GCDFP-15 blood levels for stratification of risk of breast cancer development in women with active breast gross cystic disease

TheBrerrrr(lYY7) 6, 113-119 0 I997 Pearson Professional Ltd

ORIGINAL ARTICLE

GCDFP-15 blood levels for stratification of risk of breast cancer development in women with active breast gross cystic disease D. E. Haagensen,*’ D. Kelly’ and C. A. Bodian’* *The Methodist Hospital, Department of Surgery, Sacramento, iHuagensen Research Foundation Inc., New York, *Department of Biomathematical Sciences, Mount Sinai Medical Center, New York, USA. S U M MA R Y. The risk of developing breast cancer has been shown to be elevated in women with breast gross cystic disease (GCD) confirmed by aspiration of breast cyst fluid. The GCDFP-15 protein is a major component protein of breast GCD fluid. The protein is secreted by apocrine metaplastic cells lining micro and gross cysts of the breast. It is also a constituent protein secreted by normal apocrine glandular systems, and it is present in approximately 50% of breast carcinomas. Blood levels of GCDFP-15 are measurable in the low rig/ml range in normal individuals. A proportion of patients with metastatic breast cancer have high circulating blood levels up to 70 000 rig/ml. This study evaluated the relationship between blood levels of GCDFP-15 in women with active breast gross cystic disease, demonstrated by breast cyst aspiration, and the risk of breast cancer development. A group of 13.5 patients who had active breast gross cystic disease demonstrated by cyst aspiration, also had a concomitant blood level of GCDFP- 15 determined. The patients have been subsequently followed for an average of 10.1 years and assessed for risk of breast cancer development. Twelve of these women have developed breast carcinoma and two have died of unrelated causes. For the 72 women who clinically developed fewer than 10 aspirated cysts, the relative risk of breast cancer was 1.8 if their highest GCDFP-15 plasma level was in a normal range (less than 40 rig/ml) versus a relative risk of 4.2 if the GCDFP-15 plasma level was elevated. This difference was more pronounced for the 63 women in this study with 10 or more total aspirated cysts, where the relative risk for breast cancer development was 2.0 for women with normal range GCDFP-15 plasma levels versus 7.1 for those with elevated plasma levels. Our data indicate that GDCFP- 15 blood levels may be useful in identifying groups of women with active breast gross cystic disease who are at increased risk of developing breast cancer.

of developing breast carcinoma, with estimates of relative risk ranging from 1.8 to 4.4.3-7 The report of Bodian et al7 concerned the results of an average of 18 years of followup of patients of the late Dr Cushman D. Haagensen.’ Among 1770 of his patients who had at least one gross cyst confirmed by aspiration, the risk of developing breast carcinoma was 2.9 times the risk among women of comparable ages and exposure time in the general population. This increase in risk of breast carcinoma development varied with the intensity of the breast gross cystic disease, as reflected in the total number of cyst aspirations. Patients who had 10 or more cyst aspirations developed breast cancer at almost twice the rate as women with a single aspiration.7 The fluid contained within breast gross cysts has been analysed for a number of components,xm’8 and was found to be a unique secretion. The major component proteins have been identified.8 One of these component proteins, termed GCDFP-15 for a 15K Dalton monomer sized glyco-

INTRODUCTION Human breast gross cystic disease is a common benign breast condition of middle-aged women.’ It has been shown that gross cysts evolve from microcystic apocrine metaplasia occurring in the terminal ductal lobular units of the breast.’ The stimuli causing the transformation of apocrine microcysts into macrocysts is not known. However, evidence that this is a hormonally mediated event comes from the clinical observation that gross cystic disease first appears in women in their 20s and increases in frequency up to the 40-50 age range, then essentially disappears as a disease process after the menopause.’ Several studies have followed patients with breast gross cystic disease confirmed by aspiration of cyst fluid and have found that these women have an increased risk Address

correspondence

PH II-1 102, 630 West,

mo: D. E. Haagensen, Research Foundation, 168th Street, New York, NY, 1003, USA

Inc.,

113

114

The Breast

protein, has been investigated as a marker protein with regard to its circulating blood levels in various clinical situations.8.9’2y The GCDFP-15 blood levels in ‘normal’ postmenopausal women are quite low and average 17 ng/ ml. In premenopausal women the ‘normal’ blood levels average 22 rig/ml. In women with active breast gross cystic disease a range of GCDFP-15 blood levels has been observed up to 250 ng/m1.9,2yThe highest blood levels of GCDFP-15 have been observed in women with metastatic breast carcinoma where approximately 35% of patients develop blood levels above 150 rig/ml during their disease course and approximately 10% develop blood levels above 1000 rig/ml, with the highest blood level observed being 70 000 rig/ml.’ This paper reports on a correlation of the plasma level of GCDFP-15 determined at the time of aspiration of a breast cyst with the future risk for breast carcinoma development.

MATERIALS

AND METHODS

Study group The patients in this study are part of a large group of women treated and followed by the late Dr Cushman Haagensen (CDH) as part of his ongoing research on breast diseases. Details about the patients studied and follow-up procedures have been reported elsewhere.7 Between the years of 1936 and 1983 CDH followed a total of 1770 women who developed breast gross cystic disease as defined by a cyst aspiration. Starting in 1975 all patients being seen by CDH for breast evaluation were asked if they would volunteer to donate a 7 ml peripheral sample of blood for measurement of GCDFP-15. There was over a 90% rate of concurrence with this request. This report concerns 135 patients with no prior breast cancer, who volunteered to provide a blood sample at the same time they had active gross cystic disease (GCD) confirmed by aspiration of breast cyst fluid. Follow-up information was updated in 1993-1994 by phone and by mail. All reports of breast carcinoma were confirmed by obtaining copies of pathology reports. For this study, follow-up time for each patient was measured from the date of her first blood sample obtained at the time of active GCD, to the date of diagnosis of breast cancer (12 patients), death (two patients), or last contact (121 patients), whichever came first. The average follow-up time was 10.1 years. This group of 135 patients contributed 214 blood samples at the time of a breast cyst aspiration. One hundred contributed a single sample and 35 contributed from two to seven samples, taken at the times of serial aspirations for GCD.

RESULTS The total numbers of breast gross cysts confirmed by aspiration are shown in Table 1 for the 135 women in this study group and for all the women in CDH’s series of GCD patients who had at least one cyst confirmed by aspiration. Forty-seven per cent of patients in the current study had 10 or more cysts aspirated, as compared to only 15% in the entire series. The relationship between total number of aspirated cysts and GCDFP-15 plasma level is shown in Table 2. In order to maintain comparability among patients, in this table patients with serial plasma levels are categorized according to their initial value of GCDFP-15. There is a slight, but not statistically significant, inverse association between the number of aspirated cysts and plasma level of GCDFP- 15. Table 1 Breast gross cystic disease confirmed fluid. Number of patients with aspirated cysts entire series of gross cystic disease patients Number aspirated 1 2-9 210 Totals

Chemical analysis of samples Blood samples were collected in EDTA

vacutainer tubes and were processed to plasma then frozen at -70°C until analysis. The analysis of the plasma level of GCDFP-15 was performed by a two stage RIA.8,y,“0Purified GCDFP-15 was radiolabeled with 125-Iusing Iodogen (Pierce Co., Rockford, Ill.) by the method of Freker and Speck.3’ Specific activity obtained was approximately 20 uCi/ug protein. A rabbit antiserum was utilized for GCDFP-15 detection.30z32All plasma samples were assayed in duplicate with any duplicate samples having a variation of greater than + 2% being reassayed. This assay has a sensitivity of & 5 rig/ml of GCDFP-15 and an interassay CV of 2 7%. For data analysis, the plasma levels have been grouped into three categories: normal range (less than 40 rig/ml), mildly elevated (4W9 rig/ml), and markedly elevated (50 rig/ml or greater). These ranges were chosen empirically on the basis of GCDFP-15 blood levels observed in 202 pre/perimenopausal women, without known prior breast GCD and with a normal breast examination by CDH. These ostensibly normal women had blood levels obtained and tested during the same time frame as the study group.9*29 The mean GCDFP-15 plasma level in these normal women was 22 rig/ml, with 89% of the women having plasma levels below 40 ng/ml.9X29

anticoagulated

of cysts

Number Study group 13 (9.6) 59 (43.7) 63 (46.7) 135 (100)

by aspiration in study group

of cyst versus

of patients (%)

Entire

series*

(%)

636 (35.9) 871 (49.2) 263 (14.9) 1770 (loo)

*All patients with aspirated cysts in Table 5,’ excluding 18 patients whom the total number of cysts aspirated was not known.

for

115

GCDFP- 15 blood levels in breast cystic disease Table 2 Total number of aspirated breast cysts versus the plasma of GCDFP-15 in the first blood sample concurrent with aspiration Number aspirated

of cysts

Number patients

of

Number (%) of patients with GCDFP-15 plasma levels in rig/ml <40

I

13

10 (77%)

2-9

210

59 63

39(66%) 37(59%)

Total

135

86(64%)

4049

level

Total

8(14%) 8(12%)

a50 3(23%) 12(20%) 18(29%)

100% 100% 100%

16(12%)

33(24%)

100%

(0%)

Among women with only one aspirated cyst, 77% had GCDFP-15 plasma levels in the normal range and 23% had a plasma level above 40 rig/ml; whereas the GCDFP- 15 plasma level was above 40 rig/ml in 41% of women with 10 or more aspirated cysts.

Biological variation in serial GCDFP-15 plasma levels with active breast gross cystic disease Thirty-five of the 135 women with active breast GCD had serial plasma levels of GCDFP-15 measured at the time of serial breast cyst aspirations. One hundred and fourteen blood samples were obtained. In 15 of the 35 patients all of the serial plasma levels of GCDFP-15 were below 40 rig/ml. In this subgroup no major shifts in plasma levels were seen (each individual’s serial plasma levels remained within f 5 rig/ml of one another, which is within interassay CV). Six patients had at least one of the serial plasma levels measured between 40 and 50 rig/ml. In three of these six patients all serial plasma levels were within + 5 rig/ml of one another and the other three patients showed greater variation. In 14 patients at least one serial plasma level was greater than 50 rig/ml. In only two of these 14 patients did the serial plasma levels stay within f 5 rig/ml of one another, indicating biological variation in GCDFP- 15 plasma levels occurs in most patients with values above 50 rig/ml.

Evaluation of serial plasma levels of GCDFP-15 obtained between episodes of active breast gross cystic disease In 14 of the 35 patients who had serial blood levels measured for GCDFP- 15, one or more blood samples were obtained between the time points of breast cyst aspirations when physical examination did not reveal any palpable breast cystic disease being present. These bracketted blood samples give information on GCDFP-15 plasma levels between episodes of clinically active breast gross cystic disease. In all 14 patients, the bracketted GCDFP-15 plasma levels remained in a similar range to the plasma levels detected at the time of cyst aspiration.

Evaluation of plasma levels of GCDFP-15 after menopause in patients who had active breast gross cystic disease In 25 of the 135 patients in this study, a blood sample has been obtained after the patient has passed into menopause (no menstrual period for at least 2 years) and has developed no further clinical signs of active breast gross cystic disease. Ten of these 25 patients had a GCDFP-15 plasma level above 40 rig/ml at the time of their last clinical cyst aspiration. In all 10 of these patients the postmenopausal plasma level of GCDFP-15 was lower than the last plasma level when active breast gross cystic disease was detected (Fig.). In two of these patients the postmenopausal plasma level of GCDFP-15 was still above 40 rig/ml and one of these two patients has developed breast cancer. In a prior study of 196 postmenopausal women, ostensibly normal on breast physical examination and without a prior history of breast gross cystic disease, the mean GCDFP- 15 plasma level was 17 rig/ml and only 1.5% of this study group had GCDFP-15 plasma levels above 50 ng/ml.y~‘y Also, in a prior study group of 197 postmenopausal women with a normal breast physical examination, but having a prior history of active breast gross cystic disease, only 4.1% of this patient group had GCDFP15 plasma levels above 50 ng/m1.9.‘9 Thus, it appears that the elevated plasma levels of GCDFP- 15 which occur in women

E .-

ro rL

70 60

k

50

g

40 30 20 10

0’

0

I

I

I

12345678910

I

I

I

I

I

I

1

YEARS Fig. Depicted is the GCDFP- 15 plasma level observed at the time of aspiration of the last clinically detected breast cyst (o), versus the GCDFP-I5 plasma level obtained at least 2 years later when the patient was clinically postmenopausal (0). The Y axis is the GCDFP- 15 plasma level in rig/ml. The X axis is the time in years between the GCDFP-15 measurements.

116

The Breasl

with active breast gross cystic disease are not maintained into the postmenopausal age range when breast cystic disease is no longer clinically active. It should be noted that this finding may not apply to postmenopausal women who are taking hormone replacement therapy where some patients continue to develop palpable gross cysts which require aspiration.

Relationship between GCDFP-15 plasma level and the risk of development of breast carcinoma Among this group of 135 women, 12 developed an initial carcinoma of the breast during the follow-up period. Ten of the cancers developed in women who had contributed a single blood sample, and two after serial samples. The elapsed time between entering this study and developing carcinoma ranged from less than one year to 15 years. The question of whether elevated plasma levels of GCDFP-15 are associated with an increased risk of developing breast cancer is addressed in Table 3. As part of this analysis, the follow-up for the 35 patients with serial blood samples was allocated to blood level groups according to their highest last-known value. For example, one patient contributed a blood sample in 1982 at age 40, which had a GCDFP-15 plasma level of 46 rig/ml, and another blood sample in 1984, with a plasma level of 77 rig/ml. She was last contacted in 1993, with no known cancer at that time. Two years of this patient’s follow-up were allocated to the ‘mildly elevated’ blood level group and 9 years (starting at age 42 in 1984) to the ‘markedly elevated’ group. Had the second blood level remained the same or gone down, her entire follow-up would have been allocated to the ‘mildly elevated’ group to which she was assigned by the first blood level. Eight of the 35 patients with serial blood levels changed categories in this way. Table 3 GCDFP-15

Previous work has shown that the risk of developing breast cancer is increased among women with several aspirated cysts.7 As is shown in Table 1, the patients entering this study included a disproportionate number of such women. Therefore, our data concerning cancer risk are shown separately according to whether or not the patient had a history of 10 or more breast cyst aspirations. In considering risk of breast cancer, it is also important to control for differences among groups in follow-up time, ages and calendar years of exposure. This was accomplished by calculating the relative risks of developing breast cancer for each group, that is, the ratios of the number of women in the group who developed breast cancer to the number that would have been expected according to incidence rates from the Connecticut Tumor Registry for comparable ages and calendar years of exposure. Table 3 shows that regardless of the total number of aspirated breast cysts, women with elevated levels of GCDFP-15 developed breast cancer at substantially higher rates than did women with normal blood levels. For women with fewer than 10 aspirated breast cysts, the relative risk among those with elevated blood levels of GCDFP-15 was 4.2, as compared to a relative risk of 1.8 for women with normal range blood levels. The difference in relative risks was more marked among women who had 10 or more aspirated breast cysts: 7.1 for those with elevated blood levels of GCDFP-15, versus 2.0 for those with normal range blood levels. As shown on the last two rows of Table 3, the overall relative risks of developing breast cancer among women in this study, 2.6 for those with fewer than 10 aspirated breast cysts, and 4.5 for those with 10 or more aspirated breast cysts, were virtually the same as the corresponding relative risks in the entire series of CDH patients who had at least one aspirated cyst.

Relative risk of developing breast cancer and by total number of aspirated breast Total number Less than 10

Plasma level of GCDFP- 15

Relative risk*

by highest cysts

of aspirated Numb. ptso

prior

plasma

breast

level of

cysts 10 or more

Numb. br.ca.

Relative risk*

Numb. pts”

Numb. br.ca.

<40 rig/ml

49

2

1.8

37

2

2.0

250 nglml 40-49 rig/ml

16 9

1 1

. 4.2 ;.; I

21 l1

4 2

;:; I

Study

group

total

74

4

2.5

69

8

4.4

Entire

series total

1507

152

2.6

263

40

4.5

7.1

“Number of patients who had at least part of their follow-up time allocated to corresponding categories with regard to GCDFP-15 plasma level. The follow-up of eight patients was divided into two categories of GCDFP-15 levels (see Methods). *Relative risk is the ratio of the number of patients in the group who developed breast cancer to the number of breast cancers expected in a group of women with the same ages and calender years of exposure based on incidence rates of the Connecticut Tumor Registry.

GCDFP- 15 blood levels in breast cystic disease

DISCUSSION Wellings & Alpers* demonstrated that apocrine metaplastic microcysts occuring in the TDLU’s of breast tissue are the precursor lesions which develop into clinical breast gross cystic disease. They also showed that both the prevalence of apocrine metaplasia and its intensity were greater in breast tissue from cancer related breasts than in breasts from age matched normal women.’ Apocrine metaplasia is very rare in breast tissues before the age of 30, then its presence increases dramatically in frequency until the age of 40 and beyond, where it can be found in approximately 50% of breasts.‘,‘3 Apocrine metaplasia does not disappear with the menopause but persists in breast tissue postmenopausally. Breast gross cystic disease, which evolves from apocrine metaplastic microcysts,IX2is a premenopausal disease which abates with menopause. Thus, the formation of breast gross cysts appears to require some form of stimulus in addition to the stimuli causing apocrine metaplasia and this stimulus is probably a premenopausal hormonal stimulus.‘3 The GCDFP-15 protein is one of the major component proteins present in breast gross cystic disease fluid.8,9 This protein has been well characterized with regard to its amino acid structure and the location of its gene on chromosome 7.9.‘5.3hGCDFP-15 is a normal constituent protein of all apocrine gland cells (normal and metaplastic), and of some exocrine glands with apocrine features (serous cells of mandibular salivary glands and the minor salivary glands in the major bronchi).9,‘7 The biological purpose of apocrine cells secreting GCDFP-15 remains obscure. Recent data have shown that the growth rate of certain breast cell lines increases when GCDFP-15 is added to the culture medium.38 The protein does have a binding affinity for fibrinogen and is transported in blood bound to fibrinogen.‘“,‘” However, no enzymatic effect of GCDFP-15 on fibrinogen has been determined and its purpose of binding to fibrinogen has not been established. In order to gain an understanding of the stimulatory factors which may regulate the development of breast GCD, the stimuli which modulate the secretion of GCDFP-15 have been studied.9,‘9,‘6,4M’ Androgens markedly enhance the secretion of GCDFP- 15 from both the T47D and ZR75 breast cancer cell lines.J’W3 The secretory effects of androgens on GCDFP-15 have been shown to be due to increased mRNA synthesis as a primary mechanism.3h,4’ Differences have been observed between the T47D cells and the ZR75 cells with regard to growth effects of various steroids compared to secretion effects on GCDFP-15. In the ZR75 cell line, a positive correlation has been demonstrated between increased GCDFP- 1.5 secretion and decreased growth rates.?’ This has been shown in both directions, with androgens slowing growth while increasing

117

GCDFP- 15 secretion and estrogens enhancing growth while decreasing GCDFP-15 secretion.4’ In contrast, with the T47D cell line, androgens have been shown to enhance GCDFP-15 secretion while not appreciably effecting the growth rate.4”,4’ Progestins slowed the T47D growth rate, while enhancing GCDFP- 15 secretion. However, RU486, an antiprogestin, also slowed growth in the T47D cell line while markedly inhibiting GCDFP- 15 secretion4” Thus, the interaction of various steroids on breast cancer cell lines with regard to GCDFP-15 secretion is complex, but does appear to have androgen modulated enhancement as a central effect. The data presented in this paper indicate that blood levels of GCDFP-15 in women with active breast cystic disease distinguish subgroups with differing degrees of risk for developing breast carcinoma. As shown previously,‘,7 the association between GCD and risk of breast cancer is related to the degree of GCD, as measured by the total number of cysts confimed by aspiraton of cyst fluid. This study provides evidence that the risk of breast cancer in women with active GCD may be further distinguished by blood levels of GCDFP-15. The relative risk of breast cancer was higher among women whose blood levels of GCDFP-15 exceeded 40 rig/ml then it was among women with lower blood levels. In this study, owing to the fact that patients who developed serial breast cysts were more likely to have seen CDH for breast evaluation than patients with a single breast cyst, there was a sample bias which favoured blood samples being obtained from patients who had multiple breast cysts (46.3% of the study group had a history of developing ten or more breast cysts). Thus, the data were analysed for the effect of the history of multiple breast cysts on plasma levels of GCDFP-15. Even though women who had a history of multiple breast cysts were more likely to have an elevated blood level of GCDFP-15 (Table 2) the finding of and elevated blood level of GCDFP- 15 increased the risk of breast cancer development in both women with 1 to 9 and 10 or more breast cysts (Table 3). In the latter group of women with a history of 10 or more breast cysts, the increase in risk was elevated 7.1-fold, and was similar in degree to the increase in risk seen with the occurrence of atypical epithelial hyperplasia,w or with lobular carcinoma in situ.45 Since breast gross cystic disease is clinically much more common than either atypical epithelial hyperplasia or lobular carcinoma in situ,4h it may be that the measurement of GCDFP-15 blood levels will become an important evaluation step in breast cancer risk assessment. We presume that elevated GCDFP-15 blood levels indicate both a hormonal stimulus to the breast tissue and the breast tissue’s overall content of apocrine metaplasia, as well as other associated breast epithelial processes which may be premalignant.

118

The Breast

Much work remains to be accomplished in the study of GCDFP-15 blood levels and breast cancer risk with regard to whether the duration of the elevation is important, and the age at which an elevation in GCDFP-15 blood level occurs, and whether or not elevated blood levels of GCDFP15 in women without known active breast gross cystic disease also reflect a risk for breast cancer development.

Acknowledgements We would like to thank all of the people who have donated money and time to the Haagensen Research Foundation, Inc. Dr P. Holland has kindly made research space available to the Haagensen Research Foundation at the Sacramento Blood Center. Dr Bodian was supported, in part, by grant CA-46470, and by computing resources provided by the City University of New York University Computer Center. She is a member of the Department of Biomathematical Sciences at Mount Sinai School of Medicine, New York City, New York. Dr D. Haagensen, LTC USAR-MC, received IDT credit for this research work from the NAAD. He is a member of the surgical staff at Methodist Hospital and Mercy General Hospital in Sacramento, California.

References 1. Haagensen C D. Diseases of the Breast. 3rd edn. Philadelphia: WB Saunders, 1986: 250-266. 2. Wellings S R, Alpers C E. Apocrine cystic metaplasia: subgross pathology and prevalence in cancer associated versus random autopsy breasts. Hum Path 1987; 18: 381-386. Jones B M, Bradbeer J W. The presentation and progress of macroscopic breast cysts. Brit J Surg 1980; 67: 669-671. Harrington E, Lesnick G. The association between gross cysts of the breast and breast cancer. Breast 1980; 7: 13-17. Ciatto S, Biggeri A, Del Turco M R, Bartoli D, Iossa A. Risk of breast cancer subsequent to proven gross cystic disease. Eur J Cancer 1990; 26: 555-557. 6. Bundred N J, West R R, O’Dowd J, Manse1 R E, Hughes L E. Is there an increased risk of breast cancer in women who have had a breast cvst asoirated? Brit J Cancer 1991: 64: 953-955. 7. Bodian ‘c A, Mattes R, Perzin K H. The epidemiology of gross cystic disease of the breast confirmed by biopsy or by aspiration of cyst fluid. Cancer Det Prev 1992; 16: 7-15. 8. Haagensen D E Jr, Mazoujian G M, Dilley W G, Pedersen C E, Kister S J, Wells S A Jr. Breast gross cystic disease fluid analysis. I. Isolation and radioimmunoassay for a major component protein. J Nat1 Cancer Inst 1979; 62: 239-247. 9. Haagensen D E Jr, Mazoujian G. Biochemistry and immunochemistry of gross cystic disease fluid proteins of the breast. In: Haagensen C D (ed): Diseases of the Breast. 3rd edn. Philadelphia: W B Saunders, 1986; 474-500. 10. Bradlow H L, Rosenfeld R S, Kream J, Fleisher M, O’Connor J, Schwartz M K. Steroid hormone accumulation in human breast cyst fluid. Cancer Res 1981; 41: 105-107. 11. Miller W R, Roberts M M, Creel R J, Yap P L, Kelly R W, Forrest A P M. Androgen conjugates in human breast cyst fluids. J Nat1 Cancer Inst 1982; 69: 1055-1058. 12. Vanluchene E, Vandekerckhove D, De Boever J, Sandra P. Structures and concentrations of fifteen different steroid sulfates in human breast cyst fluids. J Steroid Biochem 1984; 21: 367-371. 13. Belanger A, Caron S, Labrie F, Naldoni C, Dogliotti L, Angeli A. Levels of eighteen non-conjugated and conjugated steroids in human breast cyst fluid: relation to cyst type. Eur J Cancer 1990; 26: 277-281. 14. Pearlman W H, Gueriguian J L, Sawyer M E. A specific progesterone-binding component of human breast cyst fluid. J Biol Chem 1973; 248: 57365741.

of progesterone-binding cyst protein 15. Lea 0 A. Binding properties PBCP. Steroids 1988; 52: 333-338. C. 16. Balbin M, Freije J M P, Fueyo A, Sanchez L M, Lopez-Otin Apolipoprotein D is the major component protein in cyst fluid from women with human breast gross cystic disease. Biochem J 1990; 271: 803-807. 17. Sanchez L M, Diez-Itza I, Vizoso F, Lopez-Otin C. Cholesterol and apolipoprotein D in gross cystic disease of the breast. Clin Chem 1992; 38: 695-698. 18. Miller W R, Scott W N, Harris W H, Wang D. Using biological measurements, can patients with benign breast disease who are at high risk for breast cancer be identified. Cancer Det Prev 1992; 16: 99-106. 19. Kesner L M, Yu W, Bradlow H L, Breed C W, Fleisher M. Proteases in cyst fluids from human gross cyst breast disease. Cancer Res 1988; 48: 63796383. 20. Angeli A, Boccuzzi G, Agrimonti F, Brignardello E, Barbadoro E, Dogliotti L. Correlation between levels of DHEA-SO, and prolactin in human breast cyst fluid. Tumori 1982; 68: 393-396. 21. Smith K, Miller W R, Fennelly J A, Matthews J N S, Scott W N, Harris A L. Quantification of epidermal growth factor in human breast cyst fluids: correlation with DHEA-SO, and electrolyte concentrations. Int J Cancer 1989; 44: 229-232. 22. Takahaski K, Lai L C, Ghatei M A, James V H T, Bloom S R. Endothelin-like immunoreactivity in human breast cyst fluid. J Clin Endocrinol Metab 1990; 71: 1681-1683. 23. Collette J, Hendrick J C, Jaspar J M, Franchimont P. Presence of alpha lactalbumin, epidermal growth factor, epithelial membrane antigen and GCDFP-15 in breast cyst fluid. Cancer Res 1986; 46: 3728-3733. 24. Wang D Y, Hamed H, Fentiman I S. Epidermal growth factor and insulin-like growth factor-I in human breast cyst fluid. Ann N Y Acad Sci 1990; 586: 158-160. 25. Balbin M, Vizoso F, Sanchez L M, Venta R, Ruibal A, Fueyo A, Lopez-Otin C. GCDFP-70 protien in cyst fluid identified as albumin and used to classify cysts in women with breast gross cystic disease. Clin Chem 1991; 37: 547-551. 26. Scott W N, Miller W R. Mutagens in human breast cyst fluid. J Cancer Res Clin Oncol 1991; 117: 254258. 27. Sanchez L M, Vizoso F, Allende M J, Ruibal A, Lopez-Otin C. Quantification and molecular analysis of cathepsin D in breast cyst fluids. Eu J Cancer 1992; 78: 828-832. 28. Lai L C, Ghatei M A, Takahashi K, Pate1 K V, Schrey M P, Ghilchik M W, Bloom S R, James V H T. Mitogenic peptides in breast cyst fluid: relationship with intracystic electrolye ratios. Int J Cancer 1990; 46: 1014-1016. G, Wells S A Jr. Review 29. Haagensen D E Jr, Dilley W G, Mazoujian of GCDFP-15, an apocrine marker protein. Ann N Y Acad Sci 1990; 586: 161-173. 30. Haagensen D E Jr, Wells S A Jr, Haider M. Analysis of human breast gross cystic diease fluid protein (GCDFP-15) by isotope dilution RIA with a solid phase (Kynar) second antibody (abstract). Proc Am Assoc Clin Chem 1980; 26: 980. iodination with a 31. Freker P J, Speck J C. Protein and cell membrane sparingly chloroamine,l,3,4,6 tetra-chloro-3-alpha, 6-alpha, diphenylglycohuil. Biochem Biophys Res Comm 1978; 80: 849-857. 32. Newman E S, Moskie L A, Duggal R N, Goldenberg D M, Hansen H J. Murine monoclonal antibody absorbed onto vinylidene fluoride floccules used to eliminate antibody interference in sandwich type ._ immunoassays. Clin Chem 1989; 35: 1743-1746. J W. Melcher G W. Auchincloss H. Incidence 33. Frantz V K. Pi&en of chronic cystic breast disease in so-called “normal breast”. Cancer 1951; 4: 762-783. 34. Haagensen D E Jr. Is cystic disease related to breast carcinoma? Am J Surg Path 1991; 15: 687-694. 35. Myal Y, Gregory C, Wang H, Hamerton J L, Shiu R P C. The gene for prolactin inducible protein, uniquely expressed in exocrine organs, Maps to Chromosome 7. Somatic and Mol Genetics 1989; 15: 256-270.

GCDFP-15 blood levels in breast cystic disease 36.

37.

38.

39.

40.

41.

Myal Y, Robinson D B, Iwasiow B, Tsuyuki D, Wong P, Shiu R P C. The prolactin inducible protein (PIP/GCDFP-I 5) gene: cloning, structure and regulation. Mol Cell Endocrinol 1991; 80: 165-175. Mazoujian G, Pinkus G S, Davis S, Haagensen D E Jr. Immunohistochemistry of a breast gross cystic disease fluid protein (GCDFP- 15): a marker of aprocrine epithelium and breast carcinomas with apocrine features. Am J Path01 1983; 110: 105-121. Cassone P, Sapino A, Haagensen D E, Naldoni C. Bussolati G. Mitogenic effect of the gross cystic disease fluid protein of 15KD (GCDFP- 15) on breast cancer cell lines and on immortalized mammary cells. Int J Cancer 1995; 60: 216220. Haagensen C D, Bodian C A, Haagensen D E Jr. Breast Carcinoma: Risk and Detection. Philadelphia: W B Saunders, 198 1, 300-338. Chalbos D, Haagensen D E Jr, Parish T, Rochefort H. Identification and androgen regulation of two proteins released by T47D human breast cancer cells. Cancer Res 1987; 47: 2787-2792. Simard J, Hatton A C, Labrie C. Dauvois S, Zaho H F. Haagensen

42.

43.

44. 45. 46.

119

D E Jr, Labrie F. Inhibitory effects of estrogens on GCDFP-15 mRNA levels and secretion in ZR 75-l human breast cancer cells. Mel Endocrinology 1989; 3: 694-702. Dejardin L, Gol-Winkler R, Collette J, Delvenne C, Carlisi I, Haagensen D E, Franchimont P. Stimulatory Effect of estradiol and tamoxifen on gross cystic disease fluid protien 15000 production and mRNA levels in T47D human breast cancer cells. J Mol Endocrinology 1991; 7: 105-I 12. Haagensen D E Jr, Stewart P, Dilley W G, Wells S A Jr. Secretion of breast gross cystic disease fluid proteins by T47D breast cancer cells in culture - modulation by steriod hormones. Br Cancer Res Treat 1992; 23: 77-86. DuPont W D, Page D L. Risk factors for cancer in women with proliferative disease. New Eng J Med 1985; 312: 146-151. Bodian C A. Benign breast diseases, carcinoma in situ and cancer risk: Epidemiological Reviews 1993; 15: 177-187. Bodian C A, Perzin K H, Lattes R, Hoffman P, Abernathy T G. Prognostic significance of benign proliferative breast disease. Cancer 1993; 71: 38963907.