Fibroadenoma of the Breast: Diagnostic Pitfalls of Fine-Needle Aspiration Janine L. Benoit, M.D., Rosemin Kara, ART., S. Elizabeth McGregor, MSC., and Maire A. Duggan,
On fine-needle aspiration (FNA), fibroadenomas have a characteristic cytological appearance, although occasional cases are misinterpreted as carcinomas and vice versa. In a review of 521 breast aspirates correlated with the subsequent histology, six of 87fibroadenomas (7%) were malignant or suspicious of malignancy on FNA valse positives). Following cytological review,four were still suspicious of malignancy because of cellular dyscohesion and prominent nucleoli, while two were fibroadenomas. On FNA, four of 145 carcinomas (3 %) were diagnosed as fibroadenomas valse negatives). On review, three were malignant or suspicious of malignancy, while one was consistent with a fibroadenoma. Three false negative diagnoses were due to underappreciation of single malignant cells present between epithelial groupings typical of a fibroadenoma, while one was due to undersampling of the carcinoma. Cytologically, some fibroadenomas are sufficiently atypical that histological conjrmation is necessary to exclude a malignancy. Misinterpreting carcinomas as fibroadenomas could be avoided by careful study of the morphology of isolated cells. Diagn Cytopathol 1992;8:643-648. @ 1992 Wiley-Liss, Inc. Key Words: Fine-needle aspiration; Fibroadenomas; Cytology; Breast; False-positive diagnosis; False-negative diagnosis
The complete acceptance of fine-needle aspiration (FNA) cytology has been hampered by concerns regarding its diagnostic accuracy. This reluctance can be surmounted by identifying problematic cases and clarifying the diagnostic pitfalls. On FNA, fibroadenomas of the breast have a characteristic cytological appearance. The microscopic criteria are well defined and usually pose little diagnostic difficulty. Nevertheless, occasional cases are misinterpreted as carcinomas (false-positive cases) and conversely, some carcinomas are misinterpreted as fibroadenomas (false-negative cases). The current study was carried out to further clarify pitfalls in the cytological diagnosis of breast fibroadenomas by correlating the cytological features with the histological changes.
'
M.B., FRCPC
Materials and Methods A computer search of the cytology files at the Foothills Hospital retrieved 1,068 cases of breast FNA performed between January 1984 and May 1990. Several surgeons, physicians, and residents in training performed the aspirations. Patients were referred for further investigation of a palpable breast lump, indurated area or mammographic abnormality which was aspirated with a 10-ml syringe and either a 20-, 22-, or 25-gauge needle. The aspirated material was smeared on one or more glass slides, fixed with cytospray (Canlab Toronto, Canada) and stained by a modified Papanicolaou method. Cytological features were summarized as either (1) benign, ( 2 ) malignant, (3) suspicious for malignancy, or (4) inadequate, using published criteria. A search of the histopathology files determined that 52 1 (49%) cases had histological confirmation of the aspirated lesion, performed at the same hospital within 1 yr of the aspirate. All tissue sections were fixed in 10% phosphatebuffered formalin, routinely processed, and stained with haematoxylin and eosin (H&E). The cytological and histological diagnoses were correlated to determine the number of false-positive cases, suspicious cases that were falsely positive, and false-negative cases. Fibroadenomas that were misinterpreted as carcinomas and vice versa were selected from this group. The small study size precluded any statistical testing. A detailed cytological and histological review of these cases were conducted. For this study, a cell group was defined as having at least six epithelial cells. The cellularity of the aspirates at X 40 magnification was quantified as follows:
Cellularity: High Moderate Low
1 2 0 groups
1 2- < 20 groups < 2 groups
~
Received September 2, 1991. Accepted April 20, 1992. From the Department of Pathology, Foothills Hospital and the University of Calgary, and the Division of Epidemiology and Preventive Oncology, Tom Baker Cancer Centre, Calgary, Alberta, Canada. Address reprint requests to Dr. M.A. Duggan, Department of Pathology, Foothills Hospital, 1403 29th Street N.W., Calgary, Alberta T2N 2Y9. Canada. c~
\992 WILEY-LISS, INC
Results By histological examination, there were 87 fibroadenomas and 159 carcinomas. There were seven false positive, 20 suspicious false positive and 25 false negative cases following cytohistological correlation. From these cases, six Diagnostic Cytopathology, Vol 8, No 6
643
BENOIT ET AL.
size from 0.8 to 1.5 cm, with a mean and median of 1.2 cm. All four cases were originally diagnosed as fibroadenomas. Cytologically, dyscohesive cell groupings were seen in one case while the remainder had cohesive groupings (Fig. 3). Naked nuclei resembling myoepithelial cells were seen in half of the cases. Nucleoli were prominent in two cases and individual malignant cells were identified in the same two cases. In addition, one of these cases demonstrated mucous containing cells resembling goblet cells. One case was thickly smeared but tubular structures not typically seen in fibroadenomas were identified. On cytological review, two cases were malignant, one was suspicious for malignancy and one resembled a fibroadenoma. The histological review confirmed malignancy in all four cases. Two cases were typical ductal carcinomas, while one case was a mixed tubular and ductal carcinoma, and another was an apocrine carcinoma.
(7%) fibroadenomas were diagnosed on FNA as malignant or suspicious of malignancy (false positive cases), and four (3%) carcinomas were diagnosed as fibroadenomas (false negative cases). All the patients were female and none were pregnant.
fialse-Positive Cases (Table I) The six patients ranged in age from 35 to 44 yr, with a mean (median) of 40 (39) yr, and all were premenopausal. Data on size were available for five tumors, which ranged from 0.4 to 1.0 cm with a mean (median) of 0.7 (0.8) cm. Originally, one case was called malignant, while the remaining five were reported as suspicious for malignancy. Three of the suspicious cases were noted as having some features consistent with a fibroadenoma. Five cases were moderately cellular, and one was highly cellular. Groups of ductal cells were tightly aggregated in four cases and dyscohesive in two. Naked nuclei typical of myoepithelial cells were present in four and absent in two cases, and nucleoli were prominent in four and inconspicuous in two cases. Following cytological review, the diagnoses were revised. Two were fibroadenomas based on the presence of tightly aggregated groups of ductal cells and myoepithelial cells. Four cases were reinterpreted as suspicious for malignancy because of either cellular dyscohesion and/or nucleolar prominence (Fig. 1). Histologically, all of the cases in this group were fibroadenomas. Apocrine metaplasia was found in four and two demonstrated epithelial hlyperplasia without atypia (Fig. 2 ) .
Discussion Consequent to the increasing popularity of breast FNA, difficulties associated with a diagnosis of fibroadenoma are being recognized. 4,5 In this study, although the small number of cases did not allow a meaningful statistical analysis, clinical and cytological clues that could avoid or diminish the over and underreporting of fibroadenomas were identified. Attention to the patient's age and size of the breast mass could be of some help in distinguishing the two lesions, as the carcinoma group was slightly older, i.e., 43 vs. 40 years, and the mean (median) size of the carcinomas was almost twice that of the fibroadenomas, i.e., 1.2 cm vs. 0.7 cm. Although carcinomas and fibroadenomas occur in any age group and are of variable fibroadenoma should be cautiously diagnosed in an older female with a breast mass greater than 1.2 cm in diameter.
False-Negative Cases (Table II) The four patients ranged in age from 35 to 50 yr, with a rnean (median) of 43 (44) yr. Three were premenopausal, and one was perimenopausal. The four tumors ranged in
Table I.
False-Positive Cases: Cytological and Histological Features
Case No.
1
2
3
4
5
6
Age (yr) Size (cm)
43
35
37
37
42
44
0.7
NR
0.4
0.8
0.8
1.O
Cytological diagnosis
Malignant
Suspicious for malignancy
Suspicious for malignancy'
Suspicious for malignancy"
Suspicious for malignancy"
Suspicious for malignancy
Moderate Aggregated Absent Prominent
Moderate Aggregated Present Prominent
Moderate Disaggregated Present Inconspicuous
Moderate Aggregated Absent Prominent
High Disaggregatedb Present Prominent
Moderate Aggregated Present Inconspicuous
Review diagnosis
Suspicious for malignancy
Fibroadenoma
Suspicious for malignancya
Suspicious for malignancya
Suspicious for malignancya
Fibroadenoma
Histological diagnosis Histological features: Apocrine metaplasia Epithelial hyperplasia lntranuclear inclusions
Fibroadenoma
Fibroadenoma
Fibroadenoma
Fibroadenoma
Fibroadenoma
Fibroadenoma
Present Present Absent
Absent Absent Present
Present Absent Present
Present Absent Present
Present Present Present
Absent Absent Present
Cytological features Cellularity Cellular aggregation Naked nuclei Nucleoli
'With some features of fibroadenoma. bMostly aggregated groups.
644
Diagnostic Cytoputhology, Vol 8, No 6
FIBROADENOMA OF THE BREAST
Fig. 1. False-positive cases. A: FNA showing cellular dyscohesion (Papanicolaou, cell groups (Papanicolaou, ~ 4 0 0 ) .
x 250). B: FN A demonstrating nucleolar prominence in several
Fig. 2. False-positive cases. A Apocrine metaplasia within a fibroadenoma is arrowed (H&E, x 250). R: Areas of epithelial hyperplasia without atypia within a fibroadenoma (H&E, X400). Diagnostic Cytopaihology, Vol 8, N o 6
645
BENOIT ET AL. Table 11. False-Negative Cases: Cytological and Histological Features
-
Case No.
7
8
9
10
Age (yr)
50
44
35
43
Size (cm)
0.8
0.9
1.5
1.5
Cytological diagnosis Cytological features Cellularity Cellular aggregation Naked nuclei Nucleoli Other features
Fibroadenoma
Fibroadenoma
Fibroadenoma
Fibroadenoma
High Disaggregated Absent Prominent Individual malignant cells"
High Aggregated Present Prominent Mucus-containing cells" Individual malignant cells
High Aggregated Absent Inconspicuous Tubular structures
Low Aggregated Present Inconspicuous
Review diagnosis
Ductal carcinoma
Lobular carcinoma
Suspicious for malignancy'
Fibroadenoma
Histological diagnosis
Ductal carcinoma
Mixed tubular and ductal carcinoma
Apocrine carcinoma
Ductal carcinoma
Present
Present Mucus-containing cells
Present
Present FNA does not reflect biomv'
Histological features Features of malignancy Other features
-
-
'Features of a fibroadenoma hThick smear. 'Sampling problem.
F'ig. 3. False-negative cases. A: Malignant cells with prominent nucleoli seen in dyscohesive cell groupings (Papanicolaou, x 400). B: Malignant cells forming a cohesive group and mimicking the features of a fibroadenoma. A mucus-containing cell is arrowed (Papanicolaou, ~ 2 5 0 ) .
646
Diagnostic Cytoputhology, Vol 8, No 6
FIBROADENOMA OF THE BREAST
By FNA, fibroadenomas characteristically yield cellular smears composed of antler-shaped, monolayered groups of cohesive ductal cells, stromal tissue fragments, and single naked nuclei representing myoepithelial cells. Occasionally, only two of the three elements are present. Little reference has been made to the presence of cytological atypia in aspirates of fibroadenomas, although it is a somewhat frequent histological finding. In the current study, atypia in fibroadenomas was a feature of the older patient (mean age = 40 yr), and was occasionally of sufficient degree that it mimicked a carcinoma. Similar findings were reported by Stanley et aL4 The cause of the cytological atypia is unknown but is believed to be multifactorial and predominantly hormonal. Hormonal factors could not be implicated in this study, as all the patients were premenopausal and none were pregnant. The cytological interpretation of breast aspirate findings is generally extremely conservative because of the grave consequences of a misdiagnosis. In this study, onehalf of the false-positive cases had the usual features of a fibroadenoma. However, these cases plus one other were suspicious for malignancy because of nucleolar prominence in aggregated groups of ductal cells that also showed some cellular dyscohesion. Following cytohistological correlation, nucleolar prominence and cellular dyscohesion were seen to be a result of apocrine metaplasia and epithelial hyperplasia respectively. These correlations were also noted by Stanley et aL4 Although heightened awareness of these cytological pitfalls could prevent future misdiagnoses, the more cautious approach is to recommend biopsy of cytologically atypical fibroadenomas so as to exclude a malignancy, and the rare instance of a carcinoma arising in a fibroadenoma. l o The diagnosis of breast carcinoma by FNA is rarely problematic. The highly cellular smears are composed of groups of markedly dyscohesive cells, and individual malignant cells are nearly always present. False-negative diagnoses are mostly due to undersampling of a small neoplasm. Although some carcinomas have a benign cytological appearance, e.g., tubular carcinoma and apocrine carcinoma, ‘ , I 2 this study highlighted another area of cytological mimicry, that of carcinomas resembling fibroadenomas. The resemblance was a low to medium power phenomenon because of tightly aggregated groups of cells and naked nuclei. On review, however, two cases had frank features of malignancy because of the presence of individual malignant cells, and one was architecturally suspicious for malignancy because tubular structures are not generally seen in fibroadenomas. It was noteworthy that one tumour was a mixed tubular and ductal carcinoma, and one an apocrine carcinoma. These types of carcinomas will continue to be diagnostically challenging by virtue of their bland cytology. However, stricter attention to the criteria necessary for diagnosing fibroadenoma,
and more careful study of the morphology of individual cells at high power could avoid future misdiagnoses. The study also showed that the cellularity of the smear could be helpful in separating carcinoma from fibroadenoma. In contrast to the four carcinomas, which were mostly hypercellular, the fibroadenomas were mostly moderately cellular. The discriminating power of this parameter needs to be studied in a larger series of cases, however. In summary, this study serves to expand the literature on the diagnostic pitfalls of fibroadenomas and some carcinomas by FNA. Although the patient age, size of the lesion and cellularity of the smear could prevent some erroneous diagnoses, the cytological findings are of paramount importance. Some fibroadenomas are sufficiently atypical that biopsy will be necessary to exclude malignancy. Misinterpreting some carcinomas as fibroadenomas could be avoided by more careful cytological interpretation.
Acknowledgments The authors would like to thank Michael White for his photographic expertise and Alannah Ireland for her secretarial assistance.
References 1. Layfield LI, Glasgow BJ, Cramer H. Fine needle aspiration in management of breast masses. Pathol Annu 1989;24(2):23-62. 2. Bottles K, Chan JS, Holly EA, Chiu S-H, Miller TR. Cytologic criteria for fibroadenoma: a step-wise logistic regression analysis. Am J Clin Pathol 1988;89:707-13. 3. Oertel YC, Galblum LI. Fine needle aspiration of the breast: diagnostic criteria. Pathol Annu 1983; 18(1): 375-409. 4. Stanley MW, Edneia MT, Skoog L. Fine needle aspiration of fibroadenomas of the breast with atypia: a spectrum including cases that cytologically mimic carcinoma. Diagn Cytopathol 1990; 6: 375-82. 5 . Patel JJ, Gartell PC, Smallwood JA, et al. Fine needle aspiration cytology of breast masses: an evaluation of its accuracy and reasons for diagnostic failure. Ann R Coll Physicians Engl 1987;69: 156-9. 6. Rosai JR. Ackerman’s surgical pathology. St. Louis: CV Mosby, 1989: 1217. 7. Wilkinson S, Anderson TJ, Rifkind E, et al. Fibroadenoma of the breast: A follow-up of conservative management. Br J Surg 1989:76: 390-1. 8. Rosai JR. Ackerman’s surgical pathology. St. Louis: CV. Mosby, 1989: 1200. 9. Stanley MW, Edneia MT, Skoog L. Fine needle aspiration of fibroadenomas of the breast with atypia: a spectrum including cases that cytologically mimic carcinoma. Diagn Cytopathol 1990 6: 375-82. 10. Pick PW, Iossifides IA. Occurrence of breast carcinoma within a fibroadenoma: a review. Arch Pathol Lab Med 1984; 108: 590-4. 11. Barrows GH, Anderson TJ, Lamb JL, Dixon JM. Fine needle aspiration of breast cancer. Cancer 1986; 58: 1493-8. 12. Kline TS, Kline IK. Guides to clinical aspiration Biopsy-breast. New York: Igaku-Shoin, 1989: 32-3.
Editorial Comments Fine-needle aspiration of the breast is now a widely used procedure. The typical cytologic features of fibroadenoma Diagnostic Cytopathology, Vol 8, No 6
641
On fine-needle aspiration (FNA), fibroadenomas have a characteristic cytological appearance, although occasional cases are misinterpreted as carcinomas and vice versa. In a review of 521 breast aspirates correlated with the subsequent histology, six of 87fibroadenomas (7%) were malignant or suspicious of malignancy on FNA valse positives). Following cytological review,four were still suspicious of malignancy because of cellular dyscohesion and prominent nucleoli, while two were fibroadenomas. On FNA, four of 145 carcinomas (3 %) were diagnosed as fibroadenomas valse negatives). On review, three were malignant or suspicious of malignancy, while one was consistent with a fibroadenoma. Three false negative diagnoses were due to underappreciation of single malignant cells present between epithelial groupings typical of a fibroadenoma, while one was due to undersampling of the carcinoma. Cytologically, some fibroadenomas are sufficiently atypical that histological conjrmation is necessary to exclude a malignancy. Misinterpreting carcinomas as fibroadenomas could be avoided by careful study of the morphology of isolated cells. Diagn Cytopathol 1992;8:643-648. @ 1992 Wiley-Liss, Inc. Key Words: Fine-needle aspiration; Fibroadenomas; Cytology; Breast; False-positive diagnosis; False-negative diagnosis
The complete acceptance of fine-needle aspiration (FNA) cytology has been hampered by concerns regarding its diagnostic accuracy. This reluctance can be surmounted by identifying problematic cases and clarifying the diagnostic pitfalls. On FNA, fibroadenomas of the breast have a characteristic cytological appearance. The microscopic criteria are well defined and usually pose little diagnostic difficulty. Nevertheless, occasional cases are misinterpreted as carcinomas (false-positive cases) and conversely, some carcinomas are misinterpreted as fibroadenomas (false-negative cases). The current study was carried out to further clarify pitfalls in the cytological diagnosis of breast fibroadenomas by correlating the cytological features with the histological changes.
'
M.B., FRCPC
Materials and Methods A computer search of the cytology files at the Foothills Hospital retrieved 1,068 cases of breast FNA performed between January 1984 and May 1990. Several surgeons, physicians, and residents in training performed the aspirations. Patients were referred for further investigation of a palpable breast lump, indurated area or mammographic abnormality which was aspirated with a 10-ml syringe and either a 20-, 22-, or 25-gauge needle. The aspirated material was smeared on one or more glass slides, fixed with cytospray (Canlab Toronto, Canada) and stained by a modified Papanicolaou method. Cytological features were summarized as either (1) benign, ( 2 ) malignant, (3) suspicious for malignancy, or (4) inadequate, using published criteria. A search of the histopathology files determined that 52 1 (49%) cases had histological confirmation of the aspirated lesion, performed at the same hospital within 1 yr of the aspirate. All tissue sections were fixed in 10% phosphatebuffered formalin, routinely processed, and stained with haematoxylin and eosin (H&E). The cytological and histological diagnoses were correlated to determine the number of false-positive cases, suspicious cases that were falsely positive, and false-negative cases. Fibroadenomas that were misinterpreted as carcinomas and vice versa were selected from this group. The small study size precluded any statistical testing. A detailed cytological and histological review of these cases were conducted. For this study, a cell group was defined as having at least six epithelial cells. The cellularity of the aspirates at X 40 magnification was quantified as follows:
Cellularity: High Moderate Low
1 2 0 groups
1 2- < 20 groups < 2 groups
~
Received September 2, 1991. Accepted April 20, 1992. From the Department of Pathology, Foothills Hospital and the University of Calgary, and the Division of Epidemiology and Preventive Oncology, Tom Baker Cancer Centre, Calgary, Alberta, Canada. Address reprint requests to Dr. M.A. Duggan, Department of Pathology, Foothills Hospital, 1403 29th Street N.W., Calgary, Alberta T2N 2Y9. Canada. c~
\992 WILEY-LISS, INC
Results By histological examination, there were 87 fibroadenomas and 159 carcinomas. There were seven false positive, 20 suspicious false positive and 25 false negative cases following cytohistological correlation. From these cases, six Diagnostic Cytopathology, Vol 8, No 6
643
BENOIT ET AL.
size from 0.8 to 1.5 cm, with a mean and median of 1.2 cm. All four cases were originally diagnosed as fibroadenomas. Cytologically, dyscohesive cell groupings were seen in one case while the remainder had cohesive groupings (Fig. 3). Naked nuclei resembling myoepithelial cells were seen in half of the cases. Nucleoli were prominent in two cases and individual malignant cells were identified in the same two cases. In addition, one of these cases demonstrated mucous containing cells resembling goblet cells. One case was thickly smeared but tubular structures not typically seen in fibroadenomas were identified. On cytological review, two cases were malignant, one was suspicious for malignancy and one resembled a fibroadenoma. The histological review confirmed malignancy in all four cases. Two cases were typical ductal carcinomas, while one case was a mixed tubular and ductal carcinoma, and another was an apocrine carcinoma.
(7%) fibroadenomas were diagnosed on FNA as malignant or suspicious of malignancy (false positive cases), and four (3%) carcinomas were diagnosed as fibroadenomas (false negative cases). All the patients were female and none were pregnant.
fialse-Positive Cases (Table I) The six patients ranged in age from 35 to 44 yr, with a mean (median) of 40 (39) yr, and all were premenopausal. Data on size were available for five tumors, which ranged from 0.4 to 1.0 cm with a mean (median) of 0.7 (0.8) cm. Originally, one case was called malignant, while the remaining five were reported as suspicious for malignancy. Three of the suspicious cases were noted as having some features consistent with a fibroadenoma. Five cases were moderately cellular, and one was highly cellular. Groups of ductal cells were tightly aggregated in four cases and dyscohesive in two. Naked nuclei typical of myoepithelial cells were present in four and absent in two cases, and nucleoli were prominent in four and inconspicuous in two cases. Following cytological review, the diagnoses were revised. Two were fibroadenomas based on the presence of tightly aggregated groups of ductal cells and myoepithelial cells. Four cases were reinterpreted as suspicious for malignancy because of either cellular dyscohesion and/or nucleolar prominence (Fig. 1). Histologically, all of the cases in this group were fibroadenomas. Apocrine metaplasia was found in four and two demonstrated epithelial hlyperplasia without atypia (Fig. 2 ) .
Discussion Consequent to the increasing popularity of breast FNA, difficulties associated with a diagnosis of fibroadenoma are being recognized. 4,5 In this study, although the small number of cases did not allow a meaningful statistical analysis, clinical and cytological clues that could avoid or diminish the over and underreporting of fibroadenomas were identified. Attention to the patient's age and size of the breast mass could be of some help in distinguishing the two lesions, as the carcinoma group was slightly older, i.e., 43 vs. 40 years, and the mean (median) size of the carcinomas was almost twice that of the fibroadenomas, i.e., 1.2 cm vs. 0.7 cm. Although carcinomas and fibroadenomas occur in any age group and are of variable fibroadenoma should be cautiously diagnosed in an older female with a breast mass greater than 1.2 cm in diameter.
False-Negative Cases (Table II) The four patients ranged in age from 35 to 50 yr, with a rnean (median) of 43 (44) yr. Three were premenopausal, and one was perimenopausal. The four tumors ranged in
Table I.
False-Positive Cases: Cytological and Histological Features
Case No.
1
2
3
4
5
6
Age (yr) Size (cm)
43
35
37
37
42
44
0.7
NR
0.4
0.8
0.8
1.O
Cytological diagnosis
Malignant
Suspicious for malignancy
Suspicious for malignancy'
Suspicious for malignancy"
Suspicious for malignancy"
Suspicious for malignancy
Moderate Aggregated Absent Prominent
Moderate Aggregated Present Prominent
Moderate Disaggregated Present Inconspicuous
Moderate Aggregated Absent Prominent
High Disaggregatedb Present Prominent
Moderate Aggregated Present Inconspicuous
Review diagnosis
Suspicious for malignancy
Fibroadenoma
Suspicious for malignancya
Suspicious for malignancya
Suspicious for malignancya
Fibroadenoma
Histological diagnosis Histological features: Apocrine metaplasia Epithelial hyperplasia lntranuclear inclusions
Fibroadenoma
Fibroadenoma
Fibroadenoma
Fibroadenoma
Fibroadenoma
Fibroadenoma
Present Present Absent
Absent Absent Present
Present Absent Present
Present Absent Present
Present Present Present
Absent Absent Present
Cytological features Cellularity Cellular aggregation Naked nuclei Nucleoli
'With some features of fibroadenoma. bMostly aggregated groups.
644
Diagnostic Cytoputhology, Vol 8, No 6
FIBROADENOMA OF THE BREAST
Fig. 1. False-positive cases. A: FNA showing cellular dyscohesion (Papanicolaou, cell groups (Papanicolaou, ~ 4 0 0 ) .
x 250). B: FN A demonstrating nucleolar prominence in several
Fig. 2. False-positive cases. A Apocrine metaplasia within a fibroadenoma is arrowed (H&E, x 250). R: Areas of epithelial hyperplasia without atypia within a fibroadenoma (H&E, X400). Diagnostic Cytopaihology, Vol 8, N o 6
645
BENOIT ET AL. Table 11. False-Negative Cases: Cytological and Histological Features
-
Case No.
7
8
9
10
Age (yr)
50
44
35
43
Size (cm)
0.8
0.9
1.5
1.5
Cytological diagnosis Cytological features Cellularity Cellular aggregation Naked nuclei Nucleoli Other features
Fibroadenoma
Fibroadenoma
Fibroadenoma
Fibroadenoma
High Disaggregated Absent Prominent Individual malignant cells"
High Aggregated Present Prominent Mucus-containing cells" Individual malignant cells
High Aggregated Absent Inconspicuous Tubular structures
Low Aggregated Present Inconspicuous
Review diagnosis
Ductal carcinoma
Lobular carcinoma
Suspicious for malignancy'
Fibroadenoma
Histological diagnosis
Ductal carcinoma
Mixed tubular and ductal carcinoma
Apocrine carcinoma
Ductal carcinoma
Present
Present Mucus-containing cells
Present
Present FNA does not reflect biomv'
Histological features Features of malignancy Other features
-
-
'Features of a fibroadenoma hThick smear. 'Sampling problem.
F'ig. 3. False-negative cases. A: Malignant cells with prominent nucleoli seen in dyscohesive cell groupings (Papanicolaou, x 400). B: Malignant cells forming a cohesive group and mimicking the features of a fibroadenoma. A mucus-containing cell is arrowed (Papanicolaou, ~ 2 5 0 ) .
646
Diagnostic Cytoputhology, Vol 8, No 6
FIBROADENOMA OF THE BREAST
By FNA, fibroadenomas characteristically yield cellular smears composed of antler-shaped, monolayered groups of cohesive ductal cells, stromal tissue fragments, and single naked nuclei representing myoepithelial cells. Occasionally, only two of the three elements are present. Little reference has been made to the presence of cytological atypia in aspirates of fibroadenomas, although it is a somewhat frequent histological finding. In the current study, atypia in fibroadenomas was a feature of the older patient (mean age = 40 yr), and was occasionally of sufficient degree that it mimicked a carcinoma. Similar findings were reported by Stanley et aL4 The cause of the cytological atypia is unknown but is believed to be multifactorial and predominantly hormonal. Hormonal factors could not be implicated in this study, as all the patients were premenopausal and none were pregnant. The cytological interpretation of breast aspirate findings is generally extremely conservative because of the grave consequences of a misdiagnosis. In this study, onehalf of the false-positive cases had the usual features of a fibroadenoma. However, these cases plus one other were suspicious for malignancy because of nucleolar prominence in aggregated groups of ductal cells that also showed some cellular dyscohesion. Following cytohistological correlation, nucleolar prominence and cellular dyscohesion were seen to be a result of apocrine metaplasia and epithelial hyperplasia respectively. These correlations were also noted by Stanley et aL4 Although heightened awareness of these cytological pitfalls could prevent future misdiagnoses, the more cautious approach is to recommend biopsy of cytologically atypical fibroadenomas so as to exclude a malignancy, and the rare instance of a carcinoma arising in a fibroadenoma. l o The diagnosis of breast carcinoma by FNA is rarely problematic. The highly cellular smears are composed of groups of markedly dyscohesive cells, and individual malignant cells are nearly always present. False-negative diagnoses are mostly due to undersampling of a small neoplasm. Although some carcinomas have a benign cytological appearance, e.g., tubular carcinoma and apocrine carcinoma, ‘ , I 2 this study highlighted another area of cytological mimicry, that of carcinomas resembling fibroadenomas. The resemblance was a low to medium power phenomenon because of tightly aggregated groups of cells and naked nuclei. On review, however, two cases had frank features of malignancy because of the presence of individual malignant cells, and one was architecturally suspicious for malignancy because tubular structures are not generally seen in fibroadenomas. It was noteworthy that one tumour was a mixed tubular and ductal carcinoma, and one an apocrine carcinoma. These types of carcinomas will continue to be diagnostically challenging by virtue of their bland cytology. However, stricter attention to the criteria necessary for diagnosing fibroadenoma,
and more careful study of the morphology of individual cells at high power could avoid future misdiagnoses. The study also showed that the cellularity of the smear could be helpful in separating carcinoma from fibroadenoma. In contrast to the four carcinomas, which were mostly hypercellular, the fibroadenomas were mostly moderately cellular. The discriminating power of this parameter needs to be studied in a larger series of cases, however. In summary, this study serves to expand the literature on the diagnostic pitfalls of fibroadenomas and some carcinomas by FNA. Although the patient age, size of the lesion and cellularity of the smear could prevent some erroneous diagnoses, the cytological findings are of paramount importance. Some fibroadenomas are sufficiently atypical that biopsy will be necessary to exclude malignancy. Misinterpreting some carcinomas as fibroadenomas could be avoided by more careful cytological interpretation.
Acknowledgments The authors would like to thank Michael White for his photographic expertise and Alannah Ireland for her secretarial assistance.
References 1. Layfield LI, Glasgow BJ, Cramer H. Fine needle aspiration in management of breast masses. Pathol Annu 1989;24(2):23-62. 2. Bottles K, Chan JS, Holly EA, Chiu S-H, Miller TR. Cytologic criteria for fibroadenoma: a step-wise logistic regression analysis. Am J Clin Pathol 1988;89:707-13. 3. Oertel YC, Galblum LI. Fine needle aspiration of the breast: diagnostic criteria. Pathol Annu 1983; 18(1): 375-409. 4. Stanley MW, Edneia MT, Skoog L. Fine needle aspiration of fibroadenomas of the breast with atypia: a spectrum including cases that cytologically mimic carcinoma. Diagn Cytopathol 1990; 6: 375-82. 5 . Patel JJ, Gartell PC, Smallwood JA, et al. Fine needle aspiration cytology of breast masses: an evaluation of its accuracy and reasons for diagnostic failure. Ann R Coll Physicians Engl 1987;69: 156-9. 6. Rosai JR. Ackerman’s surgical pathology. St. Louis: CV Mosby, 1989: 1217. 7. Wilkinson S, Anderson TJ, Rifkind E, et al. Fibroadenoma of the breast: A follow-up of conservative management. Br J Surg 1989:76: 390-1. 8. Rosai JR. Ackerman’s surgical pathology. St. Louis: CV. Mosby, 1989: 1200. 9. Stanley MW, Edneia MT, Skoog L. Fine needle aspiration of fibroadenomas of the breast with atypia: a spectrum including cases that cytologically mimic carcinoma. Diagn Cytopathol 1990 6: 375-82. 10. Pick PW, Iossifides IA. Occurrence of breast carcinoma within a fibroadenoma: a review. Arch Pathol Lab Med 1984; 108: 590-4. 11. Barrows GH, Anderson TJ, Lamb JL, Dixon JM. Fine needle aspiration of breast cancer. Cancer 1986; 58: 1493-8. 12. Kline TS, Kline IK. Guides to clinical aspiration Biopsy-breast. New York: Igaku-Shoin, 1989: 32-3.
Editorial Comments Fine-needle aspiration of the breast is now a widely used procedure. The typical cytologic features of fibroadenoma Diagnostic Cytopathology, Vol 8, No 6
641
