Fax: (011) (78) 6542108
Evaluation of fine-needle aspiration cytology of breast masses in males
Article first published online: 13 MAR 2002
Copyright © 2002 American Cancer Society
Volume 96, Issue 2, pages 1–5, 25 April 2002
How to Cite
Westenend, P. J. and Jobse, C. (2002), Evaluation of fine-needle aspiration cytology of breast masses in males. Cancer, 96: 1–5. doi: 10.1002/cncr.10483
- Issue published online: 8 APR 2002
- Article first published online: 13 MAR 2002
- Manuscript Accepted: 25 SEP 2001
- Manuscript Revised: 21 SEP 2001
- Manuscript Received: 28 JUN 2001
- fine-needle aspiration cytology (FNAC);
The reliability of fine-needle aspiration cytology (FNAC) of breast masses in males could be compromised by lack of experience because breast carcinoma is rare in males and FNAC is not often used. In addition, FNAC of the more often encountered gynecomastia may lead to overdiagnosis. Therefore, in the current study the authors evaluated their experience with FNAC of breast masses occurring in males.
A total of 153 FNACs of the male breast obtained between 1985 until the end of 2000 were retrieved from the electronic files of the study institution. A total of 141 FNACs were taken from unilateral lesions in men age > 24 years, the group of men believed to be most at risk for breast carcinoma. Histologic follow-up was retrieved from the same files and was available for 72 FNACs. For specimens without histologic follow-up the nationwide pathology database was consulted and no cases of breast carcinoma were found.
The inadequate rate was 13%. When inadequate FNACs were included in the calculations, the sensitivity was found to be 87% and the specificity 78%. When the inadequate FNACs were excluded from the calculations the sensitivity was reported to be 100% and the specificity 89%. The positive predictive value of a diagnosis of malignancy was 100%. During the study period the authors' institution examined approximately 10,000 FNAC specimens from male and female breasts and 399 resection specimens from the breasts of men age > 24 years with unilateral lesions. In this last group, preoperative FNAC reduced the benign-to-malignant ratio from 19.8:1 to 3.5:1.
FNAC of the male breast is a reliable procedure in a setting in which sufficient numbers of FNACs of the breast are examined. The authors believe FNAC should be used more often in the preoperative evaluation of breast lesions occurring in males. Cancer (Cancer Cytopathol) 2002;96:000–000. © 2002 American Cancer Society.
Fine-needle aspiration cytology (FNAC) is a widely used method in the management of breast lesions in women. The technique is simple and inexpensive and good results are attainable.1, 2 However, it is used much less often in men, mainly because breast masses in males are less frequent. The incidence of breast carcinoma in men is low (0.8 per 100,000 person-years in men compared with 121.3 person-years in women in the Netherlands [data from the 1995 Dutch cancer registry]), and this may affect the reliability of FNAC. In addition, the majority of breast masses are the result of gynecomastia, a lesion that may pose some problems because under certain circumstances it can be confused with carcinoma.3, 4 To our knowledge only a limited number of articles evaluating the use of FNAC in breast masses occurring in men has been published to date.5–11 In the current study, we describe our experience with FNAC of breast lesions occurring in men over a period of 16 years. We also examined excision specimens from male breast lesions taken over the same time period irrespective of previous FNAC to document possible underuse of FNAC.
MATERIALS AND METHODS
All records of male patients undergoing FNAC of the breast between 1985 and the end of 2000 were retrieved from our electronic files (PALGA) and analyzed. The majority of aspirates were obtained on an outpatient basis and all were air-dried and stained by the May–Grünwald–Giemsa method. Results were classified as unsatisfactory, benign, atypia, suspicious, or malignant. Histologic follow-up of these aspirates was obtained from the same electronic files and correlated with these aspirates. Calculations of sensitivity, specificity, and other parameters were performed as in the National Heath Service Breast Screening Programme in the U.K..12 In these calculations, unsatisfactory samples are included; therefore, the results reflect the entire procedure including the taking of the sample. No revision of the cytology samples was performed. Sensitivity is calculated in two ways: one in which only samples with a diagnosis of malignant are included (called absolute sensitivity) and one in which samples with a diagnosis of malignant, suspicious, or atypia are included (called complete sensitivity). Specificity also is calculated in two ways: one in which only samples with a diagnosis of benign and for which histologic follow-up was available are included (called specificity [biopsy cases only]) and another (called specificity [full]) in which all samples with a diagnosis of benign with or without histologic follow-up and those samples with a diagnosis of atypia without histologic follow-up assuming a benign lesion are included.
To compare the results of the current study with those reported in the literature,5–11 we extracted the data from these reports and recalculated the sensitivity, specificity, and other parameters as we did for the current study data because the data usually were not reported in this way. To this end, FNACs with a specific benign diagnosis (e.g., “gynecomastia”) and those with a nonspecific benign diagnosis (e.g., “not malignant”) were grouped together as “benign” and FNACs with carcinoma of the breast and metastasis were grouped together as “malignant.” One report by Gupta et al.6 appears to be an expansion of previous observations,5 so only the most recent report was used. In addition, we retrieved all histologic records of male breast specimens over the same period and analyzed these data with regard to age distribution, laterality, and diagnosis.
Between 1985 and the end of 2000 a total of approximately 10,000 FNAC breast specimens was evaluated, 153 of which were from male breasts. Histologic follow-up was available for 72 FNAC samples and was not available for 1 sample with atypia, 8 unsatisfactory samples, and 72 benign samples. Of the 153 FNAC specimens, 145 were from men age ≥ 25 years; of these 145 specimens, 141 were from unilateral lesions. Histologic follow-up was available for 67 of these 141 lesions. For specimens without histologic follow-up from our own files, the nationwide electronic pathology files (PALGA) was consulted; no histologic follow-up was found in other laboratories. Therefore we assume that there were no malignancies in the group of FNAC specimens without histologic follow-up. A total of 15 malignancies were found in the follow-up data. The results of the calculations of all the relevant parameters are summarized in Table 1. Although the acceptable values in this table were for females in a breast screening program, they are included in Table 1 for reference. All values shown are above these acceptable values. When the inadequate FNAC samples were excluded from the calculations, the sensitivity was found to be 100% and the specificity 89%.
|Current study (95% CI)||Acceptable values (%)12|
|Absolute sensitivity||67 (59–74)||> 60|
|Complete sensitivity||87 (81–92)||> 80|
|Specificity (biopsy cases only)||61 (54–69)|
|Specificity (full)||78 (71–84)||> 60|
|Ppv malignant||100||> 95|
|Ppv suspicious||40 (32–48)|
|False-negative rate||0||< 5|
|False-positive rate||0||< 1|
|Inadequate rate||13 (8–18)||< 25|
|Inadequate rate from tumors||13 (8–19)|
|Suspicious rate||10 (6–15)||< 20|
In the same period, a total of 676 histologic breast specimens from male breasts were evaluated. Of these 676 specimens, 399 were from unilateral lesions in men age ≥ 25 years and all 31 malignancies, including 28 primary breast tumors, 2 melanoma metastases, and 1 malignant lymphoma, were included in this group.
The results of the current study demonstrate that FNAC of the male breast can provide data regarding sensitivity, specificity, and other parameters that can be compared with the good results that have been reported with FNAC in the female breast.1, 2 Comparing our results with the acceptable values that have been established by the National Heath Service Breast Screening Programme in the U.K. demonstrates that the results of the current study are good.12 However, these are not the best studies with which to compare the current study results because these reports concern FNAC in females, a population with a much higher rate of breast carcinoma, partly on nonpalpable lesions, and partly in screened populations. Studies in men are concerned uniformly with symptomatic palpable lesions. Unlike the situation in women, to our knowledge there are no screening programs directed at detecting asymptomatic lesions in men with an increased risk of breast carcinoma based on a strong family history or known genetic mutations. Therefore, comparing the results of the current study with those of studies regarding FNAC of the male breast is preferable. There are several studies concerning FNAC of the male breast, some of which have examined the cytologic features of gynecomastia3, 4 and others that have studied the cytologic features of carcinoma.13 However, we could find only a limited number of articles that examined the value of FNAC in the evaluation of masses found in male breasts.5–11 Because the results of FNAC in these studies were reported in different ways, we had to perform some recalculations before we were able to make a comparison (Table 2). Nevertheless, some important differences remain. There are minor differences with regard to the malignancy rate, but in 1 study7 the number of malignancies diagnosed is high because 33 of the 64 patients had an extramammary malignancy and consequently 7 of the 12 malignancies detected were metastatic neoplasms. There are major differences with regard to histologic follow-up, which has been reported to range from 6.1–47.1%. There also are major differences with regard to the follow-up of those lesions that were not excised because (to our knowledge) only one other study in addition to the current one checked for missed malignancies by consulting a cancer registry (Lilleng et al.9) or nationwide pathology database (current study). In one study,8 a review of the cytology slides was performed that may better reflect what is attainable with this technique by experts, but not what is encountered in everyday practice. Therefore, although there are some limitations when comparing these studies, it appears that acceptable values for the sensitivity, specificity, and positive predictive value of a “malignant” finding can be obtained with FNAC as shown in the studies in Table 2. In Table 2, inadequate samples were included in the calculations. The lower sensitivity found in the current study compared with the other studies was caused by two inadequate FNAC breast carcinoma samples. Differences in the sensitivity and specificity values as calculated by us from those reported by Joshi et al.11 can be explained by the fact that we included inadequate samples in our calculations and Joshi et al. did not.
|Gupta et al., 19916||Sneigel et al., 19937||Das et al., 19958||Lilleng et al., 19959||Vetto et al., 199810||Joshi et al., 199911||Current study|
|No. of FNACs||99||64||185||241||51||507||153|
|No. of malignanciesa||4||12||3||10||6||43||15|
|Histologic FU (%)||6.1||35.9||13.5||5.8||41.2||19.1||47.1|
|Complete sensitivity||100||100||100||100||100||97 (94–100)||87 (81–92)|
|Specificity (full)||85 (78–92)||90 (85–96)||85 (79–92)||87 (80–93)||96 (91–100)||69 (60–78)||78 (71–84)|
|Ppv malignant||100||100||100||89 (83–95)||100||100||100|
|Ppv suspicious||0||100||100||40 (30–50)||0||24 (16–33)c||40 (32–48)|
|False-positive rate||0||0||0||10 (4–16)||0||0||0|
|Inadequate rate||12 (6–19)||8 (5–13)||14 (7–21)||11 (5–17)||0||22 (14–31)||13 (6–15)|
Overestimating florid gynecomastia resulted in the low predictive value of a “malignant” finding and some false-positive results in one study.9 This problem has been noted previously3, 4 but appears to be a minor problem judging from the current study and the studies summarized in Table 2. The majority of malignancies in these and other studies are primary breast carcinomas although a minority is comprised of metastatic disease.7, 9, 11, 13 In two studies,6, 11 the number of breast FNAC samples was mentioned, enabling us to calculate that 1.4% and 3.8%, respectively, of the breast FNAC samples were from males, which is similar to the rate of 1.5% reported in the current series. This finding demonstrates that although breast FNAC in males is not often performed, it can be reliably performed given sufficient experience with breast FNAC in females.
The objective of performing FNAC is to select those lesions that need to be treated and to avoid unnecessary surgery in patients with benign lesions. The success of FNAC therefore is reflected by the benign:malignant ratio in the surgical specimen. In the current study FNAC was performed in 141 men age ≥ 25 years with a unilateral lesion, the population considered to be most at risk for breast carcinoma, followed by surgery in 67 of these men. In this group 15 malignancies were found, making the benign-to-malignant ratio 3.5:1. During the same period, 332 men age ≥ 25 years with a unilateral breast lesion underwent surgery without a preoperative FNAC and 16 malignancies were found, resulting in a benign-to-malignant ratio of 19.8:1. This finding demonstrates that FNAC of the male breast is an underutilized diagnostic technique. Lilleng et al.9 made similar observations. We conclude that FNAC of the male breast is a reliable diagnostic technique that should be used more often in the preoperative evaluation of breast masses occurring in males.