to the kidney) and pT1b (tumour >4 cm but 7 cm, limited to the kidney). The survival rate was calculated using the Kaplan–Meier method with the significance determined using the generalized Wilcoxon method and the Cox–Mantel method. When the difference in survival was significant by both methods, the difference was taken to be significant (with the P value obtained using the generalized Wilcoxon methods). The difference in the proportion of the clinical and pathological factors between genders was calculated using the Kruskal–Wallis H-test.
There tended to be a better prognosis in women than men, but the difference in survival rates was not quite significant (P=0.061; Table 1). In patients with clear cell carcinoma the prognosis was more favourable in women than in men, the difference in survival rates being significant (P=0.012; Table 1; Fig. 1a). In patients with papillary RCC or chromophobe cell carcinoma there was no significant difference in survival with gender (P=0.53 and 0.19; Table 1). There was only one women with cyst-associated RCC and thus it was not possible to compare the survival rates with gender.
Table 1. The survival of men and women stratified according to the subtype of RCC, factors affecting survival in those with clear cell carcinoma, and recurrence rates after nephrectomy according to pT stage
Median (range) survival, days [died from cancer]
All RCC, n
3928 (32–10 026) 
3831 (35–9145) 
Clear cell, n
3565 (39–10 026) 
3961 (35–9145) 
7726 (32–8899) 
3418 (89–7799) 
4655 (32–7081) 
7034 (568–8535) 
5956 (598–8852) 
Clear cell carcinoma, n (%)
Mean (range) age, years
Signs and symptoms at detection, n (%)
Lymph node involvement
Venous invasion (gross or microscopic invasion)
Recurrence rates after nephrectomy in patients with N0M0 clear cell carcinoma, n (%)
Other factors were assessed for their effect on the prognosis of men and women with clear cell carcinoma (Table 1). The proportion of women with stage IVB was less than that of men, and that of stage I greater than in men, both differences being significant (P<0.05). In a comparative analysis of survival with gender and adjusting for stage, women with stage I disease had a better prognosis than men with stage I (P<0.011), but there were no significant differences for stages II–IV. No other factors, e.g. grade of malignancy, age, signs and symptoms at the detection of RCC, lymph node involvement or venous invasion, were significantly different between the sexes in patients with clear cell carcinoma.
The recurrence rate in patients with N0M0 clear cell carcinoma was stratified according to the pT stage; there was no significant difference after nephrectomy with gender (Table 1), but after recurrence, the prognosis was more favourable in women than men (P=0.007; Fig. 1b).
The relationship between prognosis and gender in patients with RCC is controversial, i.e. some reports show a more favourable outcome in women than men [8–10] and others show no significant difference in survival [11–13]. In the present study the overall survival stratified by gender suggested a more favourable prognosis for women, but not significantly so (P=0.061). Lieber et al. reviewed the survival of patients with RCC categorized by gender and reported similar results to the present (P=0.06). For the present patients with clear cell carcinoma the prognosis was significantly better in women than men (P=0.012), but there were no significant differences in survival with other subtypes of RCC.
To understand why women had a better prognosis than men their characteristics must be assessed, including the proportion of stages, because stage is an important prognostic factor. In the report by Lieber et al. men tended to have a more advanced stage of disease, with 56% being stage II–IV, compared with 29% of women. In the present study the trend was similar, with more men of stage IV (30%) than women (20%). Correspondingly, more women had stage I (42%) than men (35%), with both differences being significant (P<0.05). This suggests that the most important factor affecting survival between the sexes was the different proportion of disease stages. Further investigations are needed to determine why women with clear cell carcinoma tend to have lower stage disease and a better prognosis; neither the present nor previous studies have provided a clear answer to this question.
A third factor affecting survival between the sexes was the differing survival after recurrence; women had a significantly better prognosis than men (P=0.007). However, the recurrence rates after nephrectomy, stratified according to the pT stage in N0M0 patients, were not significantly different between the sexes. In a previous report  more women tended to have slow-growing types of tumour than men, but more precise information to support this was not presented. There are reports showing no significant difference in survival between the sexes for patients with metastasis [14,15], but no data on the outcome in recurrent patients were presented. For patients with clear cell carcinoma there are no comparative data currently available.
There is an interesting interpretation of the differences in biological aggressiveness of RCC affecting men and women . Renal tubular cells and renal carcinoma cells are known to have receptors for androgens, oestrogens and progestational steroids . Further investigation into whether the sex differences reported can be accounted for by differences in steroid hormone environment depends on first identifying why women with lower stage disease have a more favourable outcome.
As for the other subtypes of RCC, there were no survival trends with gender in patients with clear cell carcinoma in the present study. Clinical characteristics, e.g. survival, sex ratio, stage and grade distribution, and clinical manifestations, differ where the subtype (variant) of RCC differs [16–18]. Therefore, each factor affecting the prognosis of RCC must be analysed and then stratified according to the subtype of RCC. Although women had a better survival rate than men in the present analysis, further investigation is required to establish data to support gender as an independent prognostic factor in patients with RCC.
T. Onishi, MD, Associated Professor.
Y. Oishi, MD, Professor and Chairman.
H. Goto, MD, Lecturer.
S. Yanada, MD, Research Assistant.
K. Abe, MD, Chief Resident.
T. Onishi, Department of Urology, Aoto University Hospital, Jikei University School of Medicine, 41–2, 6-chome, Aoto, Katsushika-ku, Tokyo 125–8506, Japan. e-mail: email@example.com