Palpable Breast Cysts Research Articles

Discriminating solitary cysts from soft tissue lesions in mammography using a pretrained deep convolutional neural network.

It is estimated that 7% of women in the western world will develop palpable breast cysts in their lifetime. Even though cysts have been correlated with risk of developing breast cancer, many of them are benign and do not require follow-up. We develop a method to discriminate benign solitary cysts from malignant masses in digital mammography. We think a system like this can have merit in the clinic as a decision aid or complementary to specialized modalities. We employ a deep convolutional neural network (CNN) to classify cyst and mass patches. Deep CNNs have been shown to be powerful classifiers, but need a large amount of training data for which medical problems are often difficult to come by. The key contribution of this paper is that we show good performance can be obtained on a small dataset by pretraining the network on a large dataset of a related task. We subsequently investigate the following: (a) when a mammographic exam is performed, two different views of the same breast are recorded. We investigate the merit of combining the output of the classifier from these two views. (b) We evaluate the importance of the resolution of the patches fed to the network. (c) A method dubbed tissue augmentation is subsequently employed, where we extract normal tissue from normal patches and superimpose this onto the actual samples aiming for a classifier invariant to occluding tissue. (d) We combine the representation extracted using the deep CNN with our previously developed features. We show that using the proposed deep learning method, an area under the ROC curve (AUC) value of 0.80 can be obtained on a set of benign solitary cysts and malignant mass findings recalled in screening. We find that it works significantly better than our previously developed approach by comparing the AUC of the ROC using bootstrapping. By combining views, the results can be further improved, though this difference was not found to be significant. We find no significant difference between using a resolution of 100 versus 200 micron. The proposed tissue augmentations give a small improvement in performance, but this improvement was also not found to be significant. The final system obtained an AUC of 0.80 with 95% confidence interval [0.78, 0.83], calculated using bootstrapping. The system works best for lesions larger than 27mm where it obtains an AUC value of 0.87. We have presented a computer-aided diagnosis (CADx) method to discriminate cysts from solid lesion in mammography using features from a deep CNN trained on a large set of mass candidates, obtaining an AUC of 0.80 on a set of diagnostic exams recalled from screening. We believe the system shows great potential and comes close to the performance of recently developed spectral mammography. We think the system can be further improved when more data and computational power becomes available.

Apocrine Cysts of the Breast

Up to one-third of women aged 30-50 years have cysts in their breasts and are presumed to be at increased risk of developing breast cancer. Here we present an extensive proteomic and immunohistochemistry (IHC) study of breast apocrine cystic lesions aimed at generating specific biomarkers and elucidating the relationship, if existent, of apocrine cysts with cancer phenotype. To this end we compared the expression profiles of apocrine macrocysts obtained from mastectomies from high risk cancer patients with those of cancerous and non-malignant mammary tissue biopsies collected from the same patients. We identified two biomarkers, 15-hydroxyprostaglandin dehydrogenase and 3-hydroxymethylglutaryl-CoA reductase, that were expressed specifically by apocrine type I cysts as well as by apocrine metaplastic cells in type II microcysts, terminal ducts, and intraductal papillary lesions. No expression of these markers was observed in non-malignant terminal ductal lobular units, type II flat cysts, stroma cells, or fat tissue as judged by IHC analysis of matched non-malignant tissue samples collected from 93 high risk patients enrolled in our cancer program. IHC analysis of the corresponding 93 primary tumors indicated that most apocrine changes have little intrinsic malignant potential, although some may progress to invasive apocrine cancer. None of the apocrine lesions examined, however, seemed to be a precursor of invasive ductal carcinomas, which accounted for 81% of the tumors analyzed. Our studies also provided some insight into the origin, development, and enlargement of apocrine cysts in mammary tissue. The successful identification of differentially expressed proteins that characterize specific steps in the progression from early benign lesions to apocrine cancer opens a window of opportunity for designing and testing new approaches for pharmacological intervention, not only in a therapeutic setting but also for chemoprevention, to inhibit cyst development as both 15-hydroxyprostaglandin dehydrogenase and 3-hydroxymethylglutaryl-CoA reductase are currently being targeted for chemoprevention strategies in various malignancies.

Outcomes of sonography-based management of breast cysts

Background Ultrasound is commonly used during diagnosis of breast lesions. Our purpose was to study the role of sonography for risk stratification of malignancy in the diagnosis and management of palpable breast cysts. Methods This was a cohort study of 176 patients with palpable breast cysts. Sonographic findings were correlated with clinical and pathologic outcomes. Results Mean cyst size was 2.0 ± 1.8 cm. Cysts were simple, complex and probably benign, and complex and suspicious for neoplasm in 82.25%, 10.25% and 7.5% of patients, respectively. Thick cyst wall ( P = 0.0001), mural tumor ( P <0.00001), eccentric mass ( P = 0.034), and internal septae ( P = 0.031) were predictive of neoplasm. Of cysts >3 cm, 33% were cancerous ( P = 0.000027). After 378 days of follow-up, 26 % of cysts had recurred. Recurrence was more frequent in patients with bilateral or multiple cysts ( P = 0.004). Conclusions Sonography is useful in risk stratification of malignancy in breast cysts. There is a high risk of recurrence after cyst aspiration.

Management of nonpalpable ultrasound-indeterminate breast lesions

Background A series of such lesions was studied to determine the validity of applying criteria routinely used to manage palpable breast cysts to the management of mammographically detected, nonpalpable breast lesions characterized as indeterminate on ultrasound (US). Methods The clinicopathologic data from a series of 134 patients who underwent US-guided fine needle aspiration biopsy for nonpalpable, mammographically detected breast lesions, categorized as indeterminate on US, were reviewed. Results Of 139 indeterminate lesions, 78 were consistent with complex cysts, whereas in 61, the cystic-versus-solid nature was indistinguishable. All 71 complex cyst lesions that contained nonbloody fluid and resolved completely were benign. Two of 7 complex cyst lesions that had incomplete resolution, bloody aspirate, or both were malignant. Of 61 cystic-versus-solid lesions, 29 and 32 were primarily solid and cystic, respectively. Three of the 29 solid lesions were malignant. Of the 32 cystic lesions, all 26 that contained nonbloody fluid and resolved completely were benign, whereas 1 of 6 lesions that had incomplete resolution, bloody aspirate, or both was malignant. Conclusion Criteria such as complete resolution and nonbloody aspirate are an effective adjunct to the management of nonpalpable, mammographically detected breast lesions categorized as indeterminate by US.

Lack of utility in clinical practice of cytologic examination of nonbloody cyst fluid from palpable breast cysts

Objective: This study was undertaken to answer the following question: Does cytologic evaluation of nonbloody fluid aspirated from breast cysts contribute to appropriate clinical management? Study Design: A retrospective review of palpable breast cyst fluid cytologic reports and associated medical records was undertaken to determine whether the cytologic findings affected patient management. Breast cyst size, fluid volume, fluid color, and patient age were abstracted from 689 medical records (1988-1999) of women whose palpable cysts had been aspirated at the Breast Diagnostic Center, Women's and Children's Hospital, Los Angeles. These observations were correlated with the fluid cytologic reports. Results: Except for frankly bloody fluid, all breast fluid cytologic reports listed the results as acellular, inadequate for cytologic diagnosis, or no malignant cells identified. Conclusion: In clinical practice only frankly bloody fluid should be submitted for cytologic analysis. All other cyst fluid should be discarded. (Am J Obstet Gynecol 2000;182:1300-5.)

Risk of breast cancer in women with palpable breast cysts

Risk of breast cancer in women with palpable breast cysts

Palpable breast cysts were associated with an increased risk for breast cancer

Source Citation Dixon JM, McDonald C, Elton RA, Miller WR, for the Edinburgh Breast Group. Risk of breast cancer in women with palpable breast cysts: a prospective study. Lancet. 1999 May 22;353:1742-5. 99275742

Palpable breast cysts were associated with an increased risk for breast cancer

Palpable breast cysts were associated with an increased risk for breast cancer

Risk of breast cancer in women with palpable breast cysts: a prospective study

7% of women in the western world develop palpable breast cysts. Studies of the relation between cysts and breast cancer have conflicting results. There are two clearly defined types of cyst. We investigated whether one cyst type is associated with a higher rate of breast-cancer development than the other. We studied 1374 women with palpable breast cysts presenting between 1981 and 1987, who had cysts aspirated between 1981 and 1989. Cysts were classified as type I if the sodium/potassium (Na+/K+) ratio in the cyst fluid was less than 3, or type II if the Na+/K+ ratio was 3 or more. Data on incidence of breast cancer were available until January, 1995, and we compared them with the expected numbers of cancers calculated from age-specific breast-cancer incidence in Scotland in 1988. 65 cancers developed during follow-up. The overall standardised incidence rate of breast cancer in patients with palpable cysts was 2.81 (95% CI 2.17-3.59). The relative incidence rate was increased for all cyst types. The standardised incidence rate of developing breast cancer among women younger than 45 years was highest at 5.94 (2.97-10.63), with a significant trend for decreasing relative incidence rate with age (p<0.05). Women older than 54 years had a standardised incidence rate of 1.73 (0.86-3.10). The standardised incidence rate of breast cancer was highest in the first year after aspiration (7.02 [3.73-12.00]) but the risk was still raised after 5 years (2.68 [1.84-3.76]). Women with breast cysts are at an increased risk of breast cancer, especially at younger ages. The type of cyst did not alter the associated relative incidence rate of breast-cancer development.

Insulin-like growth factor binding protein-3 in breast cyst fluid: relationships with insulin-like growth factors I and II and transforming growth factor-beta 1 and 2.

Women who have palpable breast cysts with intracystic Na/K > 3 may have a lower risk of developing breast cancer than those with intracystic Na/K < 3. In this study significantly higher concentrations of insulin-like growth factor-binding protein-3 (IGFBP-3), insulin-like growth factors I and II (IGF-I, IGF-II) and transforming growth factor-beta 2 (TGF-beta2) were found in the Na/K > 3 sub-group. No difference was found in transforming-growth factor-beta 1 (TGF-beta1) levels between the two sub-groups of breast cysts. A positive correlation was obtained for IGFBP-3 and TGF-beta1 in the Na/ K > 3 sub-group consistent with reports that TGF-beta1 may regulate the production of IGFBP-3. Equimolar amounts of total IGFs and IGFBP-3 in breast cyst fluid imply that most, if not all, of these IGFs are protein-bound. The significantly higher concentrations of TGF-beta2 in the Na/K > 3 sub-group may partly explain the lower risk of breast cancer in this group of women.

Angiotensin-converting enzyme and enkephalinase in human breast cyst fluid.

Palpable breast cysts with an apocrine epithelial lining (type 1) are reported to be associated with a higher risk of developing breast cancer. The composition of breast cyst fluid (BCF) might include those factors involved in this increased risk. In this study peptidase activities that were active against the substrate [125I]metenkephalin-Arg-Phe were detected in BCF. The products were identified by reversed phase high-performance liquid chromatography (HPLC) as [125I]Tyr-Gly-Gly and [125I]Met-enkephalin. This proteolysis was not inhibited by PCMB, pepstatin A, leupeptin or aprotinin but was by EDTA, showing that the activity was due to metalloproteases. The production of [125I]Try-Gly-Gly was inhibited by phosphoramidon and thiorphan, whereas that of [125I]met-enkephalin was inhibited by captopril and Bothrops jararaca peptide, indicating that these activities are enkephalinase and angiotensin-converting enzyme (ACE) respectively. A fluorometric assay for ACE demonstrated that ACE levels are significantly higher in type 2 BCF than in type 1 BCF (30.8 vs 6.1 nmol hr-1 10 microliters-1, P < 0.001). As the increased risk of cancer is linked to type 1 cysts it is possible that higher levels of peptidase in type 2 BCF reflect a protective environment in the breast in which mitogenic peptide growth factors are neutralised by proteolysis.

PP-4-14 Stereotaxic fine-needle aspiration cytology in the detection of non palpable breast cysts: An alternative to ultrasound guidance

PP-4-14 Stereotaxic fine-needle aspiration cytology in the detection of non palpable breast cysts: An alternative to ultrasound guidance

ESTERASE, ESTRADIOL AND SEX HORMONE-BINDING GLOBULIN IN BREAST CYST FLUID

Women who have palpable breast cysts lined by apocrine epithelium (intracystic Na/K < 3) may have a higher risk of developing breast cancer than those with breast cysts lined by flattened epithelium (intracystic Na/K > 3). Esterase converts fatty acid esters of E2 to unconjugated E2 and could conceivably modulate intracystic E2 concentrations. The amount of SHBG in BCF may regulate the amount of unbound, biologically active E2 present. In this study the relationships amongst E2, esterase, SHBG and Na/K in BCF were assessed. Negative correlations were found between E2 and esterase (P<0.05), E2 and SHBG (P<0.001), and E2 and Na/K (P<0.001). A positive correlation was found between SHBG and Na/K (P<0.001). No correlation was found between esterase and Na/K. The findings of this study suggest that esterase is unlikely to be important in modulating intracystic E2 concentrations.

Cystic disease and fibroadenoma of the breast: Natural history and relation to breast cancer risk

Approximately 25% of all 'discrete' breast lesions are fibroadenomas or breast cysts and they more commonly cause a breast lump than breast cancer. Despite their frequency, their natural history and relationship to subsequent breast cancer have not been clearly defined, although it would appear that palpable breast cysts, but not fibroadenomas, are associated with some increased risk of breast cancer. The diagnosis of these two entities is now possible by fine needle aspiration and excision of these lesions is only indicated in certain circumstances.

Mitogenic peptides in breast cyst fluid: Relationship with intracystic electrolyte ratios

Women with palpable breast cysts which are lined with apocrine epithelium may be at higher risk of developing breast cancer than women with breast cysts which are lined with flattened epithelium, the former group being characterized by intracystic sodium to potassium ratios below 3, while the latter group has intracystic sodium to potassium ratios above 3. In this study the distribution of intracystic concentrations of the mitogenic peptides, epidermal growth factor, endothelin and gastrin-releasing peptide in the 2 groups of breast cysts were compared to see whether differences in concentrations between the 2 cyst groups might provide an explanation for the higher risk of breast cancer observed in women with "apocrine" breast cysts. The concentrations of epidermal growth factor and gastrin-releasing peptide were significantly higher in the low electrolyte ratio group (p less than 0.001). There was no difference in endothelin concentrations between the 2 groups. Negative correlations were found between epidermal growth factor concentrations and Na+/K+ and between gastrin-releasing peptide concentrations and Na+/K+ (p less than 0.001). A positive correlation was found between gastrin-releasing peptide and epidermal growth factor concentrations in breast cyst fluid (p less than 0.001). The significantly higher intracystic concentrations of both epidermal growth factor and gastrin-releasing peptide in the low-electrolyte-ratio group may provide an explanation for the higher risk of breast cancer which has been observed in women with "apocrine" breast cysts.

Levels of androgen conjugates and oestrone sulphate in patients with breast cysts

Plasma and cyst fluid were obtained from patients with palpable breast cysts and analysed for androgen conjugates and oestrone sulphate content by radioimmunoassay. Concentrations of androgen conjugates in cyst fluids varied from 15.6 to 475.5 μmol/1. These levels were much greater than those in plasma (1.3–5.2 μmol/1) and there was no association between values in cyst aspirates and plasmas obtained from the same individuals. Levels of oestrone sulphate in breast cyst fluids (1.5–744.0 nmol/1) were also generally in excess of those in plasma (2.0–59.9 nmol/1) and again no relationship was evident between concentrations in cyst fluid and the circulation. Neither was there a relationship between levels of androgen conjugate and oestrone sulphate in plasma. In contrast, a highly significant correlation ( P < 0.001) was identified between the androgen conjugate and oestrone sulphate content of cyst fluids. Levels of both androgen conjugates and oestrone sulphate were also significantly different in groups of cysts subdivided according to electrolyte classification, cysts with low Na + :K + ratios having higher steroid concentrations than those with high Na + :K + ratios. The biological significance of the relationship between the two conjugates in cyst fluids remains unclear but it is suggested that the accumulation of these steroids involves a common mechanism.

The relationship of cyst type to risk factors for breast cancer and the subsequent development of breast cancer in patients with breast cystic disease

The frequency of epithelial hyperplasia and papillary apocrine change in patients with palpable breast cysts lined by either apocrine or flattened epithelium has been compared. Hyperplasia of any degree, severe hyperplasia with or without atypia and papillary apocrine change were all seen significantly more frequently in patients with clinically palpable apocrine cysts. Twelve patients were identified wth breast cancer with a history of cyst aspiration, in whom all cysts aspirated could be classified as apocrine or flattened on the basis of cytology or electrolyte composition of cyst fluid. Eleven patients had single or multiple apocrine cysts and one had a single flattened cyst. This represents a significantly increased preponderance of apocrine cysts as compared with that normally seen in patients with cystic disease. The mastectomy specimens of those 11 patients with a history of apocrine cyst aspiration more frequently contained hyperplastic changes and non-invasive carcinoma than age- and menopausal-matched controls who did not have a history of cystic disease. This study suggests that patients who develop cysts lined by apocrine epithelium may be at a greater risk of subsequent breast cancer than those with flattened epithelial cysts.