Breast Cancer DOI 10.1007/s12282-015-0624-9
ORIGINAL ARTICLE
Diagnostic value of fine needle aspiration and core needle biopsy in special types of breast cancer Ryuji Ohashi1 • Miyuki Matsubara1 • Yasuhiko Watarai1 • Keiko Yanagihara2 Koji Yamashita2 • Shin-ichi Tsuchiya1,3 • Hiroyuki Takei2 • Zenya Naito1,4
•
Received: 9 April 2015 / Accepted: 21 June 2015 Ó The Japanese Breast Cancer Society 2015
Abstract Background Although fine needle aspiration (FNA) biopsy is an established tool to assess breast lesions, there has been a trend toward using core needle biopsy (CNB) instead. The aim of this study was to compare the diagnostic accuracy of FNA and CNB in special types of breast cancer. Methods A retrospective review of diagnostic results of pre-operatively performed FNA or CNB, or a combination of the two, was conducted. The cases include histologically proven invasive ductal carcinoma of no special type (NST n = 159), invasive lobular carcinoma (ILC n = 65), mucinous carcinoma (MUC n = 51), and apocrine carcinoma (APO n = 25). Results The absolute diagnostic sensitivity of FNA to detect malignancy in ILC and APO patients was inferior to that of NST patients (p \ 0.001 for ILC and APO). Within each cancer type, the sensitivity of CNB was higher than that of FNA in the ILC and APO patients (p \ 0.001 and p \ 0.05, respectively). As for NST and MUC patients, FNA and CNB had equivalent sensitivity. The sensitivity
& Ryuji Ohashi [email protected] 1
Department of Diagnostic Pathology, Nippon Medical School Hospital, 1-1-5, Sendagi, Bunkyo-ku, Tokyo 113-8603, Japan
2
Division of Breast Surgery, Nippon Medical School Hospital, Tokyo, Japan
3
Department of Diagnostic Pathology, Iida Hospital, Nagano, Japan
4
Department of Pathology and Integrative Oncological Pathology, Nippon Medical School, Tokyo, Japan
of FNA alone significantly improved when combined with CNB in NST, ILC and APO patients (p \ 0.05, p \ 0.001, and p \ 0.05, respectively). Conclusions Our results suggest that FNA has less diagnostic accuracy than CNB for ILC and APO; thus, the use of CNB should be encouraged when these types of cancer are clinically suspected or when the initial FNA is inconclusive. Keywords Fine needle aspiration Core needle biopsy Special type of cancer
Introduction Fine needle aspiration (FNA) has been the main tool used to assess breast lesions for determining appropriate management over the past decades. FNA is a quick and costeffective procedure, particularly in palpable breast lesions, and is safe with a low rate of complications [1–5]. However, FNA can often be inconclusive with high false negative rates (20–50 %), and does not provide a sufficient amount of tissue for further ancillary studies such as biomarker expression [3, 6]. Due to these facts, core needle biopsy (CNB) has been replacing FNA in many circumstances. Several studies have shown that CNB is diagnostically more accurate, and has higher sensitivity and specificity as well as a lower rate of unsatisfactory samples compared to FNA [7–11]. On the other hand, CNB is relatively expensive and tissue processing can be time-consuming, causing delays in reporting diagnostic results. Thus, there remains a lack of consensus regarding which of the two biopsy methods should be routinely employed in assessing breast lesions.
123
Breast Cancer
We previously compared the absolute diagnostic sensitivity between FNA and CNB in patients with breast cancer at our institution, and found that the sensitivity of FNA was almost equivalent to that of CNB [12]. When the initial FNA diagnosis was inconclusive, combined use of CNB resulted in increased sensitivity, suggesting that a combination of FNA and CNB should be considered for such cases. In that study, however, we combined all types of breast cancer into one group of ‘‘invasive cancers’’. Therefore, the diagnostic sensitivity of FNA and CNB in particular types of breast cancer, such as invasive lobular carcinoma (ILC) and mucinous carcinoma (MUC), was not extensively studied. In the English literature, only a few studies specifically compared the diagnostic accuracy between FNA and CNB in special types of breast cancer; thus, our understanding on the efficacy of FNA and CNB in that respect remains incomplete [13, 14]. In the present study, we evaluated the diagnostic sensitivity of FNA and CNB in special types of breast cancer such as ILC, MC, and apocrine carcinoma (APO) in comparison to invasive ductal carcinoma of no special type (NST) to determine how the diagnostic value of FNA and CNB changes depending on the histological subtype of the tumor. We also assessed the efficacy of the combined use of FNA and CNB in these cases.
Materials and methods We searched the archives at the Department of Diagnostic Pathology, Nippon Medical School Hospital (Tokyo, Japan) for cases treated between January 2004 and August 2013, and identified patients with histological diagnoses of ILC (n = 65), MC (n = 51), and APO (n = 25) who had undergone either FNA or CNB examination before surgery. For comparison, we randomly selected patients with histological diagnoses of NST (n = 159) who also had undergone either FNA or CNB examination before surgery. NST cases consist of papillotubular type (n = 53), solidtubular type (n = 53) and scirrhous type (n = 53) according to the histological classification defined by General Rules for Clinical and Pathological Recording of Breast Cancer of the Japanese Breast Cancer Society, 17th Edition [15]. Before FNA or CNB biopsy, all the patients were examined by mammography and ultrasonography, and then underwent either FNA alone, CNB alone, or a combination of FNA and CNB to establish a definitive diagnosis. At our institution, FNA is generally used initially to obtain an immediate diagnosis of the lesion, whereas CNB is performed to obtain a definitive histological diagnosis and to determine additional prognostic factors required to design a management plan. For patients in whom the FNA smears were found to be nondiagnostic
123
or incompatible with clinical and/or radiological findings, CNB was additionally performed. Patients who underwent neoadjuvant chemotherapy were excluded from the study. FNA cytology was performed by a number of experienced surgeons using a 20 G needle attached to a 20-mL syringe. The FNA samples were obtained either freehand or under ultrasonographic guidance, depending on the tumor size. Two or three separate passes were made into the lesion with the needle. During each pass, the needle was moved throughout the lesion several times while aspirating. Specimens were smeared on glass slides and were immediately fixed with 95 % ethanol. The cytological slides were stained according to the routine Papanicolaou method. CNB was performed with a 16 G needle with a sample notch on a spring-loaded biopsy gun. It was also performed either freehand or under ultrasonographic guidance, depending on the tumor size. Three or four shots were made into the lesion for the CNB. The tissue specimens were fixed in 20 % formalin and stained with hematoxylin and eosin (H&E). After examination of the FNA and CNB specimens, the cytological and histological findings, respectively, were expressed in diagnostic categories according to the General Rules for Clinical and Pathological Recording of Breast Cancer of the Japanese Breast Cancer Society, 17th Edition [15, 16]. Briefly, assessment categories were generally divided into two types: inadequate if the specimen was unsuitable for diagnosis and adequate if a cytological or histological diagnosis could be made. The following four assessment categories were included in the adequate group: malignant, suspicious for malignancy, indeterminate, and normal or benign. The indeterminate category comprised cytological or histological findings that were difficult to assess as benign or malignant. Subsequently, definitive surgery was performed based on a positive FNA and/or CNB diagnosis if supported by imaging modalities such as mammography and ultrasonography. After surgery, the removed breast tissue was fixed in 20 % formalin overnight and examined for measurement of the macroscopic tumor size. Subsequently, histological sections were processed and stained with H&E, and the sizes of the whole tumor (invasive and in situ components) in addition to the invasive component alone were also measured microscopically. The histological types of the tumors were classified in accordance with the World Health Organization histological classification [17]. This study was conducted in accordance with the principles embodied in the Declaration of Helsinki (2008). Patients’ consent was obtained for the use of clinical samples for research purposes according to the regulations defined by the Ethics Committee of Nippon Medical School Hospital. The comparability among the groups was evaluated using Fisher’s exact test. Statistical differences of mean age
Breast Cancer
and tumor size between the two groups were evaluated using Student’s t test. The ‘‘absolute sensitivity’’ was calculated as the number of ‘‘malignant’’ cases/the sum of ‘‘malignant’’, ‘‘suspicious for malignancy’’, ‘‘indeterminate’’, ‘‘normal or benign’’, and ‘‘inadequate’’ cases, and expressed as percentage. Values for patients’ age and tumor size were expressed as mean ± SD. Differences were considered significant at p \ 0.05.
Results Clinical and histological characteristics of tumors Clinical and histological characteristics of tumors are summarized in Table 1. In NST cases, 123 patients underwent FNA and 64 patients underwent CNB, of whom 28 patients had undergone FNA before CNB in the same lesion (combined FNA and CNB). The ages of all patients in the NST group ranged from 37 to 85 years (mean 61.4 ± 13.1 years). The macroscopic size of all NST tumors ranged from 0.6 to 7.6 cm (mean 2.6 ± 2.1), and the microscopic size of all NST tumors (whole tumor including invasive component) ranged from 0.7 to 9.5 cm (mean 3.3 ± 2.4). There was no statistical difference in the age, breast (right or left), macroscopic or microscopic tumor size, and histological nuclear grade among FNA, CNB, and the combined FNA and CNB groups. In ILC cases, 52 patients underwent FNA and 38 patients underwent CNB, of whom 25 underwent combined FNA and CNB. The ages of all patients in the ILC group ranged from 44 to 84 years (mean 62.1 ± 11.2 years). The macroscopic size of all ILC tumors ranged from 0.3 to 11.0 cm (mean 3.5 ± 2.5), and the microscopic size of all ILC tumors (whole tumor including invasive component) ranged from 0.7 to 9.5 cm (mean 4.2 ± 3.3). The mean microscopic size of the whole tumor in the CNB group was larger than that of the FNA group (p \ 0.05). The mean macroscopic size of the tumor of the combined FNA and CNB group was larger than that of the FNA group (p \ 0.05). The mean microscopic size of the whole tumor of the combined FNA and CNB group was larger than that of the FNA group (p \ 0.05). There was no statistical difference in age, breast, and nuclear grade among the FNA, CNB, and combined FNA and CNB groups. In MUC cases, 42 patients underwent FNA and 15 patients underwent CNB, of whom 6 underwent combined FNA and CNB. The ages of all patients in the MUC group ranged from 28 to 91 years (mean 60.3 ± 16.3 years). The macroscopic size of all MUC tumors ranged from 0.4 to 16.5 cm (mean 2.7 ± 2.8), and the microscopic size of all MUC tumors (whole tumor including invasive component) ranged from 1.2 to 16.5 cm (mean 3.8 ± 3.1). The mean
microscopic size of the whole tumor of the CNB group was larger than that of the FNA group (p \ 0.05). The mean microscopic size of the whole tumor of the combined FNA and CNB group was larger than that of the FNA group (p \ 0.05). There was no statistical difference in the age, breast, and nuclear grade among the FNA, CNB, and combined FNA and CNB groups. In APO cases, 21 patients underwent FNA and 16 patients underwent CNB, of whom 12 underwent combined FNA and CNB. The ages of all patients in the APO group ranged from 40 to 83 years (mean 65 ± 11.4 years). The macroscopic size of all APO tumors ranged from 0.5 to 5.6 cm (mean 2.9 ± 2.2), and the microscopic size of all APO tumors (whole tumor including invasive component) ranged from 0.7 to 5.6 cm (mean 3.1 ± 2.2). There was no statistical difference in the age, breast, tumor size and nuclear grade among the FNA, CNB, and combined FNA and CNB groups. Comparison of diagnostic results between FNA and CNB A comparison of the diagnostic results between the FNA and CNB groups is summarized in Table 2. In NST cases, 123 patients underwent FNA, of whom 103 (83.7 %) were cytologically diagnosed as malignant, 3 (2.9 %) as suspicious for malignancy, 9 (8.7 %) as indeterminate, and 4 (3.9 %) as normal or benign (Fig. 1a). The absolute sensitivity of FNA for diagnosis of NST was 83.7 %. CNB was performed in 64 patients with NST, of whom 61 (95.3 %) were histologically diagnosed as malignant, 1 (1.6 %) as suspicious for malignancy, 1 (1.6 %) as indeterminate, and 1 (1.6 %) as normal or benign. The absolute sensitivity of CNB for the diagnosis of NST was 95.3 %. There was no significant difference in the absolute sensitivity between FNA and CNB in the diagnosis of NST (p = 0.2). In ILC cases, 52 patients underwent FNA, of whom 32 (61.5 %) were cytologically diagnosed as malignant, 10 (19 %) as suspicious for malignancy, 3 (5.8 %) as indeterminate, and 5 (9.6 %) as normal or benign (Fig. 1b). The absolute sensitivity of FNA for the diagnosis of ILC was 61.5 % and was lower than that of NST (p \ 0.001). CNB was performed in 38 patients with ILC, of whom 36 (92.9 %) were histologically diagnosed as malignant, 1 (2.6 %) as suspicious for malignancy, 1 (2.6 %) as indeterminate, and none (0 %) as normal or benign. The absolute sensitivity of CNB for the diagnosis of ILC was 92.9 %. There was a significant difference in the absolute sensitivity between FNA and CNB in the diagnosis of ILC (p \ 0.001). In MUC cases, 42 patients underwent FNA, of whom 36 (85.7 %) were cytologically diagnosed as malignant, none (0 %) as suspicious for malignancy, 4 (9.5 %) as
123
123
Age (years)
p \ 0.05 vs FNA of ILC,
20
3
b
32
15
7
13
8
3.1 ± 3.4
4.1 ± 3.6
3.1 ± 3.3
13
p \ 0.05 vs FNA of MUC
14
37
13
2.6 ± 2.6
3.5 ± 2.9
2.8 ± 2.6
32
2
8
42
3.2 ± 2.2
3.6 ± 3.0
3.6 ± 2.8
24
28
62.7 ± 11.3
1
6
31
4.1 ± 2.5
5.6 ± 3.1
a
4.2 ± 2.7
18
18
62 ± 12
CNB (n = 38)
1
5
19
3.7 ± 2.6
5.4 ± 2.9a
5.2 ± 3.0a
13
12
63.1 ± 12.6
FNA ? CNB (n = 25)
1
4
31
2.2 ± 1.4
3.4 ± 3.1
2.5 ± 2.7
20
22
60.7 ± 17
FNA (n = 42)
MUC
0
2
7
2.9 ± 2.8
5.8 ± 4.3b
4.0 ± 4.5
7
8
60.4 ± 13
CNB (n = 15)
0
1
5
1.1 ± 0.6
6.7 ± 6.1b
4.5 ± 6.7
2
4
63.5 ± 16.1
FNA ? CNB (n = 6)
4
8
9
2.0 ± 1.3
3.1 ± 2.3
3.0 ± 2.4
15
6
64.4 ± 11
FNA (n = 21)
APO
4
4
8
2.3 ± 1.3
2.6 ± 1.4
2.2 ± 1.8
9
7
63.4 ± 12
CNB (n = 16)
3
3
6
2.5 ± 1.4
2.6 ± 1.4
2.3 ± 1.6
8
4
62.1 ± 13.2
FNA ? CNB (n = 12)
IDC, NST invasive ductal carcinoma of no special type, ILC invasive lobular carcinoma, MUC mucinous carcinoma, APO apocrine carcinoma, FNA fine needle aspiration, CNB core needle biopsy
a
81
2
2.6 ± 2.4
Micro (invasive) Nuclear grade
22
3.4 ± 2.5
Micro (whole)
1
2.6 ± 2.1
Macro
Tumor size (cm)
59
64
Left
56.6 ± 11.5
57.2 ± 13
61.7 ± 13
Right
Breast (side)
FNA (n = 52)
CNB (n = 64)
FNA (n = 123)
FNA ? CNB (n = 28)
ILC
IDC, NST
Table 1 Clinical and histological characteristics of tumors
Breast Cancer
Breast Cancer Table 2 Comparison of diagnostic results between FNA and CNB IDC, NST
ILC
FNA (n = 123)
CNB (n = 64)
MUC
APO
FNA (n = 52)
CNB (n = 38)
FNA (n = 42)
CNB (n = 15)
FNA (n = 21)
CNB (n = 16)
Judgment Malignant
103
61
32
36
36
12
8
12
Suspicious for malignancy Indeterminate
3 9
1 1
10 3
1 1
0 4
1 1
3 6
2 0
Benign
4
1
5
0
2
1
2
2
Inadequate
4
Absolute sensitivity (%) a
p \ 0.001 vs FNA of NST,
83.7 (103/123) b
0 95.3 (61/64)
2 61.5 (32/52)a
0
0
92.9 (39/42)b
p \ 0.001 vs FNA of ILC, c p \ 0.001 vs FNA of NST,
85.7 (36/42) d
0 80 (12/15)
2 38 (8/21)c
0 75 (12/16)d
p \ 0.05 vs FNA of APO
IDC, NST invasive ductal carcinoma of no special type, ILC invasive lobular carcinoma, MUC mucinous carcinoma, APO apocrine carcinoma, FNA fine needle aspiration, CNB core needle biopsy
Fig. 1 Fine needle aspiration cytological findings of invasive ductal carcinoma of no special type (a), invasive lobular carcinoma (b), mucinous carcinoma (c), and apocrine carcinoma (d). Tumor cells are aggregated and focally overlapped with dissociation at the peripheral areas (a). Tumor cells with round-shaped nuclei are dispersed in a flat sheet pattern with rare cohesiveness (b). There are three-dimensional cohesive clusters within abundant mucin (c). Tumor cells are aggregated with enlarged nuclei and granular cytoplasm, as seen in apocrine metaplasia (d). Papanicolaou stain (a–d). Original magnification 9400 (a–d)
indeterminate, and 2 (4.8 %) as normal or benign (Fig. 1c). The absolute sensitivity of FNA for the diagnosis of MUC was 85.7 %. CNB was performed in 15 patients with MUC, of whom 12 (80 %) were histologically diagnosed as malignant, 1 (6.7 %) as suspicious for malignancy, 1 (6.7 %) as indeterminate, and 1 (6.7 %) as normal or benign. The absolute sensitivity of CNB for the diagnosis of MUC was 80.0 %. There was no significant difference in the absolute sensitivity between FNA and CNB in the diagnosis of MUC (p = 0.6).
In APO cases, 21 patients underwent FNA, of whom 8 (38.0 %) of were cytologically diagnosed as malignant, 3 (14 %) as suspicious for malignancy, 6 (29 %) as indeterminate, and 2 (9.5 %) as normal or benign (Fig. 1d). The absolute sensitivity of FNA for diagnosis of APO was 38.0 % and was lower than that of NST (p \ 0.001). CNB was performed in 16 patients with APO, of whom 12 (75 %) patients were histologically diagnosed as malignant, 2 (13 %) as suspicious for malignancy, none (0 %) as indeterminate, and 2 (13 %) as normal or benign. The
123
123
p \ 0.05 vs FNA of APO c
p \ 0.001 vs FNA of ILC, p \ 0.05 vs FNA of NST, a
Judgment
b
39.2 (11/28) Absolute sensitivity (%)
89.2 (25/28)a
4 Inadequate
0
4 Benign
1
6 Indeterminate
1
3 Suspicious for malignancy
1
11 Malignant
25
CNB
Table 3 Diagnostic results of FNA combined with CNB
IDC, NST invasive ductal carcinoma of no special type, ILC invasive lobular carcinoma, MUC mucinous carcinoma, APO apocrine carcinoma, FNA fine needle aspiration, CNB core needle biopsy
0
0
1
66.7 (8/12)c 66.7 (8/12)c 25 (3/12) 66.7 (4/6) 66.7 (4/6) 33.3 (2/6) 92 (23/25)b 92 (23/25)b 40 (10/25) 92.9 (26/28)a
0 1 0 0 0 0 0 2 0
2 1 0 1 2 0 0 5 1
0
2 1
6 1
1 1
0 2
0 1
1 1
1 5
3 1
2 23 26
Combined
10
23
CNB FNA (n = 25) FNA (n = 28)
0
3 4 4
CNB FNA (n = 6) Combined
MUC ILC IDC, NST
Discussion There are a significant number of reports that attempted to elucidate the efficacy of FNA in comparison to CNB in the breast cancers. Regarding diagnostic accuracy, CNB is superior to FNA in both sensitivity and specificity as evidenced by several reports [5, 8, 10, 11, 14, 18] (Table 4). Only a few studies suggested that FNA is equivalent to CNB, or is more accurate only when performed on a palpable mass [19, 20]. Because of this, CNB has increasingly become the preferred diagnostic technique even for palpable lesions [6]. However, there is still reason to consider FNA as it has advantages over CNB such as expeditiousness, cost effectiveness, and a low rate of complications. CNB, on the other hand, is expensive, requires longer tissue processing times, and has greater chance of developing complications such as bleeding and hematomas [1, 21, 22]. Thus, the issue of which procedure is superior has been under debate for decades. We previously demonstrated that FNA and CNB had overall equivalent
FNA (n = 12) Combined
APO
Diagnostic results of FNA combined with CNB are summarized in Table 3. In NST cases, 28 patients underwent FNA and CNB in the same lesion. The absolute sensitivity of FNA alone and CNB alone was 39.2 and 89.2 %, respectively. The absolute sensitivity of FNA alone improved to 92.9 % when combined with CNB (p \ 0.05 compared to FNA alone of NOS). In ILC cases, 25 patients underwent combined FNA and CNB. The absolute sensitivity of FNA alone and CNB alone was 40 and 92 %, respectively. The absolute sensitivity of FNA alone improved to 92 % when combined with CNB (p \ 0.001 compared to FNA alone of ILC). In MUC cases, 6 patients underwent combined FNA and CNB. The absolute sensitivity of FNA alone and CNB alone was 33.3 and 66.7 %, respectively. The absolute sensitivity of FNA alone did not show significant improvement even when combined with CNB (p = 0.07 compared to FNA alone of MUC). In patients with APO, 12 patients underwent combined FNA and CNB. The absolute sensitivity of FNA alone and CNB alone was 25 and 66.7 %, respectively. The absolute sensitivity of FNA alone improved to 66.7 % when combined with CNB (p \ 0.05 compared to FNA alone of APO).
8
CNB
Diagnostic results of FNA combined with CNB
8
Combined
absolute sensitivity of CNB for diagnosis of APO was 75.0 %. There was a significant difference in the absolute sensitivity between FNA and CNB in the diagnosis of APO (p \ 0.05).
3
Breast Cancer
Breast Cancer Table 4 Absolute sensitivity of FNA and CNB in the literature
Lieske et al. [10]
Number of cases
Absolute sensitivity
Total
Special histological type
FNA
CNB
763
ILC (93), ILC ? IDC (33)
65
80
ILC (4), MUC (2)
98.2
96.5
Garg et al. [11]
32
Kurita et al. [12]
182
ILC (14)
93
86
Barra et al. [14]
264
ILC (14), ILC ? IDC (5), MUC (5), MED (2)
68.5
84.7
Lam et al. [13]
37
Nagar et al. [5]
162
Liefland et al. [8]
522
NA
68
90
65
NA
57
60
Lifrange et al. [18]
MUC (37)
66.7
100
NA
89
100
Ballo et al. [19]
124
NA
97.5
90
Westenend et al. [20]
286
NA
92
88
Hatada et al. [39]
233
NA
86.9
86.2
FNA fine needle aspiration, CNB core needle biopsy, ILC invasive lobular carcinoma, IDC invasive ductal carcinoma, MUC mucinous carcinoma, MED medullary carcinoma, NA not addressed in the article Square parentheses reference number, round parentheses number of cases
diagnostic sensitivity, and proposed that CNB should be performed following inconclusive FNA results for patients in whom malignancy is highly suspected by clinical and/or radiological findings [12]. In that study, we included all tumor types, such as ILC and MUC, into a single group of ‘‘invasive cancers’’, leaving the efficacy of FNA and CNB on individual tumor types unclear. Thus far, reports on the efficacy of FNA and CNB on these special types of breast cancers are extremely limited (Table 4). For instance, Barra et al. compared the diagnostic accuracy between FNA and CNB in special types of breast cancer such as ILC (n = 14), MUC (n = 5), and medullary carcinoma (n = 2), and concluded that the presence of these cancer types did not affect the total efficacy of FNA and CNB [14] (Table 4). Other studies also mentioned that there was no special pathological subtype affecting the overall efficacy of FNA and CNB. We believe that their conclusions may be due to an insufficient number of special cases examined out of the entire cohort, and postulate that a larger scale study would clarify this issue. We therefore decided to determine the efficacy of FNA and CNB in particular types of breast cancer such as ILC, MUC, and APO. This time, the specificity and false positive rates were not evaluated because only patients with histologically confirmed diagnosis of carcinoma were retrospectively investigated for their pre-operative FNA and CNB diagnosis, which is a limitation of our study. In the present study, the diagnostic sensitivity for ILC by FNA was lower than that by CNB. In cases with combined FNA and CNB in the same lesion, the sensitivity of FNA alone significantly improved when combined by CNB. Previous reports have shown that the diagnostic sensitivity of ILC by FNA ranges from 52 to 78 %. This is lower than that of invasive ductal carcinomas, and
consistent with our results [23–28]. To our knowledge, however, no other reports have compared the diagnostic accuracy of CNB and FNA in regards to ILC. In particular, no studies have compared the sensitivity of FNA and CNB performed in the same ILC lesion. One of the reasons for the lower sensitivity for ILC may be the lower cellularity of FNA smears. Abdulla et al. indicated that the cellularity of FNA samples is disproportionally lower than that of corresponding histological sections [25]. Therefore, we suspected that the higher number of tumor cells in the CNB specimens enabled the higher diagnostic sensitivity of CNB in our ILC cases. ILC has several histological variations such as classic, solid, alveolar, tubulolobular and pleomorphic subtypes. Interestingly, FNA of pleomorphic subtype can yield a sensitivity of up to 90 %, which is higher than other subtypes of ILC. This is probably due to unique cytological features of pleomorphic subtype such as larger nuclei, distinct atypia and prominent nucleoli [27, 28]. Our ILC cases (n = 65) included 9 pleomorphic (13.8 %) and two tubulolobular (3 %) subtypes, and one solid (1.5 %) subtype. The remainder were all of the classic subtypes. We were not able to individually investigate the diagnostic accuracy for each subtype due to the small sample size, and we speculated that the diagnostic sensitivity in our ILC cases may be mostly reflective of the classic subtype, which was predominant. In our cases of MUC, the diagnostic sensitivity of total FNA was equivalent to that of total CNB of MUC and to that of total FNA in NST. Contrary to our results, Lam et al. have shown that the sensitivity of FNA in MUC was 66.7 %, which was significantly lower than the 100 % sensitivity of CNB [13]. The reason for this discrepancy is uncertain. One explanation is that diagnosis of MUC is challenged by a mild degree of nuclear atypia and abundant
123
Breast Cancer
extracellular mucin, resulting in variable cytological interpretation [29, 30]. Secondly, MUC presents with heterogeneous histological features. For instance, MUC has two histological subtypes, pure and mixed type that can affect the diagnostic sensitivity of FNA. In fact, the number of pure and mixed type of our cases was almost even (27 and 24, respectively) whereas in a previous study, pure type was predominant over mixed type (17 and 5, respectively) [31]. Moreover, MUC often expresses neuroendocrine markers that may have affected the morphology of tumor cells [32–35]. In our cases, we did not examine expression of neuroendocrine markers in all the MUC cases, and thus the effect of those proteins on the morphology of MUC tumor cells remained unknown. APO is a relatively rare neoplasm occurring in the breast, and is composed predominantly of cells with abundant eosinophilic cytoplasm reminiscent of metaplastic apocrine cells. Cytological features such as pleomorphism, nuclear crowding, increased nuclear-cytoplasmic ratio, and occasional mitotic figures are reported to be characteristic of malignant rather than benign apocrine lesions [36]. Cytological differentiation of APO from metaplastic apocrine cells is often challenging, and thus FNA of APO sometimes generates a false negative result [37, 38]. In our APO cases, the diagnostic sensitivity of FNA was significantly lower than that of NST. Concurrent FNA and CNB significantly improved the sensitivity. To our knowledge, no studies have thus far compared diagnostic sensitivity of FNA and CNB for APO. We therefore conclude that, to reach the definitive diagnosis, CNB should be performed when the FNA sample contains atypical apocrine cells that are difficult to interpret. In all the lesions except MUC cases, our results showed that combined use of FNA and CNB increased the diagnostic sensitivity. These results are consistent with our previous studies as well as other reports, but none of these studies specifically addressed the efficacy of combined use of FNA and CNB in special types of breast cancers [12, 20, 39]. For the first time, we clarified this aspect in special breast cancer types including NST, ILC, and APO. The reason that the sensitivity of MUC was not improved by combined FNA and CNB could be due to the small number of MUC cases (n = 6). Unfortunately, there are no other studies revealing the diagnostic sensitivity of combined FNA and CNB for MUC. We postulate that a large-scale study may reveal the usefulness of this combination in the future. In the 25 cases of our ILC group, the initial FNA was inconclusive, and combining with CNB eventually increased the sensitivity. Interestingly, the mean size of the tumor for these combined FNA and CNB cases was larger ([5 cm) than FNA alone cases. This result may suggest that CNB is more accurate when the ILC tumor is larger than average. We do not have an adequate explanation for
123
this phenomenon from the pathological point of view. Barra et al. showed that CNB is more suitable than FNA only for tumors ranging from 2 to 5 cm. For other sizes (less than 2 cm or more than 5 cm), FNA is equivalent to CNB [14]. Our result is not consistent with theirs, possibly because their study included 14 ILC out of 264 cases in total, and the vast majority of the remaining tumors were NST. We could not find other reports addressing the effect of tumor size on the difference between FNA and CNB results. Further studies are required to clarify this issue. In summary, we have compared diagnostic sensitivity of FNA and CNB in special types of breast cancer, including ILC, MC, APO and NST. The absolute sensitivity of CNB was higher than that of FNA for ILC and APO, but not for NST and MUC, where FNA and CNB had equivalent sensitivity. For the NST, ILC, and APO groups, combined FNA and CNB significantly improved the diagnostic sensitivity. Our results suggest that the diagnostic sensitivity of FNA and CNB is variable depending on the type of breast cancers. Thus, the use of biopsy procedures should be tailored to the type of tumor suspected to reach the accurate final diagnosis. Conflict of interest interest exists.
The authors have declared that no conflict of
References 1. He Q, Fan X, Yuan T, Kong L, Du X, Zhuang D, et al. Eleven years of experience reveals that fine-needle aspiration cytology is still a useful method for preoperative diagnosis of breast carcinoma. Breast. 2007;16:303–6. 2. Feoli F, Paesmans M, Van Eeckhout P. Fine needle aspiration cytology of the breast: impact of experience on accuracy, using standardized cytologic criteria. Acta Cytol. 2008;52:145–51. 3. Pilgrim S, Ravichandran D. Fine needle aspiration cytology as an adjunct to core biopsy in the assessment of symptomatic breast carcinoma. Breast. 2005;14:411–4. 4. Manfrin E, Mariotto R, Remo A, Reghellin D, Dalfior D, Falsirollo F, et al. Is there still a role for fine-needle aspiration cytology in breast cancer screening? Experience of the Verona Mammographic Breast Cancer Screening Program with real-time integrated radiopathologic activity (1999-2004). Cancer. 2008;114:74–82. 5. Nagar S, Iacco A, Riggs T, Kestenberg W, Keidan R. An analysis of fine needle aspiration versus core needle biopsy in clinically palpable breast lesions: a report on the predictive values and a cost comparison. Am J Surg. 2012;204:193–8. 6. Cobb CJ, Raza AS. Obituary: ‘‘alas poor FNA of breast-we knew thee well!’’. Diagn Cytopathol. 2005;32:1–4. 7. Clarke D, Sudhakaran N, Gateley CA. Replace fine needle aspiration cytology with automated core biopsy in the triple assessment of breast cancer. Ann R Coll Surg Engl. 2001;83:110–2. 8. Leifland K, Lagerstedt U, Svane G. Comparison of stereotactic fine needle aspiration cytology and core needle biopsy in 522 non-palpable breast lesions. Acta Radiol. 2003;44:387–91. 9. Oyama T, Koibuchi Y, McKee G. Core needle biopsy (CNB) as a diagnostic method for breast lesions: comparison with fine needle aspiration cytology (FNA). Breast Cancer. 2004;11:339–42.
Breast Cancer 10. Lieske B, Ravichandran D, Wright D. Role of fine-needle aspiration cytology and core biopsy in the preoperative diagnosis of screen-detected breast carcinoma. Br J Cancer. 2006;95:62–6. 11. Garg S, Mohan H, Bal A, Attri AK, Kochhar S. A comparative analysis of core needle biopsy and fine-needle aspiration cytology in the evaluation of palpable and mammographically detected suspicious breast lesions. Diagn Cytopathol. 2007;35:681–9. 12. Kurita T, Tsuchiya S, Watarai Y, Yamamoto Y, Harada O, Yanagihara K, et al. Roles of fine-needle aspiration and core needle biopsy in the diagnosis of breast cancer. Breast Cancer. 2012;19:23–9. 13. Lam WW, Chu WC, Tse GM, Ma TK, Tang AP. Role of fine needle aspiration and tru cut biopsy in diagnosis of mucinous carcinoma of breast–from a radiologist’s perspective. Clin Imaging. 2006;30:6–10. 14. Barra AeA, Gobbi H, de Rezende L, Alencar C, Gouveˆa AP, de Lucena CE, Reis JH, et al. A comparision of aspiration cytology and core needle biopsy according to tumor size of suspicious breast lesions. Diagn Cytopathol. 2008;36:26–31. 15. The Japanese Breast Cancer Society. General rules for clinical and pathological recording of breast cancer. 17th ed. Breast cytology and core needle biopsy reporting form (in Japanese) 2012;64-78. 16. Tsuchiya S, Akiyama F, Moriya T, Tsuda H, Umemura S, Katayama Y, et al. A new reporting form for breast cytology. Breast Cancer. 2009;16:202–6. 17. Lakhani SR, Cancer IAfRo, Organization WH. WHO Classification of Tumours of the Breast: International Agency for Research on Cancer; 2012. 18. Lifrange E, Kridelka F, Colin C. Stereotaxic needle-core biopsy and fine-needle aspiration biopsy in the diagnosis of nonpalpable breast lesions: controversies and future prospects. Eur J Radiol. 1997;24:39–47. 19. Ballo MS, Sneige N. Can core needle biopsy replace fine-needle aspiration cytology in the diagnosis of palpable breast carcinoma. A comparative study of 124 women. Cancer. 1996;78:773–7. 20. Westenend PJ, Sever AR. Beekman-De Volder HJ, Liem SJ. A comparison of aspiration cytology and core needle biopsy in the evaluation of breast lesions. Cancer. 2001;93:146–50. 21. Xie HB, Salhadar A, Haara A, Gabram S, Selvaggi SM, Wojcik EM. How stereotactic core-needle biopsy affected breast fineneedle aspiration utilization: an 11 year institutional review. Diagn Cytopathol. 2004;31:106–10. 22. Nassar A. Core needle biopsy versus fine needle aspiration biopsy in breast–a historical perspective and opportunities in the modern era. Diagn Cytopathol. 2011;39:380–8. 23. Sadler GP, McGee S, Dallimore NS, Monypenny IJ, DouglasJones AG, Lyons K, et al. Role of fine-needle aspiration cytology and needle-core biopsy in the diagnosis of lobular carcinoma of the breast. Br J Surg. 1994;81:1315–7. 24. Robinson IA, McKee G, Jackson PA, Cook MG, Kissin MW. Lobular carcinoma of the breast: cytological features supporting the diagnosis of lobular cancer. Diagn Cytopathol. 1995;13:196–201. 25. Abdulla M, Hombal S, al-Juwaiser A, Stankovich D, Ahmed M, Ajrawi T. Cellularity of lobular carcinoma and its relationship to
26.
27.
28.
29.
30.
31.
32.
33. 34.
35.
36.
37.
38.
39.
false negative fine needle aspiration results. Acta Cytol. 2000;44:625–32. Tan SM, Behranwala KA, Trott PA, Nasiri NA, Moskovic E, Brown G, et al. A retrospective study comparing the individual modalities of triple assessment in the pre-operative diagnosis of invasive lobular breast carcinoma. Eur J Surg Oncol. 2002;28:203–8. Hwang S, Ioffe O, Lee I, Waisman J, Cangiarella J, Simsir A. Cytologic diagnosis of invasive lobular carcinoma: factors associated with negative and equivocal diagnoses. Diagn Cytopathol. 2004;31:87–93. Menet E, Becette V, Briffod M. Cytologic diagnosis of lobular carcinoma of the breast: experience with 555 patients in the Rene Huguenin Cancer Center. Cancer. 2008;114:111–7. Gupta RK, McHutchison AG, Simpson JS, Dowle CS. Value of fine needle aspiration cytology of the breast, with an emphasis on the cytodiagnosis of colloid carcinoma. Acta Cytol. 1991;35:703–9. Ventura K, Cangiarella J, Lee I, Moreira A, Waisman J, Simsir A. Aspiration biopsy of mammary lesions with abundant extracellular mucinous material. Review of 43 cases with surgical followup. Am J Clin Pathol. 2003;120:194–202. Cyrta J, Andreiuolo F, Azoulay S, Balleyguier C, Bourgier C, Mazouni C, et al. Pure and mixed mucinous carcinoma of the breast: fine needle aspiration cytology findings and review of the literature. Cytopathology. 2013;24:377–84. Sapino A, Papotti M, Pietribiasi F, Bussolati G. Diagnostic cytological features of neuroendocrine differentiated carcinoma of the breast. Virchows Arch. 1998;433:217–22. Tse GM, Ma TK. Fine-needle aspiration cytology of breast carcinoma with endocrine differentiation. Cancer. 2000;90:286–91. Tang W, Taniguchi E, Wang X, Mori I, Kagiya T, Yang Q, et al. Loss of cell cohesion in breast cytology as a characteristic of neuroendocrine carcinoma. Acta Cytol. 2002;46:835–40. Kawasaki T, Nakamura S, Sakamoto G, Kondo T, TsunodaShimizu H, Ishii Y, et al. Neuroendocrine ductal carcinoma in situ of the breast: cytological features in 32 cases. Cytopathology. 2011;22:43–9. Yoshida K, Inoue M, Furuta S, Sakai R, Imai R, Hayakawa S, et al. Apocrine carcinoma vs. apocrine metaplasia with atypia of the breast. Use of aspiration biopsy cytology. Acta Cytol. 1996;40:247–51. Ng WK. Fine needle aspiration cytology of apocrine carcinoma of the breast. Review of cases in a three-year period. Acta Cytol. 2002;46:507–12. Haji BE, Das DK, Al-Ayadhy B, Pathan SK, George SG, Mallik MK, et al. Fine-needle aspiration cytologic features of four special types of breast cancers: mucinous, medullary, apocrine, and papillary. Diagn Cytopathol. 2007;35:408–16. Hatada T, Ishii H, Ichii S, Okada K, Fujiwara Y, Yamamura T. Diagnostic value of ultrasound-guided fine-needle aspiration biopsy, core-needle biopsy, and evaluation of combined use in the diagnosis of breast lesions. J Am Coll Surg. 2000;190:299–303.
123
ORIGINAL ARTICLE
Diagnostic value of fine needle aspiration and core needle biopsy in special types of breast cancer Ryuji Ohashi1 • Miyuki Matsubara1 • Yasuhiko Watarai1 • Keiko Yanagihara2 Koji Yamashita2 • Shin-ichi Tsuchiya1,3 • Hiroyuki Takei2 • Zenya Naito1,4
•
Received: 9 April 2015 / Accepted: 21 June 2015 Ó The Japanese Breast Cancer Society 2015
Abstract Background Although fine needle aspiration (FNA) biopsy is an established tool to assess breast lesions, there has been a trend toward using core needle biopsy (CNB) instead. The aim of this study was to compare the diagnostic accuracy of FNA and CNB in special types of breast cancer. Methods A retrospective review of diagnostic results of pre-operatively performed FNA or CNB, or a combination of the two, was conducted. The cases include histologically proven invasive ductal carcinoma of no special type (NST n = 159), invasive lobular carcinoma (ILC n = 65), mucinous carcinoma (MUC n = 51), and apocrine carcinoma (APO n = 25). Results The absolute diagnostic sensitivity of FNA to detect malignancy in ILC and APO patients was inferior to that of NST patients (p \ 0.001 for ILC and APO). Within each cancer type, the sensitivity of CNB was higher than that of FNA in the ILC and APO patients (p \ 0.001 and p \ 0.05, respectively). As for NST and MUC patients, FNA and CNB had equivalent sensitivity. The sensitivity
& Ryuji Ohashi [email protected] 1
Department of Diagnostic Pathology, Nippon Medical School Hospital, 1-1-5, Sendagi, Bunkyo-ku, Tokyo 113-8603, Japan
2
Division of Breast Surgery, Nippon Medical School Hospital, Tokyo, Japan
3
Department of Diagnostic Pathology, Iida Hospital, Nagano, Japan
4
Department of Pathology and Integrative Oncological Pathology, Nippon Medical School, Tokyo, Japan
of FNA alone significantly improved when combined with CNB in NST, ILC and APO patients (p \ 0.05, p \ 0.001, and p \ 0.05, respectively). Conclusions Our results suggest that FNA has less diagnostic accuracy than CNB for ILC and APO; thus, the use of CNB should be encouraged when these types of cancer are clinically suspected or when the initial FNA is inconclusive. Keywords Fine needle aspiration Core needle biopsy Special type of cancer
Introduction Fine needle aspiration (FNA) has been the main tool used to assess breast lesions for determining appropriate management over the past decades. FNA is a quick and costeffective procedure, particularly in palpable breast lesions, and is safe with a low rate of complications [1–5]. However, FNA can often be inconclusive with high false negative rates (20–50 %), and does not provide a sufficient amount of tissue for further ancillary studies such as biomarker expression [3, 6]. Due to these facts, core needle biopsy (CNB) has been replacing FNA in many circumstances. Several studies have shown that CNB is diagnostically more accurate, and has higher sensitivity and specificity as well as a lower rate of unsatisfactory samples compared to FNA [7–11]. On the other hand, CNB is relatively expensive and tissue processing can be time-consuming, causing delays in reporting diagnostic results. Thus, there remains a lack of consensus regarding which of the two biopsy methods should be routinely employed in assessing breast lesions.
123
Breast Cancer
We previously compared the absolute diagnostic sensitivity between FNA and CNB in patients with breast cancer at our institution, and found that the sensitivity of FNA was almost equivalent to that of CNB [12]. When the initial FNA diagnosis was inconclusive, combined use of CNB resulted in increased sensitivity, suggesting that a combination of FNA and CNB should be considered for such cases. In that study, however, we combined all types of breast cancer into one group of ‘‘invasive cancers’’. Therefore, the diagnostic sensitivity of FNA and CNB in particular types of breast cancer, such as invasive lobular carcinoma (ILC) and mucinous carcinoma (MUC), was not extensively studied. In the English literature, only a few studies specifically compared the diagnostic accuracy between FNA and CNB in special types of breast cancer; thus, our understanding on the efficacy of FNA and CNB in that respect remains incomplete [13, 14]. In the present study, we evaluated the diagnostic sensitivity of FNA and CNB in special types of breast cancer such as ILC, MC, and apocrine carcinoma (APO) in comparison to invasive ductal carcinoma of no special type (NST) to determine how the diagnostic value of FNA and CNB changes depending on the histological subtype of the tumor. We also assessed the efficacy of the combined use of FNA and CNB in these cases.
Materials and methods We searched the archives at the Department of Diagnostic Pathology, Nippon Medical School Hospital (Tokyo, Japan) for cases treated between January 2004 and August 2013, and identified patients with histological diagnoses of ILC (n = 65), MC (n = 51), and APO (n = 25) who had undergone either FNA or CNB examination before surgery. For comparison, we randomly selected patients with histological diagnoses of NST (n = 159) who also had undergone either FNA or CNB examination before surgery. NST cases consist of papillotubular type (n = 53), solidtubular type (n = 53) and scirrhous type (n = 53) according to the histological classification defined by General Rules for Clinical and Pathological Recording of Breast Cancer of the Japanese Breast Cancer Society, 17th Edition [15]. Before FNA or CNB biopsy, all the patients were examined by mammography and ultrasonography, and then underwent either FNA alone, CNB alone, or a combination of FNA and CNB to establish a definitive diagnosis. At our institution, FNA is generally used initially to obtain an immediate diagnosis of the lesion, whereas CNB is performed to obtain a definitive histological diagnosis and to determine additional prognostic factors required to design a management plan. For patients in whom the FNA smears were found to be nondiagnostic
123
or incompatible with clinical and/or radiological findings, CNB was additionally performed. Patients who underwent neoadjuvant chemotherapy were excluded from the study. FNA cytology was performed by a number of experienced surgeons using a 20 G needle attached to a 20-mL syringe. The FNA samples were obtained either freehand or under ultrasonographic guidance, depending on the tumor size. Two or three separate passes were made into the lesion with the needle. During each pass, the needle was moved throughout the lesion several times while aspirating. Specimens were smeared on glass slides and were immediately fixed with 95 % ethanol. The cytological slides were stained according to the routine Papanicolaou method. CNB was performed with a 16 G needle with a sample notch on a spring-loaded biopsy gun. It was also performed either freehand or under ultrasonographic guidance, depending on the tumor size. Three or four shots were made into the lesion for the CNB. The tissue specimens were fixed in 20 % formalin and stained with hematoxylin and eosin (H&E). After examination of the FNA and CNB specimens, the cytological and histological findings, respectively, were expressed in diagnostic categories according to the General Rules for Clinical and Pathological Recording of Breast Cancer of the Japanese Breast Cancer Society, 17th Edition [15, 16]. Briefly, assessment categories were generally divided into two types: inadequate if the specimen was unsuitable for diagnosis and adequate if a cytological or histological diagnosis could be made. The following four assessment categories were included in the adequate group: malignant, suspicious for malignancy, indeterminate, and normal or benign. The indeterminate category comprised cytological or histological findings that were difficult to assess as benign or malignant. Subsequently, definitive surgery was performed based on a positive FNA and/or CNB diagnosis if supported by imaging modalities such as mammography and ultrasonography. After surgery, the removed breast tissue was fixed in 20 % formalin overnight and examined for measurement of the macroscopic tumor size. Subsequently, histological sections were processed and stained with H&E, and the sizes of the whole tumor (invasive and in situ components) in addition to the invasive component alone were also measured microscopically. The histological types of the tumors were classified in accordance with the World Health Organization histological classification [17]. This study was conducted in accordance with the principles embodied in the Declaration of Helsinki (2008). Patients’ consent was obtained for the use of clinical samples for research purposes according to the regulations defined by the Ethics Committee of Nippon Medical School Hospital. The comparability among the groups was evaluated using Fisher’s exact test. Statistical differences of mean age
Breast Cancer
and tumor size between the two groups were evaluated using Student’s t test. The ‘‘absolute sensitivity’’ was calculated as the number of ‘‘malignant’’ cases/the sum of ‘‘malignant’’, ‘‘suspicious for malignancy’’, ‘‘indeterminate’’, ‘‘normal or benign’’, and ‘‘inadequate’’ cases, and expressed as percentage. Values for patients’ age and tumor size were expressed as mean ± SD. Differences were considered significant at p \ 0.05.
Results Clinical and histological characteristics of tumors Clinical and histological characteristics of tumors are summarized in Table 1. In NST cases, 123 patients underwent FNA and 64 patients underwent CNB, of whom 28 patients had undergone FNA before CNB in the same lesion (combined FNA and CNB). The ages of all patients in the NST group ranged from 37 to 85 years (mean 61.4 ± 13.1 years). The macroscopic size of all NST tumors ranged from 0.6 to 7.6 cm (mean 2.6 ± 2.1), and the microscopic size of all NST tumors (whole tumor including invasive component) ranged from 0.7 to 9.5 cm (mean 3.3 ± 2.4). There was no statistical difference in the age, breast (right or left), macroscopic or microscopic tumor size, and histological nuclear grade among FNA, CNB, and the combined FNA and CNB groups. In ILC cases, 52 patients underwent FNA and 38 patients underwent CNB, of whom 25 underwent combined FNA and CNB. The ages of all patients in the ILC group ranged from 44 to 84 years (mean 62.1 ± 11.2 years). The macroscopic size of all ILC tumors ranged from 0.3 to 11.0 cm (mean 3.5 ± 2.5), and the microscopic size of all ILC tumors (whole tumor including invasive component) ranged from 0.7 to 9.5 cm (mean 4.2 ± 3.3). The mean microscopic size of the whole tumor in the CNB group was larger than that of the FNA group (p \ 0.05). The mean macroscopic size of the tumor of the combined FNA and CNB group was larger than that of the FNA group (p \ 0.05). The mean microscopic size of the whole tumor of the combined FNA and CNB group was larger than that of the FNA group (p \ 0.05). There was no statistical difference in age, breast, and nuclear grade among the FNA, CNB, and combined FNA and CNB groups. In MUC cases, 42 patients underwent FNA and 15 patients underwent CNB, of whom 6 underwent combined FNA and CNB. The ages of all patients in the MUC group ranged from 28 to 91 years (mean 60.3 ± 16.3 years). The macroscopic size of all MUC tumors ranged from 0.4 to 16.5 cm (mean 2.7 ± 2.8), and the microscopic size of all MUC tumors (whole tumor including invasive component) ranged from 1.2 to 16.5 cm (mean 3.8 ± 3.1). The mean
microscopic size of the whole tumor of the CNB group was larger than that of the FNA group (p \ 0.05). The mean microscopic size of the whole tumor of the combined FNA and CNB group was larger than that of the FNA group (p \ 0.05). There was no statistical difference in the age, breast, and nuclear grade among the FNA, CNB, and combined FNA and CNB groups. In APO cases, 21 patients underwent FNA and 16 patients underwent CNB, of whom 12 underwent combined FNA and CNB. The ages of all patients in the APO group ranged from 40 to 83 years (mean 65 ± 11.4 years). The macroscopic size of all APO tumors ranged from 0.5 to 5.6 cm (mean 2.9 ± 2.2), and the microscopic size of all APO tumors (whole tumor including invasive component) ranged from 0.7 to 5.6 cm (mean 3.1 ± 2.2). There was no statistical difference in the age, breast, tumor size and nuclear grade among the FNA, CNB, and combined FNA and CNB groups. Comparison of diagnostic results between FNA and CNB A comparison of the diagnostic results between the FNA and CNB groups is summarized in Table 2. In NST cases, 123 patients underwent FNA, of whom 103 (83.7 %) were cytologically diagnosed as malignant, 3 (2.9 %) as suspicious for malignancy, 9 (8.7 %) as indeterminate, and 4 (3.9 %) as normal or benign (Fig. 1a). The absolute sensitivity of FNA for diagnosis of NST was 83.7 %. CNB was performed in 64 patients with NST, of whom 61 (95.3 %) were histologically diagnosed as malignant, 1 (1.6 %) as suspicious for malignancy, 1 (1.6 %) as indeterminate, and 1 (1.6 %) as normal or benign. The absolute sensitivity of CNB for the diagnosis of NST was 95.3 %. There was no significant difference in the absolute sensitivity between FNA and CNB in the diagnosis of NST (p = 0.2). In ILC cases, 52 patients underwent FNA, of whom 32 (61.5 %) were cytologically diagnosed as malignant, 10 (19 %) as suspicious for malignancy, 3 (5.8 %) as indeterminate, and 5 (9.6 %) as normal or benign (Fig. 1b). The absolute sensitivity of FNA for the diagnosis of ILC was 61.5 % and was lower than that of NST (p \ 0.001). CNB was performed in 38 patients with ILC, of whom 36 (92.9 %) were histologically diagnosed as malignant, 1 (2.6 %) as suspicious for malignancy, 1 (2.6 %) as indeterminate, and none (0 %) as normal or benign. The absolute sensitivity of CNB for the diagnosis of ILC was 92.9 %. There was a significant difference in the absolute sensitivity between FNA and CNB in the diagnosis of ILC (p \ 0.001). In MUC cases, 42 patients underwent FNA, of whom 36 (85.7 %) were cytologically diagnosed as malignant, none (0 %) as suspicious for malignancy, 4 (9.5 %) as
123
123
Age (years)
p \ 0.05 vs FNA of ILC,
20
3
b
32
15
7
13
8
3.1 ± 3.4
4.1 ± 3.6
3.1 ± 3.3
13
p \ 0.05 vs FNA of MUC
14
37
13
2.6 ± 2.6
3.5 ± 2.9
2.8 ± 2.6
32
2
8
42
3.2 ± 2.2
3.6 ± 3.0
3.6 ± 2.8
24
28
62.7 ± 11.3
1
6
31
4.1 ± 2.5
5.6 ± 3.1
a
4.2 ± 2.7
18
18
62 ± 12
CNB (n = 38)
1
5
19
3.7 ± 2.6
5.4 ± 2.9a
5.2 ± 3.0a
13
12
63.1 ± 12.6
FNA ? CNB (n = 25)
1
4
31
2.2 ± 1.4
3.4 ± 3.1
2.5 ± 2.7
20
22
60.7 ± 17
FNA (n = 42)
MUC
0
2
7
2.9 ± 2.8
5.8 ± 4.3b
4.0 ± 4.5
7
8
60.4 ± 13
CNB (n = 15)
0
1
5
1.1 ± 0.6
6.7 ± 6.1b
4.5 ± 6.7
2
4
63.5 ± 16.1
FNA ? CNB (n = 6)
4
8
9
2.0 ± 1.3
3.1 ± 2.3
3.0 ± 2.4
15
6
64.4 ± 11
FNA (n = 21)
APO
4
4
8
2.3 ± 1.3
2.6 ± 1.4
2.2 ± 1.8
9
7
63.4 ± 12
CNB (n = 16)
3
3
6
2.5 ± 1.4
2.6 ± 1.4
2.3 ± 1.6
8
4
62.1 ± 13.2
FNA ? CNB (n = 12)
IDC, NST invasive ductal carcinoma of no special type, ILC invasive lobular carcinoma, MUC mucinous carcinoma, APO apocrine carcinoma, FNA fine needle aspiration, CNB core needle biopsy
a
81
2
2.6 ± 2.4
Micro (invasive) Nuclear grade
22
3.4 ± 2.5
Micro (whole)
1
2.6 ± 2.1
Macro
Tumor size (cm)
59
64
Left
56.6 ± 11.5
57.2 ± 13
61.7 ± 13
Right
Breast (side)
FNA (n = 52)
CNB (n = 64)
FNA (n = 123)
FNA ? CNB (n = 28)
ILC
IDC, NST
Table 1 Clinical and histological characteristics of tumors
Breast Cancer
Breast Cancer Table 2 Comparison of diagnostic results between FNA and CNB IDC, NST
ILC
FNA (n = 123)
CNB (n = 64)
MUC
APO
FNA (n = 52)
CNB (n = 38)
FNA (n = 42)
CNB (n = 15)
FNA (n = 21)
CNB (n = 16)
Judgment Malignant
103
61
32
36
36
12
8
12
Suspicious for malignancy Indeterminate
3 9
1 1
10 3
1 1
0 4
1 1
3 6
2 0
Benign
4
1
5
0
2
1
2
2
Inadequate
4
Absolute sensitivity (%) a
p \ 0.001 vs FNA of NST,
83.7 (103/123) b
0 95.3 (61/64)
2 61.5 (32/52)a
0
0
92.9 (39/42)b
p \ 0.001 vs FNA of ILC, c p \ 0.001 vs FNA of NST,
85.7 (36/42) d
0 80 (12/15)
2 38 (8/21)c
0 75 (12/16)d
p \ 0.05 vs FNA of APO
IDC, NST invasive ductal carcinoma of no special type, ILC invasive lobular carcinoma, MUC mucinous carcinoma, APO apocrine carcinoma, FNA fine needle aspiration, CNB core needle biopsy
Fig. 1 Fine needle aspiration cytological findings of invasive ductal carcinoma of no special type (a), invasive lobular carcinoma (b), mucinous carcinoma (c), and apocrine carcinoma (d). Tumor cells are aggregated and focally overlapped with dissociation at the peripheral areas (a). Tumor cells with round-shaped nuclei are dispersed in a flat sheet pattern with rare cohesiveness (b). There are three-dimensional cohesive clusters within abundant mucin (c). Tumor cells are aggregated with enlarged nuclei and granular cytoplasm, as seen in apocrine metaplasia (d). Papanicolaou stain (a–d). Original magnification 9400 (a–d)
indeterminate, and 2 (4.8 %) as normal or benign (Fig. 1c). The absolute sensitivity of FNA for the diagnosis of MUC was 85.7 %. CNB was performed in 15 patients with MUC, of whom 12 (80 %) were histologically diagnosed as malignant, 1 (6.7 %) as suspicious for malignancy, 1 (6.7 %) as indeterminate, and 1 (6.7 %) as normal or benign. The absolute sensitivity of CNB for the diagnosis of MUC was 80.0 %. There was no significant difference in the absolute sensitivity between FNA and CNB in the diagnosis of MUC (p = 0.6).
In APO cases, 21 patients underwent FNA, of whom 8 (38.0 %) of were cytologically diagnosed as malignant, 3 (14 %) as suspicious for malignancy, 6 (29 %) as indeterminate, and 2 (9.5 %) as normal or benign (Fig. 1d). The absolute sensitivity of FNA for diagnosis of APO was 38.0 % and was lower than that of NST (p \ 0.001). CNB was performed in 16 patients with APO, of whom 12 (75 %) patients were histologically diagnosed as malignant, 2 (13 %) as suspicious for malignancy, none (0 %) as indeterminate, and 2 (13 %) as normal or benign. The
123
123
p \ 0.05 vs FNA of APO c
p \ 0.001 vs FNA of ILC, p \ 0.05 vs FNA of NST, a
Judgment
b
39.2 (11/28) Absolute sensitivity (%)
89.2 (25/28)a
4 Inadequate
0
4 Benign
1
6 Indeterminate
1
3 Suspicious for malignancy
1
11 Malignant
25
CNB
Table 3 Diagnostic results of FNA combined with CNB
IDC, NST invasive ductal carcinoma of no special type, ILC invasive lobular carcinoma, MUC mucinous carcinoma, APO apocrine carcinoma, FNA fine needle aspiration, CNB core needle biopsy
0
0
1
66.7 (8/12)c 66.7 (8/12)c 25 (3/12) 66.7 (4/6) 66.7 (4/6) 33.3 (2/6) 92 (23/25)b 92 (23/25)b 40 (10/25) 92.9 (26/28)a
0 1 0 0 0 0 0 2 0
2 1 0 1 2 0 0 5 1
0
2 1
6 1
1 1
0 2
0 1
1 1
1 5
3 1
2 23 26
Combined
10
23
CNB FNA (n = 25) FNA (n = 28)
0
3 4 4
CNB FNA (n = 6) Combined
MUC ILC IDC, NST
Discussion There are a significant number of reports that attempted to elucidate the efficacy of FNA in comparison to CNB in the breast cancers. Regarding diagnostic accuracy, CNB is superior to FNA in both sensitivity and specificity as evidenced by several reports [5, 8, 10, 11, 14, 18] (Table 4). Only a few studies suggested that FNA is equivalent to CNB, or is more accurate only when performed on a palpable mass [19, 20]. Because of this, CNB has increasingly become the preferred diagnostic technique even for palpable lesions [6]. However, there is still reason to consider FNA as it has advantages over CNB such as expeditiousness, cost effectiveness, and a low rate of complications. CNB, on the other hand, is expensive, requires longer tissue processing times, and has greater chance of developing complications such as bleeding and hematomas [1, 21, 22]. Thus, the issue of which procedure is superior has been under debate for decades. We previously demonstrated that FNA and CNB had overall equivalent
FNA (n = 12) Combined
APO
Diagnostic results of FNA combined with CNB are summarized in Table 3. In NST cases, 28 patients underwent FNA and CNB in the same lesion. The absolute sensitivity of FNA alone and CNB alone was 39.2 and 89.2 %, respectively. The absolute sensitivity of FNA alone improved to 92.9 % when combined with CNB (p \ 0.05 compared to FNA alone of NOS). In ILC cases, 25 patients underwent combined FNA and CNB. The absolute sensitivity of FNA alone and CNB alone was 40 and 92 %, respectively. The absolute sensitivity of FNA alone improved to 92 % when combined with CNB (p \ 0.001 compared to FNA alone of ILC). In MUC cases, 6 patients underwent combined FNA and CNB. The absolute sensitivity of FNA alone and CNB alone was 33.3 and 66.7 %, respectively. The absolute sensitivity of FNA alone did not show significant improvement even when combined with CNB (p = 0.07 compared to FNA alone of MUC). In patients with APO, 12 patients underwent combined FNA and CNB. The absolute sensitivity of FNA alone and CNB alone was 25 and 66.7 %, respectively. The absolute sensitivity of FNA alone improved to 66.7 % when combined with CNB (p \ 0.05 compared to FNA alone of APO).
8
CNB
Diagnostic results of FNA combined with CNB
8
Combined
absolute sensitivity of CNB for diagnosis of APO was 75.0 %. There was a significant difference in the absolute sensitivity between FNA and CNB in the diagnosis of APO (p \ 0.05).
3
Breast Cancer
Breast Cancer Table 4 Absolute sensitivity of FNA and CNB in the literature
Lieske et al. [10]
Number of cases
Absolute sensitivity
Total
Special histological type
FNA
CNB
763
ILC (93), ILC ? IDC (33)
65
80
ILC (4), MUC (2)
98.2
96.5
Garg et al. [11]
32
Kurita et al. [12]
182
ILC (14)
93
86
Barra et al. [14]
264
ILC (14), ILC ? IDC (5), MUC (5), MED (2)
68.5
84.7
Lam et al. [13]
37
Nagar et al. [5]
162
Liefland et al. [8]
522
NA
68
90
65
NA
57
60
Lifrange et al. [18]
MUC (37)
66.7
100
NA
89
100
Ballo et al. [19]
124
NA
97.5
90
Westenend et al. [20]
286
NA
92
88
Hatada et al. [39]
233
NA
86.9
86.2
FNA fine needle aspiration, CNB core needle biopsy, ILC invasive lobular carcinoma, IDC invasive ductal carcinoma, MUC mucinous carcinoma, MED medullary carcinoma, NA not addressed in the article Square parentheses reference number, round parentheses number of cases
diagnostic sensitivity, and proposed that CNB should be performed following inconclusive FNA results for patients in whom malignancy is highly suspected by clinical and/or radiological findings [12]. In that study, we included all tumor types, such as ILC and MUC, into a single group of ‘‘invasive cancers’’, leaving the efficacy of FNA and CNB on individual tumor types unclear. Thus far, reports on the efficacy of FNA and CNB on these special types of breast cancers are extremely limited (Table 4). For instance, Barra et al. compared the diagnostic accuracy between FNA and CNB in special types of breast cancer such as ILC (n = 14), MUC (n = 5), and medullary carcinoma (n = 2), and concluded that the presence of these cancer types did not affect the total efficacy of FNA and CNB [14] (Table 4). Other studies also mentioned that there was no special pathological subtype affecting the overall efficacy of FNA and CNB. We believe that their conclusions may be due to an insufficient number of special cases examined out of the entire cohort, and postulate that a larger scale study would clarify this issue. We therefore decided to determine the efficacy of FNA and CNB in particular types of breast cancer such as ILC, MUC, and APO. This time, the specificity and false positive rates were not evaluated because only patients with histologically confirmed diagnosis of carcinoma were retrospectively investigated for their pre-operative FNA and CNB diagnosis, which is a limitation of our study. In the present study, the diagnostic sensitivity for ILC by FNA was lower than that by CNB. In cases with combined FNA and CNB in the same lesion, the sensitivity of FNA alone significantly improved when combined by CNB. Previous reports have shown that the diagnostic sensitivity of ILC by FNA ranges from 52 to 78 %. This is lower than that of invasive ductal carcinomas, and
consistent with our results [23–28]. To our knowledge, however, no other reports have compared the diagnostic accuracy of CNB and FNA in regards to ILC. In particular, no studies have compared the sensitivity of FNA and CNB performed in the same ILC lesion. One of the reasons for the lower sensitivity for ILC may be the lower cellularity of FNA smears. Abdulla et al. indicated that the cellularity of FNA samples is disproportionally lower than that of corresponding histological sections [25]. Therefore, we suspected that the higher number of tumor cells in the CNB specimens enabled the higher diagnostic sensitivity of CNB in our ILC cases. ILC has several histological variations such as classic, solid, alveolar, tubulolobular and pleomorphic subtypes. Interestingly, FNA of pleomorphic subtype can yield a sensitivity of up to 90 %, which is higher than other subtypes of ILC. This is probably due to unique cytological features of pleomorphic subtype such as larger nuclei, distinct atypia and prominent nucleoli [27, 28]. Our ILC cases (n = 65) included 9 pleomorphic (13.8 %) and two tubulolobular (3 %) subtypes, and one solid (1.5 %) subtype. The remainder were all of the classic subtypes. We were not able to individually investigate the diagnostic accuracy for each subtype due to the small sample size, and we speculated that the diagnostic sensitivity in our ILC cases may be mostly reflective of the classic subtype, which was predominant. In our cases of MUC, the diagnostic sensitivity of total FNA was equivalent to that of total CNB of MUC and to that of total FNA in NST. Contrary to our results, Lam et al. have shown that the sensitivity of FNA in MUC was 66.7 %, which was significantly lower than the 100 % sensitivity of CNB [13]. The reason for this discrepancy is uncertain. One explanation is that diagnosis of MUC is challenged by a mild degree of nuclear atypia and abundant
123
Breast Cancer
extracellular mucin, resulting in variable cytological interpretation [29, 30]. Secondly, MUC presents with heterogeneous histological features. For instance, MUC has two histological subtypes, pure and mixed type that can affect the diagnostic sensitivity of FNA. In fact, the number of pure and mixed type of our cases was almost even (27 and 24, respectively) whereas in a previous study, pure type was predominant over mixed type (17 and 5, respectively) [31]. Moreover, MUC often expresses neuroendocrine markers that may have affected the morphology of tumor cells [32–35]. In our cases, we did not examine expression of neuroendocrine markers in all the MUC cases, and thus the effect of those proteins on the morphology of MUC tumor cells remained unknown. APO is a relatively rare neoplasm occurring in the breast, and is composed predominantly of cells with abundant eosinophilic cytoplasm reminiscent of metaplastic apocrine cells. Cytological features such as pleomorphism, nuclear crowding, increased nuclear-cytoplasmic ratio, and occasional mitotic figures are reported to be characteristic of malignant rather than benign apocrine lesions [36]. Cytological differentiation of APO from metaplastic apocrine cells is often challenging, and thus FNA of APO sometimes generates a false negative result [37, 38]. In our APO cases, the diagnostic sensitivity of FNA was significantly lower than that of NST. Concurrent FNA and CNB significantly improved the sensitivity. To our knowledge, no studies have thus far compared diagnostic sensitivity of FNA and CNB for APO. We therefore conclude that, to reach the definitive diagnosis, CNB should be performed when the FNA sample contains atypical apocrine cells that are difficult to interpret. In all the lesions except MUC cases, our results showed that combined use of FNA and CNB increased the diagnostic sensitivity. These results are consistent with our previous studies as well as other reports, but none of these studies specifically addressed the efficacy of combined use of FNA and CNB in special types of breast cancers [12, 20, 39]. For the first time, we clarified this aspect in special breast cancer types including NST, ILC, and APO. The reason that the sensitivity of MUC was not improved by combined FNA and CNB could be due to the small number of MUC cases (n = 6). Unfortunately, there are no other studies revealing the diagnostic sensitivity of combined FNA and CNB for MUC. We postulate that a large-scale study may reveal the usefulness of this combination in the future. In the 25 cases of our ILC group, the initial FNA was inconclusive, and combining with CNB eventually increased the sensitivity. Interestingly, the mean size of the tumor for these combined FNA and CNB cases was larger ([5 cm) than FNA alone cases. This result may suggest that CNB is more accurate when the ILC tumor is larger than average. We do not have an adequate explanation for
123
this phenomenon from the pathological point of view. Barra et al. showed that CNB is more suitable than FNA only for tumors ranging from 2 to 5 cm. For other sizes (less than 2 cm or more than 5 cm), FNA is equivalent to CNB [14]. Our result is not consistent with theirs, possibly because their study included 14 ILC out of 264 cases in total, and the vast majority of the remaining tumors were NST. We could not find other reports addressing the effect of tumor size on the difference between FNA and CNB results. Further studies are required to clarify this issue. In summary, we have compared diagnostic sensitivity of FNA and CNB in special types of breast cancer, including ILC, MC, APO and NST. The absolute sensitivity of CNB was higher than that of FNA for ILC and APO, but not for NST and MUC, where FNA and CNB had equivalent sensitivity. For the NST, ILC, and APO groups, combined FNA and CNB significantly improved the diagnostic sensitivity. Our results suggest that the diagnostic sensitivity of FNA and CNB is variable depending on the type of breast cancers. Thus, the use of biopsy procedures should be tailored to the type of tumor suspected to reach the accurate final diagnosis. Conflict of interest interest exists.
The authors have declared that no conflict of
References 1. He Q, Fan X, Yuan T, Kong L, Du X, Zhuang D, et al. Eleven years of experience reveals that fine-needle aspiration cytology is still a useful method for preoperative diagnosis of breast carcinoma. Breast. 2007;16:303–6. 2. Feoli F, Paesmans M, Van Eeckhout P. Fine needle aspiration cytology of the breast: impact of experience on accuracy, using standardized cytologic criteria. Acta Cytol. 2008;52:145–51. 3. Pilgrim S, Ravichandran D. Fine needle aspiration cytology as an adjunct to core biopsy in the assessment of symptomatic breast carcinoma. Breast. 2005;14:411–4. 4. Manfrin E, Mariotto R, Remo A, Reghellin D, Dalfior D, Falsirollo F, et al. Is there still a role for fine-needle aspiration cytology in breast cancer screening? Experience of the Verona Mammographic Breast Cancer Screening Program with real-time integrated radiopathologic activity (1999-2004). Cancer. 2008;114:74–82. 5. Nagar S, Iacco A, Riggs T, Kestenberg W, Keidan R. An analysis of fine needle aspiration versus core needle biopsy in clinically palpable breast lesions: a report on the predictive values and a cost comparison. Am J Surg. 2012;204:193–8. 6. Cobb CJ, Raza AS. Obituary: ‘‘alas poor FNA of breast-we knew thee well!’’. Diagn Cytopathol. 2005;32:1–4. 7. Clarke D, Sudhakaran N, Gateley CA. Replace fine needle aspiration cytology with automated core biopsy in the triple assessment of breast cancer. Ann R Coll Surg Engl. 2001;83:110–2. 8. Leifland K, Lagerstedt U, Svane G. Comparison of stereotactic fine needle aspiration cytology and core needle biopsy in 522 non-palpable breast lesions. Acta Radiol. 2003;44:387–91. 9. Oyama T, Koibuchi Y, McKee G. Core needle biopsy (CNB) as a diagnostic method for breast lesions: comparison with fine needle aspiration cytology (FNA). Breast Cancer. 2004;11:339–42.
Breast Cancer 10. Lieske B, Ravichandran D, Wright D. Role of fine-needle aspiration cytology and core biopsy in the preoperative diagnosis of screen-detected breast carcinoma. Br J Cancer. 2006;95:62–6. 11. Garg S, Mohan H, Bal A, Attri AK, Kochhar S. A comparative analysis of core needle biopsy and fine-needle aspiration cytology in the evaluation of palpable and mammographically detected suspicious breast lesions. Diagn Cytopathol. 2007;35:681–9. 12. Kurita T, Tsuchiya S, Watarai Y, Yamamoto Y, Harada O, Yanagihara K, et al. Roles of fine-needle aspiration and core needle biopsy in the diagnosis of breast cancer. Breast Cancer. 2012;19:23–9. 13. Lam WW, Chu WC, Tse GM, Ma TK, Tang AP. Role of fine needle aspiration and tru cut biopsy in diagnosis of mucinous carcinoma of breast–from a radiologist’s perspective. Clin Imaging. 2006;30:6–10. 14. Barra AeA, Gobbi H, de Rezende L, Alencar C, Gouveˆa AP, de Lucena CE, Reis JH, et al. A comparision of aspiration cytology and core needle biopsy according to tumor size of suspicious breast lesions. Diagn Cytopathol. 2008;36:26–31. 15. The Japanese Breast Cancer Society. General rules for clinical and pathological recording of breast cancer. 17th ed. Breast cytology and core needle biopsy reporting form (in Japanese) 2012;64-78. 16. Tsuchiya S, Akiyama F, Moriya T, Tsuda H, Umemura S, Katayama Y, et al. A new reporting form for breast cytology. Breast Cancer. 2009;16:202–6. 17. Lakhani SR, Cancer IAfRo, Organization WH. WHO Classification of Tumours of the Breast: International Agency for Research on Cancer; 2012. 18. Lifrange E, Kridelka F, Colin C. Stereotaxic needle-core biopsy and fine-needle aspiration biopsy in the diagnosis of nonpalpable breast lesions: controversies and future prospects. Eur J Radiol. 1997;24:39–47. 19. Ballo MS, Sneige N. Can core needle biopsy replace fine-needle aspiration cytology in the diagnosis of palpable breast carcinoma. A comparative study of 124 women. Cancer. 1996;78:773–7. 20. Westenend PJ, Sever AR. Beekman-De Volder HJ, Liem SJ. A comparison of aspiration cytology and core needle biopsy in the evaluation of breast lesions. Cancer. 2001;93:146–50. 21. Xie HB, Salhadar A, Haara A, Gabram S, Selvaggi SM, Wojcik EM. How stereotactic core-needle biopsy affected breast fineneedle aspiration utilization: an 11 year institutional review. Diagn Cytopathol. 2004;31:106–10. 22. Nassar A. Core needle biopsy versus fine needle aspiration biopsy in breast–a historical perspective and opportunities in the modern era. Diagn Cytopathol. 2011;39:380–8. 23. Sadler GP, McGee S, Dallimore NS, Monypenny IJ, DouglasJones AG, Lyons K, et al. Role of fine-needle aspiration cytology and needle-core biopsy in the diagnosis of lobular carcinoma of the breast. Br J Surg. 1994;81:1315–7. 24. Robinson IA, McKee G, Jackson PA, Cook MG, Kissin MW. Lobular carcinoma of the breast: cytological features supporting the diagnosis of lobular cancer. Diagn Cytopathol. 1995;13:196–201. 25. Abdulla M, Hombal S, al-Juwaiser A, Stankovich D, Ahmed M, Ajrawi T. Cellularity of lobular carcinoma and its relationship to
26.
27.
28.
29.
30.
31.
32.
33. 34.
35.
36.
37.
38.
39.
false negative fine needle aspiration results. Acta Cytol. 2000;44:625–32. Tan SM, Behranwala KA, Trott PA, Nasiri NA, Moskovic E, Brown G, et al. A retrospective study comparing the individual modalities of triple assessment in the pre-operative diagnosis of invasive lobular breast carcinoma. Eur J Surg Oncol. 2002;28:203–8. Hwang S, Ioffe O, Lee I, Waisman J, Cangiarella J, Simsir A. Cytologic diagnosis of invasive lobular carcinoma: factors associated with negative and equivocal diagnoses. Diagn Cytopathol. 2004;31:87–93. Menet E, Becette V, Briffod M. Cytologic diagnosis of lobular carcinoma of the breast: experience with 555 patients in the Rene Huguenin Cancer Center. Cancer. 2008;114:111–7. Gupta RK, McHutchison AG, Simpson JS, Dowle CS. Value of fine needle aspiration cytology of the breast, with an emphasis on the cytodiagnosis of colloid carcinoma. Acta Cytol. 1991;35:703–9. Ventura K, Cangiarella J, Lee I, Moreira A, Waisman J, Simsir A. Aspiration biopsy of mammary lesions with abundant extracellular mucinous material. Review of 43 cases with surgical followup. Am J Clin Pathol. 2003;120:194–202. Cyrta J, Andreiuolo F, Azoulay S, Balleyguier C, Bourgier C, Mazouni C, et al. Pure and mixed mucinous carcinoma of the breast: fine needle aspiration cytology findings and review of the literature. Cytopathology. 2013;24:377–84. Sapino A, Papotti M, Pietribiasi F, Bussolati G. Diagnostic cytological features of neuroendocrine differentiated carcinoma of the breast. Virchows Arch. 1998;433:217–22. Tse GM, Ma TK. Fine-needle aspiration cytology of breast carcinoma with endocrine differentiation. Cancer. 2000;90:286–91. Tang W, Taniguchi E, Wang X, Mori I, Kagiya T, Yang Q, et al. Loss of cell cohesion in breast cytology as a characteristic of neuroendocrine carcinoma. Acta Cytol. 2002;46:835–40. Kawasaki T, Nakamura S, Sakamoto G, Kondo T, TsunodaShimizu H, Ishii Y, et al. Neuroendocrine ductal carcinoma in situ of the breast: cytological features in 32 cases. Cytopathology. 2011;22:43–9. Yoshida K, Inoue M, Furuta S, Sakai R, Imai R, Hayakawa S, et al. Apocrine carcinoma vs. apocrine metaplasia with atypia of the breast. Use of aspiration biopsy cytology. Acta Cytol. 1996;40:247–51. Ng WK. Fine needle aspiration cytology of apocrine carcinoma of the breast. Review of cases in a three-year period. Acta Cytol. 2002;46:507–12. Haji BE, Das DK, Al-Ayadhy B, Pathan SK, George SG, Mallik MK, et al. Fine-needle aspiration cytologic features of four special types of breast cancers: mucinous, medullary, apocrine, and papillary. Diagn Cytopathol. 2007;35:408–16. Hatada T, Ishii H, Ichii S, Okada K, Fujiwara Y, Yamamura T. Diagnostic value of ultrasound-guided fine-needle aspiration biopsy, core-needle biopsy, and evaluation of combined use in the diagnosis of breast lesions. J Am Coll Surg. 2000;190:299–303.
123
