SEARCH

SEARCH BY CITATION

Keywords:

  • fiberoptic ductoscopy;
  • fluorescence in situ hybridization;
  • breast carcinoma

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

BACKGROUND

Breast carcinoma and precancer are believed to start in the lining of the milk duct or lobule. Ductography and fiberoptic ductoscopy (FDS) are used to identify abnormal intraductal lesions, although it is difficult to distinguish malignant from benign cases. Therefore, we studied the clinical usefulness of fluorescence in situ hybridization (FISH) analysis of a numerical aberration of chromosomes (aneusomy) using ductal lavage from patients with nipple discharge.

METHODS

We applied ductography and FDS to 90 women who had nipple discharge. Ductal lavages obtained from patients with positive ductography and/or FDS findings were subjected to cytology and FISH analysis using centromere probes for chromosomes 1, 11, and 17. Patients with samples that showed aneusomy in at least one of the three chromosomes were diagnosed as positive.

RESULTS

Histologic evaluation revealed 54 benign lesions and six malignancies. The sensitivity, specificity, and diagnostic accuracy were 33.3%, 88.9%, and 83.3%, respectively, for cytology and 100%, 100%, and 100%, respectively, for FISH.

CONCLUSION

The results demonstrated that FISH has a diagnostic accuracy comparable to cytology. This technique has 100% specificity is making a definitive diagnosis of malignancy in patients with indeterminate cytologic results, suggesting that FISH diagnosis can be a good adjunct to cytology. Cancer 2003;97:690–4. © 2003 American Cancer Society.

DOI 10.1002/cncr.11091

Abnormal nipple discharge is one of the frequent complaints of patients who visit breast clinics.1, 2 Most patients with nipple discharge have fibrocystic change, a benign disease in middle-aged women. However, 7.4–20.9% of patients with abnormal nipple discharge also have breast carcinoma.1, 3–5 Therefore, it is important to distinguish between discharge in malignant and benign cases. Generally, the presence of a malignant lesion in the breast is suspected if the discharge is bloody and observed unilaterally in postmenopausal women. The Methods used to evaluate the discharge include ductography, cytologic study, and measurement of the carcinoembryonic antigen level.6–11 However, these methods have high rates of false-negative and false-positive results, requiring pathologic study following surgical biopsy.1 A noninvasive and more reliable diagnostic modality is needed to evaluate nipple discharge.

Fiberoptic ductoscopy (FDS) is an emerging technique. It allows direct visual access to the breast ductal system through nipple orifice cannulation and exploration.12–14 The technique identifies abnormal lesions, but does not readily distinguish between benign and malignant cases.15, 16 Therefore, histopathologic evaluation is currently the most reliable method for obtaining a final diagnosis. To avoid segmentation of the mammary glands in patients with benign lesions fluorescence in situ hybridization (FISH) is being used to evaluate nipple discharge.

The FISH technique evaluates metaphase or interphase chromosomes to detect genetic aberrations in a variety of solid tumors.17, 18 A physical deletion or aneusomy (monosomy or polysomy) involving a specific chromosome is detected easily by FISH as a loss or gain of fluorescence signals. In addition, FISH analysis of interphase nuclei detects genetic aberrations that are difficult to identify by cytology.

Genetic alterations have been reported at specific sites on chromosomes 1p, 1q, 3p, 11p, 13q, 16q, and 17p. Several putative tumor suppressor genes located on these chromosomes are involved in the development and progression of breast carcinoma.19–21 Studies have shown that chromosome 1 is most frequently affected by rearrangements and that chromosomal changes are a precursor to clonal aberrations in some breast carcinoma patients.19, 22–24 The imbalance occurs as the result of either an i(1)(q10) or der(1;16)(q10;p10). Conversely, frequent loss of heterozygosity (LOH) has been detected in chromosomes 1p, 1q, 11p, 17p, and 17q in breast carcinoma patients, indicating the presence of putative tumor suppressor genes in these chromosomal regions. Molecular genetic analyses support the role of these chromosomes in the carcinogenesis and development of breast carcinoma.22–24

We investigated FISH analysis using centromere probes (chromosomes 1, 11, and 17) in the ductal lavage of patients who had positive findings on ductography and/or FDS screening.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Patients

All patients in this study presented with abnormal nipple discharge. Informed patient consent and certification from the institutional human subject protection committee of the Kansai Medical University were obtained before all studies. Between February 2000 and March 2002, 90 female patients with nipple discharge provided consent and were enrolled in the study. Their mean age was 55.2 years. These patients underwent FDS and ductography as described previously.15 Excisional biopsies were performed on 60 patients after positive ductography or FDS findings.

Pathologic Analysis.

Patients underwent breast surgery (duct-lobular segmentectomy). Detailed histologic analysis of the extirpated tissues was carried out to evaluate intraductal abnormalities found during FDS and ductograms. As surgically resected specimens were totally observed with serial sections (< 3 mm), the abnormal lesions of the ductal tree were identified.

Ductal lavage and Cytology

The outer air channel of the fiberscope permits the installation and irrigation of saline washings and retrieval of cells from the breast ductal system. After the fiberscope was retracted, instilled saline (1–2 mL) was retrieved and the samples were obtained for cytology and FISH examination. Smears were prepared from centrifuged specimens. Then, smears on glass slides were Pap stained for alcohol-fixed material. The cytologic diagnoses were classified as benign, atypical/indeterminate, suspicious/probably malignant, malignant, and unsatisfactory.25

In Situ Hybridization

The samples were placed 75-mM potassium chloride and incubated at 37 °C for 10 minutes. The cells were fixed by adding an equal volume of methanol and acetic acid (3:1). After repeated washing in a methanol-acetic acid solution, the cell suspension was dropped onto a slide, air dried, and baked at 65 °C for 4 hours. The sample was denatured by exposure to a denaturing solution (70% formamide, 2 × SSC) for 2 minutes at 70 °C. The denatured sample was immediately cooled in 70% ethanol for 2 minutes and dehydrated in 85% and 100% ethanol. Finally, it was stored on a hot plate until used for hybridization.

Two-color FISH was carried out to detect chromosomes 11 (11 alpha satellite DNA probe labeled with Spectrum Green, VYS-32-132011) and 17 (17 alpha satellite DNA probe labeled with Spectrum Orange, VYS-32-130017) simultaneously. Single-color FISH was used to detect chromosome 1 (1 alpha satellite DNA probe labeled with Spectrum Orange, VYS-32-130001). The samples were exposed to a DNA probe for 5 minutes at 70 °C and hybridization was allowed to proceed overnight. The glass slides were washed with a solution (50% formamide, 2 × SSC) at 45 °C and washed again in 2 × SSC (1 × SSC: 0.15 M NaCl, 15 mM sodium citrate). A diamidinophenylindole solution (Vysis, Downers Grove, IL) was used for nuclear staining. The samples were examined under a fluorescence microscope.

FISH Analysis

A FISH analysis was performed with a microscope (Olympus BX50, Tokyo, Japan). Hybridization signals were evaluated in more than 100 interphase nuclei. In the triple hybridization, each color signal was recorded separately. Each slide was evaluated independently by two examiners. Average counts defined the result in each case. According to previous studies,25 we defined the criteria as follows: 1) chromosomal monosomy: greater than 15% of nuclei with one signal specific for (peri)centromeres of chromosomes 1, 11, and 17 or 2) chromosomal polysomy: greater than 20% of nuclei with three signals specific for chromosomes 1, 11, and 17, respectively.25

Statistical Analysis

Differences in sensitivity, specificity, and diagnostic accuracy were compared by chi-square test. Differences with a probability value of less than 0.05 were considered significant.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Ninety women with nipple discharge participated in the study. The mean age of all patients was 55.2 years (range, 30–60 years). Before screening, the average duration of nipple discharge had been 6 months. Bloody discharge was the primary complaint in 10 of 90 (11.1%) women. Nonphysiologic serous discharge was present in the remaining 80 (88.9%) patients.

Breast carcinoma and precancer start in the lining of the milk duct or lobule. Using FDS, the appearance of the intraductal papillary lesions was either red, yellow, or ash-gray.14 However, it was difficult to distinguish malignant from benign lesions by ductographic findings and/or FDS screening. Histologic evaluation of the resected specimens revealed 54 benign lesions (38 intraductal papillomas, 12 diffuse papillomatoses, and 4 ductal ectasias), 4 ductal carcinomas in situ (DCIS), and 2 invasive ductal carcinomas (Table 1).

Table 1. Relationship between Cytology and FISH on Histology Results
ResultsHistology
Carcinoma (n = 6)Benign (n = 54)
  • FISH: fluorescence in situ hybridization.

  • a

    Indeterminate and benign results combined.

Cytology  
 Positive26
 Negativea448
FISH  
 Positive60
 Negative054

Aneusomy of Chromosomes 1, 11, and 17 Detected by FISH

Representative results of FISH analysis are shown in Figure 1. There were plenty of cells to review in each of the 60 cases due to ductal lavage. The incident rate of having aneusomy in at least one of the three chromosomes (1, 11, and 17) was 100% (all six samples) for invasive ductal carcinoma and DCIS (Table 1). Conversely, aneusomy was not observed in any of the 54 benign lesions. Therefore, FISH samples showing at least one of the three chromosomes were diagnosed as malignant.

thumbnail image

Figure 1. Fluorescence in situ hybridization (FISH) analysis of ductal lavage. Two-color FISH (chromosome 11 [green] and chromosome 17 [orange]) was conducted on samples from a benign lesion (A) and a malignant lesion (B). Two green signals and two orange signals are observed in each nucleus from the benign lesion, but more than two green signals and two orange signals are seen in each nucleus from the cancerous lesion.

Download figure to PowerPoint

Comparison of Diagnostic Utility between FISH and Cytology

Preoperative diagnosis by FISH and cytology were compared with subsequent histology in Table 2. The sensitivity, specificity, and diagnostic accuracy were 33.3%, 88.9%, and 83.3%, respectively, for cytology. There were four false-negative cases and six false-positive cases using cytology. A representative false-negative case is shown in Figure 2. Cytologic evaluation revealed overlapping clusters with irregular nuclei (suspicious/probably malignant), histopathology revealed invasive ductal carcinoma, and the results by FISH analysis were positive.

Table 2. FISH Results of Breast Lesions According to Histology
Numerical aberration of chromosomes (chromosomes 1, 11, 17)aBenign lesions (n = 54)DCIS (n = 4)Invasive ductal carcinoma (n = 2)
  • FISH: fluorescence in situ hybridization; DCIS: ductal carcinoma in situ.

  • a

    −: disomy; +: aneusomy.

−/−/−5400
−/−/+000
−/+/−000
+/−/−010
−/+/+010
+/−/+000
+/+/−011
+/+/+011
thumbnail image

Figure 2. Representative false-negative case by cytology showed overlapping clusters with irregular nuclei (suspicious/probably malignant). This case was identified as invasive ductal carcinoma by histopathologic evaluation and FISH analysis was positive for disease.

Download figure to PowerPoint

Conversely, the sensitivity, specificity, and diagnostic accuracy were 100%, 100%, and 100% for FISH analysis. The differences between cytology and FISH analysis in the sensitivity, specificity, and diagnostic accuracy were statistically significant (P < 0.05).

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

The ductal system of the breast illustrates a significant degree of branching morphogenesis. Intraductal papillary lesions can exist in any branch. The combined utility of ductography and FDS screening may be the best approach to identify intraductal abnormal lesions, although it was difficult to distinguish between malignant and benign lesions.15

Exfoliative cytology via endoscopic biopsy was much more accurate than simple discharge cytology.10, 13, 14 In addition, microsatellite alterations were detected in nipple discharge accompanied by breast carcinoma.26 In that study, at least one LOH using microsatellite markers (D11S1818, D11S2000, D16S402, D16S504, D16S518, D17S520, and D17S786) was found in either the supernatant or cell pellet in 7 of 10 breast carcinoma patients (70%). Therefore, histopathologic evaluation is used to reach a final diagnosis. These results show that more reliable examinations are needed to avoid an excisional biopsy.

We investigated the numerical aberrations of chromosomes 1, 11, and 17 by FISH analysis using selected probes to cell nuclei derived from ductal lavage. We also compared FISH analysis with cytologic evaluation. Aneusomy of chromosomes 1, 11, and 17 was seen in all breast malignancies, including DCIS and invasive ductal carcinoma, but never in benign breast lesions. Therefore, FISH analysis of ductal lavage exhibited higher sensitivity, specificity, and diagnostic accuracy than cytology.

An important problem inherent in cytology is subjective diagnosis. That is, variation in diagnostic accuracy exists among cytologists, depending on their skills.27 Conversely, FISH diagnosis is objective and is not likely to be affected by examiner skill to the extent seen in cytology. In FISH diagnosis, an examiner has only to count the number of clear signals, limiting interobserver variation.25 In addition, FISH diagnosis can be determined without the aid of experienced cytologists.

This is a small pilot trial and the results need to be confirmed in larger definitive studies with substantially more cancers cases. However, the accuracy of FISH analysis limits unnecessary surgery and spares as much involved breast tissue as possible.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

The authors are indebted to Dr. Y. Tamaki (Osaka University.).

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES