Fractal dimensions of breast lesions on cytology smears

In the present study, box-counting fractal dimensions of benign and malignant cells of breast tumors on cytology material were measured and compared. We selected fine-needle aspiration cytology smears of 14 cases of histopathology-proven infiltrating duct carcinomas and 7 cases of fibroadenoma of the breast. Five cells were randomly selected from each case. Box-counting fractal dimensions of all cells were measured with the help of an image cytometer (Leica, Cambridge, UK), using Quantimet 600 software (Leica). In total, 70 malignant cells and 35 benign cells were studied. The mean fractal dimensions of malignant cells and benign cells were 0.9571 +/- 0.1265 and 0.8354 +/- 0.1367, respectively. There was significant difference in the fractal dimension of malignant and benign cells (P = 0.006, Mann-Whitney nonparametric test). The measurement of fractal dimension may be helpful in discriminating malignant from benign cells. This may be another discriminating feature of malignant cells, along with classic image morphometry based on Euclidean geometry.     

Myoepithelial cells in needle aspirations of two cases of unusual breast lesions: an aid in differential diagnosis

The observation of myoepithelial cells bordering abnormal epithelial cells in fine-needle aspiration smears of two unusual breast lesions helped to prevent misdiagnosis of these lesions as malignant. The presence of myoepithelial cells has been a well-established criterion for histologically benign lesions. Their presence or absence appears to be equally valuable in fine-needle aspiration smears of breast lesions.     

Fibroadenoma mimicking papillary carcinoma on ThinPrep of fine-needle aspiration of the breast

OBJECTIVE: To compare and contrast benign and malignant lesions of the breast that have similar appearances on fine-needle aspiration cytology and that constitute diagnostic pitfalls. DESIGN: The cytology files (dated November 1995 through May 1998) of the Beth Israel Deaconess Medical Center were searched to identify cases of breast fine-needle aspiration biopsies that were highly cellular and composed of bland-appearing spindle/columnar cells and that could represent either epithelial or stromal cells; these cases were reported as indeterminate (atypical/suspicious) and had subsequent excisional biopsies taken. RESULTS: Four such cases were found. Two were fibroadenomas and 2 were papillary carcinomas. Their appearances were strikingly similar on aspiration cytology. All cases were prepared with the ThinPrep method. On microscopic examination, all 4 cases were hypercellular and had many single cells and clusters of columnar/elongate cells. Immunocytochemistry proved these cells to be of epithelial origin. At least occasional bipolar stromal cells were seen in the background. The only appreciable difference between the benign and malignant cases was more significant nuclear atypia, which was barely discernible, in the malignant cases. Immunocytochemistry for smooth muscle actin was helpful in 2 cases that had sufficient material. CONCLUSIONS: Some cases of fibroadenomas and papillary carcinomas can be very difficult, if not impossible, to distinguish on fine-needle aspiration cytology. Immunocytochemistry may be helpful if sufficient material is available. To avoid false-negative or false-positive diagnosis on cytology, it is best to report such cases as atypical or suspicious with final diagnosis pending excisional biopsy.     

Papillary neoplasms of the breast: Fine-needle aspiration findings in cystic and solid cases

Papillary neoplasms of the breast are uncommon; at the time of needle aspiration, the diagnostic yield from such lesions can be initially classified as cystic or solid. We describe the fine-needle aspiration findings in four cystic papillary neoplasms (three intracystic papillary carcinomas and one intracystic papilloma) and three solid masses (two sclerosing ductal lesions and one infiltrating ductal carcinoma with prominent papillary component). The smears were examined with respect to the following features: cellularity, architectural pattern in cell groups, cytologic pleomorphism, degree of cohesiveness, morphology and size of individual cells, anisonucleosis, nuclear-cytoplasmic ratio, irregularity of nuclear contour, chromatin texture, macronucleoli, the presence of bipolar nuclei, apocrine cells, and multinucleated giant cells. A combination of cytologic and clinical characteristics may be helpful in distinguishing benign from malignant papillary lesions. © 1994 Wiley-Liss, Inc.     

Are epithelial cells in fat or connective tissue a reliable indicator of tumor invasion in fine-needle aspiration of the breast?

The diagnosis of breast carcinoma tumor invasion by fine-needle aspiration (FNA) cytology continues to be controversial. To assess the reliability of predicting tumor invasion by FNA, we examined the cytologic smears of 183 FNAs of benign and malignant solid epithelial lesions of the breast for which histologic follow-up was available. The study group consisted of 94 invasive carcinomas, eight pure ductal carcinomas in situ (DCIS), and 81 benign lesions (fibroadenoma, fibrocystic changes, papilloma, adenosis). Epithelial cellularity, presence of epithelial cells in dispersed fat droplets and presence of epithelium within intact fragments of fibrofatty connective tissue were tabulated. Epithelial cellularity in dispersed fat was semiquantitatively scored. The cytologic diagnosis of the epithelial cells in all cases was recorded as benign, malignant, or indeterminant for malignancy. Findings showed that 95.5% of invasive carcinomas, 100% of DCIS, and 68.1% of benign lesions contained epithelial cells in dispersed fat; 80.8% of invasive carcinomas, 66.7% of DCIS, and 60.7% of benign lesions contained epithelial cells in intact fibrofatty connective tissue. Corrected score of epithelium within fat was 0.781 for invasive carcinoma, 0.727 for DCIS, and 0.562 for benign lesions. The difference in values for all parameters was not statistically significant between invasive carcinoma and DCIS, but reached significance between invasive carcinoma and benign lesions. Eighteen cases (7/94 invasive carcinoma, 5/8 DCIS, 6/81 benign lesions) contained atypical epithelial cells indeterminant for malignancy, all of which had epithelial cells present in dispersed fat when dispersed fat was present on the slides, indicating that this criterion was not helpful in discriminating between a benign and malignant diagnosis. We conclude that the presence of epithelial cells either admixed within dispersed fatty droplets or seemingly within fragments of fibrofatty connective tissue is not a reliable indicator of tumor invasion in FNA of the breast, and is frequently found in both benign and malignant breast lesions. The presence of epithelial cells in intact or dispersed fat is most likely a mechanical artifact of aspiration and/or smear preparation. Diagn. Cytopathol. 16:137–142, 1997. © 1997 Wiley-Liss, Inc.     

Fine-needle aspiration of fibromatosis of the breast

Percutaneous fine-needle aspiration of breast fibromatosis yielded clusters of uniform spindle cells. The differential diagnosis of spindle stromal cells obtained by fine-needle aspiration of the breast is discussed. This includes a wide spectrum of both benign and malignant breast lesions that have to be considered prior to suggesting the diagnosis of fibromatosis.     

Fine-needle aspiration cytology of mesenchymal tumors of the breast

This report describes the fine-needle aspiration (FNA) cytologic findings of 15 cases of sarcomas involving the breast out of a combined series of 2,064 breast FNA biopsies, including 580 malignancies, thereby accounting for 2.6% of all the malignant breast tumors. The series consisted of 14 women and one man with a mean age of 48.4 yr (range, 29-63). There were eight cases of cystosarcoma phyllodes, including one malignant cystosarcoma phyllodes. Three benign cystosarcoma phyllodes had a significant concomitant atypical epithelial hyperplasia, which lead to a misdiagnosis of carcinoma in two of the cases. The third case was correctly identified as recurrent cystosarcoma phyllodes. In retrospect, features suggestive for cystosarcoma phyllodes and unusual for breast carcinoma include increased numbers of naked nuclei and hypercellular stromal fragments. Sarcomatous patterns in our four metaplastic carcinomas included chondrosarcoma (two cases), malignant fibrous histiocytoma (MFH) (one case), and fibrosarcoma (one case). Two additional pure primary MFHs (both of which had electron microscopic confirmation) and one metastatic fibrosarcoma to the breast were encountered. Recognition of unusual cytologic patterns for breast carcinoma should suggest the possibility of a primary or metastatic sarcoma to the breast. Potential pitfalls for misdiagnosis include the presence of atypical epithelial hyperplasia in some cases of cystosarcoma phyllodes, along with occasional cases having patterns indistinguishable from a fibroadenoma. The pleomorphic and bizarre cellular features can suggest the diagnosis of metaplastic and pure sarcomas of the breast, although the potential exists for confusion with very poorly differentiated carcinoma. FNA diagnosis of sarcomatous lesions of the breast is essential in order to insure proper surgical treatment.     

Breast carcinoma simulating fibroadenoma or fibrocystic change by fine-needle aspiration. A study of 16 cases.

The cytologic features of fine-needle aspiration specimens from 16 breast carcinomas that closely simulated benign lesions were analyzed and compared to smears from fibroadenomas and fibrocystic change. No combination of features was found that accurately separated all benign and malignant cases. Many nuclei with discernible small nucleoli in smears with many single bipolar nuclei indicated a benign lesion, whereas nuclear hyperchromasia indicated a malignant one.     

Expression of Sialyl-Tn in fine-needle aspirates from mammographically detected breast lesions: A marker of malignancy?

Malignant transformation is frequently associated with abnormal expression of cell surface carbohydrates. Sialyl-Tn (STn) is a core carbohydrate antigen of tumor-associated mucin formed by the premature 2-6 sialylation of N-acetylgalactosamine. In an attempt to verify whether this antigen is restricted to malignant cells, we studied 30 cases of fine-needle aspiration (FNA) cytology from mammographically detected breast lesions. The rationale for choosing this material was the acknowledged difficulty in diagnosing cytologically small breast lesions, especially epithelial intraductal proliferations. The cases were divided in benign lesions (two fibroadenomas and ten ductal hyperplasias) and malignant lesions (16 ductal carcinomas). Ten of sixteen malignant cases (62.5%) were positive for STn. Five of fourteen benign cases (35.7%) were also positive for STn (two fibroadenomas and three ductal hyperplasias). The most consistent positive results in benign lesions resulted from cases that displayed apocrine metaplasia, although positivity has also been observed in ductal cells without metaplasia. We did not find statistical significant differences among STn expression in benign and malignant breast lesions detected by FNA (P = 0.14). Thus, we conclude that STn is neither specific nor sensitive for detection of malignancy in FNA from mammographically detected breast lesions. Diagn. Cytopathol. 1998;18:325–329. © 1998 Wiley-Liss, Inc.     

Fine-needle aspiration biopsy of invasive micropapillary carcinoma of the breast: a report of five cases

Invasive micropapillary carcinoma of the breast is an uncommon variant of infiltrating ductal carcinoma. Observing its distinctive cytologic appearance and aggressive behavior is important for early diagnosis by fine-needle aspiration cytology (FNAC). There are only a few reported cases in the literature. Five women presented with breast masses. FNAC showed malignant epithelial tumors, and mastectomy materials showed invasive micropapillary carcinoma for all of them. Three patients had axillary lymph node metastases. Invasive micropapillary carcinoma, with its angulated papillary clusters lacking a fibrovascular core, and irregular crowded nuclei, has a distinctive cytologic appearance which correlates with its histological features. A differential diagnosis from other primary or metastatic papillary lesions of the breast may be possible using immunohistochemistry and some cytologic features. The limited experience with invasive micropapillary carcinoma should not discourage others from undertaking further studies.     

Comparative features of ductal carcinoma in situ and infiltrating ductal carcinoma of the breast on fine-needle aspiration biopsy

To evaluate the usefulness of fine-needle aspiration biopsy of the breast in separating ductal carcinoma in situ (DCIS) from infiltrating ductal carcinoma, the authors reviewed 16 preoperative fine-needle aspiration biopsies from biopsy-proven exclusive DCIS and 39 fine-needle aspiration biopsies from infiltrating ductal carcinomas with or without an in situ component. Seven (44%) of the DCIS and eight (21%) of the infiltrating ductal carcinomas had inadequate material for diagnosis on the aspiration biopsy. Five (32%) of the DCIS and 29 (74%) of the infiltrating ductal carcinomas caused suspicion or had positive results for malignancy. Four (25%) of the DCIS and two (5%) of the infiltrating ductal carcinomas showed atypical cells. Morphologic features of the atypical or malignant cells in the adequate specimens from these two lesions were similar except that the cells from the infiltrating ductal carcinomas showed more irregular nuclear spacing (94% vs. 44%, P less than 0.01) and more pronounced nuclear overlapping (65% vs. 33%) than those from the DCIS. In addition, the fine-needle aspiration biopsies of the DCIS tended to be hypocellular (less than 10 cells/10X) (44% vs. 6.5%, P less than 0.05) and to contain benign epithelial cells (22% vs. 6.5%) and macrophages (33% vs. 13%). Although the suspicion of DCIS might be raised when hypocellularity, benign epithelial cells, and macrophages are noted in a fine-needle aspiration biopsy of the breast that has positive results or causes suspicion for malignancy, fine-needle aspiration biopsy cannot be relied upon to distinguish DCIS from infiltrating ductal carcinoma.     

Nucleolar organizer regions in FNAC smears of breast lesions.

The argyrophil technique for staining proteins associated with nucleolar organizing regions was applied to the fine-needle aspiration cytology smears of 45 cases of breast disease. These included 14 malignant and 31 benign lesions. A correlation with histology sections was done in 12 cases. There was a significant difference between AgNOR count of benign and malignant breast disease with one case out of 45 falling in the "gray" zone of overlap. Infiltrating lobular carcinoma was found to have the lowest NOR counts among the malignant lesions. The smear staining was superior to that on sections, the positive features being lack of background staining and better dispersion of NORs in the nucleoli.     

Interpretation of diagnostic cytology with likelihood ratios

Sensitivity, specificity, and predictive values are commonly used to report the performance characteristics of a diagnostic test. With tests that by their nature have more than two classes of results, calculation of sensitivity and specificity demands the combination of categories into two classes, leading to a loss of diagnostic information. Fine-needle aspiration cytology of the breast is usually reported using four categories: definitely malignant, suspicious, benign, and unsatisfactory. We used likelihood ratios to show that fine-needle aspiration cytology of the breast appreciates each of these four categories fully. This approach is preferred above the usual one, which combines "definitely malignant" and "suspicious" categories as abnormal and often discards unsatisfactory aspirations from calculation. Knowledge of the prior probability of breast cancer and the likelihood ratio for a particular class of results of fine-needle aspiration cytology of the breast allows for easy calculation of the posttest probability of the disease. Application of this method should lead to more rational support from fine-needle aspiration cytology for clinical decision making.     

Fine-needle aspiration cytology of hematopoietic lesions from multiple sites

We reviewed 130 fine-needle aspiration (FNA) biopsies from 118 patients with a variety of benign and malignant hematopoietic lesions. There were 74 (57%) malignant, 45 (35%) benign, and 11 (8%) atypical diagnoses. Immunocytochemistry of the aspirated material was performed in 47 (36%) and electron microscopy in 4 (3%) of the cases. FNA cytology was utilized to make a primary hematopoietic malignant diagnosis in approximately half of the cases and to confirm recurrence in the remainder. The malignant cases included non-Hodgkin's lymphoma. Hodgkin's disease, medullary and extramedullary plasmacytoma, and granulocytic sarcoma. Forty-two malignant cases had either previous or follow-up surgical biopsy with no false-positive diagnoses. Of the 11 atypical cases, seven had surgical confirmation with five malignant and two benign diagnoses. The benign hematopoietic lesions correctly identified included acute and chronic lymphadenitis, granulomatous processes, and eosinophilic granuloma. Only 5 of the 45 benign FNA biopsies had surgical pathology follow-up, with no false-negative diagnoses. The most commonly aspirated sites were lymph nodes (71%), although hematopoietic lesions were correctly identified in a number of extranodal locations, including soft tissue (8%), abdominal viscera (6%), lungs (5%), mediastinum (2.5%), bone (3%), and thyroid, salivary gland, and breast (1.5% each). This study demonstrates the clinical utility and diagnostic accuracy of FNA cytology in the evaluation of benign and malignant hematopoietic disorders from multiple sites. Ancillary studies performed on the aspirated material aided in making a specific and accurate diagnosis.     

ThinPrep vs. conventional smear cytologic preparations in analyzing fine-needle aspiration specimens from palpable breast masses.

Limited data exist concerning the cellular features of the ThinPrep (Cytyc Corp., Boxborough, MA) technique in the analysis of breast fine-needle aspiration specimens. Therefore, we analyzed a series of 75 surgically excised palpable breast masses and compared ThinPrep and conventional smear fine-needle aspiration preparations. Each mass was aspirated twice. The first sample was used for two alcohol-fixed conventional smears, and the second sample was rinsed into CytoLyt (Cytyc Corp., Boxborough, MA) solution for processing into a ThinPrep slide. The paired slides were separated and independently analyzed for adequacy, overall cellularity, single epithelial cells (absent, rare, moderate, or numerous), epithelial architecture (sheets or three-dimensional clusters), myoepithelial cells and stripped bipolar nuclei (present or absent), and nuclear detail (poor, satisfactory, or excellent). Each sample was classified as negative, negative consistent with fibroadenoma, atypical favoring benign, atypical favoring malignant, or positive for malignant cells. The 75 breast masses included 32 carcinomas and 43 benign lesions. Four conventional smears and one ThinPrep were unsatisfactory. Significantly, more conventional smears were limited by drying artifact (9 vs. 0). ThinPrep aspirates of carcinomas had better nuclear detail (P = 0.03) and greater cellularity (P = 0.05). ThinPrep aspirates of benign masses had greater epithelial cellularity (P = 0.007) and better nuclear detail (P < 0.001), and more specimens had myoepithelial cells (P = 0.007). The ThinPrep interpretation classified 29 of 32 carcinomas (91%) as positive and three as atypical favoring malignant (sensitivity = 100%). The conventional smear interpretation classified 28 of 31 carcinomas (90%) as positive and three as atypical favoring malignant (sensitivity = 100%). The ThinPrep interpretation classified 42 benign lesions as negative (23 cases), negative consistent with fibroadenoma (8 cases), atypical favoring benign (10 cases), and atypical favoring malignant (1 case) (specificity = 74%). The conventional smear interpretation classified 40 benign lesions as negative (25 cases), negative consistent with fibroadenoma (12 cases), and atypical favoring benign (3 cases) (specificity = 93%). ThinPrep was less specific, but the difference was not statistically significant (P = 0.065). In summary, ThinPrep aspirates had greater cellularity and better nuclear detail than conventional smears, and were just as sensitive in identifying the carcinomas. The difference in specificity between the two techniques was not statistically significant (P = 0.065). Diagn. Cytopathol. 1999;21:137-141.     

Fine-needle aspiration cytology of malignant lymphoma of the thyroid

Our experience with fine-needle aspiration cytology in five cases of malignant lymphoma of the thyroid is reported. Only one case was correctly diagnosed as malignant lymphoma from the original cytologic reports. Three cases were misdiagnosed as anaplastic carcinoma of small-cell type and two of these three cases were subsequently revised as malignant lymphoma after a second aspiration. The final case was misdiagnosed as chronic thyroiditis at both the first and second aspirations; following a third aspiration, 8 mo later, it was correctly diagnosed as malignant lymphoma. Cytological findings of malignant lymphoma are as follows: many malignant cells are distributed as isolated cells and show monotonous features. Malignant cells are slightly larger than normal lymphocytes. Sometimes cleaved cells or large nucleoli are visible. The differential diagnosis of cytologic findings of malignant lymphoma and the other thyroid diseases is discussed in detail.     

Fractal signature and lacunarity in the measurement of the texture of trabecular bone in clinical CT images

Fractal analysis is a method of characterizing complex shapes such as the trabecular structure of bone. Numerous algorithms for estimating fractal dimension have been described, but the Fourier power spectrum method is particularly applicable to self-affine fractals, and facilitates corrections for the effects of noise and blurring in an image. We found that it provided accurate estimates of fractal dimension for synthesized fractal images. For natural texture images fractality is limited to a range of scales, and the fractal dimension as a function of spatial frequency presents as a fractal signature. We found that the fractal signature was more successful at discriminating between these textures than either the global fractal dimension or other metrics such as the mean width and root-mean-square width of the spectral density plots. Different natural textures were also readily distinguishable using lacunarity plots, which explicitly characterize the average size and spatial organization of structural sub-units within an image. The fractal signatures of small regions of interest (32x32 pixels), computed in the frequency domain after corrections for imaging system noise and MTF, were able to characterize the texture of vertebral trabecular bone in CT images. Even small differences in texture due to acquisition slice thickness resulted in measurably different fractal signatures. These differences were also readily apparent in lacunarity plots, which indicated that a slice thickness of 1 mm or less is necessary if essential architectural information is not to be lost. Since lacunarity measures gap size and is not predicated on fractality, it may be particularly useful for characterizing the texture of trabecular bone.     

Mucocele-like tumors of the breast. Cytologic findings in two cases.

Mucocele-like tumors of the breast originally were reported by Rosen in 1986 as benign lesions that histologically resembled colloid carcinoma of the breast. The authors document two cases of mucocele-like tumors to illustrate the difficulty in separating these lesions from colloid carcinoma on the basis of fine-needle aspiration biopsy. Cytologically, mucocele-like tumors contained abundant mucin, few clusters, and sheets of regular epithelium that lacked nuclear atypia, and they contained no intact single cells. The authors recommend open surgical biopsy when fine-needle aspiration biopsy findings in such cases are equivocal.     

Ductal adenoma of the breast: The cytological features of six cases

Ductal adenoma of the breast may simulate malignancy by both clinical, histological, and cytological examination. Fine-needle aspirations from breast lesions are now daily routine and preoperative recognition of this benign lesion is desirable. This study describes the cytologic features of six additional cases of ductal adenoma. The smears from all the lesions were highly cellular with epithelial cells in sheets. Numerous large fragments of purple stroma in tight connection with epithelial cells often making finger-like hyaline structures or globules between cells were seen. Naked oval nuclei in the background were observed in all cases indicating benignancy. We find the cytologic picture in fine-needle aspirations from ductal adenomas sufficiently characteristic for preoperative diagnosis. The differential diagnoses toward other benign and malignant conditions are discussed. © 1994 Wiley-Liss, Inc.     

Cytospins—an alternative method for fine-needle aspiration cytology of the breast: A study of 148 cases

The aim of this study was to evaluate the cytospin technique as an alternative method to prepare fine-needle aspiration (FNA) specimens of the breast. To do so, the cytology of 148 breast FNAs that had been prepared by the cytospin technique and that had histologic correlation, was reviewed. All the cases that were diagnosed as malignant by cytology were proved malignant after surgical excision, and there were no false-positive results. All but two cases diagnosed as benign by cytology proved to be benign on excision. The two false-negative cases were missed due to sampling error. The cytological features seen on cytospins were similar to those seen on conventional direct smears. The major advantage of this method is that no aspirate is unsatisfactory due to unskilled direct smear technique. This, along with its good correlation with histology, proves that the cytospin method is an effective alternative to conventional direct smears for breast FNA. © 1995 Wiley-Liss, Inc.