[269] Unnecessary Subsequent Surgery Might Be Avoided for a Diagnosis of Atypical Ductal Hyperplasia on a Core Needle Biopsy by Evaluating Histologic Features: A Study of 71 Cases.

Bradley M Turner, Prasanna Kumar, Rameela Chandraseekhar, Gregory Wilding, Thaer Khoury. Roswell Park Cancer Institute, Buffalo, NY

Background: Significant discordance has been reported between the image guided core needle biopsy (CNB) diagnosis of atypical ductal hyperplasia (ADH) and the diagnosis of a malignant lesion (ductal carcinoma in situ [DCIS] or invasive carcinoma) on subsequent excisional biopsy (EB). Subsequent EB which only identifies ADH or a more benign lesion results in greater risk for the patient and increased healthcare costs. The purpose of our study is to further examine the histologic features of ADH that might predict a malignant lesion so that unnecessary subsequent surgery might possibly be avoided.
Design: The database at Roswell Park Cancer Institute, Buffalo, NY, was examined between 1996 and 2010 for all diagnoses of ADH on CNB. Seventy-one cases of ADH with subsequent EB were identified. Histologic features of ADH were evaluated according to pattern-type (micropapillary, cribriform or both), presence of Ca++ in ducts, number of cores sampled, number of cores and ducts involved with ADH, size of ADH focus, and multifocality. These histologic features were evaluated between benign vs. malignant outcomes on subsequent EB, using Fisher's exact test.
Results: There were 43 (61%) cases with benign results and 28 (39%) with malignant results on subsequent EB. The presence of calcium in benign ducts was a statistically significant marker for benign outcome on EB. The presence of ADH in more than one core, size greater than 2 mm, and multifocality were all statistically significant predictors of a malignant outcome on EB (Table I).
Conclusions: Our results indicate that there are histologic features that might help in predicting malignant findings on a subsequent EB. These histologic features, along with imaging studies can help clinicians to subcategorize patients into high and low risk groups for malignancy, thus possibly avoiding unnecessary subsequent surgery.

Table I : Histologic features predicting benign vs. malignant findings on subsequent EB
Histologic feature ResultsBenignMalignant*** p-value
Ca++ IN BENIGN DUCTS* Positive14(32.6)**17(65.4) 0.01
 Negative29(67.4)9(34.6)
    
NUMBER OF INVOLVED CORES 131(72.1)11(39.3) 0.03
 2-38(18.6)11(39.3)
 ≥44(9.3)6(21.4)
    
LARGEST SIZE-MM <229(67.4)6(21.4) 0.0002
 ≥214(32.6)22(78.6)
    
MULTIFOCALITYPositive29(67.4)11(39.3) 0.03
 Negative14(32.6)17(60.7)
*Two CNB's with malignancy on EB had no benign ducts; **N (%); ***Invasive carcinoma or DCIS.


Category: Breast

Monday, February 28, 2011 8:15 AM

Platform Session: Section C, Monday Morning

 

Close Window