Confocal microscopy of unfixed breast needle core biopsies: a comparison to fixed and stained sections

Interesting study looking at confocal scanning laser microscopy and experience with comparison to standard H&E sections.  Will be interesting to see if results can be reproduced on more cases by other groups.  May be very helpful in the future for the ever troublesome distinctions between low grade ductal carcinoma in-situ and atypical ductal hyperplasia lesions.

Linda M Schiffhauer, J. Neil Boger, Thomas A Bonfiglio, James M Zavislan, Margarita Zuley and Christi Alessi Fox

BMC Cancer 2009, 9:265doi:10.1186/1471-2407-9-265

Abstract (provisional)

Introduction

Needle core biopsy, often in conjunction with ultrasonic or stereotactic guided techniques, is frequently used to diagnose breast carcinoma in women. Confocal scanning laser microscopy (CSLM) is a technology that provides real-time digital images of tissues with cellular resolution. This paper reports the first step in developing techniques to rapidly screen needle core breast biopsy and surgical specimens at the point of care. CSLM requires minimal tissue processing and has the potential to reduce the time from excision to diagnosis. Following imaging, specimens can still be submitted for standard histopathological preparation.

Methods

Needle core breast specimens from 49 patients were imaged at the time of biopsy. Their lesions had been categorized as Breast Imaging Reporting And Data System (BI-RADS) 3, 4 or 5. The core biopsies were imaged with the CSLM before fixation. Samples were treated with 5% citric acid and glycerin USP to enhance nuclear visibility in the reflectance confocal images. Immediately following imaging, the specimens were fixed in buffered formalin and submitted for histological processing and pathological diagnosis. CSLM images were then compared to the standard histology.

Results

The pathologic diagnoses were 7 invasive ductal carcinomas, 2 invasive lobular carcinomas, 3 ductal carcinomas in-situ, 21 fibrocystic changes/proliferative conditions, 9 fibroadenomas, and 5 other/benign; two were excluded due to imaging difficulties. Morphologic and cellular features of benign and cancerous lesions were identified in the confocal images and were comparable to standard histologic sections of the same tissue.

Conclusions

CSLM is a technique with the potential to screen needle core biopsy specimens in real-time. The confocal images contained sufficient information to identify stromal reactions such as fibrosis and cellular proliferations such as intra-ductal and infiltrating carcinoma, and were comparable to standard histologic sections of the same tissue. Morphologic and cellular features of benign and cancerous lesions were identified in the confocal images. Additional studies are needed to 1) establish correlation of confocal and traditional histologic images for the various diseases of the breast; 2) validate diagnostic use of CSLM and; 3) define and potentially minimize borderline situations such as some cases of well-differentiated ductal CIS vs. atypical hyperplasia.