Significance of Epidermal Growth Factor Receptor in Advanced Ovarian Cancer By Giovanni Scambia, Pierluigi Benedetti Panici, Francesco Battaglia, Gabriella Ferrandina, Gabriela Baiocchi, Stefano Greggi, Rosa De Vincenzo, and Salvatore Mancuso Purpose: The purpose of this study was to investigate the significance of epidermal growth factor receptor (EGF-R) expression in a group of advanced ovarian carcinomas. Patients and Methods: The study was conducted on 72 previously untreated patients with International Federation of Gynecology and Obstetrics (FIGO) stage Ill-IV disease. The median follow-up was 24 months (range, 4 to 75 months). EGF-Rwas measured by a radioreceptorial assay. A cutoff of 1.5 fmol per milligram of protein was chosen to define EGF-R positivity. Medians and life tables obtained with the Kaplan and Meier method were analyzed by the log-rank test. The risk of progression was estimated by Cox's proportional hazards model. Results: EGF-R was detected in 54%of primary tumors. When EGF-R was analyzed in different tissue specimens of the same tumor, consistent findings were noted in 88% (seven of eight) of cases. A lower concordance rate (nine
L
ONG-TERM SURVIVAL of patients with advanced ovarian carcinoma remains poor.' Progno-
sis of these patients appears to be influenced by clinical
and histopathologic factors that are frequently interrelated.2' 3 Thus, the identification of new biologic factors
related to an individual tumor's aggressiveness would be extremely useful to allow individualized and possibly more effective treatments. Epidermal growth factor (EGF) is a single-chain
polypeptide with growth-promoting effects in several normal and neoplastic cell lines.4 The EGF-stimulating
action is mediated by a specific transmembrane receptor (EGF-R) that shows close similarity to the product of
the v-erb-B oncogene.' Consistent data have suggested that EGF-R expression might represent a parameter of poor prognosis in breast,6' 7 bladder,' esophageal,' and cervical"' cancer. The presence of EGF-R in ovarian
cancer specimens-''4 and the growth stimulatory effect 6 of EGF on several ovarian cancer cell lines'5 ," have
recently been reported. However, although a relation-
of 15; 60%) was found between primary tumors and omental metastases, with a tendency toward higher EGF-R levels in the latter. The EGF-R expression did not significantly correlate with age, stage, grading, and residual tumor after primary surgery. In the univariate analysis, stage IV disease, postoperative residual tumor diameter greater than 2 cm, presence of ascites, and EGF-R positivity were found to be significantly associated with a greater risk of disease progression. In the multivariate analysis, only the postoperative residual tumor and the EGF-R expression remained significantly associated with a high risk of progression. Conclusion: Data reported here suggest that the presence of EGF-R in advanced ovarian tumor at the time of the primary surgery identifies a subset of patients with a particularly poor prognosis. J Clin Oncol 10:529-535. © 1992 by American Society of Clinical Oncology. PATIENTS AND METHODS Between 1986 and 1990, 76 previously untreated patients with histologically confirmed stage III-IV (International Federation of Gynecology and Obstetrics [FIGO] classification17 ) ovarian carcinoma were admitted to the study. The World Health Organization (WHO) histologic typing of ovarian tumors"8 was adopted. Four patients were excluded from the analysis because of inadequate follow-up data. The clinicopathologic features of the remaining cases are listed in Table 1. Chemotherapy was started 2 to 3 weeks after surgery. All patients received cisplatin-containing regimens. Thirty-seven patients received high-dose cisplatin (200 mg/im2) for three courses." Gynecologic examination, abdominopelvic ultrasonography, CA125 assay, and radiologic investigations, if necessary, were performed monthly for the clinical assessment of response, which was recorded according to WHO criteria. 20 About 28 days after the last course, clinical complete responders underwent second-look laparoscopy. In laparoscopy-negative cases, second-look laparotomy was performed for the assessment of pathologic response. During laparotomy and after peritoneal washings and careful inspection of the abdominal cavity, biopsy of all suspicious lesions was performed, and, in case of no evidence of disease, at least 20 random biopsies were taken. Patients who initially had only an explorative laparotomy underwent a second laparotomy after
ship between EGF-R expression and poor outcome in
patients with ovarian cancer has been reported,14 the prognostic value of EGF-R in a multivariate analysis has
been never investigated. We report here the results of a large, single-institution, prospective study conducted at the Catholic University in Rome, Italy, which aims to assess the clinical significance of EGF-R in advanced ovarian carcinoma.
From the Department of Gynecology and Obstetrics, Catholic University, Rome, Italy. Submitted June 3, 1991; accepted November 19, 1991. Address reprint requests to Professor Salvatore Mancuso, Department of Gynecology, Catholic University, Largo A. Gemelli 8, 00168 Rome, Italy. © 1992 by American Society of Clinical Oncology. 0732-183X/9211004-0005$3.00/0
Journalof Clinical Oncology, Vol 10, No 4 (April), 1992: pp 529-535
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 20, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
529
530
SCAMBIA ET AL Table 1. Patient Characteristics
stopped by the addition of 3 mL of 25 mmol/L TRIS, 20% glycerol, lcrl RS 0 f25mo/ f3m stopped" byteadto 5
Patients
Entered Assessable Age (years) < 50 > 50 FIGO stage III IV Grade of differentiation Well-differentiated (G1) Moderately-poorly differentiated (G2-G3) Histology Serous Mucinous Endometrioid Undifferentiated Others Ascites No Yes Surgical debulking Optimally debulked (RT < 2 cm) Suboptimally debulked (RT > 2 cm) Clinical response to chemotherapy Complete response Partial response-no change
%
mmol/L
obtained
^'"'^" coUUedU
76
NaN3,
by
and
0.1%
bovine
at
centrifugation
n1 a gamlma LcounterI
serum
for
2,000g
I mlllru. 1
albumin.
minutes
20
Pellets
at
were
00C
and
KesulLS were expressed
19 53
26 74
57
79
13• 59
18
as femtomoles per milligram of membrane protein. Protein concen23 tration was measured using the method reported by Bradford. Tissue weight permitting, Scatchard analysis was performed with concentrations of 125-I-EGF ranging from 0.28 to 3.4 nmol/L, either alone or in the presence of unlabeled EGF (810 limol/L). As 7 used for breast cancer, 1.5 fmol per milligram of protein was considered the cutoff for EGF-R-positive tumors to distinguish ovarian tumors with high (> 1.5 fmol/mg protein) and low (< 1.5 fmol/mg protein) EGF-R content.
82
StatisticalAnalysis
50 4 9 6 3
69 5 13 8 4
19 53
26 74
43
60
29
40
The Wilcoxon rank-sum test was used to compare receptor 2 levels. x analysis was used to analyze the relationship between EGF-R expression and tumor characteristics or response to chemotherapy. All medians and life tables were computed using the 24 product-limit estimate by Kaplan and Meier, and the curves were examined by means of the log-rank test.2 The risk of progression 26 was estimated by Cox's proportional hazards model. Multivariate 27 analysis was performed with BMDP statistical software. A backward stepwise procedure was used to identify the major prognostic factors. Progression-free survival (PFS) was calculated from the date of the first surgery to the date of clinical or pathologic progression or death. The median follow-up was 24 months (range, 4 to 75 months; as of January 1991).
48 24
67 33
Abbreviation: RT, postoperative residual tumor.
chemotherapy, and a second cytoreduction was attempted. Pathologic complete responders received no further therapy, and all other patients were treated according to ongoing phase II studies.2.22
RESULTS
Expression of EGF-R Figure 1 shows a representative example of a saturation analysis plot of the specific binding of 125-I-EGF in ovarian cancer specimens. A linear regression fit (r = .98)
Iodine-125-EGFBinding Measurement Tumor specimens, collected during primary surgery, were frozen 0 on dry ice shortly after surgical removal and stored at -80 C until processed. A representative section of specimens was retained for histopathologic examination. The membrane fraction and cytosol were prepared as described elsewhere.7 Briefly, tumor specimens were finely minced and homogenized in 5-volume ice-cold buffer consisting of 25 mmol/L trometamol (TRIS), 1.5 mmol/L edathamil [EDTA], 5 mmol/L NaN3, 0.1% monothioglycerol, and 20% glycerol (TENMG), by applying several intermittent bursts of an Ultra-Turrax homogenizer (Janke & Kunkel, IKA, Labortechnik, Staufen, Germany). The crude homogenate was centrifuged at 7,000g for 20 minutes at 00 C. The supernatants were then centrifuged at 105,000g for 75 minutes at O0C. The membrane pellet was resuspended in 25 mmol/L TRIS, 1.5 mmol/L EDTA, 5 mmol/L NaN3, 20% glycerol, and 10 mmol/L magnesium chloride (MgCI2) (TENG + MgCI2). Aliquots of the suspension (100 IL containing 300 to 500 pLgprotein) were incubated with iodine-125 (125-I)-EGF (2.6 nmol/L) in the presence or absence of unlabeled EGF (1 mmol/L) for 16 hours at room temperature in a final volume of 400 IL. Binding was
z
a 9
2..
o
u;
o o e: m
o_
r
OUND98(pM)
0 SOUND(pM)
"5I-EGF
concentrations (nM)
Fig 1. Scatchard analysis of 125-1-EGF binding to ovarian tumor membranes. Specifically bound 125-1-EGF was measured as detailed in Patients and Methods.
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 20, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
531
EGF-R IN ADVANCED OVARIAN CANCER
indicated one specific high-affinity binding site (dissociation constant [KI] = 0.65 nmol/L) with an x-intercept on the Scatchard plot of 14.78 fmol/mg protein. Among the ovarian cancer specimens tested, K, values ranged from 0.58 to 1.6 nmol/L. The specificity of EGF binding in ovarian cancer was evaluated by testing the ability of various peptidic hormones at the same concentration of unlabeled EGF to prevent binding of 125-I-EGF. Luteinizing hormone-releasing hormone (LH-RH), thyrotropinreleasing hormone (TRH), insulin, follicle-stimulating hormone (FSH), growth hormone (GH), and endothelial cell growth factor all failed to compete effectively for 125-I-EGF (< 10% displacement of 125-I-EGF binding), whereas unlabeled EGF displaced 66% of the bound 125-I-EGF (data not shown). We also investigated the heterogeneity of EGF binding in ovarian cancer (Table 2). In eight cases EGF-R concentrations were simultaneously measured in different specimens of the same tumor. Although differences in the concentrations of EGF-R within each tumor were found, in only one case we observed high (> 1.5 fmol/mg protein) and low values coexisting in the same tumor. This analysis revealed that EGF-R is a stable feature within primary ovarian cancer. On the other hand, a lower concordance rate (nine of 15; 60%) was found between primary tumors and omenTable 2. EGF-R Content in Primary Epithelial Ovarian Carcinomas and in Specimens From Metastases Obtained Concomitantly Patient
No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
tal metastases, with the latter exhibiting a tendency toward a higher percentage of specimens with high EGF-R levels. Overall EGF-R content was higher in metastatic deposits (median, 2.68; range, 0 to 10.8 fmol/mg protein) than in primary tumors (median, 1.40; range, 0 to 10.30 fmol/mg protein), although this difference did not reach a statistical significance. Of the 72 primary tumors examined, 39 (54%) exhibited high values (> 1.5 fmol/mg protein) of EGF-R. Table 3 shows the distribution of EGF-R according to the clinical and histopathologic characteristics of the on-study population. EGF-R expression was not significantly associated with age, stage, histopathologic grading, postoperative residual tumor, and response to chemotherapy. EGF-R Expression and Risk of Progression During follow-up, disease progression was observed in 38 patients. High values of EGF-R were present in 28 of the 38 progressive cases. Figure 2 shows the PFS curves in relation to EGF-R status. A highly significant association between patients with low EGF-R expression and a longer PFS compared with patients with high EGF-R expression (P = .0004) was Table 3. EGF-R Content in Ovarian Cancer by Clinicopathologic Characteristics
Ovarian Primary Tumor 1.6,1.8* 5.8,9.8* 2.5,0.7* 10.9, 6.3* 0, 0* 15.9,12.0, 12.8* 1.6, 3.6* 0, 1.3* 0 0 0 2.2 3.1 10.3 1.3 6.6 0 5.3 4.3 1.6 1.0
*Different specimens of the primary tumor.
EGF-R Content 1.5 fmol/mg Protein
No. of
EGF-R Content (fmol/mg protein) Metastases -
1.9 0.8 8.7 0 9.3 6.6 10.8 3.6 2.7 2.7 0 2.5 0.1 0 9.2
Variable Age (years) 2 cm Clinical response to chemotherapy Complete response Partial response-no change
Cases
No. of Cases
%
19 53
8 31
42 58
57 15
29 10
51 67
13 59
8 31
61 52
50 4 9 6 3
29 2 5 2 1
58 50 55 33 33
43 29
21 18
49 62
48 24
22 17
46 71
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 20, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
SCAMBIA ET AL
532 -----
LL,
Table 5. Multivariate Analysis of Prognostic Variables for PFS in Patients With Advanced Ovarian Carcinoma
I L I__
Prognostic Variable z o
FIGO stage III IV Surgical debulking RT < 2 cm RT > 2 cm Ascites No Yes EGF-R status (fmol/mg protein) < 1.5 > 1.5
o.
z
MONTHS
Fig 2. PFS rate by EGF-R content. EGF-R+ (> 1.5 fmol/mg protein, ---- ; 68%; 95% CI, 42% to 94%): 39 patients entered, 28 progressed in a median of 15 months. EGF-R- (< 1.5 fmol/mg protein, -; 23%; 95% CI, 8% to 38%): 33 patients entered, 10 progressed in a median of NR. P = .0004 by log-rank test.
observed. The 36-month PFS was 68% (95% confidence interval [CI], 42% to 94%) for patients with EGF-R less than 1.5 fmol/mg protein, compared with 23% (95% CI, 8% to 38%) for those with EGF-R > 1.5 fmol/mg protein. The relative risk of disease progression was estimated in a univariate analysis and then in a multivariate analysis using a backward stepwise procedure. In the univariate analysis (Table 4), stage IV disease, postoperative residual tumor diameter greater than 2 cm, presence of ascites, and high EGF-R content were found to be significantly associated with a greater risk of disease Table 4. Univariate Analysis of Prognostic Variables for PFS in Patients With Advanced Ovarian Carcinoma
Age (years) 50 FIGO stage III IV Histopathologic grading G1 G2-G3 Surgical debulking RT < 2 cm RT > 2 cm Ascites No Yes EGF-R status (fmol/mg protein) < 1.5 > 1.5
No. of Cases
Median PFS (months)
19 53
18 16
1* 1.1
57 15
20 11
1 " 2.6
.006
13 59
16 19
1* 0.8
.41
43 29
47 13
1* 2.7
.002
19 53
49 15
1" 3.2
.002
33 39
50 15
1* 3.4
RR1
P
No. of Cases
Median PFS (months)
57 15
20 11
-
43 29
47 13
1 2.8
19 53
49 15
-
33 39
50 15
1* 3.8
RR2
P .004 .0005
Abbreviation: RR2, adjusted relative risk of progression taking into account all the factors of the table. *Reference category.
progression. In the multivariate analysis (Table 5), only postoperative residual tumor and EGF-R expression remained significantly associated with a high risk of progression. Figure 3 shows the PFS curves according to EGF-R
status and clinical response to chemotherapy. Among patients who achieved a clinical complete response (CR) to chemotherapy at 3 years, those with low values of EGF-R had a 75% PFS with a relative risk of progression of 0.3, compared with 33% and a relative risk of 1.4 for patients with high values of EGF-R. The worst outcome was observed in the subset of patients with high EGF-R expression and less than CR to chemotherapy; in fact, all of these patients progressed within 24 months.
.92
z 0 z
MONTHS
Abbreviation: RR1, unadjusted relative risk of progression. *Reference category.
.0005
Fig 3. PFS rate and relative risk (RR) of progression by EGF-R content and clinical CR to chemotherapy. EGF-R- and 75% with CR (a), RR = 0.3; EGF-R- and 43% with no CR (b), RR = 1.2; EGF-R+ and 33% with CR (c), RR = 1.4; and EGF-R+ and no CRs (d), RR = 5.3.
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 20, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
533
EGF-R IN ADVANCED OVARIAN CANCER DISCUSSION
Our data indicate that a significant proportion of ovarian tumors contain specific EGF-binding proteins. In our series the percentage of tumor with high EGF-R expression was higher than that previously reported by Bauknecht et al."3 On the other hand, Berchuck et al' 4 using an immunohistochemical method reported 82% of ovarian tumors expressing EGF-R. Methodologic differences and/or differences in the cutoff values may explain these discrepancies. Interestingly, when EGF-R was analyzed in multiple sites of the same tumor, consistent findings were noted in 88% of cases. Therefore, it appears that in most cases EGF-R status can be adequately assessed using a single tumor sample. However, omental metastases express high EGF-R levels more frequently than their respective primary tumors, although the difference was not statistically significant. Nevertheless, it can be hypothesized that, as in breast cancer, 7,28 metastatic clones express EGF-R more frequently, thereby suggesting that EGF-R expression can be related to the metastases potential. In our experience, the expression of EGF-R was not related to any clinical and/or pathologic tumor characteristic. This is consistent with the data reported by Baucknecht et al"3 and Berchuck et al.' 4 Interestingly, in spite of the lack of correlation with clinicopathologic features, EGF-R status demonstrated a strong prognostic value in patients with advanced ovarian cancer. In fact, after multivariate analysis the association of high EGF-R levels with worse PFS was independent from the other known prognostic factors such as residual disease and stage. Although it is difficult to reconcile this finding with the fact that EGF-R and its ligands play a role in the growth and differentiation of normal ovarian surface epithelium,'4 some considerations should be taken into account. In normal cells, EGF-R levels are regulated by several factors, such as steroid hormones and other autocrine peptide factors, and are differently expressed according to different phases of cellular metabolism.29"'3 In cancer cells, EGF-R overexpression may result from gene amplification or, more frequently, from the loss of normal regulatory mechanisms of transcription.32,33 Therefore, cancer cells may be continuously exposed to the stimulatory effect of EGF/transforming growth factor-alpha. The presence of higher amounts of EGF-R may also result in an enhanced proliferation for a given amount of autocrinereleased growth factors. This may explain the association of enhanced EGF-R expression with more aggres-
sive tumor behavior, which has been reported for various epithelial cancers. 710 The current prognostic characterization of patients with advanced ovarian cancer is still inadequate. In fact, about 50% of patients after an optimal surgical debulking and a pathologic CR to primary chemotherapy will recur and die of disease.34 Therefore, the identification of new variables correlating with the malignant potential of the cancer cells would be clinically valuable in the selection of therapy for individual patients. Data from the present study suggest that the assessment of EGF-R status at the time of initial surgery may allow the identification of a subset of patients with a particularly poor prognosis. These patients could be candidates for more aggressive and/or experimental primary treatments than those conventionally used. However, in our series, 25% of patients with EGF-Rnegative tumors and CR to chemotherapy progressed within 3 years. Hopefully similar patients could be identified through other prognostic parameters. Among the new prognostic factors proposed for ovarian cancer, the amplification/overexpression of erbB2/neu seems promising. 35' 36 The product of the neu proto-oncogene is a 185-kd glycoprotein defined as p185 that, like the EGF-R, is a transmembrane protein with tyrosine-kinase activity.3 7 These properties and the close
structural homology of p18537 and EGF-R suggest that p185 is itself a growth factor receptor. The finding that EGF stimulates the phosphorylation of p18538 suggests that both EGF-R and erb-B2/neu may be involved in the same pathway leading to cell proliferation and tumor progression. Therefore, it is conceivable that the simultaneous evaluation of EGF-R and erb-B2/neu may increase the prognostic characterization of ovarian cancer patients. In this context, it is interesting that Wright et a19 reported an apparent additive effect of p185 and EGF-R expression on the prognosis of patients with breast cancer. Data presented here, together with the recently reported in vitro data',"16 indicating that ovarian cancer cell proliferation is modulated by the EGF/EGF-R system, further confirm that this system may play a role in the control of ovarian cancer onset and spread. Therefore, drugs and/or biologic response modifiers able to interfere with the EGF-mediated control of cell proliferation may be useful in ovarian cancer therapy. In this context, it is notable that monoclonal antibodies, anti-EGF-R,4 and synthetic, EGF-like peptides41 have been demonstrated to inhibit the growth of cancer cells.
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 20, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
534
SCAMBIA ET AL REFERENCES
1. Ozols RF, Young RC: Chemotherapy of ovarian cancer. Semin Oncol 11:251-263, 1984 2. Dembo AJ, Bush RS: Choice of postoperative therapy based on prognostic factors. Int J Radiat Oncol Biol Phys 8:893-897, 1982 3. Malkasian GD, Melton LJ, O'Brien PC, et al: Prognostic significance of histologic classification and grading of epithelial malignancies of the ovary. Am J Ostet Gynecol 149:274-282, 1984 4. Carpenter G: Receptors for epidermal growth factor and other polypeptide mitogens. Ann Rev Biochem 56:881-884, 1987 5. Downward J, Yarden Y, Mayes E, et al: Close similarity of epidermal growth factor receptor and v-erb-B oncogene protein sequences. Nature 307:521-527, 1984 6. Sainsbury JRC, Farndon JR, Needham JK, et al: Epidermal growth factor receptor status as predictor of early recurrence of and death from breast cancer. Lancet 1:1398-1401, 1987 7. Battaglia F, Scambia G, Rossi S, et al: Epidermal growth factor receptor in human breast cancer: Correlation with steroid hormone receptors and axillary lymph-node involvement. Eur J Cancer Clin Oncol 24:1685-1690, 1988 8. Neal DE, Marsh C, Bennet MK, et al: Epidermal growth factor receptors in human bladder cancer: Comparison of invasive and superficial tumors. Lancet 1:366-368, 1985 9. Ozawa S, Ueda M, Ando N, et al: Prognostic significance of epidermal growth factor receptor in esophageal squamous cell carcinomas. Cancer 63:2169-2173, 1989 10. Pfeiffer D, Stellwag B, Pfeiffer A, et al: Clinical implications of epidermal growth factor receptor in the squamous cell carcinoma of the uterine cervix. Gynecol Oncol 3:146-150, 1989 11. Battaglia F, Scambia G, Benedetti Panici P, et al: Epidermal growth factor receptors in gynecological malignancies. Gynecol Obstet Invest 27:42-44, 1989 12. Gullick WJ, Marsden JJ, Whittle N, et al: Expression of epidermal growth factor receptors on human cervical, ovarian, and vulval carcinomas. Cancer Res 46:285-292, 1986 13. Bauknecht T, Runge M, Schwall M, et al: Occurrence of epidermal growth factor receptors in human adnexal tumors and their prognostic value in advanced ovarian carcinomas. Gynecol Oncol 29:147-157, 1988 14. Berchuck A, Rodriguez GC, Kamel A, et al: Epidermal growth factor receptor expression in normal ovarian epithelium and ovarian cancer. Correlation of receptor expression with prognostic factors in patients with ovarian cancer. Am J Obstet Gynecol 164:669-674, 1991 15. Berchuck A, Olt GJ, Everitt L, et al: The role of peptide growth factors in epithelial ovarian cancer. Obstet Gynecol 75:255262, 1990 16. Scambia G, Benedetti Panici P, Battaglia F, et al: Presence of epidermal growth factor (EGF) receptor and proliferative response to EGF in six human ovarian carcinoma cell lines. Int J Gynecol Cancer (in press) 17. International Federation of Gynecology and Obstetrics: Changes in definitions of clinical staging for carcinoma of the cervix and ovary. Am J Obstet Gynecol 156:263-264, 1987 18. Serov SF, Scully RE: Histological typing of ovarian tumors, in International Histological Classification of Tumors, no. 9. Geneva, Switzerland, World Health Organization, 1973
19. Benedetti Panici P, Greggi S, Scambia G, et al: High-dose (200 mg/mz) cisplatin-induced neurotoxicity in primary advanced ovarian cancer patients. Cancer Treat Rep 71:669-670, 1987 20. World Health Organization: WHO Handbook for Reporting Results of Cancer Treatment, no. 48. Geneva, Switzerland, WHO, 1979, pp 16-21 21. Benedetti Panici P, Scambia G, Greggi S, et al: Recombinant interleukin-2 continuous infusion in ovarian cancer patients with minimal residual disease at second-look. Cancer Treat Rev 16:123-127, 1988 22. Benedetti Panici P, Scambia G, Greggi S, et al: Doxorubicin and cyclophosphamide, alternated with bleomycin and mitomycin C as a second line regimen in advanced ovarian carcinoma resistant to cis-platin based chemotherapy. Oncology 47:296-298, 1990 23. Bradford MM: A rapid and sensitive method for the quantitation of microgram quantities of protein using the principle of protein dye-binding. Anal Biochem 72:248-254, 1976 24. Kaplan E, Meier P: Nonparametric estimation from incomplete observation. J Am Stat Assoc 53:457-481, 1958 25. Mantel N: Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep 50:163-170, 1966 26. Cox DR: Regression models and life tables. J R Stat Soc 34:197-220, 1972 27. Dixon WS: BMDP statistical software. Berkeley, CA, University of California Press, 1981 28. Macias A, Azevedo E, Perez R, et al: Receptors for epidermal growth factor in human mammary carcinomas and their metastases. Anticancer Res 6:849-852, 1986 29. Gladhaug IP, Christoffersen T: n-Butyrate and dexamethasone synergically modulate the surface expression of epidermal growth factor receptors in cultured rat hepatocytes. FEBS Lett 243:21-24, 1989 30. Clark AJL, Ishii S, Richert N, et al: Epidermal growth factor regulates the expression of its own receptor. Proc Natl Acad Sci USA 82:8374-8378, 1985 31. Gardner RM, Verner G, Kirland JL, et al: Regulation of uterine epidermal growth factor (EGF) receptors by estrogen in the mature rat and during the estrous cycle. J Ster Biochem 32:339-343, 1989 32. Ulrich A, Coussens JS, Hayflick JS, et al: Human epidermal growth factor receptor cDNA sequence and aberrant expression of the amplified gene in A431 epidermoid carcinoma cells. Nature 309:418-425, 1984 33. Ozanne B, Richards CS, Hendler F, et al: Over-expression of the EGF receptor is a hallmark of squamous cell carcinomas. J Pathol 194:9-14, 1986 34. Copeland U, Gershenson DM: Ovarian cancer recurrences in patients with no macroscopic tumor at second-look laparotomy. Obstet Gynecol 68:873-874, 1986 35. Slamon SD, Godolphin W, Yones LA, et al: Studies of the HER-2/neu proto-oncogene in human breast and ovarian cancer. Science 244:707-712, 1989 36. Berchuck A, Kamel A, Whitaker R, et al: Overexpression of HER-2/neu is associated with poor survival in advanced epithelial ovarian cancer. Cancer Res 50:4087-4091, 1990 37. Bargman CI, Hung MC, Weinberg RA: The neu oncogene encodes an epidermal growth factor receptor-related protein. Nature 319:226-230, 1986
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 20, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
535
EGF-R IN ADVANCED OVARIAN CANCER 38. Stern DF, Kamps MP: EGF-stimulated tyrosine phosphorilation of p185/neu: A potential model for receptor interactions. EMBO J 7:995-1001, 1988 39. Wright C, Angus B, Nicholson S, et al: Expression of c-erbB-2 oncoprotein: A prognostic indicator in human breast cancer. Cancer Res 49:2087-2090, 1989
40. Masui H, Kawamoto T, Sato JD, et al: Growth inhibition of human tumor cells in athymic mice by antiepidermal growth factor receptor monoclonal antibodies. Cancer Res 44:1002-1007, 1988 41. Eppstein DA, Marsh YV, Schryver BB, et al: Inhibition of epidermal growth factor stimulated cell growth by a synthetic peptide. J Cell Physiol 141:420-430, 1989
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 20, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
L
ONG-TERM SURVIVAL of patients with advanced ovarian carcinoma remains poor.' Progno-
sis of these patients appears to be influenced by clinical
and histopathologic factors that are frequently interrelated.2' 3 Thus, the identification of new biologic factors
related to an individual tumor's aggressiveness would be extremely useful to allow individualized and possibly more effective treatments. Epidermal growth factor (EGF) is a single-chain
polypeptide with growth-promoting effects in several normal and neoplastic cell lines.4 The EGF-stimulating
action is mediated by a specific transmembrane receptor (EGF-R) that shows close similarity to the product of
the v-erb-B oncogene.' Consistent data have suggested that EGF-R expression might represent a parameter of poor prognosis in breast,6' 7 bladder,' esophageal,' and cervical"' cancer. The presence of EGF-R in ovarian
cancer specimens-''4 and the growth stimulatory effect 6 of EGF on several ovarian cancer cell lines'5 ," have
recently been reported. However, although a relation-
of 15; 60%) was found between primary tumors and omental metastases, with a tendency toward higher EGF-R levels in the latter. The EGF-R expression did not significantly correlate with age, stage, grading, and residual tumor after primary surgery. In the univariate analysis, stage IV disease, postoperative residual tumor diameter greater than 2 cm, presence of ascites, and EGF-R positivity were found to be significantly associated with a greater risk of disease progression. In the multivariate analysis, only the postoperative residual tumor and the EGF-R expression remained significantly associated with a high risk of progression. Conclusion: Data reported here suggest that the presence of EGF-R in advanced ovarian tumor at the time of the primary surgery identifies a subset of patients with a particularly poor prognosis. J Clin Oncol 10:529-535. © 1992 by American Society of Clinical Oncology. PATIENTS AND METHODS Between 1986 and 1990, 76 previously untreated patients with histologically confirmed stage III-IV (International Federation of Gynecology and Obstetrics [FIGO] classification17 ) ovarian carcinoma were admitted to the study. The World Health Organization (WHO) histologic typing of ovarian tumors"8 was adopted. Four patients were excluded from the analysis because of inadequate follow-up data. The clinicopathologic features of the remaining cases are listed in Table 1. Chemotherapy was started 2 to 3 weeks after surgery. All patients received cisplatin-containing regimens. Thirty-seven patients received high-dose cisplatin (200 mg/im2) for three courses." Gynecologic examination, abdominopelvic ultrasonography, CA125 assay, and radiologic investigations, if necessary, were performed monthly for the clinical assessment of response, which was recorded according to WHO criteria. 20 About 28 days after the last course, clinical complete responders underwent second-look laparoscopy. In laparoscopy-negative cases, second-look laparotomy was performed for the assessment of pathologic response. During laparotomy and after peritoneal washings and careful inspection of the abdominal cavity, biopsy of all suspicious lesions was performed, and, in case of no evidence of disease, at least 20 random biopsies were taken. Patients who initially had only an explorative laparotomy underwent a second laparotomy after
ship between EGF-R expression and poor outcome in
patients with ovarian cancer has been reported,14 the prognostic value of EGF-R in a multivariate analysis has
been never investigated. We report here the results of a large, single-institution, prospective study conducted at the Catholic University in Rome, Italy, which aims to assess the clinical significance of EGF-R in advanced ovarian carcinoma.
From the Department of Gynecology and Obstetrics, Catholic University, Rome, Italy. Submitted June 3, 1991; accepted November 19, 1991. Address reprint requests to Professor Salvatore Mancuso, Department of Gynecology, Catholic University, Largo A. Gemelli 8, 00168 Rome, Italy. © 1992 by American Society of Clinical Oncology. 0732-183X/9211004-0005$3.00/0
Journalof Clinical Oncology, Vol 10, No 4 (April), 1992: pp 529-535
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 20, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
529
530
SCAMBIA ET AL Table 1. Patient Characteristics
stopped by the addition of 3 mL of 25 mmol/L TRIS, 20% glycerol, lcrl RS 0 f25mo/ f3m stopped" byteadto 5
Patients
Entered Assessable Age (years) < 50 > 50 FIGO stage III IV Grade of differentiation Well-differentiated (G1) Moderately-poorly differentiated (G2-G3) Histology Serous Mucinous Endometrioid Undifferentiated Others Ascites No Yes Surgical debulking Optimally debulked (RT < 2 cm) Suboptimally debulked (RT > 2 cm) Clinical response to chemotherapy Complete response Partial response-no change
%
mmol/L
obtained
^'"'^" coUUedU
76
NaN3,
by
and
0.1%
bovine
at
centrifugation
n1 a gamlma LcounterI
serum
for
2,000g
I mlllru. 1
albumin.
minutes
20
Pellets
at
were
00C
and
KesulLS were expressed
19 53
26 74
57
79
13• 59
18
as femtomoles per milligram of membrane protein. Protein concen23 tration was measured using the method reported by Bradford. Tissue weight permitting, Scatchard analysis was performed with concentrations of 125-I-EGF ranging from 0.28 to 3.4 nmol/L, either alone or in the presence of unlabeled EGF (810 limol/L). As 7 used for breast cancer, 1.5 fmol per milligram of protein was considered the cutoff for EGF-R-positive tumors to distinguish ovarian tumors with high (> 1.5 fmol/mg protein) and low (< 1.5 fmol/mg protein) EGF-R content.
82
StatisticalAnalysis
50 4 9 6 3
69 5 13 8 4
19 53
26 74
43
60
29
40
The Wilcoxon rank-sum test was used to compare receptor 2 levels. x analysis was used to analyze the relationship between EGF-R expression and tumor characteristics or response to chemotherapy. All medians and life tables were computed using the 24 product-limit estimate by Kaplan and Meier, and the curves were examined by means of the log-rank test.2 The risk of progression 26 was estimated by Cox's proportional hazards model. Multivariate 27 analysis was performed with BMDP statistical software. A backward stepwise procedure was used to identify the major prognostic factors. Progression-free survival (PFS) was calculated from the date of the first surgery to the date of clinical or pathologic progression or death. The median follow-up was 24 months (range, 4 to 75 months; as of January 1991).
48 24
67 33
Abbreviation: RT, postoperative residual tumor.
chemotherapy, and a second cytoreduction was attempted. Pathologic complete responders received no further therapy, and all other patients were treated according to ongoing phase II studies.2.22
RESULTS
Expression of EGF-R Figure 1 shows a representative example of a saturation analysis plot of the specific binding of 125-I-EGF in ovarian cancer specimens. A linear regression fit (r = .98)
Iodine-125-EGFBinding Measurement Tumor specimens, collected during primary surgery, were frozen 0 on dry ice shortly after surgical removal and stored at -80 C until processed. A representative section of specimens was retained for histopathologic examination. The membrane fraction and cytosol were prepared as described elsewhere.7 Briefly, tumor specimens were finely minced and homogenized in 5-volume ice-cold buffer consisting of 25 mmol/L trometamol (TRIS), 1.5 mmol/L edathamil [EDTA], 5 mmol/L NaN3, 0.1% monothioglycerol, and 20% glycerol (TENMG), by applying several intermittent bursts of an Ultra-Turrax homogenizer (Janke & Kunkel, IKA, Labortechnik, Staufen, Germany). The crude homogenate was centrifuged at 7,000g for 20 minutes at 00 C. The supernatants were then centrifuged at 105,000g for 75 minutes at O0C. The membrane pellet was resuspended in 25 mmol/L TRIS, 1.5 mmol/L EDTA, 5 mmol/L NaN3, 20% glycerol, and 10 mmol/L magnesium chloride (MgCI2) (TENG + MgCI2). Aliquots of the suspension (100 IL containing 300 to 500 pLgprotein) were incubated with iodine-125 (125-I)-EGF (2.6 nmol/L) in the presence or absence of unlabeled EGF (1 mmol/L) for 16 hours at room temperature in a final volume of 400 IL. Binding was
z
a 9
2..
o
u;
o o e: m
o_
r
OUND98(pM)
0 SOUND(pM)
"5I-EGF
concentrations (nM)
Fig 1. Scatchard analysis of 125-1-EGF binding to ovarian tumor membranes. Specifically bound 125-1-EGF was measured as detailed in Patients and Methods.
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 20, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
531
EGF-R IN ADVANCED OVARIAN CANCER
indicated one specific high-affinity binding site (dissociation constant [KI] = 0.65 nmol/L) with an x-intercept on the Scatchard plot of 14.78 fmol/mg protein. Among the ovarian cancer specimens tested, K, values ranged from 0.58 to 1.6 nmol/L. The specificity of EGF binding in ovarian cancer was evaluated by testing the ability of various peptidic hormones at the same concentration of unlabeled EGF to prevent binding of 125-I-EGF. Luteinizing hormone-releasing hormone (LH-RH), thyrotropinreleasing hormone (TRH), insulin, follicle-stimulating hormone (FSH), growth hormone (GH), and endothelial cell growth factor all failed to compete effectively for 125-I-EGF (< 10% displacement of 125-I-EGF binding), whereas unlabeled EGF displaced 66% of the bound 125-I-EGF (data not shown). We also investigated the heterogeneity of EGF binding in ovarian cancer (Table 2). In eight cases EGF-R concentrations were simultaneously measured in different specimens of the same tumor. Although differences in the concentrations of EGF-R within each tumor were found, in only one case we observed high (> 1.5 fmol/mg protein) and low values coexisting in the same tumor. This analysis revealed that EGF-R is a stable feature within primary ovarian cancer. On the other hand, a lower concordance rate (nine of 15; 60%) was found between primary tumors and omenTable 2. EGF-R Content in Primary Epithelial Ovarian Carcinomas and in Specimens From Metastases Obtained Concomitantly Patient
No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
tal metastases, with the latter exhibiting a tendency toward a higher percentage of specimens with high EGF-R levels. Overall EGF-R content was higher in metastatic deposits (median, 2.68; range, 0 to 10.8 fmol/mg protein) than in primary tumors (median, 1.40; range, 0 to 10.30 fmol/mg protein), although this difference did not reach a statistical significance. Of the 72 primary tumors examined, 39 (54%) exhibited high values (> 1.5 fmol/mg protein) of EGF-R. Table 3 shows the distribution of EGF-R according to the clinical and histopathologic characteristics of the on-study population. EGF-R expression was not significantly associated with age, stage, histopathologic grading, postoperative residual tumor, and response to chemotherapy. EGF-R Expression and Risk of Progression During follow-up, disease progression was observed in 38 patients. High values of EGF-R were present in 28 of the 38 progressive cases. Figure 2 shows the PFS curves in relation to EGF-R status. A highly significant association between patients with low EGF-R expression and a longer PFS compared with patients with high EGF-R expression (P = .0004) was Table 3. EGF-R Content in Ovarian Cancer by Clinicopathologic Characteristics
Ovarian Primary Tumor 1.6,1.8* 5.8,9.8* 2.5,0.7* 10.9, 6.3* 0, 0* 15.9,12.0, 12.8* 1.6, 3.6* 0, 1.3* 0 0 0 2.2 3.1 10.3 1.3 6.6 0 5.3 4.3 1.6 1.0
*Different specimens of the primary tumor.
EGF-R Content 1.5 fmol/mg Protein
No. of
EGF-R Content (fmol/mg protein) Metastases -
1.9 0.8 8.7 0 9.3 6.6 10.8 3.6 2.7 2.7 0 2.5 0.1 0 9.2
Variable Age (years) 2 cm Clinical response to chemotherapy Complete response Partial response-no change
Cases
No. of Cases
%
19 53
8 31
42 58
57 15
29 10
51 67
13 59
8 31
61 52
50 4 9 6 3
29 2 5 2 1
58 50 55 33 33
43 29
21 18
49 62
48 24
22 17
46 71
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 20, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
SCAMBIA ET AL
532 -----
LL,
Table 5. Multivariate Analysis of Prognostic Variables for PFS in Patients With Advanced Ovarian Carcinoma
I L I__
Prognostic Variable z o
FIGO stage III IV Surgical debulking RT < 2 cm RT > 2 cm Ascites No Yes EGF-R status (fmol/mg protein) < 1.5 > 1.5
o.
z
MONTHS
Fig 2. PFS rate by EGF-R content. EGF-R+ (> 1.5 fmol/mg protein, ---- ; 68%; 95% CI, 42% to 94%): 39 patients entered, 28 progressed in a median of 15 months. EGF-R- (< 1.5 fmol/mg protein, -; 23%; 95% CI, 8% to 38%): 33 patients entered, 10 progressed in a median of NR. P = .0004 by log-rank test.
observed. The 36-month PFS was 68% (95% confidence interval [CI], 42% to 94%) for patients with EGF-R less than 1.5 fmol/mg protein, compared with 23% (95% CI, 8% to 38%) for those with EGF-R > 1.5 fmol/mg protein. The relative risk of disease progression was estimated in a univariate analysis and then in a multivariate analysis using a backward stepwise procedure. In the univariate analysis (Table 4), stage IV disease, postoperative residual tumor diameter greater than 2 cm, presence of ascites, and high EGF-R content were found to be significantly associated with a greater risk of disease Table 4. Univariate Analysis of Prognostic Variables for PFS in Patients With Advanced Ovarian Carcinoma
Age (years) 50 FIGO stage III IV Histopathologic grading G1 G2-G3 Surgical debulking RT < 2 cm RT > 2 cm Ascites No Yes EGF-R status (fmol/mg protein) < 1.5 > 1.5
No. of Cases
Median PFS (months)
19 53
18 16
1* 1.1
57 15
20 11
1 " 2.6
.006
13 59
16 19
1* 0.8
.41
43 29
47 13
1* 2.7
.002
19 53
49 15
1" 3.2
.002
33 39
50 15
1* 3.4
RR1
P
No. of Cases
Median PFS (months)
57 15
20 11
-
43 29
47 13
1 2.8
19 53
49 15
-
33 39
50 15
1* 3.8
RR2
P .004 .0005
Abbreviation: RR2, adjusted relative risk of progression taking into account all the factors of the table. *Reference category.
progression. In the multivariate analysis (Table 5), only postoperative residual tumor and EGF-R expression remained significantly associated with a high risk of progression. Figure 3 shows the PFS curves according to EGF-R
status and clinical response to chemotherapy. Among patients who achieved a clinical complete response (CR) to chemotherapy at 3 years, those with low values of EGF-R had a 75% PFS with a relative risk of progression of 0.3, compared with 33% and a relative risk of 1.4 for patients with high values of EGF-R. The worst outcome was observed in the subset of patients with high EGF-R expression and less than CR to chemotherapy; in fact, all of these patients progressed within 24 months.
.92
z 0 z
MONTHS
Abbreviation: RR1, unadjusted relative risk of progression. *Reference category.
.0005
Fig 3. PFS rate and relative risk (RR) of progression by EGF-R content and clinical CR to chemotherapy. EGF-R- and 75% with CR (a), RR = 0.3; EGF-R- and 43% with no CR (b), RR = 1.2; EGF-R+ and 33% with CR (c), RR = 1.4; and EGF-R+ and no CRs (d), RR = 5.3.
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 20, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
533
EGF-R IN ADVANCED OVARIAN CANCER DISCUSSION
Our data indicate that a significant proportion of ovarian tumors contain specific EGF-binding proteins. In our series the percentage of tumor with high EGF-R expression was higher than that previously reported by Bauknecht et al."3 On the other hand, Berchuck et al' 4 using an immunohistochemical method reported 82% of ovarian tumors expressing EGF-R. Methodologic differences and/or differences in the cutoff values may explain these discrepancies. Interestingly, when EGF-R was analyzed in multiple sites of the same tumor, consistent findings were noted in 88% of cases. Therefore, it appears that in most cases EGF-R status can be adequately assessed using a single tumor sample. However, omental metastases express high EGF-R levels more frequently than their respective primary tumors, although the difference was not statistically significant. Nevertheless, it can be hypothesized that, as in breast cancer, 7,28 metastatic clones express EGF-R more frequently, thereby suggesting that EGF-R expression can be related to the metastases potential. In our experience, the expression of EGF-R was not related to any clinical and/or pathologic tumor characteristic. This is consistent with the data reported by Baucknecht et al"3 and Berchuck et al.' 4 Interestingly, in spite of the lack of correlation with clinicopathologic features, EGF-R status demonstrated a strong prognostic value in patients with advanced ovarian cancer. In fact, after multivariate analysis the association of high EGF-R levels with worse PFS was independent from the other known prognostic factors such as residual disease and stage. Although it is difficult to reconcile this finding with the fact that EGF-R and its ligands play a role in the growth and differentiation of normal ovarian surface epithelium,'4 some considerations should be taken into account. In normal cells, EGF-R levels are regulated by several factors, such as steroid hormones and other autocrine peptide factors, and are differently expressed according to different phases of cellular metabolism.29"'3 In cancer cells, EGF-R overexpression may result from gene amplification or, more frequently, from the loss of normal regulatory mechanisms of transcription.32,33 Therefore, cancer cells may be continuously exposed to the stimulatory effect of EGF/transforming growth factor-alpha. The presence of higher amounts of EGF-R may also result in an enhanced proliferation for a given amount of autocrinereleased growth factors. This may explain the association of enhanced EGF-R expression with more aggres-
sive tumor behavior, which has been reported for various epithelial cancers. 710 The current prognostic characterization of patients with advanced ovarian cancer is still inadequate. In fact, about 50% of patients after an optimal surgical debulking and a pathologic CR to primary chemotherapy will recur and die of disease.34 Therefore, the identification of new variables correlating with the malignant potential of the cancer cells would be clinically valuable in the selection of therapy for individual patients. Data from the present study suggest that the assessment of EGF-R status at the time of initial surgery may allow the identification of a subset of patients with a particularly poor prognosis. These patients could be candidates for more aggressive and/or experimental primary treatments than those conventionally used. However, in our series, 25% of patients with EGF-Rnegative tumors and CR to chemotherapy progressed within 3 years. Hopefully similar patients could be identified through other prognostic parameters. Among the new prognostic factors proposed for ovarian cancer, the amplification/overexpression of erbB2/neu seems promising. 35' 36 The product of the neu proto-oncogene is a 185-kd glycoprotein defined as p185 that, like the EGF-R, is a transmembrane protein with tyrosine-kinase activity.3 7 These properties and the close
structural homology of p18537 and EGF-R suggest that p185 is itself a growth factor receptor. The finding that EGF stimulates the phosphorylation of p18538 suggests that both EGF-R and erb-B2/neu may be involved in the same pathway leading to cell proliferation and tumor progression. Therefore, it is conceivable that the simultaneous evaluation of EGF-R and erb-B2/neu may increase the prognostic characterization of ovarian cancer patients. In this context, it is interesting that Wright et a19 reported an apparent additive effect of p185 and EGF-R expression on the prognosis of patients with breast cancer. Data presented here, together with the recently reported in vitro data',"16 indicating that ovarian cancer cell proliferation is modulated by the EGF/EGF-R system, further confirm that this system may play a role in the control of ovarian cancer onset and spread. Therefore, drugs and/or biologic response modifiers able to interfere with the EGF-mediated control of cell proliferation may be useful in ovarian cancer therapy. In this context, it is notable that monoclonal antibodies, anti-EGF-R,4 and synthetic, EGF-like peptides41 have been demonstrated to inhibit the growth of cancer cells.
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 20, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
534
SCAMBIA ET AL REFERENCES
1. Ozols RF, Young RC: Chemotherapy of ovarian cancer. Semin Oncol 11:251-263, 1984 2. Dembo AJ, Bush RS: Choice of postoperative therapy based on prognostic factors. Int J Radiat Oncol Biol Phys 8:893-897, 1982 3. Malkasian GD, Melton LJ, O'Brien PC, et al: Prognostic significance of histologic classification and grading of epithelial malignancies of the ovary. Am J Ostet Gynecol 149:274-282, 1984 4. Carpenter G: Receptors for epidermal growth factor and other polypeptide mitogens. Ann Rev Biochem 56:881-884, 1987 5. Downward J, Yarden Y, Mayes E, et al: Close similarity of epidermal growth factor receptor and v-erb-B oncogene protein sequences. Nature 307:521-527, 1984 6. Sainsbury JRC, Farndon JR, Needham JK, et al: Epidermal growth factor receptor status as predictor of early recurrence of and death from breast cancer. Lancet 1:1398-1401, 1987 7. Battaglia F, Scambia G, Rossi S, et al: Epidermal growth factor receptor in human breast cancer: Correlation with steroid hormone receptors and axillary lymph-node involvement. Eur J Cancer Clin Oncol 24:1685-1690, 1988 8. Neal DE, Marsh C, Bennet MK, et al: Epidermal growth factor receptors in human bladder cancer: Comparison of invasive and superficial tumors. Lancet 1:366-368, 1985 9. Ozawa S, Ueda M, Ando N, et al: Prognostic significance of epidermal growth factor receptor in esophageal squamous cell carcinomas. Cancer 63:2169-2173, 1989 10. Pfeiffer D, Stellwag B, Pfeiffer A, et al: Clinical implications of epidermal growth factor receptor in the squamous cell carcinoma of the uterine cervix. Gynecol Oncol 3:146-150, 1989 11. Battaglia F, Scambia G, Benedetti Panici P, et al: Epidermal growth factor receptors in gynecological malignancies. Gynecol Obstet Invest 27:42-44, 1989 12. Gullick WJ, Marsden JJ, Whittle N, et al: Expression of epidermal growth factor receptors on human cervical, ovarian, and vulval carcinomas. Cancer Res 46:285-292, 1986 13. Bauknecht T, Runge M, Schwall M, et al: Occurrence of epidermal growth factor receptors in human adnexal tumors and their prognostic value in advanced ovarian carcinomas. Gynecol Oncol 29:147-157, 1988 14. Berchuck A, Rodriguez GC, Kamel A, et al: Epidermal growth factor receptor expression in normal ovarian epithelium and ovarian cancer. Correlation of receptor expression with prognostic factors in patients with ovarian cancer. Am J Obstet Gynecol 164:669-674, 1991 15. Berchuck A, Olt GJ, Everitt L, et al: The role of peptide growth factors in epithelial ovarian cancer. Obstet Gynecol 75:255262, 1990 16. Scambia G, Benedetti Panici P, Battaglia F, et al: Presence of epidermal growth factor (EGF) receptor and proliferative response to EGF in six human ovarian carcinoma cell lines. Int J Gynecol Cancer (in press) 17. International Federation of Gynecology and Obstetrics: Changes in definitions of clinical staging for carcinoma of the cervix and ovary. Am J Obstet Gynecol 156:263-264, 1987 18. Serov SF, Scully RE: Histological typing of ovarian tumors, in International Histological Classification of Tumors, no. 9. Geneva, Switzerland, World Health Organization, 1973
19. Benedetti Panici P, Greggi S, Scambia G, et al: High-dose (200 mg/mz) cisplatin-induced neurotoxicity in primary advanced ovarian cancer patients. Cancer Treat Rep 71:669-670, 1987 20. World Health Organization: WHO Handbook for Reporting Results of Cancer Treatment, no. 48. Geneva, Switzerland, WHO, 1979, pp 16-21 21. Benedetti Panici P, Scambia G, Greggi S, et al: Recombinant interleukin-2 continuous infusion in ovarian cancer patients with minimal residual disease at second-look. Cancer Treat Rev 16:123-127, 1988 22. Benedetti Panici P, Scambia G, Greggi S, et al: Doxorubicin and cyclophosphamide, alternated with bleomycin and mitomycin C as a second line regimen in advanced ovarian carcinoma resistant to cis-platin based chemotherapy. Oncology 47:296-298, 1990 23. Bradford MM: A rapid and sensitive method for the quantitation of microgram quantities of protein using the principle of protein dye-binding. Anal Biochem 72:248-254, 1976 24. Kaplan E, Meier P: Nonparametric estimation from incomplete observation. J Am Stat Assoc 53:457-481, 1958 25. Mantel N: Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep 50:163-170, 1966 26. Cox DR: Regression models and life tables. J R Stat Soc 34:197-220, 1972 27. Dixon WS: BMDP statistical software. Berkeley, CA, University of California Press, 1981 28. Macias A, Azevedo E, Perez R, et al: Receptors for epidermal growth factor in human mammary carcinomas and their metastases. Anticancer Res 6:849-852, 1986 29. Gladhaug IP, Christoffersen T: n-Butyrate and dexamethasone synergically modulate the surface expression of epidermal growth factor receptors in cultured rat hepatocytes. FEBS Lett 243:21-24, 1989 30. Clark AJL, Ishii S, Richert N, et al: Epidermal growth factor regulates the expression of its own receptor. Proc Natl Acad Sci USA 82:8374-8378, 1985 31. Gardner RM, Verner G, Kirland JL, et al: Regulation of uterine epidermal growth factor (EGF) receptors by estrogen in the mature rat and during the estrous cycle. J Ster Biochem 32:339-343, 1989 32. Ulrich A, Coussens JS, Hayflick JS, et al: Human epidermal growth factor receptor cDNA sequence and aberrant expression of the amplified gene in A431 epidermoid carcinoma cells. Nature 309:418-425, 1984 33. Ozanne B, Richards CS, Hendler F, et al: Over-expression of the EGF receptor is a hallmark of squamous cell carcinomas. J Pathol 194:9-14, 1986 34. Copeland U, Gershenson DM: Ovarian cancer recurrences in patients with no macroscopic tumor at second-look laparotomy. Obstet Gynecol 68:873-874, 1986 35. Slamon SD, Godolphin W, Yones LA, et al: Studies of the HER-2/neu proto-oncogene in human breast and ovarian cancer. Science 244:707-712, 1989 36. Berchuck A, Kamel A, Whitaker R, et al: Overexpression of HER-2/neu is associated with poor survival in advanced epithelial ovarian cancer. Cancer Res 50:4087-4091, 1990 37. Bargman CI, Hung MC, Weinberg RA: The neu oncogene encodes an epidermal growth factor receptor-related protein. Nature 319:226-230, 1986
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 20, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
535
EGF-R IN ADVANCED OVARIAN CANCER 38. Stern DF, Kamps MP: EGF-stimulated tyrosine phosphorilation of p185/neu: A potential model for receptor interactions. EMBO J 7:995-1001, 1988 39. Wright C, Angus B, Nicholson S, et al: Expression of c-erbB-2 oncoprotein: A prognostic indicator in human breast cancer. Cancer Res 49:2087-2090, 1989
40. Masui H, Kawamoto T, Sato JD, et al: Growth inhibition of human tumor cells in athymic mice by antiepidermal growth factor receptor monoclonal antibodies. Cancer Res 44:1002-1007, 1988 41. Eppstein DA, Marsh YV, Schryver BB, et al: Inhibition of epidermal growth factor stimulated cell growth by a synthetic peptide. J Cell Physiol 141:420-430, 1989
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 20, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
