RETURN TO ABSTRACT LISTING


Objective:
To determine imaging and patient factors predictive of risk for atypical ductal hyperplasia (ADH) upgrade to malignancy in a hybrid academic-community breast imaging practice.

Materials and Methods:
We retrospectively reviewed image guided ADH biopsies (stereotactic, ultrasound, MRI) from 1/1/12 to 12/31/22 at our academic-community hybrid practice with standardized management recommendations for high-risk lesions. Patients with concurrent ipsilateral cancer and patients without follow-up with a breast surgeon were excluded (all ADH biopsies are referred for a surgical consult). Data collected: patient age, risk factors (personal or family history of breast cancer, genetic mutations, etc.), lesion type/size, biopsy modality, and upgrade rate (UGR) (DCIS/invasive carcinoma). We compared the rates of imaging surveillance and surgical excision management at the two practices. Statistical analyses included chi-squared and t-tests.

Results:
Of the 594 cases of ADH, 337 cases were excluded due to ipsilateral cancer or lack of follow-up. Analysis is on 251 patients with 257 ADH biopsies. 47.4% were performed at the academic center (AC) and 52.5% at the community center (CC). AC patients were older (57.8 vs 53.8 years, <em>p</em> = 0.002) and more likely to have an ultrasound biopsy compared to CC (<em>p</em> = 0.007). Based on lesion type, asymmetries with ADH diagnosis were more often biopsied at the AC (7.4% vs 0.7%; <em>p</em> = 0.006), and calcifications more often at the CC (82.2% vs 68.0%; <em>p</em> = 0.008). Overall UGR was 15.2%, with CC 13.4% and AC 17.3%, (<em>p</em> = 0.426). There were no significant differences in lesion size, excision rate, or UGR between centers. 40 ADH cases had imaging observation and 217 underwent excision. Image observed patients were older (58.7 vs 55.1 years, <em>p</em> = 0.045), with larger lesions (14.2 vs. 10.5 mm; <em>p</em> = 0.035), and without personal/family history of breast cancer (<em>p</em> = 0.009). Masses represented 16.7% of the lesions and were more likely to be excised (<em>p</em> = 0.009), but not to upgrade (UGRs: masses 23.6% vs calcifications 12.7%; <em>p</em> = 0.076). Among excised masses, there was no association between patient age, risk factors, and lesion size when comparing upgraded lesions versus not. Calcifications were the most common lesion (88.9%) but did not demonstrate a difference in management. Larger extent of surgically excised calcifications was more likely to upgrade (14.2 vs. 8.7 mm, <em>p</em> = 0.011). There was no difference in UGR related to patient age or underlying risk factors for calcifications.

Conclusion:
Despite differences in patient populations, there was no statistical difference in management or UGRs between practice sites. It remains challenging to risk stratify ADH biopsies based on individual patient factors. Masses were more likely to be excised, but larger extent of calcification was more likely to upgrade. Hybrid academic-community practices can function similarly in the management of ADH when we use our expertise and standardize our management recommendations.