Core Needle Biopsy and Preoperative Risk Factors as Predictors of Upstaging Atypical Ductal Hyperplasia on Surgical Excision

Abstract

Background

Patients with atypical ductal hyperplasia (ADH) diagnosed on core needle biopsy (CNB) are recommended for surgical excision. However, the rate of upstaging to malignancy on final pathology varies widely. This study sought to determine the rate of upstaging for ADH diagnosed on CNB and to evaluate risk factors associated with upstaging.

Methods

Patients diagnosed with ADH by CNB from 1/2014 to 9/2024 were identified from a prospectively maintained database. Pathology, imaging, demographics, family history of breast cancer, genetic testing, and risk of future malignancy were collected. Univariate and multivariable analysis were performed to evaluate predictors of upstaging.

Results

456 patients with ADH on CNB met inclusion criteria. Median age was 53 years (range 19-89 years). Overall, 110 (24.1%) patients were upstaged on surgical excision. The highest risk of upstaging was observed in patients with multifocal ADH (71.4%). Upstaged patients were older (P < .001), more likely to have a mammographic mass or architectural distortion (P = .03), abnormal MRI (P = .001), ADH bordering on DCIS or multifocal ADH on biopsy (P < .001). Multivariable analysis identified abnormal MRI as an independent predictor of higher risk (P = .046) and focal ADH as a predictor of lower upstage risk (P = .03). At a median follow-up of 52.5 months, 21 (4.6%) patients developed subsequent DCIS, and 20 (4.4%) patients developed subsequent invasive cancer.

Conclusion

ADH on CNB was upstaged on surgical excision in 24.1% of patients. Multifocal ADH and abnormal MRI findings strongly predict upstaging to malignancy. The 9.6% rate of subsequent malignancy necessitates long-term surveillance for all patients with ADH.

Keywords

atypical ductal hyperplasia risk of malignancy predictors of atypical ductal hyperplasia upstaging

Introduction

Atypical ductal hyperplasia (ADH) is a high-risk, non-malignant breast lesion that confers a four-to-five-fold increase in lifetime risk of developing invasive breast carcinoma. Pathologically, ADH is characterized by epithelial proliferation in breast ducts and shares cellular and molecular characteristics with low-grade ductal carcinoma in situ (DCIS). The distinction between ADH and DCIS is primarily quantitative: ADH is limited to lesions measuring <2 mm or involving no more than two ductal spaces, whereas low-grade DCIS exceeds both criteria.¹ Importantly, distinguishing these entities on core needle biopsy (CNB) can be difficult due to the intrinsic limitations of tissue sampling. This limitation is compounded by the tendency for ADH to occur incidentally within coexisting breast tumors, such that its presence on CNB may indicate an undersampled component of more extensive disease. Accordingly, the resulting diagnostic ambiguity has historically led to the recommendation for surgical excision of ADH to mitigate the risk of missing an underlying carcinoma. Rates of upstaging on surgical excision from ADH to DCIS or invasive breast carcinoma vary widely in the contemporary literature, ranging from 4.9% to 59.1%, underscoring the significant heterogeneity and ongoing uncertainty in management.^2–13 Prior studies have attempted to define predictors of ADH upgrade on surgical excision but have been constrained by small sample sizes and the inclusion of patients with mixed or overlapping diagnoses, such as atypical lobular hyperplasia (ALH) and ADH with coexisting lobular carcinoma in situ (LCIS). We performed a comprehensive analysis of lesions with ADH on CNB and evaluated preoperative, radiologic, and pathologic risk factors that are predictive of upstaging to DCIS or invasive breast carcinoma upon surgical excision.

Methods

This study was performed with the approval of our Institutional Review Board. Patients with a diagnosis of ADH on CNB from January 2014 to September 2024 were identified from a prospectively maintained institutional database using the Deep6 AI software. Patients with ADH and concomitant cancer on CNB were excluded from analysis. Demographics, histopathological features on CNB, including focality and the presence of concomitant diagnoses, mammographic, ultrasound, and MRI findings, family history of breast cancer, genetic testing, and subsequent pathological findings on surgical excision were reviewed. Student’s t-test and Pearson’s chi-square tests were used to compare continuous and categorical variables for statistical significance, respectively. Multivariable logistic regression was used to identify independent predictors of upstage. A P-value <0.05 was considered significant. Statistical analyses were performed using STATA software.

Results

A total of 456 women were identified with ADH on CNB, of whom 110 (24.1%) were upstaged to DCIS or invasive cancer at the time of surgical excision. Seventy-one (15.6%) patients were found to have DCIS at the time of surgical excision, while 39 (8.6%) were upstaged to invasive cancer. Patients with focal ADH on biopsy had the lowest rate of upstaging, with 26 of 168 patients (15.5%) upstaged to either DCIS or invasive cancer (Table 1). Moderate upstaging rates were seen in patients with pure ADH (45/214 patients, 21.0%), as well as ADH with concomitant ALH (21/103 patients, 20.4%), and ADH with concomitant LCIS (3/20 patients, 15.0%). Patients with multifocal ADH on biopsy had the highest rates of upstaging, with 5 of 7 patients (71.4%) upstaged to either DCIS or invasive cancer, followed by patients with ADH bordering on DCIS (32/61 patients, 52.5%).

Table 1.

Patients Upstaged to DCIS or Invasive Cancer by ADH Histology on CNB

ADH histology	Not upstagedN = 346	Upstaged to DCISN = 71	Upstaged to invasive breast cancerN = 39
Pure ADH (n = 214)	169 (79.0%)	32 (15.0%)	13 (6.1%)
Focal ADH (n = 168)	142 (84.5%)	12 (7.1%)	14 (8.3%)
ADH bordering on DCIS (n = 61)	29 (47.5%)	22 (36.1%)	10 (16.4%)
Multifocal ADH (n = 7)	2 (28.6%)	4 (57.1%)	1 (14.3%)

Compared to patients who were not upstaged at the time of surgical excision, those who were upstaged tended to be older (median 57 vs 52 years, P < .001); however, there were no significant differences in family history of breast or ovarian cancers, presence of a pathogenic genetic mutation, or prior breast biopsies between the two groups (Table 2). Patients who were upstaged were more likely to present with a mass (11.8% vs 5.2%, P = .03) or distortion seen on mammography (5.5% vs 1.7%, P = .03) and were also more likely to have abnormal findings on MRI (61.1% vs 32.6%, P = .001). There were no significant differences in the prevalence of concurrent ADH with ALH or ADH with LCIS between patients who were upstaged and those who were not. Similarly, there were no differences in the type of surgery performed between the two groups (P = .27).

Table 2.

Characteristics of Patients With ADH by Upstage Status

	Not upstaged (n = 346)	Upstaged (n = 110)	P-value
Age, median (IQR)	52 (46,63)	57 (50,70)	<.001
Family history breast or ovarian	105 (30.4%)	36 (32.7%)	.64
Genetic mutation (n = 139)	12 (12.8%)	7 (15.6%)	.65
Prior breast biopsy	106 (30.6%)	38 (34.6%)	.44
Abnormal mammogram	304 (87.9%)	92 (83.6%)	.25
Mammographic findings			.03
Benign	29 (8.4%)	11 (10.0%)
Asymmetry	17 (4.9%)	6 (5.5%)
Calcifications	263 (76.0%)	67 (60.9%)
Mass	18 (5.2%)	13 (11.8%)
Distortion	6 (1.7%)	6 (5.5%)
Ultrasound correlate	81 (48.2%)	43 (55.8%)	.27
Abnormal MRI	29 (32.6%)	22 (61.1%)	.003
ADH histology			<.001
Pure ADH (n = 214)	169 (49.4%)	45 (41.7%)
Focal ADH (n = 168)	142 (41.5%)	26 (24.1%)
ADH bordering on DCIS (n = 61)	29 (8.5%)	32 (29.6%)
Multifocal ADH (n = 7)	2 (0.6%)	5 (4.6%)
>1 lesion biopsied in ipsilateral breast	11 (3.2%)	7 (6.4%)	.14
ADH with ALH (n = 103)	82 (23.7%)	21 (19.1%)	.31
ADH with LCIS (n = 25)	17 (4.9%)	3 (2.7%)	.33
Surgery			.27
Partial mastectomy	342 (99.1%)	107 (97.3%)
Unilateral mastectomy	0 (0%)	1 (0.9%)
Bilateral mastectomy	3 (0.9%)	2 (1.8%)

At a median follow-up of 52.5 months, among patients with ADH on CNB who underwent surgical excision, 44 (9.6%) developed a subsequent DCIS or invasive cancer. 24 patients (5.3%) had DCIS and 20 (4.4%) had invasive cancer (Table 3). There were no differences in subsequent malignancy rates, laterality of the subsequent DCIS or invasive cancer, or vital status with or without disease between patients who were upstaged during the index operation and those who were not.

Table 3.

Outcomes of Patients With ADH by Upstage Status

	Not upstaged (n = 346)	Upstaged (n = 110)	P-value
Subsequent DCIS	17 (4.9%)	7 (6.4%)	.55
Ipsilateral	12 (3.5%)	3 (2.7%)
Contralateral	5 (1.5%)	4 (3.6%)
Subsequent invasive cancer	17 (4.9%)	3 (2.7%)	.33
Ipsilateral	4 (1.2%)	1 (0.9%)
Contralateral	13 (3.8%)	2 (1.8%)
Vital status			.15
Alive without disease	407 (99.0%)	107 (97.3%)
Alive with disease	1 (0.4%)	2 (2.1%)
Dead	1 (0.4%)	1 (1.0%)

On multivariable logistic regression analysis, focal ADH was significantly associated with a lower rate of upgrade on surgical excision (OR 0.23, 95% CI: 0.06-0.84, P = .03, Table 4). Multifocal ADH, ADH bordering on DCIS, and the presence of more than one biopsied lesion in the ipsilateral breast appeared to confer an increased risk of upgrade from ADH to DCIS or invasive cancer; however, this was not statistically significant. The presence of abnormal findings on MRI was an independent predictor of upgrade to DCIS or invasive cancer on surgical excision (OR 3.21, 95% CI: 1.02-10.11, P = .046).

Table 4.

Multivariable Logistic Regression of Risk of Upstaging Among Patients With ADH on Core Needle Biopsy

	Odds ratio	CI	P-value
Age	1.03	0.98-1.08	.30
Family history	1.14	0.37-3.49	.82
Prior biopsy	0.51	0.16-1.65	.26
Abnormal mammogram	0.94	0.64-1.37	.74
Abnormal ultrasound	1.47	0.46-4.69	.51
Abnormal MRI	3.21	1.02-10.11	.046
ADH histology (ref: pure ADH)
Focal ADH	0.23	0.06-0.84	.03
Multifocal ADH	2.04	0.06-65.47	.69
ADH bordering DCIS	2.39	0.59-9.77	.22
>1 lesion biopsied in ipsilateral breast	8.09	0.24-273.5	.25

Discussion

The prevalence of DCIS or invasive carcinoma in ADH specimens varies widely in the literature, complicating efforts to identify which patients are most likely to benefit from surgical excision. Prior studies have reported widely variable upstaging rates ranging from 4.9% to 59.1%; however, these analyses are largely limited by small sample sizes and the inclusion of heterogeneous CNB diagnoses, often combining ADH with additional high-risk or atypical lesions.^2–13 Our study provides a comprehensive analysis of ADH diagnosed on CNB and demonstrates an overall upstage rate of 24.1%, with substantially higher rates observed among patients with multifocal ADH or ADH bordering on DCIS.

Histologically, ADH and DCIS are essentially indistinguishable, with the differentiating factor being the size of the lesion. The amount of sampling on the CNB can therefore be the only distinction between ADH and DCIS. This breadth of range underscores the difficulty in defining a universally accepted low-risk group. Among our patient cohort, approximately one in four patients diagnosed with ADH on CNB had a malignancy requiring excision, mirroring a national average of 29% and validating the representativeness of our study population for further risk analysis.¹³ This upstage rate, coupled with the fact that an ADH diagnosis increases a patient’s risk for breast cancer by 4 to 5 times, is the rationale for the current recommendation of precautionary surgical excision.^12,14 While this data can support surgery, it simultaneously indicates that >70% of patients undergo unnecessary surgery. The challenge lies in safely identifying the large subset of patients who are not upstaged and who could benefit from less invasive active surveillance.

The most recent established guidelines from organizations such as the National Cancer Comprehensive Network and the American Society of Breast Surgeons have identified key pathological, radiological, and medical history factors that categorize patients as low-risk for upstaging to DCIS or invasive cancer upon surgical excision. These include the absence of other high-risk lesions, the presence of solely mammographic calcifications rather than a mass or architectural distortion, smaller lesion size on imaging (<1 cm) with fewer foci, more biopsy cores obtained, and smaller volume of ADH on biopsy specimens.^15–17 Nevertheless, the recommendations remain cautious, generally advocating for surgical excision while placing considerable emphasis on interdisciplinary consultation and patient-led shared decision-making. This article seeks to shed light on additional, previously underappreciated characteristics that should be considered when planning treatment, thereby refining the selection of patients for observation vs surgical management.

In our study, focal ADH was an independent predictor of a low upstage rate, with an overall rate of 15.5%, which was lower than that observed in pure ADH (21.0%) and dramatically lower than the 71.4% seen in multifocal ADH. The low odds ratio of 0.23 suggests that focal ADH represents a distinct pathological entity less likely to harbor underlying malignancy and, when considered alongside other patient-specific risk factors, should be incorporated into low-risk stratification frameworks to guide individualized management. Conversely, ADH bordering on DCIS represents a high-risk pathological diagnosis, demonstrating an upstage rate of 52.5% in our cohort, and therefore should continue to be regarded as a strong indication for surgical excision. Multifocal ADH identified on biopsy emerged as the highest-risk subtype, with an upstage rate of 71.4% on univariate analysis. This finding was further supported by multivariable analysis, in which multifocal ADH demonstrated a large odds ratio of 2.04, indicating an increased likelihood of upstaging compared with pure ADH. In addition, patients who underwent more than one biopsy in the ipsilateral breast due to the presence of multiple lesions had the highest likelihood of upstaging, with an odds ratio of 8.09. However, neither of these results reached statistical significance, likely due to the small sample size. Nonetheless, a diagnosis of multifocal ADH on CNB or the presence of multiple lesions in the ipsilateral breast warranting biopsy should be considered a high preoperative predictor of upstaging, necessitating surgical excision.

Abnormal breast MRI findings were identified as a significant independent predictor of upstaging in our cohort, with an odds ratio of 3.21 (P = .046), indicating that patients with abnormal MRI findings were more than three times as likely to be upstaged compared with those with unremarkable MRIs. In our study, patients with abnormal MRI findings demonstrated an upstage rate of 43.1%, consistent with prior reports suggesting that contrast-enhanced imaging improves the prediction of ADH upstaging and underscoring the important role of advanced imaging in contemporary risk stratification.^18,19

Our study also offers insight into long-term risk, with 9.6% of patients developing subsequent DCIS or invasive breast cancer during follow-up, without a predilection for ipsilateral vs contralateral disease. This finding reinforces that a diagnosis of ADH confers a persistent lifetime risk of breast cancer that warrants ongoing surveillance, irrespective of whether surgical excision was deferred or malignancy was identified at the time of surgical excision.

This study has several important limitations that warrant consideration. Foremost, the relatively small overall sample size limits statistical power and may have constrained our ability to detect significant associations on multivariable analysis, particularly for higher-risk subgroups such as multifocal ADH and ADH bordering on DCIS. Although both entities demonstrated markedly elevated upstage rates and large effect sizes, these findings did not reach statistical significance on multivariable analysis, likely reflecting limited event numbers rather than a true absence of association. In addition, the retrospective design introduces the potential for selection bias, variability in imaging interpretation, and heterogeneity in pathologic assessment across cases. These limitations underscore the need for larger, prospective studies to validate risk stratification frameworks and refine criteria for safely omitting surgical excision in patients with ADH.

Footnotes

ORCID iD

Farin Amersi

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

References

Amin

Wagner

. Contemporary management of atypical breast lesions identified on percutaneous biopsy: a narrative review. Annals of Breast Surgery. 2021;5:9. doi:10.21037/abs-20-117.

Styblo

Arciero

Wang

Cohen

. Management of high-risk breast lesions diagnosed on core biopsies and experiences from prospective high-risk breast lesion conferences at an academic institution. Breast Cancer Res Treat. 2021;185(3):573-581. doi:10.1007/s10549-020-05977-9.

Vicks

Mason

Perez

, et al. Increased risk of upstage when combinations of breast lesions of uncertain malignant potential are found on core needle biopsy: the need for surgical excision. Am J Surg. 2024;227:6-12. doi:10.1016/j.amjsurg.2023.10.004.

Casaubon

Niakan

Vicks

Perez Coulter

Jacobbe

Mason

. The effect of delay of excisional biopsy on upstage rate for atypical ductal hyperplasia, flat epithelial atypia, intraductal papilloma, and radial scar. Breast Cancer Res Treat. 2022;196(3):527-534. doi:10.1007/s10549-022-06745-7.

Karwowski

Lumley

Stokes

, et al. Atypical ductal hyperplasia on core needle biopsy: surgical outcomes of 200 consecutive cases from a high-volume breast program. Breast J. 2021;27(3):287-290. doi:10.1111/tbj.14170.

Menen

Ganesan

Bevers

, et al. Long-term safety of observation in selected women following core biopsy diagnosis of atypical ductal hyperplasia. Ann Surg Oncol. 2017;24(1):70-76. doi:10.1245/s10434-016-5512-9.

Grabenstetter

Brennan

Sevilimedu

, et al. Is surgical excision of focal atypical ductal hyperplasia warranted? Experience at a tertiary care center. Ann Surg Oncol. 2023;30(7):4087-4094. doi:10.1245/s10434-023-13319-4.

Lustig

Guo

Liu

, et al. Development and prospective validation of a risk calculator that predicts a low risk cohort for atypical ductal hyperplasia upstaging to malignancy: evidence for a watch and wait strategy of a high-risk lesion. Ann Surg Oncol. 2020;27(12):4622-4627. doi:10.1245/s10434-020-08881-0.

Pawloski

Christian

Knezevic

, et al. Atypical ductal hyperplasia bordering on DCIS on core biopsy is associated with higher risk of upgrade than conventional atypical ductal hyperplasia. Breast Cancer Res Treat. 2020;184(3):873-880. doi:10.1007/s10549-020-05890-1.

10.

Peña

Shah

Fazzio

, et al. Multivariate model to identify women at low risk of cancer upgrade after a core needle biopsy diagnosis of atypical ductal hyperplasia. Breast Cancer Res Treat. 2017;164(2):295-304. doi:10.1007/s10549-017-4253-1.

11.

Nguyen

Albarracin

Whitman

Lopez

Sneige

. Atypical ductal hyperplasia in directional vacuum-assisted biopsy of breast microcalcifications: considerations for surgical excision. Ann Surg Oncol. 2011;18(3):752-761. doi:10.1245/s10434-010-1127-8.

12.

Hartmann

Radisky

Frost

, et al. Understanding the premalignant potential of atypical hyperplasia through its natural history: a longitudinal cohort study. Cancer Prev Res. 2014;7(2):211-217. doi:10.1158/1940-6207.CAPR-13-0222.

13.

Schiaffino

Calabrese

Melani

, et al. Upgrade rate of percutaneously diagnosed pure atypical ductal hyperplasia: systematic review and meta-analysis of 6458 lesions. Radiology. 2020;294(1):76-86. doi:10.1148/radiol.2019190748.

14.

Breast hyperplasia (ductal or lobular) | benign conditions. https://www.cancer.org/. https://www.cancer.org/cancer/types/breast-cancer/non-cancerous-breast-conditions/hyperplasia-of-the-breast-ductal-or-lobular.html

15.

Klassen

Fraker

Pruthi

. Updates on management of atypical hyperplasia of the breast. Mayo Clin Proc. 2025;100(6):1051-1057. doi:10.1016/j.mayocp.2025.01.029.

16.

Resource Guide Resource Guide . Surgical Management of Benign or High-Risk Lesions Purpose. https://www.breastsurgeons.org/docs/statements/asbrs-high-risk-lesions.pdf

17.

Bevers

Niell

Baker

, et al. NCCN guidelines® insights: breast cancer screening and diagnosis, version 1.2023: featured Updates to the NCCN guidelines. J Natl Compr Cancer Netw. 2023;21(9):900-909. doi:10.6004/jnccn.2023.0046.

18.

Johnson

Stanczak

Winblad

Amin

. Breast MRI assists in decision-making for surgical excision of atypical ductal hyperplasia. Surgery. 2023;173(3):612-618. doi:10.1016/j.surg.2022.07.036.

19.

Bertani

Urbani

La Grassa

, et al. Atypical ductal hyperplasia: breast DCE-MRI can be used to reduce unnecessary open surgical excision. Eur Radiol. 2020;30(7):4069-4081. doi:10.1007/s00330-020-06701-3.