Flow cytometric dna content analysis of paraffin block embedded endometrial carcinomas

0360-x)16/91 $3.00 + .OO Copyright 0 1991 Pergamon Press

Inf. 3. Radiation Oncology Bid. Phys. Vol. 21, pp. 1033-1039 F’rintcd in the U.S.A. All rights reserved.

plc

??Brief Communication

FLOW CYTOMETRIC

DNA CONTENT ANALYSIS OF PARAFFIN BLOCK EMBEDDED ENDOMETRIAL CARCINOMAS

ANDREA. KONSKI,M.D. ,* JONATHAN L. MYLES, M.D. ,t THOMASSAWYER, M.S. ,jJAMESNEISLER, M.D. ,$ GARTH~WBS, M.D. ,$ SUZETTELEININGER,B.S. ,t Kmu Km, M.D.? ANDRALPH R. DOBELBOWER JR., M.D., PH.D.* Medical College of Ohio, 3000 Arlington Ave., Toledo, OH 43699 Flow cytometry is being used as an aid in planning the treatment of patients with various malignancies. We report our experience with DNA content analysis on paraffkembedded carcinomas. Hospital, radiation therapy, clinic, and pathology records were reviewed in 139 cases of endometrial carcinoma diagnosed between December 1980 and December 1986. Patients having Stage IV tumors, endometrial sarcomas, dual primary tumors, or incomplete records were eliminated from the analysis, which left 98 evahtabie patients. This report outlines our experience with the first 20 patients. Five of 29 (25%) specimens demonstrated DNA content consistent with aneuploidy, median coeffkient of variance of 5.3%. The median survival time of these five patients is 55 months, with three dying of cancer and one patient dying of other causes but with metastatic disease. The median %S phase was 3.7% in the 15 patients comprising the DNA content diploid population, median coefftcient of variance 5.4%. No patient whose tumor showed S-phase cells below 3.7% died of endometrial cancer. Four of 7 patients developed recurrent cancer with 3 of the 4 patients dying of disease in the high %S phase group. The median patients survival time in the DNA content diploid population was 73 (range: 17-98) months. Patients with 3.7% or below S-phase cells had a median survival time of 75 (range: 40-9@ months whereas the median survival time was 48 (range: 17-g9) months for patients having a %S phase fraction above 3.7%. Although the number of patients studied is small, it appears that DNA content aneupioid tumors are frequently “upstaged” at surgery. These patients may not benefit from preoperative irradiation. Accurate determination of the %S phase fraction in DNA content dipioid tumors may possibly identify patients with a poorer prognosis who may benefit from adjuvant therapy. Flow cytometric DNA content analysis, Endometrial carcinomas.

INTRODUCTION

METHODS AND MATERIALS

Flow cytometry is being used as an aid in cancer management both in diagnosis and aiding in treatment decisions (1, 4, 7, 8, 9, 16, 18, 20, 23, 24, 25, 32). The majority of reports deal with DNA content analysis of fresh tissue, with a minority reporting results from paraffin-embedded tissue. Performing DNA content analysis on paraffin embedded tissue gives investigators the opportunity to study archival pathologic material where patients have a known outcome. Hedley et al. were among the fast groups to report DNA content analysis from archival paraffin-embedded tissue (14). We report results from 20 patients who had histologically proven endometrial carcinoma and have well documented clinical follow-up.

Hospital, radiation therapy, and clinic charts of 139 patients with endometrial carcinoma diagnosed at the Medical College of Ohio between December 1980 and December 1986 were reviewed. Patients having clinical Stage IV disease, endometrial sarcomas, dual primary tumors, or incomplete records were eliminated from the analysis, leaving 98 evaluable patients. This report details the initial 20 patients having flow cytometric DNA content analysis of paraffin-embedded tissue. All of the histologic slides were reviewed by two surgical pathologists (J.M. and K.K.) and graded according to FIG0 criteria (5). They selected blocks that best represented the neoplasm. Deparaffinization and rehydration of the tissue was performed as described by Hedley et al. (12, 14). In an effort to mini-

Presented in part at the 32nd Annual Meeting of the American Society for Therapeutic Radiology and Oncology, Miami, FL, 15-20 October 1990. * Dept. of Radiation Therapy. t Dept. of Pathology.

$ Dept. of Obstetrics and Gynecology. Reprint requests to: Andre A. Konski, M.D., Toledo Radiological Associates, LaVoy Office Building, 4841 Monroe St., Toledo, OH 43623. Accepted for publication 15 April 1991. 1033

I. J. Radiation Oncology 0 Biology 0 Physics

1034

September 1991, Volume 2 1, Number 4

Table 1. DNA content aneuploid patients Patient

Cl. St.

Path. St.

Radiation

Myo.

1 2 3 4 5

III I I I I

IV II IV II I

Pre-op Pre-op Pre-op None None

> > < <

l/3 l/3 l/3 l/3

Grade

Specimen

Recurrence

StattlS

III III III II II

Biopsy Biopsy Operative Operative Operative

Abd. Dist. Abd. and dist. Local None

DOD DOD DWD DOD None

Cl. St. = Clinical Stage; Path St. = Pathologic Stage; Pre-op = Preoperative; Post-op = Postoperative; Myo. = Myomettial invasion; Abd. = Abdominal; Dist. = Distant; DOD = Dead of disease; DWD = Dead with disease but with other causes; NED = No evidence of disease; DsD = Dead without disease; Vag. Obt. = Vaginal obturator.

mize debris, and thus increase the accuracy of S phase measurements, nuclei were isolated by mincing the entire piece of tissue in pepsin rather than extracting from tissue sections. The block was deparaffinized in xylene then rehydrated with graded ethanol, The tissue was then washed in deionized water twice, minced, and digested in 0.5% pepsin for 30 min at 37°C in HCL, pH 1.5. The tissue was then filtered through nylon mesh and washed twice. Nuclei were stained with propidium iodide using the method of Vindelov et al. (30) DNA content analysis was performed on 10,000 nuclei per specimen using an EPICS-C flow cytometer* equipped with a 5W argon-ion laser. The laser wavelength was 488 nm, regulated to a 500 mW constant light output. Histograms were gated on peak versus integral fluorescence for doublet discrimination. The DNA index was defined as the ratio of the mean peak channel of the GO/G1 aneuploid population divided by the mean peak channel of the 2C population. It is very difficult to determine the mean peak channel of the 2C population when using paraffin-embedded tissue. Routine internal standards used for DNA content analysis in fresh tissue cannot be used in paraffin embedded tissue. The DNA index from paraffin embedded tumors is calculated by assuming that the Gl peak with the least DNA content is the diploid peak. Hypodiploid tumors, therefore, cannot be determined in paraffin embedded tumors. A DNA content diploid tumor would have a DNA index of 1.O and hyperdiploid tumors would have a DNA index greater than one. The Coefficient of Variance (CV) is an expression of the width of the GO/G1 peak. This is used as measure of the technical quality of the analysis. Because of software limitations, %S phase could not be calculated for the DNA content aneuploid tumors. The majority of patients with Stage 1A grade 1 tumors had a total abdominal hysterectomy and bilateral salpingo-oophorectomy (TAWBSO), peritoneal washings, and lymph node sampling. If greater than one-third myometrial invasion or tumor grade higher than grade 1 was found in the hysterectomy specimen, postoperative external beam irradiation with or without a vaginal Cesium insertion was delivered. The external beam irradiation was usually given with a four-field technique, but arc rotation * Coulter Electronics, Hialeah, FL.

was used in obese patients. The majority of patients received u5 Gy in 1.8-2.0 Gy fractions with external beam therapy followed by a single vaginal cesium application of 40 Gy to the Vaginal mucosa. The majority of patients having Stage 1A or lB, grade 2 or 3 tumors received whole pelvis irradiation preoperatively (40-45 Gy) followed by an intracavitaty Cesium application usually delivering 20 Gy to point A. Point A was defined in a plane perpendicular to the uterine tandem 2 cm superior to the cervical OSand 2 cm lateral to the midline of the tandem. This was followed by TAH/BSO with lymph node evaluation and peritoneal washings. The importance of histopathologic factors in the hysterectomy specimen is now recognized, and most patients will undergo surgery with adjuvant irradiation based upon the hysterectomy specimen. Survival was calculated from the date of diagnosis using the Kaplan-Meier method. The log-rank and Wilcoxon methods of analysis were used to evaluate differences in survival between the DNA content aneuploid and diploid populations and high %S diploid and low %S diploid populations . RESULTS As the Medical College of Ohio is a regional referral center, paraffin blocks were not available for all patients or were not of sufficient quality to allow DNA content analysis. Of the first 20 patients, 5 were found to be DNA content aneuploid or thought to be DNA content aneuploid. The latter category was created to describe those specimens that were probably DNA content aneuploid but had a large CV. The large CV was secondary to the debris that was formed from the isolation of the nuclei from the paraffin blocks. The median CV for the DNA content aneuploid group was 5.3 (range: 4.9-10.2)%. Table 1 details the stage, grade, myometrial invasion, treatment, and outcome of treatment in the DNA content aneuploid tumors. “Specimen” refers to the specimen, either biopsy or resection, that was used for cytometric analysis. One of the pathologic Stage IV patients was found to have upper abdominal disease before surgery and a TAI-I/

Endometrial carcinoma flow cytometry ??A. A. KONSKI et al.

1035

Table 2. Low %S phase patients Cl. St.

Path. St.

Radiation

Myo.

Grade

Specimen

Recurrence

status

6

I

None

I I

I I I

None

7 8

Pre-op None

< l/3 None

II II I

Operative Biopsy Operative

None None None

NED NED

9 10

I I

I I

None None

< 113

I

Biopsy

None

NED NED

11 12 13

I I I

I II I

Post-op Post-op None

None > l/3 > l/3 < l/3

I I I I

Biopsy Biopsy Operative Biopsy

None None None None

NED NED NED DSD

Patient

Cl. St. = Clinical Stage; Path. St. = Pathologic Stage; Pre-op = Preoperative; Post-op = Postoperative; Myo. = Myometrial invasion; Abd. = Abdominal; Dist. = Distant; DOD = Dead of disease; DWD = Dead with disease but with other causes; NED = No evidence of disease; DsD = Dead without disease; Vag. Obt. = Vaginal obturator.

BSO was never performed. Only one patient with a DNA content aneuploid content neoplasm had positive peritoneal washings. In four patients the cancer recurred (one local only, one abdominal only, one abdominal and distant, and one distant only) at a median time of 11.5 (range: 3-28) months post-resection. The median survival time of the group was 55 (range: 3-87) months. Three patients died of cancer; one patient died of other causes but with metastatic disease and one patient is alive without evidence of cancer. There were 15 DNA content diploid tumors with a median CV of 5.5 (range: 3.7-9.0)%. The median %S phase in the DNA content diploid group was 3.7 (range: 2. l12.6)%. For the purpose of this analysis we compare those patients with tumors at or below the median %S phase (n = 8) with those with tumors above the median %S phase (n = 7). There were eight patients with tumors at or below the median %S phase (low %S group) and seven patients with tumors above the median (high %S group). Table 2 outlines the patient characteristics in the low %S group. One patient received preoperative irradiation (because of a grade II tumor) and two patients received postoperative irradiation (both >1/3 myometrial invasion). No patient had recurrence of neoplasm in this group and all but one patient

(who died of lung cancer) are alive. In the case of the patient having preoperative irradiation, the DNA content analysis was performed on non-irradiated tissue. The median CV for the 198 1 specimens in this group was 5.49 (range: 5.0-6.3)%, whereas the median CV for the 1982 group was 6.1 (4.2-6.5)%. The median CV for the entire group is 5.5%. Table 3 lists the patient characteristics in the high %S group. Three patients received preoperative irradiation; in two of those cases the disease was “upstaged” at surgery, with disease found outside the radiation portals. The nuclei used for the DNA content analysis in those cases were obtained from the irradiated specimen. Four of the seven patients in this group developed recurrent cancer (one local, one local and abdominal, one abdominal, and one distant). Three of those patients died of cancer and one patient was salvaged and is presently alive without evidence of disease. The median CV in this group was 5.2 (range: 3.79.0)%. The median CV was 6.4 (range: 4.7-7.7)% in the 1981 patients and 5.2 (range: 3.7-9.0)% in the 1982 patients . The median patient survival time in the DNA content aneuploid group was 55 (range: 3-87) months. The overall median patient survival time in the DNA content dip-

Table 3. High %S phase patients Patient

Cl. St.

Path. St.

Radiation

Myo.

Grade

Specimen

Recurrence

Status

14 15 16 17 18 19 20

I II I I I I I

I II II I IV I IV

None Pre-op None Post-op* Pre-op None Pre-op

< l/3 > 113 < l/3 > l/3 None < l/3 < l/3

II II II I III I II

Biopsy Operative Operative Biopsy Operative Biopsy Operative

None Local and abd. None Local Abd. and dist. None Distant

NED DOD DSD NED DOD NED DOD

*Patient received vaginal obturator only. Cl. St. = Clinical Stage; Path. St. = Pathologic Stage; Pre-op = Preoperative; Post-op = Postoperative; Myo. = Myometrial invasion; Abd. = Abdominal; Dist. = Distant; DOD = Dead of disease; DWD = Dead with disease but with other causes; NED = No evidence of disease; DsD = Dead without disease; Vag. Obt. = Vaginal obturator.

I. J. Radiation Oncology 0 Biology 0 Physics

60

September 1991, Volume 21, Number 4

L

L-----o I

I 40

..

20

..

L-0 ??

0, 4

0

2

6 Tk o= ?? =

+

10

CyrsP)

Diploid Aneuploid

o+t 0

2

6

+

10

i&e &.I ?=Low ? XS Diploid w= High %S Diploid

Fig. 1. Kaplan-Meier survival plot comparing DNA content aneuploid (n = 5) and DNA content diploid (n = 15) populations. There is a trend toward improved survival in the DNA content diploid population, but because of the small numbers, no statistical difference in survival is noted.

Fig. 2. Kaplan-Meier survival plot comparing low %S DNA content diploid (n = 8) and high %S DNA content diploid (n = 7) patients. No statistical difference in survival was seen, but the low %S group tended to have improved survival compared to the high %S group.

loid group was 73 (range: 17-98) months, but it was only 48 (range: 17-89) months in the high %S group as compared to 75 (range: 40-98) months in the low %S group. Figure 1 depicts Kaplan-Meier survival curves of the DNA content diploid versus the DNA content aneuploid patients. There is no statistical difference between the two curves, log-rank p = 0.09 and Wilcoxon p = 0.1, probably because of the small numbers of patients in each group. Figure 2 outlines Kaplan-Meier survival plots of the high and low %S diploid patients. The low %S diploid patients tend to fare better than the high %S diploid patients, but once again the difference is not statistically different, log-rank p = 0.16 and Wilcoxon p = 0.10.

adenocarcinoma (15). They found that if the progesterone receptor levels were L 100 fmol/mg, the 3-year diseasefree survival was 93% compared to 36% if the receptor level was < 100 fmol/mg. Performing DNA content analysis on paraffin embedded tissues is advantageous because results of a certain treatment regimen or the natural history of a disease may be correlated. Statistical analysis can then be performed with certain prognostic information being identified. Flow cytometric DNA content analysis is being used more frequently to aid in the diagnosis of cancer and to help plan treatment once a diagnosis is made. Some DNA content aneuploid tumors have been shown to be more aggressive than DNA content diploid tumors. The %S phase in the DNA content diploid tumors is also being used to identify further those patients who are at a higher risk for recurrence or developing metastatic disease.

DISCUSSION Clinicians have tried to identify factors that will predict for relapse-free and overall survival in hope of tailoring therapy for individual patients’ needs. Patients with less aggressive cancers would warrant less aggressive treatment in comparison to more aggressive cancers. Histopathologic factors such as histologic grade, depth of myometrial invasion, and stage have been found to provide valuable prognostic information and to aid in determining treatment in patients with endometrial adenocarcinoma (3, 10, 21). Recently, there have been attempts to identify other factors that would predict more aggressive cancers (6, 19, 22, 26). Kacinski et al. reported that pathologic specimens contaming high levels of expression of fms complimentary mRNA oncogene had endometrial cancers with aggressive clinical behavior and poor outcome (17). Ingram et al., among others, have reported on the effect of progesterone receptor status and survival in endometrial

Fig. 3. This DNA histogram shows significant cellular debris which is highlighted between both the bold and narrow arrows respectively. This makes identification of the aneuploid peak difficult.

Endometrial carcinoma flow cytometry 0 A. A. KONSKI et al.

Peak Rod Flourorconce

Fig. 4. This histogram is relatively clear of debris. The area between the bold and narrow arrows respectively shows very little signal in comparison to Figure 1.

DNA content analysis in paraffin embedded tissue, however, is technically more difficult and the results are more difficult to analyze when compared to DNA content analysis performed on fresh tissue (12, 13, 14, 28, 29). The wider the CV, the poorer the technical quality of the specimen and the less reliable the results. Rosenberg et al. reported a mean CV of 5.7% in their study (27). Britton ef al., on the other hand, reported a mean CV of 7.25% for their GO/G1 peak (2). They also classified tumors with GO/G1 CV’s less than 14% as DNA content diploid or normal, and specimens resembling non-tumor controls with CV’s of the GO/G1 peak larger than 14% and failing to fulfill criteria to be classified otherwise were designated as DNA content “diploid by default.” Figure 3 is a DNA histogram of a 198 1 specimen showing significant debris whereas Figure 4 is a DNA histogram of a 1982 specimen that is relatively free of cellular debris. We used the entire block to prepare the nuclei to reduce the proportion of sectioned nuclei and reduce cellular debris. Tissue processed in different fixatives can also give variations in the CV (12). Specimens from 1981 tended to have larger CV’s than specimens from 1982. This may be related to changes in the paraffin blocks occurring during storage. We compared our method of preparing nuclei from archival paraffin blocks to results obtained from fresh tissue and recent paraffin-embedded tumors. We could not show any difference in DNA content analysis between the fresh tumor and paraffin-embedded tumor with our method. DNA content analysis has been used to help identify poor prognostic features of tumors in a number of different anatomical sites. Moberger et al. reported that patients with near diploid or tetraploid endometrial tumors had a significantly lower death rate than aneuploid tumors (25). They, however, did not report the results of DNA content diploid neoplasms which had high %S phase. Rosenberg et al. found a correlation between FIG0 grade and DNA content aneuploid tumors when they analyzed paraffin-embedded endometrial carcinomas: the higher the grade, the greater the percentage of DNA content aneuploid cancers (27). They also found that the higher the %S phase, the

1037

shorter the survival time, which is in agreement with our results. Britton et al. reported that although DNA nondiploid paraffin-embedded endometrial carcinomas only accounted for 16% of their Stage I patients, these patients accounted for 50% of the relapses (2). DNA content analysis was also predictive of survival. They, however, did not further divide the diploid population into high or low %S phase. Geisinger et al., using fresh tissue, found that ploidy did not carry any prognostic significance in their series (11). This probably resulted from the fact that they used a supradiploid value as a threshold. Friedlander et al. reported on DNA content analysis in Stage III and IV ovarian tumors (8). Multivariate analysis showed that cellular DNA content and FIG0 stage were the only significant independent prognostic variables. The median survival was only 13 months for DNA content aneuploid tumors and 60 months for DNA content diploid tumors. DNA content aneuploid tumors in our series usually contained other poor prognostic features such as > l/3 myometrial invasion (2/4) and grade 2 (2/5) or grade 3 (3/5) differentiation. DNA content analysis was helpful in dividing the DNA content diploid tumors into “high risk” and “low risk” groups. We arbitrarily picked the median %S phase value to divide the patients for analysis. Clarke et al. used the median %S phase to distinguish high and low %S phase tumors (4). Using the median value, they found that survival time was statistically different in the high and low %S phase tumors. They then looked for a %S phase value that yielded the best statistical difference. Rosenberg et al. analyzed 120 paraffin-embedded tumors and found a mean %S phase of 5% (27). They then divided the analysis into %S phase < 5%, 5-lo%, and > 10%. They did not explain why the 10% value was chosen, but they found a lo-fold increase in cancer mortality with a > 10% %S phase. In our 20 cases, no patient in the low %S phase group died of endometrial cancer whereas 3 of 7 patients in the high %S phase group died of endometrial cancer. The majority of patients in the high %S phase group, however, contained other poor prognostic features. Five of 7 patients in the high %S group had grade 2 tumors and 3 of 7 patients had “upstaging” at surgery with 2 of 3 having disease outside of the radiation portals. Abnormalities in the DNA content analysis paralled the well-known poor prognostic histopathologic criteria. The majority of patients (3/5) with DNA content aneuploid tumors received preoperative irradiation because of high histopathologic grade. Only 1 of 8 patients in the low %S group received preoperative irradiation and none of these patients died of endometrial cancer, which confirms the selection criteria for adjuvant irradiation. None of the 8 patients in the low %S phase group, 3 of 7 patients in the high %S phase group, and 1 of 5 patients in the DNA content aneuploid group had the DNA content analysis performed on irradiated tissue. All of the patients

1038

I. J. Radiation Oncology ??Biology 0 Physics

in the high 9% phase group having the analysis performed on irradiated tissue died of disease. Theoretically, irradiated tissue may have a higher rate of cell division and may have a higher S-phase fraction (31). We feel that the DNA content analysis in patients receiving preoperative irradiation was accurate reflection of the biology of the tumor itself rather than of the effect of the irradiation on the tumor, since they had other poor prognostic histopathologic features and they all died of endometrial cancer.

September 1991, Volume 21, Number 4

Although the number of patients is small, with this analysis we feel that DNA content analysis may be used to help identify patients with a probability of disease outside typical preoperative radiation portals. This will require study of a larger number of cases. DNA content analysis also may help to identify patients with disease outside typical preoperative irradiation portals and thus would not benefit from preoperative irradiation. This will require study of a larger number of cases.

REFERENCES 1. Baisch, H.; Otto, V.; Konig, K. DNA content of human kidney carcinoma cells in relation to histologic grading. Br. J. Cancer 45878-886; 1982. 2. B&ton, L. C.; Wilson, T. 0.; Gaffey, T. A.; Lieber, M. M.; Wieand, H. S.; Podartz, K. C. Flow cytometric DNA analysis of Stage 1 endometrial carcinoma. Gynecol. Oncol. 34:317-322; 1989. 3. Christopherson, W. M.; Connelly, P. J.; Alberhasky, R. C. Carcinoma of the endometrium. An analysis of prognosticators in patients with favorable subtypes and Stage 1 disease. Cancer 51:1705-1709; 1983. 4. Clark, G. M.; Dressler, L. G.; Owens, M. A.; Pounds, G.; Oldaher, T.; McGuire, W. L. Prediction of relapse or survival in patients with node-negative breast cancer by DNA flow cytometry. N. Eng. J. Med. 320:627633; 1989. 5. Classification and staging of malignant tumors in the female pelvis (accepted by the General Assembly of FIG0 in New York, 12 April, 1970). Acta Obstet. Gynecol. &and. 50:1-7; 1971. 6. Ehrlich, C. E.; Young, P. C.; Stehman, F. B.; Sutton, G. P.; Alford, W. M. Steroid receptors and clinical outcome in patients with adenocarcinoma of the endometrium. Am. J. Obstet. Gynecol. 158:79&807; 1988. 7. Fenan, P. J.; Quirke, P.; Dixon, M. F. Is DNA aneuploidy a good prognostic indicator in patients with advanced colorectal cancer? Br. J. Cancer 54:327-330; 1986. 8. Friedlander, M. L.; Hedley, D. W.; Swanson, C.; Russell, P. Prediction of long-term survival by flow cytometric analysis of cellular DNA content in patients with advanced ovarian cancer. J. Clin. Oncol. 6:282-290; 1988. 9. Freidlander, M. L.; Hedley, D. W.; Taylor, I. W. Clinical and biological significance of aneuploidy in human tumors. J. Clin. Pathol. 37:961-974; 1984. 10. Frich, H. C.; Munnell, E. W.; Richart, R. M.; Berger, A. P.; Lawry, A. B. Carcinoma of the endometrium. Am. J. Obstet. Gynecol. 115:663-676; 1973. 11. Geisinger, K. R.; Homesley, H. D.; Morgan, T. M.; Kute, T. E.; Marshall, R. B. Endomettial adenocarcinoma: a multiparameter clinicopathologic analysis including the DNA profile and the sex steroid hormone receptors. Cancer 58: 1518-1525; 1986. 12. Hedley, D. W. Flow cytometry using paraffin-embedded tissue: five years on. Cytometry 10:229-241; 1989. 13. Hedley, D. W.; Freidlander, M. L.; Taylor, I. W. Application of DNA flow cytometry to paraffin-embedded archival material for the study of aneuploidy and its clinical significance. Cytometry 6:327-333; 1985. 14. Hedley, D. W.; Freidlander, M. L.; Taylor, I. W.; Rugg, C. A.; Musgrove, E. A. Method for analysis of cellular DNA content of paraffin-embedded pathological material using flow cytometry. J. Histochem. Cytochem. 31:1333-1335; 1983. 15. Ingram, S. S.; Rosenman, J.; Heath, R.; Morgan, T. M.;

16. 17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

Moore, P.; Varia, M. The predictive value of progesterone receptor levels in endometrial cancer. Int. J. Radiat. Oncol. Biol. Phys. 17:21-27; 1989. Jacobson, A. Prognostic impact of ploidy level in carcinoma of the cervix. Am. J. Clin. Oncol. 7:475-480; 1984. Kacinski, B. M.; Carter, D.; Mittal, K.; Kahom, E. I.; Bloodgood, R. S.; Donahue, J.; Donofrio, L.; Edwards, R.; Schwartz, P. E.; Chambers, J. T.; Chambers, S. K. High level expression of FMS proto oncogene mRNA is observed in clinically aggressive human endometrial adenocarcinomas. Int. J. Radiat. Oncol. Biol. Phys. 15:823-824; 1988. Kreichergs, A.; Zetterberg, A.; Soderberg, G. The prognostic significance of nuclear DNA content in chondrosarcoma. Anal. Quant. Cytol. 2~272-278; 1980. Liao, B. S.; Twiggs, C. B.; Leung, B. S.; Yu, W. C. Y.; Potish, R. A.; Prems, K. A. Cytoplasmic estrogen and progesterone receptors as prognostic parameters in primary endometrial carcinoma. Obstet. Gynecol. 67:463467; 1986. Look, A. T.; Hayes, M. D.; Nitschke, R. Cellular DNA content as a predictor of response to chemotherapy in infants with unresectable neuroblastoma. N. Eng. J. Med. 311:231235; 1984. Lotochi, R. J.; Copeland, C. J.; DePetrillo, A. D.; Muirhead, W. Stage I endometrial adenocarcinoma treatment results in 835 patients. Am. J. Obstet. Gynecol. 146:141-145; 1983. Martin, J. D.; Hahnel, R.; McCartney, A. J.; Woodings, T. C. The effect of estrogen receptor status on survival in patients with endometrial cancer. Am. J. Obstet. Gynecol. 147: 322-324; 1983. McGuire, W. L.; Dressler, L. G. Emerging impact of flow cytometry in predicting recurrence and survival in breast cancer patients. JNCI 75405-410; 1985. Merkel, D. E.; Dressler, L. G.; McGuire, W. L. Flow cytometry, cellular DNA content and prognosis in human malignancy. J. Clin. Oncol. 5:1690-1703; 1987. Moberger, B.; Auer, G.; Forsslund, G.; Moberger, G. The prognostic significance of DNA measurements in endometrial carcinoma. Cytometry 5:430-436; 1984. Palmer, D. C.; Muir, IM.; Alexander, A. I.; Cauchi, M.; Bennett, R. C.; Quinn, M. A. The prognostic importance of steroid receptors in endometrial carcinoma. Obstet. Gynecol. 72:388-393; 1988. Rosenberg, P.; Wingren, S.; Simonsen, F.; Stat, 0.; Risberg, B.; Nordenskzold, B. Flow cytometric measurements of DNA index and S-phase on paraffin-embedded early stage endometrial cancer: an important prognostic indicator. Gynecol. Oncol. 35:50-54; 1989. Schutte, B.; Reynders, M.; Bosman, F. T.; Blijhan, G. H. Flow cytometric determinations of DNA ploidy level in nuclei isolated from paraffin embedded tissue. Cytometry 626 30; 1985. Stephenson, R. A.; Gay, H.; Fair, W. R.; Melamed, M. R.

Endometrial carcinoma flow cytometry 0 A. A. KONSKI et al. Effect of section thickness on quality of flow cytometric DNA content determination in paraffin-embedded tissue. Cytometry 7:414I; 1986. 30. Vindelov, L. L.; Christensen, I. J.; Nissen, W. I. A detergent trypsin method for the preparation of nuclei for flow cytometric DNA analysis. Cytometry 3:323-327; 1983.

1039

31. Withers, H. R. Biological Basis of Radiation Therapy. In: Perez, C. A.; Brady, L. W., eds. Principles and practice of radiation oncology. Philadelphia, PA: Lippincott Co.; 1987: 67-98. 32. Zetterberg, A.; Esposti, P. L. Prognostic significance of nuclear DNA levels in prostatic carcinoma. Stand. J. Urol. Nephrol. 5553-58; 1980.