Fax: (011) 81 3 5771 0512
Intraductal biopsy for diagnosis and treatment of intraductal lesions of the breast
Article first published online: 13 OCT 2004
Copyright © 2004 American Cancer Society
Volume 101, Issue 10, pages 2164–2169, 15 November 2004
How to Cite
Matsunaga, T., Kawakami, Y., Namba, K. and Fujii, M. (2004), Intraductal biopsy for diagnosis and treatment of intraductal lesions of the breast. Cancer, 101: 2164–2169. doi: 10.1002/cncr.20657
- Issue published online: 29 OCT 2004
- Article first published online: 13 OCT 2004
- Manuscript Accepted: 16 AUG 2004
- Manuscript Revised: 4 AUG 2004
- Manuscript Received: 13 APR 2004
- breast carcinoma;
- nipple discharge;
- intraductal papilloma;
- ductal carcinoma;
Bloody nipple discharge is a significant clue in the detection of ductal carcinoma of the breast. In the past, pathologic diagnoses were obtained exclusively via excision, but recently developed mammoscopic techniques have been found to yield valuable information relating to the diagnosis of intraductal lesions.
Mammary duct endoscopy (i.e., mammoscopy) was performed a combined total of 407 times for 295 patients who experienced nipple discharge. Intraductal breast biopsy (IDBB) under mammoscopic observation was performed in 193 intraductal papillomas (from a total of 107 patients) and 30 ductal carcinomas (from a total of 27 patients); IDBB was performed a combined total of 36 times in the 27 patients who had breast carcinoma and yielded 21 diagnostic specimens (58.3%). In addition, the therapeutic value of IDBB was assessed in 70 patients with intraductal papilloma who had undergone more than 3 years of follow-up; these 70 patients harbored a combined total of 75 intraductal papillomas.
IDBB correctly identified the presence of carcinoma in 9 of 27 patients (33.3%); 7 other lesions (25.9%) were placed in the suspected carcinoma (i.e., atypical papillary lesion) category, and 5 (18.5%) were identified as intraductal papillomas. Using IDBB, it was difficult to collect diagnostic specimens from patients with breast carcinoma, because of the location and weak tissue cohesiveness of these lesions compared with intraductal papillomas. The 193 intraductal biopsies performed on intraductal papillomas yielded only 20 specimens that were insufficient for diagnosis. IDBB exhibited therapeutic efficacy in 54 of 70 patients with intraductal papilloma (77.6%) who had more than 3 years of clinical follow-up. Therapeutic results tended to be less favorable for patients who had intraductal lesions in multiple duct lobular units.
Mammoscopy can contribute not only to the diagnosis of cases of nipple discharge but also to the treatment of intraductal papilloma. Cancer 2004. © 2004 American Cancer Society.
Since the mammary duct endoscopy (i.e., mammoscopy) procedure was first reported in 1988,1 it has been applied in the clinical setting for the detection of intraductal lesions.2–8 Intraductal breast biopsy (IDBB) under mammoscopic observation is used to obtain pathologic diagnoses of intraductal lesions that are accompanied by nipple discharge. The introduction of IDBB under direct observation has ushered in a new era in the pathologic diagnosis and treatment of secreting benign intraductal lesions. In the current study, we investigated the usefulness of this technique with regard to the differential diagnosis of ductal carcinoma and intraductal papilloma and also with regard to the treatment of secreting intraductal papillomas.
MATERIALS AND METHODS
Between October 1993 and December 2002, mammoscopy was performed a combined total of 407 times in 295 patients who had nipple discharge in the absence of a breast tumor at the Tokyo Metropolitan Cancer Detection Center (Tokyo, Japan). The average patient age was 47.7 years (standard deviation, 11.2 years; range, 23–88 years). A single draining duct was observed in 240 patients, 2 draining ducts were observed in 35 patients, 3 draining ducts were observed in 11 patients, and ≥ 4 draining ducts were observed in 9 patients. In cases in which multiple orifices were present, the draining ducts to be observed on mammoscopy were identified by their bloody appearance.
Regardless of whether bloody material was present, mammography, ultrasonography, and nipple discharge cytology were routinely performed at the Tokyo Metropolitan Cancer Detection Center. The presence of a cluster of ductal cells was considered indicative of an intraductal papillary lesion. Galactography was performed if cytologic findings suggested the presence of a papillary lesion or if a bloody background was readily observable, and indications for mammoscopy were based on galactographic findings. IDBB under mammoscopic observation was performed a combined total of 223 times in 134 patients.
In the current study, the usefulness of IDBB in the diagnosis of intraductal lesions was investigated. In addition, the therapeutic value of IDBB was assessed in 70 patients (harboring a combined total of 75 intraductal papillomas) for whom clinical follow-up lasting longer than 3 years was possible.
In most cases, IDBB was performed using a metallic tube with a side aperture near the tip (JN Biopsy Needle; Hakko Company, Tokyo, Japan) (Figs. 1, 2).9 In other cases, when a papillary lesion was present in a duct that diverged from the main duct at a sharp angle, a 19-gauge elastic tube with a diagonally transected tip was used. Pancreatic duct forceps with an outer diameter of 1 mm (FB-44D; Olympus, Tokyo, Japan) were used when the intraductal papilloma was located in a lactiferous sinus and the caliber of the duct was large enough to use forceps. There were very few of these cases.
Mammoscopes measuring 0.55, 0.7, and 0.72 mm in outer diameter (Fiber Tech Company, Tokyo, Japan) were used in the IDBB procedure. All were semirigid endoscopes.
In the 295 patients for whom mammoscopy was performed, the intraductal lesions that were diagnosed included 154 intraductal papillomas (in a total of 140 patients) and 67 breast carcinomas (in a total of 66 patients). In the remaining 89 patients, mammoscopy yielded no significant findings. When multiple lesions were found in the same duct-lobular unit, they were counted as a single lesion. IDBB under mammoscopic observation was performed in a total of 193 intraductal papillomas (from 107 patients) and in a total of 30 ductal carcinomas (from 27 patients). The therapeutic efficacy of IDBB was evaluated in 70 patients (with a combined total of 75 intraductal papillomas) for whom clinical follow-up lasting longer than 3 years was possible.
Accuracy of IDBB in the Diagnosis of Ductal Carcinoma
IDBB under mammoscopic observation was performed a total of 36 times, yielding 21 diagnostic specimens (58.3%), in the 27 patients who had breast carcinoma. Breast carcinoma was diagnosed in 9 of these specimens, a diagnosis of suspected carcinoma (i.e., atypical papillary neoplasia) was rendered in 7 cases, and intraductal papilloma was diagnosed in the remaining 5 specimens; thus, IDBB was successful in 16 of these 21 cases. In addition, IDBB yielded accurate diagnoses for all 107 patients who had intraductal papillomas. Overall, IDBB had a sensitivity of 76.2%, a specificity of 100%, a true-positive rate of 100%, and a true-negative rate of 95.5%.
In five patients diagnosed with intraductal papilloma, ductal lavage cytology from peripheral ducts (performed using an epidural catheter) indicated the presence of carcinoma, and duct-lobular segmentectomy subsequently was performed under local anesthesia. Eight patients had intraductal papilloma in the proximal duct coexisting with carcinoma in peripheral ducts. The coexistence of papilloma and carcinoma was confirmed on the first IDBB in one of these eight cases. Among the remaining seven patients, IDBB yielded a diagnosis of atypical papillary neoplasia in five and a diagnosis of intraductal papilloma in one; in a seventh patient, the amount of diagnostic material available was insufficient. In one patient with coexisting papilloma and carcinoma, IDBB was performed 4 times over the course of 8 months, with the final biopsy yielding a diagnosis of atypia.
With regard to the 27 patients who had breast carcinoma, IDBB was performed because it was thought to be the most reliable method for collecting diagnostic specimens. Other diagnostic findings included microcalcifications, which were documented in five patients, and sonographic abnormalities (e.g., irregularly shaped low-echoic areas and irregularly shaped dilated ducts), which were found in nine patients. Sonography-guided fine-needle aspiration cytology was performed in four patients who had negative findings on IDBB, yielding breast carcinoma diagnoses in three of these four.
Breast carcinoma was detected in another 2 patients approximately 3 years after IDBB; one of these patients initially had insufficient material for pathologic diagnosis but had benign findings on ductal lavage cytology, and the other initially was diagnosed with an atypical papillary lesion. In both of these patients, nipple discharge ceased after IDBB, and consequently, they believed that their lesions had resolved. After 3 years, however, both patients self-detected breast induration and thus returned to our clinic. The locations of the carcinomas detected in these patients coincided with the duct-lobular units previously visualized on mammary ductography.
Efficacy of IDBB in the Treatment of Intraductal Papilloma
IDBB yielded sufficient diagnostic material in 173 of 193 cases (89.6%). Of the 70 patients (harboring a combined total of 75 lesions) who were followed for more than 3 years, 36 experienced cessation of nipple discharge after their first IDBB, and another 13 experienced cessation of nipple discharge after a subsequent IDBB. Thus, nipple discharge was halted in a total of 49 patients (70.0%). In another five patients in whom intraductal masses could not be detected on either ductography or mammoscopy, the presence of a bloody background and clusters of ductal cells did not persist after IDBB, despite the fact that nipple discharge continued to occur. Thus, IDBB was found to have therapeutic efficacy in 54 patients (77.1%). In 15 patients, bloody nipple discharge did not cease after IDBB; IDBB was performed twice in one of these patients, 3 times in another, 4 times in 2 patients, 5 times in 3 patients, and more than 5 times in 8 patients. Recurrent intraductal papilloma developed a year after IDBB in one patient (Fig. 3).
Factors influencing the therapeutic efficacy of IDBB were examined in the 70 patients with intraductal papilloma who were followed for more than 3 years. IDBB was considered to have been effective in the 49 patients who experienced cessation of nipple discharge and in the 5 patients in whom intraductal masses could not be detected. For all other patients, including the one who experienced a recurrence of intraductal papilloma 1 year after biopsy, IDBB was considered to have been ineffective.
Papillomas were located in major (nondivergent) lactiferous ducts in 31 of the 54 patients for whom IDBB was therapeutically effective (57.4%) and in 6 of the 16 patients for whom IDBB was therapeutically ineffective (37.5%); this difference was not significant (chi-square test: P = 0.164). Multiple lesions were observed in 12 patients in whom IDBB exhibited therapeutic efficacy (22.2%) and in 8 patients in whom IDBB did not exhibit therapeutic efficacy (50.0%) (chi-square test: P = 0.031). Multiple nipple discharge orifices were present in 19 patients for whom IDBB was effective (35.2%) and in 11 patients for whom IDBB was ineffective (68.8%) (chi-square test: P = 0.017). Fifteen patients for whom IDBB was therapeutically effective (27.8%) had bilateral nipple discharge, compared with 8 patients for whom IDBB was therapeutically ineffective (50.0%) (chi-square test: P = 0.098). With regard to lesion shape, intraductal papillomas were most commonly classified as hemispheric or papillary, whereas ductal carcinomas were most commonly classified as flat protrusions8; flat protruding intraductal masses were noted in 3 of the 54 patients in whom IDBB exhibited therapeutic efficacy (5.6%) and in 3 of the 16 patients in whom IDBB did not exhibit therapeutic efficacy (18.8%) (chi-square test: P = 0.098). Overall, our findings suggest that IDBB is less likely to exhibit therapeutic efficacy in patients with multiple lesions (Table 1).
|Factor||Treatment effective (n = 54)||Treatment ineffective (n = 16)||Chi-square P value|
|No. of divergences||0.164|
|Presence of multiple lesions||12||8||0.031|
|Presence of multiple orifices||19||11||0.017|
|Hemispheric or papillary||51||13|
Bloody nipple discharge is an important clue in the detection of ductal carcinoma of the breast. For the diagnosis of ductal lesions, ductography and nipple discharge cytology typically are performed. In the past, pathologic diagnosis was performed exclusively via excision. Although microdochectomy remains an effective option for the diagnosis and treatment of nipple discharge,10 mammoscopy is an epoch-making method for the assessment of intraductal lesions accompanied by nipple discharge, as this technique yields new information on such lesions and delivers the tissue specimens that are necessary for IDBB-based diagnosis.8 In addition to serving as a diagnostic tool, IDBB also enabled the removal of intraductal papillomas and led to the elimination of bloody nipple discharge in a number of patients.
Mammoscopy is not useful in the identification of the terminal duct-lobular unit from which the breast carcinoma originates. In their study of 42 mastectomy specimens, Diets et al.11 reported that mammoscopy into the distal ducts was successful in 43% of all cases. Mammoscopy facilitates the rendering of a diagnosis by allowing observation of the proximal portion of the intraductal carcinoma and also by generating diagnostic specimens. IDBB was less effective in the collection of diagnostic material from breast carcinomas, possibly because these lesions were located in peripheral duct-lobular units and exhibited weak tissue cohesiveness compared with intraductal papillomas. In general, peripheral ductal carcinomas were difficult to access for the collection of diagnostic material. Only the proximal portion of a ductal carcinoma can be detected mammoscopically, with this finding indicating the extension of the lesion beyond terminal duct-lobular units in large ducts.12 IDBB-based pathologic diagnosis was unsuccessful in five patients with breast carcinoma. In one of these five cases, there was a coexisting intraductal papilloma on the proximal side, and consequently, the distal intraductal lesion could not be biopsied. In the remaining four cases, the diagnosis of malignancy was difficult because the specimens contained myoepithelial cells, making the proliferation of ductal cells less evident. Another factor that was thought to be responsible for misdiagnosis in these cases was small specimen size.
Mammoscopy also was effective in diagnosing cases of coexisting intraductal papilloma and carcinoma in the same duct-lobular unit. When such lesions are present, the intraductal papilloma typically is located in the proximal portion of the duct-lobular unit, and consequently, it is possible for the breast carcinoma to go undetected on discharge cytology and ductography. In these cases, a mammoscope can be inserted into the periphery beyond the proximal lesion, allowing visualization of the peripheral lesion; in addition, the proximal mass can be removed mammoscopically, allowing the peripheral lesion to be biopsied.8 Thus, mammoscopy is capable of contributing to the pathologic diagnosis of intraductal masses and exhibits limited invasiveness compared with excisional biopsy.
Breast carcinoma was detected 3 years after IDBB in 2 patients. In general, surgical excision or close follow-up should be recommended for patients with insufficient diagnostic material or atypical papillary lesions diagnosed on IDBB. In addition, biannual follow-up ultrasonography and testing for nipple discharge (via palpation), along with annual mammography, are recommended for all patients following mammoscopy. Among the 89 patients who had negative mammoscopic findings, breast carcinoma was detected in 3 during follow-up. In all three of these patients, nipple discharge ceased after mammoscopy, and in each case, the detected breast carcinoma was located in a different area from the area previously identified on galactographic examination. Clinical follow-up is believed to be critical for patients with bloody nipple discharge.13
Surgical excision was recommended for patients with atypical papillary lesions diagnosed on IDBB and for patients with multiple intraductal lesions in peripheral ducts, due to the poor therapeutic efficacy of IDBB and the elevated risk of malignancy in such patients. For these patients, an epidural catheter with an outer diameter of 0.8 mm was advanced into the peripheral ducts, and hand-applied suction was used to collect peripheral intraductal fluid. As an alternative to excision, follow-up performed every 3 months during the first year and every 6 months during the following year was believed to be appropriate for patients with multiple lesions in peripheral ducts, provided that IDBB was performed on all mammoscopically observed papillomas and that ductal lavage cytology using an epidural catheter revealed benign findings.
In the past, surgical excision was the only method available for the pathologic diagnosis and treatment of nipple discharge10,14–20; however, the introduction of mammoscopy has heralded the beginning of a new era, in which IDBB performed under direct observation is used to diagnose and treat secreting intraductal papillomas. Mammoscopy can serve as an important adjunct intervention in breast-conservation therapy,21–23 and in the percutaneous endoscopic approach, mammoscopy allows direct visualization and biopsy of the lesion of interest.24 Mammoscopic evaluation also assists in the elimination of unnecessary surgical excision when benign lesions are present. With regard to the treatment of solitary intraductal papillomas, surgical excision has been supplanted by IDBB under mammoscopic observation as the preferred intervention.
The presence of multiple lesions was associated with poor therapeutic efficacy in patients undergoing IDBB. It also was suggested that the multifocal generation of intraductal papillomas was associated with the presence of multiple nipple orifices. Duct-lobular unit distribution patterns are infinitely complex, and it is not always possible to visualize all mammary duct branches using ductography. When an intraductal papilloma was located in a branch that diverged at a sharp angle, only the proximal margin in a major duct could be detected mammoscopically. In our previous report,8 neither the amount of specimen collected nor the amount of lesion remaining after IDBB was found to be significantly associated with the efficacy of treatment for nipple discharge. Morphologic changes have been reported following mammoscopy,25 and there is speculation that the biopsy procedure could lead to necrosis of the intraductal papilloma being examined. Overall, our findings indicate that mammoscopy contributes not only to the diagnosis of lesions in patients with nipple discharge but also to the treatment of intraductal papilloma.
The authors thank Dr. J. Patrick Barron of the International Medical Communications Center, Tokyo Medical University (Tokyo, Japan), for his review of the current article.
- 1A new concept in breast investigation: echo-histological acino-ductal analysis or analytic echography. Biomed Pharmacother. 1988; 42: 282–296..
- 5Intraductal biopsy of the breast (IDBB) for intraductal lesions: a case of breast cancer diagnosed by IDBB. Jpn J Breast Cancer. 1989; 4: 253–258., , , et al.
- 9Endoscopic biopsy in the intraductal lesions of the breast for the mammary duct fiberscope. Jpn J Breast Cancer. 1995; 10: 405–409., .
- 13Breast cancer screening for nipple discharge: diagnostic procedure for cancer detection. J Jpn Assoc Breast Cancer Screen. 2002; 11: 281–288., , , et al.
- 14A simple operation for the discharge nipple. Surgery. 1938; 4: 914–916..
- 16Preoperative galactography increased the diagnostic yield of major duct excision for nipple discharge. Cancer. 1997; 82: 1874–1880., , , .