(CANCER RESEARCH 52. 1791-1795. April I. 1992]
Association of Cyst Type with Risk Factors for Breast Cancer and Relapse Rate in
Women with Gross Cystic Disease of the Breast1
Carlo Naldoni, Massimo Costantini, Luigi Dogliotti, Paolo Bruzzi,2 Lauro Succhi, Gianfranco Buzzi, M¡rcllaTorta,
and Alberto Angeli
Cancer Prevention Center, St. Maria delle Croci Hospital, Via Missiroli, 10, 48100 Ravenna [C. N., L. B., G. B.]; Unit of Clinical Epidemiology and Trials, National
Cancer Institute, Viale Benedetto XV 10, 16132, Genoa [M. C., P. B.]; and Department of Clinical and Biological Sciences, University of Turin at St. Luigi Hospital,
10043, Orbassano (Torino), Italy [L. D., M. T., A. A.].
ABSTRACT
The concentration of potassium (K' ) and sodium (Na+) was measured
in breast cyst fluid (BCF) from 611 cysts >3 ml aspirated in 520 women
with gross cystic disease of the breast. These women were enrolled, from
1983 on, in a cohort study aimed at assessing the relationship between
cyst type, as defined by the K*/Na+ ratio in BCF, and the risk of breast
cancer. The inverse relationship between k* and Nu* and the bimodal
distribution of the K+/Na+ ratio in BCF were confirmed. Type I cysts
were defined as cysts with a K* Vi* >1.5 in BCF. Among women with
type I cysts, a higher proportion of women with one or no births, of
women with a history of apocrine cysts, of current smokers, and of women
who do not drink coffee was found, as compared to women with other
types of cysts. The risk of cyst relapse was significantly higher among
women with type I cysts than among women with other types of cysts
and among women with multiple cysts at presentation. These findings
indicate that type I BCF is a marker of "active" gross cystic disease of
the breast and suggest that it may be associated with increased breast
cancer risk.
INTRODUCTION
GCD1 is a benign condition of the breast which is reported
to affect 7% of women in western countries (1) and its incidence
is highest in the premenopausal decade (2). Although cysts are
not considered premalignant lesions, several epidemiological
studies indicate that women with GCD have an increased risk
of subsequent breast cancer (3-5). Conversely, no association
was found in several other studies, including the large retro
spective cohort study of Dupont and Page based on biopsy
specimens (6, 7).
In recent years, novel approaches to this issue are being
offered by studies on the composition of the aspirated BCF (8).
Studies on the biochemical composition of this medium con
sistently indicate that the concentration in BCF of most sub
stances shows a bimodal distribution. These include several
cations such as Na+, K* and Cl~ (9-12), hormones (12-16),
and growth factors (17-20). The consistency of these results
has led to identification of two types of cysts. Type I cysts,
often referred to as secretory (apocrine) cysts, have a BCF
electrolyte composition with high levels of potassium ions, low
levels of Na+ and Cl~, and large concentrations of androgen
conjugates, epidermal growth factor, and gross cystic disease
fluid protein; while type II cysts, also referred to as transudative
(flattened) cysts, have an electrolyte composition quite similar
to that found in plasma (high Na+ levels) and lower concentra
tions of androgen conjugates, epidermal growth factor, and
Received 8/1/91; accepted 1/24/92.
The costs of publication of this article were defrayed in part by the payment
of page charges. This article must therefore be hereby marked advertisement in
accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1This work is supported by a grant from the Istituto Oncologico Romagnolo,
Forli, Italy, Contract 90133.1, and by a grant of the National Research Council,
Special Project Oncology. Contract 880051.44.
2To whom requests for reprints should be addressed.
' The abbreviations used are: GCD, gross cystic disease; BCF, breast cyst fluid.
gross cystic disease fluid protein (7, 8, 21-23). A third group
of cysts with intermediate features is often identified. Prelimi
nary evidence suggests that women bearing type I cysts have an
increased breast cancer risk (24). However, the overall evidence
is circumstantial and based on studies suffering from major
limitations (25-27).
In 1983 a cohort study was undertaken to evaluate the rela
tionship between the risk of breast cancer and the biochemical
characteristics of BCF in women affected by GCD. Preliminary
results of this study are presented here, focused on: (a) classi
fication of breast cysts; (b) factors associated with cyst type; (c)
cyst relapse by cyst type at presentation.
PATIENTS
AND METHODS
Patients. Between February 15, 1983, and March 31, 1990, 968
women ages 30-69 years underwent aspiration of breast cysts at the
Cancer Prevention Center of Ravenna. Forty-nine of these women did
not meet eligibility criteria: 27 women had a history of cancer of any
site including breast, 8 women had a breast cancer diagnosed at the
initial examination, and 14 women were not living in the area. In
addition, 91 women were excluded because their BCF samples were
lost. The present report is focused on 520 patients with BCF >3 ml,
since, according to the original protocol, electrolyte concentrations
were not measured in 308 patients with BCF <3 ml. Six hundred eleven
samples of breast cyst fluid were obtained through fine-needle aspiration
from these 520 women; 194 were aspirated on the same day from 87
patients (16.7%) bearing multiple cysts in the same or in the contralateral breast. All women were interviewed by trained personnel using a
structured questionnaire focused on known or suspected risk factors
for breast cancer. The follow-up program includes physical examination
and ultrasound scan every year and mammography every other year.
Thus far, the proportion of women lost to follow-up has been very low
(<1%). In all samples Na* and K* were determined by an ion-selective
electrode technique.
Statistical Methods. To assess whether agreement in cyst type among
multiple cysts aspirated simultaneously or relapsed was in excess of
that expected by chance alone, the Kstatistic, a measure of the frequency
of agreement, was used (28). Cyst type distributions in the subgroups
of each variable were compared using the Mantel-Haenszel test (x2 test
for trend). For categorical variables, such as age, menopausa! status,
and previous diagnosis of BBD, differences in the cyst type distribution
were tested by means of the ordinary x2 test for heterogeneity. The
probability of type I cysts was then modeled as a function of all variables
under study in a multivariate logistic model by means of a step-down
procedure starting from the model with all variables included (29).
Relapse rates were studied as time from enrollment to first relapse,
using usual life table techniques, i.e., the Kaplan-Meyer product limit
estimator and the log-rank test. At enrollment, all palpable cysts were
aspirated. A relapse was defined as any cyst with BCF >3 ml aspirated
in either breast on any examination following enrollment. Relapse rates
in different subgroups identified according to K*/Na" ratio in BCF
were compared as follows. First, the relapse rate in the entire group of
women with solitary cysts was computed by dividing the number of
relapses observed in the whole cohort for the total follow-up experi
enced by the cohort from enrollment through the date of first relapse
or through March 31, 1990. The expected number of relapses in each
1791
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1992 American Association for Cancer Research.
CYST TYPE AND BREAST CANCER RISK FACTORS IN CYSTIC DISEASE
subgroup was computed by multiplying this rate for the person years
of follow-up experienced in each subgroup. Ratios of the observed to
the expected number of relapses in each subgroup were then compared.
RESULTS
Electrolytes in BCF and Classification of Cysts. The K+, Na+,
KYNa* ratio distribution in BCF showed the typical bimodal
distribution. Na+ appeared inversely correlated with K+ (r =
0.98, P < 0.01). The scatter plot of log K+ versus log Na+ of
BCF from solitary cysts (Fig. 1) suggests that cyst fluids can be
naturally separated into three subgroups: type I cysts with high
K* (KVNa* > 1.5); type II cysts with high Na+ (K+/Na+ <
0.25); and type III cysts (K+/Na+ between 0.25 and 1.5), as an
intermediate group.
Cyst Type and Number of Cysts. The distribution of cysts by
type and number of cysts per patient is reported in Table 1: 363
cysts (57.9%), type I; 154 cysts (24.6%), type II; and 110 cysts
(17.5%), type III. Eighty-seven patients (16.7%) presented ini
tially with multiple cysts. The proportion of type I cysts signif
icantly increases with the increasing number of cysts per patient
(x: for trend = 11.52, P< 0.001).
Agreement in Cyst Type. Agreement in cyst type among
multiple cysts aspirated simultaneously did not significantly
exceed that expected by chance alone (Table 2). Through March
31, 1990, 113 women relapsed, 61 with 1 cyst, 25 with 2 cysts,
and 27 with 3 or more cysts. A significantly greater concordance
than expected (P < 0.05) was observed among the 61 patients
with 1 relapsed cyst (observed = 36, expected = 28.18, «=
0.24, P < 0.05), but not among the 25 women with 2 relapses.
Cyst Type and Risk Factors for Breast Cancer. The univariate
association between type I cysts and known or suspected risk
factors for breast cancer in patients with single cyst at enroll
ment is shown in Table 3. Women with type I cysts, when
compared to women with other types of cysts, more frequently
reported a history of apocrine cysts but not of other types of
benign breast diseases (P = 0.018, test for heterogeneity). An
inverse correlation between coffee consumption and probability
Table 2 Agreement in cyst type among multiple cysts aspirated simultaneously
(A) or relapsed (B) in the same patient
"Observed" is the number of patients with all cysts of the same type; "expected"
is the number expected by chance alone based on marginal totals.
No. of patients with all cysts of the
same type
A. No. of patients with simulta
neous cysts at enrollment
(no. of cysts)
70(2)
15(3)
B. No. of patients with cyst re
lapse (no. of cysts at relapse)
61 (1)
25(2)
"/XO.OS.
Observed
Expected
35
6
33.69
5.50
0.04
0.11
36
9
28.18
6.56
0.24"
0.15
of type I cysts relative to other types of cysts was observed (P
= 0.036, test for trend). Type I cysts decrease in relative
frequency with increasing number of births (P = 0.016, test for
trend). No association between cyst type and age at first birth
was observed. The multivariate analysis (Table 4) confirms the
association of type I cysts with low parity, history of apocrine
cysts, and coffee consumption.
Relapses. One hundred thirteen women (26.1%) with solitary
cysts at enrollment and 34 women (39.1%) with multiple cysts
developed 1 or more cysts in the follow-up period. Observed
and expected numbers of relapses in various subgroups of
patients with solitary cysts, defined on the basis of K+/Na+
ratio in BCF, are shown in Table 5. A 2-fold increase in the
relapse rate is observed when all subgroups with K+/Na+ in
BCF above the cutoff value of 1.5 are pooled and compared
with women with Na+/K+ in BCF <1.5. However, no associa
tion between K+/Na+ ratio and relapse rate is observed within
the subgroups below 1.5 and within the subgroups above 1.5.
The highest relapse rate was observed among women with
multiple cysts (P < 0.05) (Fig. 2).
DISCUSSION
2
Fig. 1. Distribution of concentration of log K and log Na in breast cyst fluids
of 433 solitary' cysts.
Table 1 Distribution of 627 cysts by type and number of cysts per patient
X2= 11.52 (P = 0.0007). test for trend in the porportion of type I cysts.
ICysts/patient
Type
Despite its frequency and its supposed association with an
increased breast cancer risk (3-7), cystic disease of the breast
is ill characterized both epidemiologically and clinically (3031). This study describes one of the largest series of consecutive
patients with gross cystic disease. Virtually all eligible patients
were prospectively enrolled in the study. Furthermore, the
Cancer Prevention Center of Ravenna sees the overwhelming
majority of breast diseases in the area (32). Therefore, this
series, unlike many others, can be considered highly represent
ative of cases of GCD in the area and free of any kind of
selection bias. It must be emphasized, however, that the present
study is limited to women with large cysts (>3 cm in diameter).
Therefore, generalization to other types of cystic disease of the
breast is unwarranted.
The bimodal distribution of cation concentration in breast
cyst fluids is confirmed, according to previous reports (10, 12,
17). Various cutoff values in K*/Na+ ratio have been empirically
IIIn%812081110N%18.714.317.825.017.5N4331404545627%100100100100100100
IIn1212751154%27.919.311.120.024.6Type
proposed to subdivide cysts in two
%12345Total23193323436353.466.471.175.080.057.9Type
n
subgroups (Table 6): those
with high levels of K* and low levels of Na+ (type I); and those
with high levels of Na* and low levels of K+ (type II). A third
group (type III cysts), representing a group with intermediate
KVNa* ratio, has been often defined in order to overcome the
difficulty of identifying a single cutoff value. The scatter plot
of log K+ versus log Na+ in Table 1 suggests that values of the
1792
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1992 American Association for Cancer Research.
CYST TYPE AND BREAST CANCER RISK FACTORS IN CYSTIC DISEASE
Table 3 Univariate association between cysts type (type I versus type II and III)
and various risk factors among patients with single cyst at enrollment
,No. cysts
Typen-,,
VariablesAge
%35
%
No.
(yr)
<40
41-45
46-50
>50No.69
50.7
34
49.3
82
62.6
49
37.4
131
82
49.7
83
50.3
165
47.1
36
52.9
68Type,32
P = 0.368 (test for heteroge
neity)
Menopausa! status
Premenopausal
Perimenopausal
Hysterectomy"
Menopausa! age <50 yr
Menopausa! age >50 yr
360
16
24
18
15
Menarche
<11
12
13
>13
105
85
115
127
Abortions
No
1
291
101
41
Quetelet
I
II
III
IV
107
108
110
108
Oral contraceptives (total mos. of
use)
No
<12
12-36
>36
345
44
27
16
Smoking habit
No
Ex-smokers
Smokers <6 cigarettes/day
Smokers 6-10 cigarettes/day
Smokers >10 cigarettes/day
249
31
25
49
72
199
5
12
7
8
44.7
55.3
161
68.7
31.3
11
50.0
12 50.0
38.9
61.1
7
46.7
53.3
8
=0.258 (test for heteroge
neity)
47.6
50
56.5
48
60.9
70
49.6
63
P = 0.684 (test
55
52.4
37
43.5
45
39.1
64
50.4
for trend)
160
55.0
131
45.0
54
53.5
47
46.5
17
41.5
24
58.5
P = 0.097 (test for trend)
64
59.8
53
49.1
61
55.5
53
49.1
P = 0.232 (test
43
40.2
55
50.9
49
44.5
55
50.9
for trend)
178
51.6
167
48.4
27
60.0
17
40.0
17
63.0
10
37.0
9
56.3
7
43.7
P = 0.205 (test for trend)
125
16
18
29
39
48.8
50.2
124
48.4
51.6
15
72.0
28.0
7
40.8
59.2
20
54.2
45.8
33
P = 0.226 (test for trend)
CoffeeNoYes,
cups/dayYes,
<3
>3 cups/day10128043621441961.4
trend)AlcoholNoYesNo.
3951.4
13644.2
2438.648.655.8P
= 0.036 (test for
Table 3 Continued
TyPceyNo.
I cysts
its%47.238.551.028.646.461.144.450.0P=
No.21516129553075552.8
%
VariablesFamily
cancerNoYesPrevious
history of breast
19261.5
10P
0.50949.0
=
breastdiseaseNoGCD,
diagnosis of benign
apocrineGCD,
apocrineFibrocystic
not
diseaseFibroadenomaNodulectomy
n.o.s."No.40726263775618910Type
13471.4
2253.6
2638.9
155.6
450.0
1
5H-III
0.018 (testheteroge-neity)
for
' Presumably without ovariectomy.
*n.o.s., not otherwise specified.
Table 4 Multivariate association between cyst type (type I versus type II and III)
and known or suspected risk factors for breast cancer
Menopausa! status, number of previous abortions, age at menarche, age at first
birth. Quetelet index, family history of breast cancer, alcohol consumption, and
oral contraceptive use, included in the initial model, were deleted from the
regression equation as nonsignificantly (/" > 0.1) contributing to its likelihood
using a step-down procedure.
OR"
Variables
Age
<40
41-45
46-50
>50
Ref.
1.82
1.13
0.94
0.0854 (3 d.f.)
No. of births
0
1
2
>2
Ref.
0.61
0.40
0.34
0.0067(1 d.f.)
Previous benign breast disease
No
Apocrine GCD
Not apocrine GCD
Others
2.78
1.26
0.80
0.0058 (3 d.f.)
Smoking habit
No
Exsmokers
Smokers
Ref.
1.04
1.45
0.0756(1 d.f.)
Ref.
0.62
0.50
0.0187(1 d.f.)
Coffee consumption
No
Yes, <3 cups/day
Yes, >3 cups/day
°OR, odds ratio.
Ref.
K+/Na+ ratio of 1.5 and 0.25 represent the natural cutoff values
152.7
11
78P
0.77164.8
=
births012>22451655413218958134873576932754.7
of
trend)Age
1957.6
5649.2
9646.6
3145.347.335.242.450.853.4P
= 0.016 (test for
birthNulliparous15-2021-2526-30>305466176112253537856212P
at 1st
1956.1
2948.3
9155.4
5048.0
1335.243.951.744.652.00.458
=64.8 (test for heteroge
neity)
for identifying these three subgroups of cysts. It remains to be
determined whether type III cysts are more related to type I or
II cysts in their behavior. In our series, the risk of further cyst
development according to the subgroups identified from various
authors (Table 5) clearly identifies a K+/Na+ ratio of 1.5 as the
appropriate cutoff value between cysts at low and high risk of
relapse. The high relapse rate observed for cysts with high K+/
Na+ ratio is consistent with the results of other studies (33-36)
while the intermediate group of cysts, behaves, at least from
this point of view, like cysts with a low K*/Na+ ratio (type II).
A significant agreement in cyst type, above that expected by
chance alone, was not observed among multiple cysts aspirated
in the same patient. The frequency of exact agreement in cyst
type between solitary cysts at enrollment and relapsed cysts was
slightly greater than expected. These findings partly contrast
1793
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1992 American Association for Cancer Research.
CYST TYPE AND BREAST CANCER RISK FACTORS IN CYSTIC DISEASE
Table 5 Observed and expected number of relapses by K*/Na* ratio in BCF of
the index cyst among 433 patients with solitary cyst at presentation
et al.) (34, 37) is sufficient to identify a patient "at risk"
independently of the simultaneous presence of other cysts of
K*/Na* in
of
different type. The studies of Ebbs and Bates (35) and Molina
patients972438202326617371202231RelapsesObserved17582192230193380Expected"27.975.7312.244.815.946.6314.3917.6517.6456.6956.31Observed/expected0.610.870.650
BCF<0.100.10-0.240.25-0.660.67-1.001.01-1.501.51-2.002.01-3.003.01-4.00>4.00<1.50>1.50No.
et al. (22) indicated a poor stability in cyst type from multiple
cysts aspirated simultaneously and at the following attendance.
On the other hand, in our study, the proportion of type I cysts
increases with increasing number of aspirated cysts and the
probability of type I cyst is increased in women with a previous
history of apocrine GCD. This observation lends support to the
hypothesis that a type I BCF is a marker of "active" GCD.
Little is known of the epidemiológica!determinants of benign
breast disease in general (30, 38) and specifically of GCD. The
interpretation of our findings is hampered by the lack of a
(relative risk = 2.45, P< 0.0001)
control group. The comparison of the distribution of several
Total
433
113
factors among women presenting with solitary cysts of different
' Expected number are computed using the relapse rate in the whole group.
types provides interesting clues and is free from selection and
recall biases frequently affecting case-control studies of chronic
conditions. However, due to the lack of a control group, any
association observed in this study between a risk factor and the
0.9 presence of type I cyst is equally consistent with a positive
0.8association between that factor and the incidence of type I cysts
and with a negative association between the same factor and
0.7the incidence of cysts of different types, and vice versa. In our
06study, the relative proportion of type I cysts decreased with
0.5increasing number of births. The association between reproduc
tive history and breast cancer risk is well established, the risk
0.4being inversely related with the number of births and directly
03related to age at first birth (39). In our data, no association
with age at first birth was present. Women with type I cysts
0?
more frequently referred a previous diagnosis of apocrine GCD.
0I
A significant association was present with coffee consumption
and, with borderline significance, with smoking habits. The
i.:;
th
M
association between smoking habits or coffee consumption and
the breast cancer risk has been investigated with controversial
Fig. 2. Relapse-free survival among 520 patients by number of cysts at pres
results (40-41). A possible role of methylxanthines in benign
entation and, among patients with solitary cysts, by type of cyst. OBS, observed;
breast disease has been suggested (42). As a consequence, these
EXP, expected.
findings must be considered cautiously.
Table 6 Cutoff values ofK*/Na* ratio used by various authors to identify cysts
The most interesting result of this study is the demonstration
that cysts with a BCF K+/Na+ ratio above 1.5 are associated
subgroups
valuesTypeil<0.25<0.33<0.66<0.33<0.2K+
Ratio cutoff
with a 2-fold increased risk of relapse. The observation that
when women with cysts below and above the cutoff of 1.5 are
111IntermediategroupIntermediategroupIntermediategroupIntermediategroupIntermediategroupIntermediategroup°
I>1>0.33>1.5>0.33>3K*
Ref.Miller,
evaluated separately, no relationship exists between the K+/
1983(10)Dixon.rto/.,
et al.,
Na+ ratio and the risk of relapse confirms the existence of two
1983(37)Dogliotti,
distinct types of cysts. As could be expected on merely statistical
1986(11)Boccardo,
et al.,
grounds, women with multiple cysts at presentation show the
1988(17)Angeli,
et al.,
highest rate of relapse.
1990(8)Bradlow,
et al.,
The major limitation of our study derives from the fact that
Na*and
<20
Na*or
>75 and
1983(21)Vizoso,
et al.,
only cysts >3 ml were considered. This limitation accounts for
<25">2>1.5'
CICr>75"<0.1<0.25Type
the lower relapse rate as compared to previous reports. How
1990(12)Present
et al.,
ever, the use of clearly established and easily reproducible
selection criteria such as this one prevents biases in the assess
reportKVNa'Type
ment of relapses.
mmol/liter.
In conclusion, the results of this study confirm that two
distinct types of breast cysts can be identified on the basis of
with those reported by Miller et al. (10) and Dogliotti et al. BCF composition and indicate that these two groups differ in
(11) and challenge the hypothesis that it is possible to classify their epidemiological determinants and clinical behavior. The
patients according to cyst type. It should be noted that this striking observation that, thus far, 9 of 9 incident cases of breast
hypothesis provides the basis for all cohort studies (including cancer (one in situ and 8 invasivecancers) were observed in this
the present one) of women with breast cysts classified according cohort among women with type I cyst at enrollment is in
to BCF composition. Alternatively, one may speculate that the agreement with incidental findings from previous studies (24)
presence of type I cysts (or apocrine cysts as classified by Dixon but requires confirmation after a longer follow-up.
1794
SOLITARY-TYPE
SOLITARY-TYPE
MULTIPLE
P VALUE <0.0001
11-11
I
(T.lt
fi
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1992 American Association for Cancer Research.
CYST TYPE AND BREAST CANCER RISK FACTORS IN CYSTIC DISEASE
ACKNOWLEDGMENTS
The following researchers and institutions took part in the study:
Dr. L. Tavolazzi (Laboratorio Analisi, Ospedale S. M. delle Croci,
USL 35, Ravenna); Dr. A. Mazzetti (Laboratorio RIA, Ospedale S. M.
delle Croci); Ing. L. Gogioso (Servizio di Biostatistica Sperimentale ed
Elaborazione Dati, Istituto Nazionale per la Ricerca sul Cancro,
Genoa).
REFERENCES
1. Haagensen, C. D., Bodian, C., and Haagensen, D. E., Jr. (eds.). Breast
Carcinoma—Risk and Detection, pp. 55-80. London: W. B. Saunders Co.,
1981.
2. Skidmore, F. D. The epidemiology of breast cyst disease in two British
populations and the incidence of breast cancer in these groups. Ann. NY
Acad. Sci., 586: 276-287, 1990.
3. Haagensen, C. D. Disease of the Breast, Ed. 3, Chap. 16. Philadelphia: W.
B. Saunders Co., 1986.
4. Ciatto, S., Biggeri, A., Rosselli Del Turco, M., Bartoli, D., and lossa, A.
Risk of breast cancer subsequent to proven gross cystic disease. Eur. J.
Cancer, 26:555-557, 1990.
5. Davis, H. II., Simons, M., and Davis, J. B. Cystic disease of the breast:
relationship to carcinoma. Cancer (Phila.), 17: 957-958. 1964.
6. Hutchinson, W. B., Thomas, D. B., Hamlin, W. B., Roth, G. J., Peterson,
A. V., and Williams, B. Risk of breast cancer in women with benign breast
disease. J. Nati. Cancer Inst., 65: 13-20, 1980.
7. Dupont, W. D., and Page, D. L. Risk factors for breast cancer in women
with proliferate breast disease. N. Engl. J. Med., 312: 146-151, 1985.
8. Angeli, A., Bradlow, H. L., Bodian, C. A.. Chasalow, F. I., Dogliotti, L., and
Haagensen, D. E., Jr. Criteria for classifying breast cyst fluids. Ann. NY
Acad. Sci., 586:49-52, 1990.
9. Bradlow, H. L., Skidmore, F. D., Schwartz, M. K., and Fleisher, M. Cation
levels in human breast cyst fluid. Clin. Oncol., 7: 338-390, 1981.
10. Miller, W. R., Dixon, J. M., Scott, W. N., and Forrest, A. P. M. Classification
of human breast cysts according to electrolyte and androgen conjugate
composition. Clin. Oncol.. 9: 227-232. 1983.
11. Dogliotti, L., Orlandi, F., Torta, M., Buzzi, G., Naldoni, C., Mazzetti, A.,
and Angeli, A. Cations and dehydroepiandrosterone-sulfate
in cyst fluid of
pré-and menopausal patients with gross cystic disease of breast. Evidence
for the existence of subpopulations of cysts. Eur. J. Cancer & Clin. Oncol.,
22:1301-1307, 1986.
12. Vizoso, F., Fueyo, A., Allende, M. T., Fernandez, J., Garcia-Moran, M.. and
Ruibal. A. Evaluation of human breast cysts according to their biochemical
and hormonal composition, and cytologie examination. Eur. J. Surg. Oncol.,
16: 209-214, 1990.
13. Miller, W. R., Scott, W. N., Kelly, R. W., and Hawkins, R. A. Steroid
hormones in breast cyst fluids. Ann. NY Acad. Sci., 586: 60-69, 1990.
14. Bradlow, H. L., Rosenfeld, R. S., Kream. J., Fleisher, M., O'Connor, J., and
Schwartz, M. K. Steroid hormone accumulation in human breast cyst fluid.
Cancer Res., 41: 105-107, 1981.
15. Orlandi, F., Caraci. P., Puligheddu, B., Torta, M., Dogliotti, L., Del Monte,
I., and Angeli, A. Relationship to cation-related cyst subpopulations. Ann.
NY Acad. Sci., 586: 79-82, 1990.
16. Castagnetta, L., Granata, O. M., Brignone, G., Blasi, L., Arcuri, F., Mesiti,
M., D'Aquino. A., and Preitano, W. Steroid patterns of benign breast disease.
Ann. NY Acad. Sci., 586: 121-136, 1990.
17. Boccardo, F., Valenti, G., Zanardi, S., Cerniti, G., Passio, T., Bruzzi, P., De
Franchis, V., Barreca, A„Del Monte, P., and Minuto, F. Epidermal growth
factor in breast cyst fluid: relationship with intracystic cation and androgen
conjugate content. Cancer Res., 45: 5860-5863, 1988.
18. Collette, J., Van Cauwenberge, J-R., Dejardin, L., Cariisi, A., Jaspar, J-M.,
and Franchimont. P. Epidermal growth factor and prolactin in human breast
cyst fluid. Ann. NY Acad. Sci., 586: 146-157, 1990.
19. Lai, C. L., Dunkley, S. A., Reed, M. J., Ghilchik, M. W., Shaikh, N. A., and
James. V. H. T. Epidermal growth factor and oestradiol in human breast
cyst fluid. Eur. J. Cancer., 26: 481-484, 1990.
20. Wang, D. Y., Hamed, H., and Fentiman, I. S. Epidermal growth factor and
insulinlike growth factor-I in human breast cyst fluid. Ann. NY Acad. Sci.,
586: 158-160, 1990.
21. Bradlow, H. L., Skidmore, F. D., Schwartz, M. K., Fleisher, M., and
Schwartz, D. Cations in breast cyst fluid. In: A. Angeli, H. L. Bradlow, and
L. Dogliotti (eds.), Endocrinology of Cystic Breast Disease, pp. 197-201.
New York: Raven Press, 1983.
22. Molina, R., Fuella, X., Herranz, M., Prats, M., Velasco, A., Zanon, G.,
Martinez-Osaba, M. J., and Ballesta, A. Biochemistry of cyst fluid in libro
cystic disease of the breast. Ann. NY Acad. Sci., 586: 29-42, 1990.
23. Haagensen. D. E., Jr., Mazoujian, G., Dilley, W. G., Pederson, C. E., Kister,
S. J., and Wells. S. A. Breast gross cystic disease fluid analysis. Isolation and
radioimmunoassay for a major component protein. J. Nati. Cancer Inst., 62:
239-247, 1979.
24. Dixon, J. M., Lumsden, A. B., and Miller, W. R. The relationship of cyst
type to risk factors for breast cancer and the subsequent development of
breast cancer in patients with breast cystic disease. Eur. J. Cancer & Clin.
Oncol., 21: 1047-1050, 1985.
25. Bodian, C. A. Some limitations on studies about the relationship between
gross cystic disease and risk of subsequent breast cancer. Ann. NY Acad.
Sci., 586: 259-265, 1990.
26. Bruzzi, P., Bucchi, L., Costantini, M., and Naldoni, C. Follow-up studies of
patients with categorized breast cysts. Ann. NY Acad. Sci., 586:43-48, 1990.
27. Page, D. L., and Dupont, W. D. Are breast cysts a premalignant marker?
Eur. J. Cancer & Clin. Oncol., 22:635-636, 1986.
28. Fleiss, J. L. Statistical Methods for Rales and Proportion, Ed. 2, pp. 212236. New York: John Wiley and Sons, 1981.
29. Halperin, M., Blackwelder, W. C., and Verter, J. Estimation of the multivariate logistic risk function: a comparison of the discriminant function and
maximum likelihood approaches. J. Chronic Dis., 24: 125-131, 1971.
30. Ernster, V. The epidemiology of benign breast disease. Epidemiol. Rev., 3:
184-202, 1981.
31. Hughes, L., Mansel, R., and Webster, D. J. T. Aberrations of normal
development and involution (ANDI): a new perspective of pathogenesis and
nomenclature of benign breast disorders. Lancet, 2: 1316-1319, 1987.
32. IARC. Cancer Incidence in Five Continents, Vol 6. Lyon, France: Interna
tional Agency for Research on Cancer, in press, 1992.
33. Jones, B. M., and Bradbeer, J. W. The presentation and progress of macro
scopic breast cysts. Br. J. Surg., 67: 669-671, 1980.
34. Dixon, J. M., Scott, W. N., and Miller, W. R. Natural history of cystic
disease: the importance of cyst type. Br. J. Surg., 72:190-192, 1985.
35. Ebbs, S. R., and Bates, T. Breast cyst type does not predict the natural history
of cyst disease of breast cancer risk. Br. J. Surg., 75: 702-704, 1988.
36. Mansel, R. E., Harrison, B. J., Melhuish, J., Sheridan, W., Pye, J. K.,
Pritchard, G., Maddox, P. R., Webster, D. J. T., and Hughes, L. E. A
randomized trial of dietary intervention with essential fatty acids in patients
with categorized cysts. Ann. NY Acad. Sci., 586: 288-294. 1990.
37. Dixon, J. M., Miller, W. R., Scott, W. N., and Forrest, A. P. M. The
morphological basis of human breast cyst population. Br. J. Surg., 70: 604606, 1983.
38. Brinton, L. A., Vessey, M. P., Flavel, R., and Yeates, D. Risk factors for
benign breast disease. Am. J. Epidemiol., 3: 203-214, 1981.
39. Negri, E., La Vecchia, C., Bruzzi, P., Dardanoni, G., Decarli, A., Palli, D.,
Parazzini, F., and Rosselli Del Turco, M. Risk factors for breast cancer:
pooled results from three Italian case-control studies. Am. J. Epidemiol.,
128: 1207-1215, 1988.
40. Baron, J. A. Smoking and oestrogen related diseases. Am. J. Epidemiol.,
119:9-22, 1984.
41. Brinton, L. A. Relationship of benign breast disease to breast cancer. Ann.
NY Acad. Sci., 586: 266-270, 1990.
42. Wanh, D. Y., and Fentiman, I. S. Epidemiology and endocrinology of benign
breast disease. Breast Cancer Res. Treat.. 6: 5-36, 1985.
1795
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1992 American Association for Cancer Research.
Association of Cyst Type with Risk Factors for Breast Cancer
and Relapse Rate in Women with Gross Cystic Disease of the
Breast
Carlo Naldoni, Massimo Costantini, Luigi Dogliotti, et al.
Cancer Res 1992;52:1791-1795.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/52/7/1791
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1992 American Association for Cancer Research.