Self-Selection for Mammography and Breast Cancer Incidence by Stage

Aims and Background

In the field of breast cancer control, obtaining population-based data on spontaneous mammography (MG) screening should be a priority. This study focuses on the breast cancer incidence and stage distribution in relation to MG use in Ravenna (Italy), 1987-88.

Methods

We estimated the MG rates, expected (E) incidence based on mortality data, observed (O) incidence, predicted excess incidence based on MG rates, observed excess incidence, and stage distribution according to the MG history.

Results

The highest MG rate (37%) was found among residents aged 40-44 but none of these had a T1a-bN0 breast cancer diagnosed nor was the predicted excess incidence demonstrated. Between 45 and 64 years, 80% of self-selected screenees had repeat (“incidence”) MG and the O:E incidence ratio was 1.32 (95% Cl 1.09-1.58). The observed excess incidence was 3.8-fold (95% Cl 2.56-5.16) greater than that predicted. Advanced (T2+ and/or N1+) cases accounted for 42% of patients diagnosed within 3 years of their last MG, for 55% of those diagnosed more than 3 years after their last MG, and for 70% of those with no previous MG. Above age 70, a significant worsening of stage at diagnosis was associated with a clear-cut drop in the proportion of breast cancer patients with previous self-referral for MG.

Conclusions

The results indicate that self-selection and its implications are major features of spontaneous screening practice.

Effects of the Bethesda System on the Rate of Unsatisfactory Pap Smears in Spontaneous Cervical Screening

Aims

In 1990, The Bethesda System (TBS) was introduced into spontaneous cervical screening practice in Ravenna, Italy. Negative/benign reports with the recommendation for early repeat smears (RERS) due to some limitation in sample adequacy were considered no longer acceptable. A monitoring program for the rate of unsatisfactory smears (UNS) was implemented. The aim of the present study was to evaluate the effects of such changes in the screening procedure.

Methods

The frequency of UNS in 1990 was compared with that of UNS+RERS in 1988 (assumed as a baseline year) by the calculation of the standardized rate ratio with the 95% confidence interval (CI). The trend in the standardized rate of UNS from 1990 to 1994 was evaluated by the calculation of the average annual variation with the 95% CI.

Results

The immediate effect of TBS (1990:1988 comparison) was a significant increase in the rate of UNS attributable to scant cellularity, poor fixation and thick areas (rate ratio, 2.35; 95% CI, 2.18 to 2.53) and to the absence of endocervical component (1.45; 95% CI, 1.30 to 1.60). The rate of UNS attributable to the presence of cytolysis, inflammation, blood and foreign material decreased by about 6 times (0.16; 95% CI, 0.13 to 0.19). The midterm effect of TBS (trend from 1990 to 1994) was a decrease in the total rate of UNS by an average of 2.3% per year. The downward trend was significant for smears showing scant cellularity, poor fixation and thick areas (-1.5% per year) and the absence of endocervical component (-0.7% per year). UNS attributable to the presence of cytolysis, inflammation, blood and foreign material stabilized.

Conclusions

TBS led to a substantial change in the type of information provided by the cytology report (Immediate effect). The monitoring program according to TBS led to a reduction in UNS attributable to sample taker (midterm effect).

A First Survey of Organized Cervical Cancer Screening Programs in Italy

In Italy, where no national screening program for cervical cancer exists, organized programs have developed on a local basis. We performed the first survey of existing organized programs by mailing a standard questionnaire to a large network of possibly involved services. For the present survey, a program was defined as organized if personal invitations were sent. We identified 29 already active organized programs and 4 others in a starting phase. The target population of active programs included 2,074,820 women in the age range 25-64 years, corresponding to 13.5% of the Italian female population of the same age. The situation is rapidly evolving since many regional programs are being implemented. Most programs followed Italian and European recommendations as regards the age limits, interval between screening rounds, presence of a fail-safe system for women referred for colposcopy, presence of protocols for diagnostic workup and treatment, and presence of referral centers for such phases. However, many programs did not meet national guidelines as regards the size of laboratories interpreting smears, which were frequently small. Second-level referral centers also frequently had a very small activity. The average (weighted for size of the invited population) compliance to invitation and coverage (proportion of women with at least one test in the last 3 years) was 32.6% and 66.0%, respectively, therefore needing to be improved. Compliance to colposcopy (weighted for number of referred women) was 81.4%. We found a very high variability in the proportion of women referred for colposcopy that could only be partly explained by different referral protocols and could depend on different criteria of smear interpretation: the average (weighted for number of tested women) was 2.01%. A need for improvement in the process of evaluation and for homogenization of criteria of cytology interpretation was identified: work in this regard is on-going.

"CADEAT": considerations on the use of CAD (computer-aided diagnosis) in mammography

Computer-aided diagnosis (CAD) has been extensively reported to increase sensitivity by about 10% when added to a single reading while increasing recall rate by 12%, and its current use can be safely recommended in clinical practice. CAD has been suggested as a possible alternative to conventional double reading in screening. Uncontrolled comparison is consistent and suggests that CAD is comparable to double reading in incremental cancer detection rate (CAD +10.6%, double reading +9.1%) and possibly better in recall rate (CAD +12.5%, double reading +28.8%). However, controlled studies comparing single reading + CAD to conventional double reading are not consistent and on average suggest a lower cancer detection rate (-5.1%) and a lower recall rate (-9.8%) for CAD. Scientific evidence is not sufficient for a safe recommendation of single reading + CAD as a current alternative to conventional double reading.

Effectiveness of service screening: a case-control study to assess breast cancer mortality reduction

The aim of this study was the evaluation of the impact of service screening programmes on breast cancer mortality in five regions of Italy. We conducted a matched case-control study with four controls for each case. Cases were defined as breast cancer deaths occurred not later than 31 December 2002. Controls were sampled from the local municipality list and matched by date of birth. Screening histories were assessed by the local, computerised, screening database and subjects were classified as either invited or not-yet-invited and as either screened or unscreened. There were a total of 1750 breast cancer deaths within the 50 to 74-year-old breast cancer cases and a total of 7000 controls. The logistic conditional estimate of the cumulative odds ratios comparing invited with not-yet-invited women was 0.75 (95% CI: 0.62-0.92). Restricting the analyses to invited women, the odds ratio of screened to never-respondent women corrected for self-selection bias was 0.55 (95% CI: 0.36-0.85). The introduction of breast cancer screening programmes in Italy is associated with a reduction in breast cancer mortality attributable to the additional impact of service screening over and above the background access to mammography.

Mastectomy rates are decreasing in the era of service screening: a population-based study in Italy (1997-2001)

We enrolled all 2162 in situ and 21 148 invasive cases of breast cancer in 17 areas of Italy, diagnosed in 1997–2001. Rates of early cancer increased by 13.7% in the screening age group (50–69 years), and breast conserving surgery by 24.6%. Advanced cancer rates decreased by 19.4%, and mastectomy rates by 24.2%. Service screening did not increase mastectomy rates in the study population.

Probability of early repeat Pap smear in an integrated cervical screening programme

The integration of routine Pap smear practice based on self-referral into an organized programme based on personal invitation is the recommended approach to cervical cancer screening. In this study, the cytology archive of the integrated cervical screening programme implemented in the Faenza district (northern Italy) was used to compare the population of women responding to invitation with that of self-referred women for the cumulative probability of early (< 36 months) repeat Pap smear after a negative result. Between 1995 and 1998, women aged 25-64 living in five municipalities of the district were targeted by the first round of the programme. Eligible for the study were 2356 women responding to invitation for screening and 2221 women not invited because of self-referral for Pap smear at any other setting. The probability of early repeat Pap smear was estimated by the life table method. Differences were evaluated by the Gehan test. Multivariate determinants were assessed by the Cox regression analysis. At 36 months' follow-up, the cumulative probability of early repeat Pap smear was 44% among self-referred women and 6% among women responding to invitation (relative hazard = 4.8). For self-referred women, the probability was related to age (with a peak at 35-44 years), previous Pap smear history and municipality of residence. Among women responding to invitation, only an inverse association with age was demonstrated. In conclusion, the observed differences in overall probability and determinants of early repeat Pap smear between the two screening populations provided important information on Pap smear usage in the total target population.

Do patients with rectosigmoid adenomas 5 mm or less in diameter need total colonoscopy?

BACKGROUND

The need for colonoscopy in the care of patients with rectosigmoid adenoma 5 mm or less in diameter is still debatable.

METHODS

We estimated the prevalence of proximal adenomas among 3052 consecutive subjects undergoing total colonoscopy. Rectosigmoid adenoma was classified as diminutive (5 mm), small (6 to 10 mm), or large (>/=11 mm). Advanced proximal adenoma was 10 mm in diameter or larger, or with a villous component, severe dysplasia, or infiltrating adenocarcinoma.

RESULTS

Proximal adenoma was found in 212 of 2483 patients (8.5%, 95% CI [7.5, 9.7]) without distal neoplastic polyps, 49 of 214 (22.9%, 95% CI [17.6, 29.2]) with diminutive, 44 of 174 (25.3%, 95% CI [19.1, 32.5] with small, and 70 of 181 (38.7%, 95% CI [31.6, 46.2]) with large distal adenoma. Advanced proximal adenoma was found in 49 (2.0%, 95% CI [1.5, 2.6]), 8 (3.7%, 95% CI [1.7, 7.5]), 17 (9.8%, 95% CI [6.0, 15.4]), and 29 patients (16.0%, 95% [11.2, 22.4]), respectively. In patients with distal adenoma risk for proximal lesions increased with increasing age, size, and number of distal adenomas (p = 0.01). Size of distal adenoma was the strongest predictor of the presence of proximal advanced adenoma (multivariate analyses).

CONCLUSIONS

In a clinical setting, the decision to perform colonoscopy should take into account proximal lesions of clinical interest, life expectancy, costs, and risks associated with the procedure. When detection of advanced proximal adenoma is the goal, presence of distal diminutive adenoma alone might not be an indication for total colonoscopy.

Distal hyperplastic polyps do not predict proximal adenomas: results from a multicentric study of colorectal adenomas

BACKGROUND

The association between distal hyperplastic polyps and proximal adenomas is still a matter of debate. We investigated this association while taking into account patient characteristics.

METHODS

After exclusion of patients with inflammatory bowel diseases, familial adenomatous polyposis, or any cancer, 3088 eligible consecutive subjects aged 18 to 69 years underwent total colonoscopy in four gastroenterology units. The odds ratios (OR) of having proximal adenomas according to patient characteristics (age, sex, medical center, year of endoscopy, reasons for referral, and distal findings) were estimated in univariate and multivariate analyses.

RESULTS

Patients with distal polyps of any type showed an adjusted OR of 2.5 (95% CI [1.9, 3.1] p < .001) of having proximal adenomas as compared with those without distal polyps. When distal adenomas and distal hyperplastic polyps were included in the multivariate model as independent factors, the presence of adenomas significantly increased the risk of proximal adenomas (OR = 2.8: 95% CI [2.2, 3.6] p < .001), whereas the presence of hyperplastic polyps did not (OR = 1.1: 95% CI [0.8, 1.5] p = .64). No association with number, size, or location of distal hyperplastic polyps was seen.

CONCLUSIONS

Our data show that the presence of hyperplastic polyps should not be the sole indication for total colonoscopy because they are not associated with proximal adenomas when adjusting for patient characteristics and presence of distal adenomas.

Cohort study of association of risk of breast cancer with cyst type in women with gross cystic disease of the breast

OBJECTIVE

To assess correlation between type of breast cyst and risk of breast cancer in women with gross cystic disease of the breast.

DESIGN

Cohort study of women with breast cysts aspirated between 1983 and 1993 who were followed up until December 1994 for occurrence of breast cancer.

SETTING

Major cancer prevention centre.

SUBJECTS

802 women with aspirated breast cysts.

MAIN OUTCOME MEASURES

Type of breast cyst based on cationic content of cyst fluid: type I (potassium:sodium ratio > 1.5), type II (potassium:sodium ratio < 1.5), or mixed (both types). Subsequent occurrence and type of breast cancer.

RESULTS

After median follow up of six years (range 2-12 years) 15 cases of invasive breast cancer and two ductal carcinomas in situ were diagnosed in the cohort: 12 invasive cancers (and two carcinomas in situ) among the 417 women with type I cysts, two cancers among the 325 women with type II cysts, and one among the 60 women with mixed cysts. The incidence of breast cancer in women with type I cysts was significantly higher than that in women with type II cysts (relative risk 4.62 (95% confidence interval 1.26 to 29.7)). These results were confirmed after adjustment for several risk factors for breast cancer (relative risk 4.24 (1.12 to 27.5)).

CONCLUSIONS

The increased risk of breast cancer of women with breast cysts seems to be concentrated among women with type I breast cysts.

Wolfe's mammographic patterns in women with gross cystic disease of the breast

The epidemiologic determinants of the mammographic pattern were studied in 710 patients with aspirated gross (> 1 ml) cysts of the breast. The prevalence of the mammograms classified as P2-DY, that are considered to be associated with an increased breast cancer risk, was 636/710 or 89.6%. No relationship between mammographic patterns and characteristics of breast cyst fluid such as K+/Na+ ratio, apocrine changes and dehydroepiandrosterone sulfate concentration was observed. A significant decrease in the proportion of these patterns with increasing age (p = 0.006), Quetelet Index (p < 0.001), parity (p = 0.001), and in postmenopausal women (p = 0.026) was found. Conversely, P2-DY patterns were significantly associated with a later age at menarche (p = 0.023) and alcohol consumption (p = 0.001). In multivariate analysis, an independent association with age was not observed whereas the associations with age at menarche, parity, and relative weight were confirmed. In conclusion, the epidemiologic determinants of mammographic patterns are the same in Gross Cystic Disease patients as in unaffected women, and the lack of correlation between mammographic pattern and cyst type suggests that the latter may represent an independent predictor of breast cancer risk.

Mitogenic effect of the 15-kDa gross cystic disease fluid protein (GCDFP-15) on breast-cancer cell lines and on immortal mammary cells

The biological significance of a major protein component in the fluid of gross cystic breast disease and a recognized marker of apocrine metaplasia, i.e. the 15-kDa glycoprotein (GCDFP-15), is presently unknown. We have added GCDFP-15 to cell culture medium and tested its effect on proliferation of 4 human breast-cancer cell lines (MCF7, BT474, MDA-MB231 and T47D) and a "normal" human immortal breast-cell line (MCF10A). These breast-cell lines showed a mitogenic response to GCDFP-15 (10 micrograms/ml). GCDFP-15 enhanced cell growth of the MCF10A, MCF7, BT474 and MDA-MB231 cell lines at both 48 and 96 hr of exposure. The glycoprotein exerted a mitogenic effect on the T47D cell line at 48 hr but not at 96 hr. This may be due to an auto-regulatory effect of endogenous GCDFP-15 synthesized by the T47D cells. GCDFP-15 was ineffective on 2 colon-cancer cell lines (HT29 and NIC-H716), on the IMR32 neuroblastoma cell line and on the NIC-H209 small-cell lung carcinoma cells. A separate major breast cystic disease fluid protein of 24 kDa (GCDFP-24) was tested, following the same experimental design, on the 5 breast-cell lines, and showed no mitogenic activity. The mitogenic effect of GCDFP-15 observed in this study in both "normal" and malignant breast epithelial cells suggests a possible relationship between apocrine metaplasia in breast cystic disease and the development of breast epithelial hyperplasia. In addition, a possible role of GCDFP-15 in breast-cancer progression should be considered.

Steroid biochemistry and categorization of breast cyst fluid: relation to breast cancer risk

Patients bearing macrocysts of the breast are at higher risk of later developing cancer. The fluid filling the cysts (breast cysts fluid, BCF) contains unusual amounts of steroid conjugates, first androgen and estrogen sulfates. Measuring BCF cations (K+,Na+) allows categorization of cysts into two major subsets (type I and type II) that are associated with a different degree and/or turnover of apocrine metaplastic cells in the lining epithelium. Type I cysts (high K+/Na+ ratio) accumulate hugh amounts of dehydroepiandrosterone sulfate, estrone sulfate, androstane-3 alpha,17 beta-diol glucuronide, androsterone glucuronide and contain more testosterone and dihydrotestosterone than type II. Conversely, type II cysts (low K+/Na+ ratio) contain more progesterone and pregnenolone. A cohort study was started in 1983 at the Cancer Prevention Center, Ravenna, Italy, with the aim of evaluating the relationships between the biochemistry of BCF and the incidence of breast cancer in women with gross cystic disease (GCD) of the breast. The bimodal distribution of the cationic pattern has been confirmed from data obtained in 798 patients aspirated. The risk of cyst relapse was significantly higher among women with type I cysts or with multiple cysts at presentation. Twelve incident cases of breast cancer have been diagnosed among women whose BCF was categorized. Eleven out of 12 cases had type I or multiple cysts. The cumulative incidence of breast cancer among patients bearing type I cysts was 2.5%. We conclude that women with GCD bearing type I cysts have an increased breast cancer risk when compared with the counterpart bearing type II cysts or the general population.

Performance of Fineneedle Aspiration Cytology of the Breast. Clinical Experience in Ravenna (Italy)

Aims and background

Fineneedle aspiration cytology (FNAC) is a routine test in the evaluation of breast lesions. We assessed the diagnostic accuracy of mammography (MG), physical examination (PE), ultrasonography (US) and FNAC in 1064 histologically confirmed breast lesions (638 malignant, 426 benign) observed consecutively at the Cancer Prevention Center of Ravenna (Italy).

Methods

The performance of each test and the additional contribution of FNAC were determined.

Results

FNAC was done in 69.6 % of cancers and 39.7 % of benign lesions (P = 0.00000), the frequency of aspiration being significantly associated with severity at MG, PE, and US. For FNAC, the true positive rate was 95.1 % and the true negative rate 67.4 %. Only one breast cancer case was detected by FNAC alone (additional true positive rate 0.2 %). The positive predictive value of FNAC in the absence of other abnormalities was 5 %. The negative predictive value of a benign report at MG, PE, US and FNAC was 100 %.

Conclusions

All breast lesions should be evaluated by all available techniques, especially FNAC, and open biopsy should be avoided for those reported as benign at all tests.

Association of cyst type with risk factors for breast cancer and relapse rate in women with gross cystic disease of the breast

The concentration of potassium (K+) and sodium (Na+) was measured in breast cyst fluid (BCF) from 611 cysts greater than 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 Na+ and the bimodal distribution of the K+/Na+ ratio in BCF were confirmed. Type I cysts were defined as cysts with a K+/Na+ greater than 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.

Levels of eighteen non-conjugated and conjugated steroids in human breast cyst fluid: relationships with cyst type

The present study investigates the levels of a large series of 18 non-conjugated or conjugated steroids in 71 samples of human breast cyst fluid (BCF) as divided into three groups corresponding to different electrolyte composition. In the type 1 group, the K+/Na+ ratio was higher than 1.5, while in type 2 it was lower than 0.66 and finally type 3 had an intermediate ratio. Pregnenolone (PREG) and progesterone (PROG) levels were approximately 2-fold higher (P less than 0.05) in the type 2 than in the type 1 group while both 17-OH-pregnenolone (17-OH-PREG) and 17-OH-progesterone (17-OH-PROG) concentrations were similar in these two groups. Most of the C-19 steroids analyzed, namely dehydroepiandrosterone sulfate (DHEAS), androst-5-ene-3 beta, 17 beta-diol (5-ene-DIOL), testosterone (TESTO), dihydrotestosterone (DHT), androstane-3 alpha, 17 beta-diol (3 alpha-DIOL), androsterone (ADT), androstane-3 alpha, 17 beta-diol glucuronide (3 alpha-DIOL-G) and androsterone glucuronide (ADT-G) were 180-360% (P less than 0.05) higher in type 1 than in type 2 cysts while no difference in C-18 steroid and C-18 steroid glucuronide levels was observed. A small or no difference was seen in steroid levels between types 2 and 3. We conclude that the arbitrary division according to the electrolyte composition of BCF permits identification of different patterns of steroid concentrations in BCF. However, the mechanism responsible for both sets of parameters remain unclear.

Apocrine cells in breast cyst fluid and their relationship to cyst type: a morphometric study

The morphology of apocrine cells exfoliated in breast cyst fluid (BCF) was studied in 78 BCF samples obtained from 39 premenopausal patients with gross cystic disease who were bearing two simultaneously aspirated cysts. 57/78 samples showed cell clusters suitable for computer-assisted cytometry. This was performed on 5820 cells using a Leitz Texture Analysis System (TAS). We measured the surface areas of cytoplasm, nucleus and nucleolus; we also calculated the nuclear/cytoplasmic (N/C), nuclear/nucleolar (N/n) ratios and the nuclear roundness factor (RF). Cysts were divided according to the cationic pattern of BCF: Type I, K+/Na+ greater than 1.5; Type II, K+/Na+ less than 0.66. The cytometric analysis was made on 47 samples of Type I and 10 samples of Type II. At the light microscope, no difference was apparent between the apocrine cells coming from Type I or Type II cysts. Cytometric measurements showed significant differences for the apocrine cells aspirated from Type I vs. Type II cysts for the mean cytoplasmic area (97.13 +/- 24.28 S.D. mu2 vs. 59.66 +/- 14.90 S.D. mu2, respectively) and the mean nucleolar area (4.35 +/- 0.99 S.D. mu2 vs. 2.75 +/- 0.71 S.D. mu2, respectively). Our data do not allow the inference of apocrine changes in the epithelium lining the cysts simply from the cationic pattern of BCF. The significantly wider cytoplasm and nucleoli of the apocrine cells aspirated from Type I cysts could reflect different functional stages of these particular cells.

Shifted distribution of cation-related cyst types in post-menopausal patients with gross cystic disease of the breast

Gross cystic disease of the breast (GCD) is rarely seen after the menopause. Recent work has shown that by measuring electrolytes in the breast cyst fluid (BCF) it is possible to identify two principal classes of cyst, designated Type 1 (K+/Na+ greater than 1.5) and Type 2 (K+/Na+ less than 0.66). A smaller, intermediate class (Type 3) also appears to exist. We measured K+, Na+ and dehydroepiandrosterone sulphate (DHA-S) in 38 BCF samples aspirated from 33 women with GCD who had undergone spontaneous menopause at least 1 yr previously. Statistically significant correlations were found between DHA-S and cations (positive in relation to K+, P less than 0.001; negative in relation to Na+, P less than 0.001). The distribution of cyst types was shifted with respect to that characteristic of cases that occur at an earlier age: whereas Type 1 cysts predominate in menstruating women, Type 2 cysts proved more numerous in the post-menopausal subjects. The difference was statistically validated (P less than 0.001). The results seem to indicate a sharp reduction in high K+, high DHA-S cysts after the menopause, which may be paralleled by a decrease in the associated apocrine metaplasia. In view of the major biochemical differences between Type 1 and Type 2 cysts and of the suggested differences as regards cancer risk, classification of post-menopausal patients with GCD by cyst type is critical prior to any clinical trial or follow-up.

Cations and dehydroepiandrosterone-sulfate in cyst fluid of pre- and menopausal patients with gross cystic disease of the breast. Evidence for the existence of subpopulations of cysts

Cations (K+ and Na+) content was evaluated in 444 breast cyst fluid (BCF) specimens, aspirated from 391 patients with gross cystic disease of the breast (GCD), a benign form admittedly at major risk of cancer. In 306/444 BCF, dehydroepiandrosterone-sulfate (DHA-S) content was also evaluated. A positive correlation (P less than 0.001) was observed between log K+ vs. log DHA-S whereas a negative correlation was found between log Na+ and log DHA-S (P less than 0.001). Cysts were subdivided in three types according to their cationic concentration: most were of type I (K+/Na+ greater than 1.5) and type II (K+/Na+ less than 0.66) whereas only 10% was of the type III (intermediate). No statistical difference in subtype distribution was apparent when considering patients aspirated in the follicular vs. luteal phase of the menstrual cycle; on the contrary, a significant difference (P less than 0.001) was found between menstruating vs. menopausal patients (type I = 54.8% vs. 32.2%; type II = 34.5% vs. 58.1%, respectively). Ninety-four BCF samples were aspirated simultaneously in 41 patients bearing multiple cysts: the same cationic subtype was present in 29/41 patients. Our data confirm and extend previous observations, and provide conclusive evidence that breast macrocysts can be divided on the basis of their electrolyte composition into different types. Accordingly, the composition of BCF should always be assessed for prospective studies on GCD and breast cancer risk.