Int. J. Cancer: 55,566-570 (1993) 0 1993 Wiley-Liss, Inc.

Publication of the International Union Against Cancer Publication de I'Union lnternationale Contre le Cancer

TUMOR-CELL PROLIFERATION AND PROGNOSIS IN RENAL-CELL CARCINOMA Per LARSSON'.3, Goran ROOS2, Roger STENLINC~ and Borje LJUNGBERG1,4 Departments of 'Urology and Andrology and 2Pathology, University of UrneB, UmeB; and 3Department of Urology, Karolinska Institute, Sodersjukhuset, Stockholm, Sweden.

In the present study the prognostic value of tumor S-phase fraction was evaluated in 69 patients with renal-cell carcinoma. The proportion of S-phase cells was calculated from DNA histograms by flow cytometry in multiple samples from each tumor. The mean tumor S-phase percentage varied between 1.0 and 17.0%, mean 7.5%, with a significant difference be- tween diploid and non-diploid tumors. Stage-I tumors showed significantly lower S-phase values than tumors of stages 11/111 and IV. Histopathological grade correlated with fraction of cells in S-phase. Nineteen tumors were homogeneously diploid, I was tetraploid and 49 were aneuploid. Heterogeneity concern- ing S-phase values was found in 46 of the 69 tumors and concerning DNA ploidy in 34 tumors. Survival time was longer for patients with diploid tumors than for those with aneuploid tumors. Patients with S-phase values <7.5% had a longer survival time than patients having tumors with S-phase values >7.5%. Within the group of aneuploid tumors, patients with S-phase values <7.5% had a significantly better prognosis. In multivariate analysis, only tumor stage and S-phase gave signifi- cant independent prognostic information. The S-phase fraction seems to be an additional prognostic parameter for patients with renal-cell carcinoma. o 1993 WiCey-Liss, Inc.

The clinical behavior of renal-cell carcinoma varies consider- ably. It is therefore important to identify parameters predict- ing the clinical outcome for individual patients. Throughout the years, numerous factors have been suggested to carry prognostic information, clinical stage and histopathologic grade representing the basis for prognostic assessment in renal-cell carcinoma (Medeiros et al., 1988; Stroke1 et al., 1989). Flow cytometric analysis of deoxyribonucleic acid content (DNA) in renal-cell carcinoma has been the subject of reports emphasiz- ing the advantage of DNA analysis, as a supplement to the grading and staging systems, to attain a more reliable predic- tion of disease progression and survival (Hoppel et al., 1986; Ljungberg et al., 1986; Grignon et al., 1989).

The S-phase percentage determined from a DNA histogram provides an indication of the tumor proliferation rate, and has been shown to be of prognostic relevance in various epithelial tumors, such as bladder, prostate and breast carcinoma (Tri- bukait et al., 1982; Barlogie et al., 1983; Arnerlov et al., 1992). Concerning renal-cell carcinoma, earlier reports have de- scribed a correlation between aneuploid tumors and high S-phase fractions (Baisch et al., 1982; Chin et al., 1985; Ljungberg et al., 198.5). The aim of this study was to evaluate the proportion of cells in S-phase as a prognostic factor for patients with renal-cell carcinoma.

MATERIAL AND METHODS Patients

Determination of DNA content and calculation of S-phase cells was performed in 82 patients with renal-cell carcinoma, who had undergone perifascial nephrectomy at the University Hospital, Umel, Sweden, between December 1985 and March 1991. There were 53 men and 29 women, mean age 64.7 5 12.1 years (range 25-85 years). Follow-up times for surviving patients ranged from 12-73 months (median 37 months). The patients were examined preoperatively and followed according to a program including clinical examination, pulmonary X-ray, bone scintigram and computerized tomography.

Tumors The nephrectomized kidney with tumor was divided, then 1

kidney cortex sample and 4-6 tumor samples were taken in a schematic manner as described (Ljungberg et al., 1985). Each sample was divided into 2 parts, one for histopathologic examination and one for DNA flow cytometry.

Clinical stage and histopathological examination Clinical stage at time of surgery was defined according to

Robson et al. (1969). Histopathological grade was defined according to Skinner et al. (1971). In the grading process the lowest degree of differentiation in the tumor was taken into account. All specimens were re-examined by one pathologist (R.S.).

DNA and S-phase analysis Methods used for flow cytometric DNA analysis have been

described (Ljungberg et al., 1985). Briefly, the fresh samples were minced and stained using a propidium iodide solution according to Vindel~v el al. (1983). The samples were filtered and run in a FACScan flow cytometer (Becton-Dickinson, San Jose, CA). The coefficient of variation (CV) in the FACScan was less than 3% for normal lymphocytes.

The individual tumor samples were designated as diploid (DNA index = 1.0) when only 1 peak was found and aneuploid when 2 separate peaks were detected, since it was assumed that all tumor samples contained normal as well as tumor cells. The kidney cortex tissue samples were used as standards for diploidy. Cell populations with DNA indices between 1.95 and 2.05 were judged to be tetraploid if a definite S-phase could be visualized and the main peak represented more than 15% of all cells. Heterogeneity was defined as intratumoral cell popula- tions with a difference in DNA index of at least 0.2. A tumor was designated as diploid when all samples were diploid, and aneuploid when at least one of the 4 to 6 analysed samples was aneuploid/ tetraploid.

The proportion of S-phase cells was calculated from the DNA histograms using a mathematical model according to Baisch et al. (1975). The different cell-cycle compartments were calculated with a computer program using 2 decimals, i.e. the percentage. Histograms with multiple peaks, near-diploid peaks, background debris or too few cells were excluded. The histograms were corrected for background noise. In 3 tumors, S-phase fractions were not evaluable in any sample and in 10 none of the aneuploid samples could be evaluated. In these 13 tumors the DNA histograms did not allow adequate S-phase analysis and thus these 13 patients were excluded from the study. The S-phase values obtained from all samples of an individual tumor were used to calculate the mean S-phase of a tumor. In tumors with both diploid and aneuploid cell clones, the mean S-phase percentage used for statistical comparison of the tumor was calculated from the aneuploid samples.

4To whom correspondence and reprint requests should be sent, at the Department of Urology and Andrology, University of Umel, S-901 85 Umei, Sweden.

Received: April 14,1993 and in revised form June 21,1993.

TUMOR-CELL PROLIFERATION IN RENAL-CELL CARCINOMA 567

Intratumoral heterogeneity in S-phase was defined as an S-phase difference of 3% or more between the tumor samples.

Statistical analysis For statistical analysis the Mann-Whitney U-test was used.

Survival curves were obtained with the Kaplan-Meier method and survival analysis using the log-rank test.

RESULTS

Among the 69 patients, 22 had tumors in stage I (32%), 2 in stage I1 (3%), 10 in stage IIIa (14%), 9 in stage IIIb (13%) and 26 in stage IV (38%). Twelve tumors were grade 2 (17%), 33 grade 3 (48%) and 24 were grade 4 (35%).

DNA analysis Nineteen (28%) tumors were homogeneously diploid. Aneu-

ploidy was evident in 49 tumors (71%). One tumor (1%) was tetraploid and was referred to as an aneuploid tumor in subsequent evaluation. Heterogeneity was shown in 34 (68%) of the 50 non-diploid tumors, 28 had both diploid and aneuploid cell clones and 6 showed multiple aneuploid cell populations. The remaining 16 aneuploid tumors were homo- geneous concerning ploidy. The DNA indices in non-diploid samples varied between 0.8 and 2.5, mean value 1.7.

The distribution of diploid and aneuploid tumors in relation to stage and histopathological grade is shown in Tables I and 11.

S-phase analysis

TABLE I11 - S-PHASE FRACTIONS IN RELATION T O TUMOR DNA CONTENT, CLINICAL STAGE AND HISTOPATHOLOGICAL GRADE IN 69

PATIENTS WITH RENAL-CELL CARCINOMA

S-phase percentage Mean SD Range Number

Kidney cortex samples 2.5 1.3 1.0 - 7.0 59 Tumors 7.5 4.2 1.0- 17.5 69 Diploid tumors 3.1 1.4 1.0-6.0 19 Aneuploid tumors

Aneuploid samples 9.1 3.7 2.0 - 17.5 50 Diploid samples 2.9 1.5 1.0-7.0 28

cal stage 4.5 2.8 1.3 - 10.5 22

Clinii T 1

I1 IIIa IIIb

8.6 4.8 5.3 - 12.0 2 9.9 4.7 3.5- 17.5 10 8.7 3.3 3.3- 13.8 9

IV 8.5 4.2 1.0- 17.0 26 Histopathologic grade

1 n 2 3.5 2.7 1.0- 10.5 12 3 7.3 3.8 1.8- 16.3 33 4 9.5 4.0 3.0- 17.5 24

Diploid DNA content

10

The mean S-phase percentage in 69 tumors varied between 1.0 and 17.5%, mean value 7.5% and median 7.0% (Table 111). Among the 19 diploid tumors, the mean S-phase value was 3.1%, significantly different from 9.1% in the 50 aneuploid tumors 0, < 0.001). The 16 homogeneously aneuploid tumors had a mean S-phase value of 9.4 k 4.2% (range 2.3-17.5) not different from 8.3 2 2.9% (range 6.0-13.0) in the 5 tumors with

8

Q)

2 6 - multiple aneuploid clones or 9.1 I ~ I 3.6% (range 2.0-17.0) in the aneuploid clones in the 26 aneuploid tumors containing

tumors compared with the diploid samples in heterogeneous

Heterogeneity concerning S-phase values was found in 46

values were 2.5 k 1.4% (0-5.0%) in diploid tumors and 6.0 k

c 0)

d) a Q)

c

2 diploid tumor-cell clones. There was no statistical difference between S-phase values of diploid tumor samples in diploid

aneuploid tumors.

tumors, 11 of the 19 diploid (57%) and 35 of the 50 aneuploid tumors (70%). The mean intratumoral difference in S-phase

4.8% (0-15%) in aneuploid tumors, as illustrated in Figures 1 and 2. There was a clear correlation between mean and maximal S-phase values in the individual tumors (r = 0.927,

1 4

e v)

2

TABLE I - CLINICAL STAGE ACCORDING T O ROBSON ETAL. (1969) IN RELATION TO DNA CONTENT IN 69 RENAL-CELL CARCINOMAS lndlvldual cases

Clinical stage I I1 IIIa IIIb IV

DNA content Total

Diploid 11 1 2 1 4 19 Aneuploid 11 1 8 8 22 50 Total 22 2 10 9 26 69

TABLE I1 - HISTOPATHOLOGICAL GRADE T O SKINNER ETAL. (1971) IN RELATION TO DNA CONTENT IN 69 RENAL-CELL CARCINOMAS

Histopathological grade 1 2 1 4

DNA content Total ~~ ~

Diploid - 9 8 2 19 Aneuploid - 3 25 22 50 Total - 12 33 24 69

FIGURE 1 - Heterogeneity of S-phase values in multiple samples from 19 diploid renal-cell carcinomas; data represent the mean and range for each tumor.

p < 0.001). The maximal S-phase values were 9.0% in diploid and 21.0% in aneuploid tumors. The mean value of each tumor was further used for statistical analysis since the maximal S-phase values gave no additional information.

S-phase values in relation to stage, grade and DNA content are given in Table 111, showing a difference between stage-I compared with stage-11-111 and stage-IV tumors, p < 0.001 respectively. The S-phase fractions in grade-2 tumors were statistically different compared with both grade-3 and grade-4 tumors ( p = 0.002 andp < 0.001 respectively). Also, between grade 3 and 4 tumors, a significant difference was present ( p = 0.034) (Table 111).

LARSSON ET AL. 568 22

20

18

16

14

€ i 12

Anruploid DNA content

FIGURE 2 - Heterogeneity of S-phase values in multiple samples from 50 aneuploid renal-cell carcinomas; data represent the mean and range for each tumor. The mean value represents the mean of the aneuploid cell populations of each tumor when the range represents S-phases of both aneuploid and diploid tumor samples.

In 59 patients S-phase analysis on kidney cortex tissue samples could be performed, showing a mean S-phase percent- age of 2.5% (Table 111). No difference in S-phase was found between kidney cortex tissues from kidneys with diploid and aneuploid tumors. S-phase values of the kidney cortex tissue differed significantly from those of all the diploid tumor cell samples ( p < 0.008).

Survival During the follow-up period, 42 patients had died, 35 of

renal cell carcinoma, with a mean survival time of 16.4 months (range 1-55 months), and 7 due to intercurrent diseases, with a mean survival time of 23.1 months (range 1-60 months). Upon dividing the patient material into 2 groups with a breakpoint at the mean S-phase value of the tumors (7.5%), we observed a significant difference in survival time between these 2 groups ( p < 0.001, log-rank test, Fig. 3). Statistically significant differences were also achieved using breakpoints varying from 5.5 to 9.5%. No difference in survival time was shown for

p' 0 . 0 0 1

0 1 5 3 0 6 5 6 0 75 I(0wrns

FIGURE 3 - Kaplan-Meier survival curves comparing outcome of 37 atients with S-phase values <7.5%, the mean S-phase value, wit[ outcome of 32 patients with S-phase values > 7.5%.

---- - I

0 1 5 30 4 5 60 75

M O N T H S

FIGURE 4 - Kaplan-Meier survival curves of 26 patients with distant metastases, stage IV, comparing survival of 11 patients with S-phase values <7.5% with outcome of 15 patients with S-phase values > 7.5%.

patients with distant metastases ( p = 0.27, Fig. 4) in contrast to patients without metastases ( p < 0.001, Fig. 5).

There was a significant correlation between DNA content and survival for all patients ( p = 0.002) and also in the group of patients without metastases ( p = 0.014, Fig. 6). For patients with distant metastases we found no difference in survival time between patients with diploid and aneuploid tumors, using the log rank test. However, 4 patients with diploid tumors had a 50% survival rate a t 22 months compared with 6 months for 26 patients with aneuploid tumors. Patients with aneuploid tu- mors and S-phase values < 7.5% survived significantly longer than patients having S-phase values > 7.5% ( p = 0.012, Fig. 7). In tumors with S-phase below 7.5%, no statistical survival difference between diploid and aneuploid tumors ( p = 0.30) was found.

Multivariate prognostic analysis The relation between survival time, age, sex, tumor grade,

stage, DNA-content and S-phase fraction was analysed using Cox's proportional hazard regression model. Only stage and S-phase fraction were identified as independent prognostic parameters with risk estimation of 7.93 ( p < 0.001) and 3.41 ( p = 0.002), respectively.

TUMOR-CELL PROLIFERATION IN RENAL-CELL CARCINOMA 569

100

80

I- = W "

6 0 P 2 < 2 4 0 > = 10 a

20

0

26

stage 1 - 1 1 1 , S-phase < 7 . 5 %

p' 0.001

r . I

0 15 S O 4 5 60 75 M O N T H S

FIGURE 5 - Kaplan-Meier survival curves of 43 patients without distant metastases, stage 1-111, comparing outcome of 26 patients with S-phase values <7.5% with outcome of 17 patients with S-phase values > 7.5%.

1 5 d i p l o i d , s t a g e 1 - 1 1 1

8 0 .'

60 .'

40 .'

20 "

0 15 30 4 5 60 75 M O N T t l S

FIGURE 6 - Kaplan-Meier survival curves of 37 patients with diploid tumors in stages 1-111, 15 patients with aneuploid tumors in stages 1-111, 4 patients with diploid tumors in stage IV and 26 patients with aneuploid tumors in stage IV.

DISCUSSION

In recent years several studies have focused on flow cytomet- ric assessment of tumor DNA, but only a few have addressed the issue of S-phase calculation in renal-cell carcinoma (Baisch et al., 1982; Chin et a/., 1985; Ljungberg et al., 1985). To our knowledge, only one study has evaluated the prognostic information given by this parameter; however, in that study DNA analysis and S-phase evaluation were performed on archival paraffin-embedded material (Masters et al., 1992).

In the present study, aneuploidy was evident in 72% of the tumors. Previous reports describe the percentage of aneuploid tumors in renal-cell carcinoma to be 50 to 72%, with an average around 63% (Chin et al., 1985; Kloppel et al., 1986; Ljungberg et al., 1986; Grignon et al., 1989; Currin et al., 1990). In accordance with previous reports, aneuploidy correlated to pathological stage and grade (Ljungberg et al., 1985; Currin et al., 1990). Our data also revealed a significant univariate correlation between DNA content and survival, demonstrating a survival advantage for patients with diploid tumors compared with aneuploid tumors, as reported previously (Ljungberg et aZ., 1986, 1991; Grignon et al., 1989; Currin et af., 1990). The low number of patients with diploid tumors and metastases

0 t . I

0 15 30 45 60 75 M O N T H S

FIGURE 7 - Kaplan-Meier survival curves of 50 patients with aneuploid tumors, comparing outcome of 18 patients with S-phase values <7.5% with outcome of 32 patients with S-phase values > 7.5%.

might explain the lack of prognostic significance of DNA content in the present study. In previous studies on DNA content, about one-third of diploid as well as of aneuploid renal-cell carcinomas showed metastatic spread, indicating that a larger amount of patient material might be needed to illustrate this survival difference (Ljungberg et al., 1986, 1988).

The percentage of S-phase fractions in aneuploid-cell popu- lations was significantly higher than in diploid tumors, confirm- ing previous findings (Chin et al., 1985; Ljungberg et al., 1985; Grignon et al., 1989). The S-phase values obtained for diploid tumors were in agreement with data presented by Grignon et al. (1989) presenting a mean S-phase percentage of 2.9% in 52 diploid tumors. There was no difference in S-phase values of diploid cell clones in heterogeneously aneuploid tumors com- pared with homogeneously diploid tumors. In our study the mean S-phase value for aneuploid tumors was 9.1%, lower than the 13.4% published by Frankfurt et al. (1984). We found no difference in S-phase values in the aneuploid cell clones, whether the aneuploid tumors were heterogeneous or homoge- neously aneuploid. We were able to demonstrate a correlation between S-phase and the prognostic indicators, clinical stage and grade. These findings support the theories that cell proliferation is an important parameter of the malignant potential of a tumor.

Patients having tumors with low S-phase values showed a significantly longer survival time than tumors with high values. Our results confirm previous findings in other tumors, includ- ing bladder, prostate, breast and endometrial carcinoma (Tri- bukait et al., 1982; Barlogie et al., 1983; Tribukait, 1987; Arnerlov et al., 1992; Wagenius et al., 1992). In bladder carcinoma, Tribukait (1987) showed decreased survival rates accompanying increased proportions of S-phase cells, with a very high risk of death when tumor S-phase values exceeded 15%. We observed a significant difference when using a cut-off at 7.5%, the mean S-phase value, but also when using S-phase values between 5.5 to 9.5%. S-phase analysis gave no prognos- tic information for tumors with distant metastases. Our data need to be confirmed in larger amounts of patient material.

The correlation between higher S-phase values and aneu- ploid DNA content might be one explanation for the more malignant behavior of aneuploid compared with diploid renal- cell carcinoma. Furthermore, by combining S-phase value and ploidy, we also found that, for patients with aneuploid tumors, there was a difference in survival time depending on the tumor S-phase values and thus the proliferation rate. Furthermore, no statistical difference in survival time was found when

570 LARSSON E T A L .

REFERENCES

comparing diploid tumors with aneuploid tumors having low S-phase values. Thus, S-phase values gave additional prognos- tic information to that obtained by DNA content. In addition, for patients without distant metastases, the S-phase value also differentiated survival. When using Cox’s proportional analy- sis, stage and S-phase fraction were identified as independent prognostic parameters comparable with multivariate results in cervical carcinoma (Strang et af , 1991). These findings are in contrast to the results presented by Masters et al. (1992). In their study, both DNA-ploidy and S-phase fraction had prog- nostic information, but this was lost when tumor grade was taken into account. However, that study was retrospectively performed on 13- to 37-year-old paraffin-embedded archival material, known to give less satisfactory DNA histograms than with fresh tissue. With the acceptance of a CV up to 11%, interpretation of the S-phase value is even more difficult. Despite these drawbacks, the study of Masters et al. had a similar proportion of tumors with unsuccessful evaluations of S-phase values as compared with our study, performed on fresh tissue samples mostly giving adequate evaluable DNA histograms with a low CV.

The frequent heterogeneity of DNA content in the non- diploid renal-cell carcinomas (68%) supports our previous

findings (Ljungberg et al., 1985). In the present study we also observed a frequent intratumoral heterogeneity of the S-phase fractions, both in diploid and in aneuploid tumors, but more pronounced in the latter. So far, no correlation between clinical behavior of the disease and heterogeneity of the tumors has been found.

In conclusion, the proportion of cells in S-phase seems to give prognostic information in renal-cell carcinoma. In this study, S-phase values gave independent prognostic informa- tion while tumor DNA ploidy lost its prognostic value. The study indicates that more detailed cell kinetic analysis might give complementary information regarding prognosis and tu- mor biology. Further studies in this field appear to be war- ranted in renal-cell carcinoma.

ACKNOWLEDGEMENTS

This work was supported by grants from the Swedish Cancer Society, the Swedish Society of Medicine, the Lions Research Foundation and the Medical Faculty, University of Umel, Sweden.

ARNERLOV, C., EMDIN, S.O., LUNDGREN, B., Roos, G., SODERSTROM, J., BJERSING, L., NORBERG, C. and ANGQVIST, K.A., Mammographic growth rate, DNA ploidy, and S-phase fraction analysis in breast carcinoma. A prognostic evaluation in a screened population. Cancer, 70,1935-1942 (1992). BAISCH, H., GOHDE, W. and LINDEN, W., Analysis of PCP-data to determine the fraction of cells in the various phases of cell cycle. Rad. environm. Biophys., 12,31-39 (1975). BAISCH, H., OTTO, U., KONIG, K., KLOPPEL, G., KOLLERMANN, M. and LINDEN, W.A., DNA content of human kidney carcinoma cells in relation to histological grading. Brit. J. Cancer, 45,878-886 (1982). BARLOGIE, B., RABER, M.N., SCHUMANN, J., JOHNSON, T.S., DREWINKO, B., SWARTZENDRUBER, D., GOHDE, W., ANDREEFF, M. and FREIREICH, E.J., Flow cytometry in clinical cancer research. Cancer Rex, 43,

CHIN, J.L., PONTES, E. and FRANKFURT, O., Flow cytometric deoxyribo- nucleic acid analysis of primary and metastatic human renal cell carcinoma. J. UroL, 133,582-585 (1985). CURRIN, S.M., LEE, S.E. and WALTHER, P.J., Flow cytometric assess- ment of deoxyribonucleic acid content in renal adenocarcinoma, does ploidy status enhance prognostic stratification over stage alone? J. Urol., 143,458-463 (1990). FRANKFURT, O.S., GRECO, W.R., SCLOCUM, H.K., ARBUCK, S.G., GAMARRA, M., PAVELIC, Z.P. and RUSTUM, Y.M., Proliferative charac- teristics of primary and metastatic human solid tumors by DNA flow cytometry. Cytonzefry, 5,629-635 (1984). GRIGNON, D.J., AYALA, A.G., EL-NAGGAR, A., WISHNOV, K.I., Ro, J.Y., SWANSON, D.A., MCLEMORE, D., GIACCO, G.G. and GUINEE, U.F., Renal cell carcinoma. A clinico-pathologic and DNA flow cytometric analysis of 103 cases. Cancer, 64,2133-2140 (1989). KLOPPEL, G., KNOFEL, W.T., BAISCH, H. and OTTO, U., Prognosis of renal cell carcinoma related to nuclear grade, DNA content and Robson stage. Europ. UroL, 12,42&431 (1986). LJUNGBERG, B., LARSON, P., STENLING, R. and Roos, G., Flow cytometric deoxyribonucleic acid analysis in stage I renal cell carci- noma. J. Urol., 146,697-699 (1991). LJUNGBERG, B., STENLING, R. and Roos, G., DNA content in renal cell carcinoma with reference to tumor heterogeneity. Cancer, 56,503-508 (1985).

3982-3997 (1983).

LJUNGBERG, B., STENLING, R. and Roos, G., Prognostic value of deoxyribonucleic acid content in metastatic renal cell carcinoma. J. UroL, 136,801-804 (1986). LJUNGBERG, B., STENLING, R. and Roos, G., Tumor spread and DNA content in human renal cell carcinoma. Cancer Res., 48, 3165-3167 (1988). MASTERS, J.R.W., CAMPLEJOHN, R.S., PARKINSON, M.C., WOODHOUSE, C.R.J. and O’REILLY, S.M., Does DNA flow cytometry give useful prognostic information in renal parenchymal adenocarcinoma? Brit. J. UroL, 70,364-369 (1992). MEDEIROS, L.J., GEB, A.B. and WEISS, L.M., Prognostic significance of morphological parameters in 121 cases. Cancer, 61, 1639-1651 (1988). ROBSON, C.J., CHURCHILL, B.M. and ANDERSON, W., The results of radical nephrectomy for renal cell carcinoma. J. Urol., 101, 297-301 (1969). SKINNER, D.G., COLVIN, R.B., VERMILLION, C.D., PFISTER, R.C. and LEADBETTER, W.F., Diagnosis and management of renal cell carci- noma. Cancer, 28,1165-1177 (1971). STRANG, P., STENDAHL, U., BERGSTROM, R., FRANKENDAL, B. and TRIBUKAIT, B., Prognostic flow cytometric information in cervical squamous-cell carcinoma: a multivariate analysis of 307 patients. Gynecol. Oncol., 43,3-8 (1991). STROKEL, S., THOENES, W., JACOBI, G.H. and LIPPOLD, R., Prognostic parameters in renal cell carcinoma-a new approach. Europ. Urol., 16, 416-422 (1989). TRIBUKAIT, B., Flow cytometry in assessing the clinical aggressiveness of genito-urinary neoplasms. Wld J. UroL, 5,108-122 (1987). TRIBUKAIT, B., GUSTAFSSON, H. and ESPOSTI, P.L., The significance of ploidy and proliferation in the clinical and biological evaluation of bladder tumors: a study of 100 untreated cases. Brit. J. Urol., 54, 130-135 (1982). VINDEL~V, L.L., CHRISTENSSON, I . J . and NISSEN, N.I., A detergent trysin method for the reparation of nuclei for flow cytometric DNA analysis. Cytometry, 3, !23-327 (1983). WAGENIUS, G., BERGSTR~M, R., STRANG, P., GERDES, U., ROGO, K., TRIBUKAIT, B. and STENDAHL, U., Prognostic significance of flow cytometric and clinical variables in endometrial adenocarcinoma stages I and KAnticancerRes., 12,725-732 (1992).