We identified 323 patients in this retrospective study, among them, 138 patients had AH at the margins in BCS, while 185 patients did not have this pathological feature at the margins. The median follow-up was 48 months (range 13–117). The comparison of clinicopathological characteristics and neoadjuvant chemotherapy response rate between the two groups is presented in Table 1. Patients with AH or without AH did not differ significantly by age, tumor histologic type, initial clinic T stage, initial nodal stage, receptor status, post-NAC pathologic T stage, post-NAC pathologic nodal status or overall pCR rate.
Table 1
Clinicopathlogical characteristics of two groups
| Atypical hyperplasia | No atypical hyperplasia | p value | |
Median age | 43 years (rang 24–63) | 43 years (rang 19–71 ) | |
Age (years) | 138 | 185 | 0.433 |
≤ 40 | 50 (36.2%) | 75 (40.5%) | |
༞40 | 88 (63.8%) | 110 (59.5%) | |
Tumor histologic | | | 0.880 |
Invasive ductal carcinoma | 130 (94.2%) | 175 (94.6%) | |
Other type of invasive carcinoma | 8 (5.8%) | 10 (5.4%) | |
Initial clinic T stage | | | 0.586 |
T1 | 17 (12.3%) | 27 (14.6%) | |
T2 | 109 (79.0%) | 143 (77.3%) | |
T3 | 12 (8.7%) | 15 (8.1%) | |
Initial nodal stage | | | 0.511 |
N0 | 42 (30.4%) | 52 (28.1%) | |
N1 | 86 (62.3%) | 114 (62.7%) | |
N2༆N3 | 10 (7.2%) | 17 (9.2%) | |
Receptor status | | | 0.169 |
HR+/HER2- | 64 (46.4%) | 102 (55.1%) | |
HR+/HER2+ | 40 (29.0%) | 42 (22.7%) | |
HR-/HER2+ | 12 (8.7%) | 21 (11.4%) | |
HR-/HER-- | 22 (15.9%) | 20 (10.8%) | |
Post-NAC pathologic T stager | | | 0.825 |
T0 | 28 (20.3%) | 42 (22.7%) | |
T1 | 82 (59.4%) | 104 (56.2%) | |
T2 | 28 (20.3%) | 39 (21.1%) | |
Post-NAC pathologic nodal status | | | 0.492 |
Negative | 71 (51.4%) | 88 (47.6%) | |
Positive | 67 (48.6%) | 97 (52.4%) | |
Overall pCR | | | 0.686 |
yes | 28 (20.3%) | 41 (22.2%) | |
no | 110 (79.7%) | 144 (77.8%) | |
Bold value is statistically significant when p < 0.05 |
HR Hormone receptor, HER2 human epidermal growth factor receptor 2, pCR pathologic complete response, NAC Neoadjuvant chemotherapy |
During the follow-up period, 8 (5.8%) patients in the AH group, and 8 (4.3%) patients in the non-AH group experienced ipsilateral breast tumor recurrence (IBTR). The 5-year rates of IBTR were 6.7% (95% CI, 4.4%~9.0%) and 4.8% (95% CI, 3.1%~6.5%) in patinets with and without AH, respectively. Distant-metastasis-free survival (DMFS) at 5 years was 86.3% (95% CI, 82.2%~90.4%) in the AH group, and 89.8% (95% CI, 87.4%~92.2%) in the non-AH group, respectively. Additionally, the 5-years overall survival (OS) rate of the patients with or without AH was 93.1% (95% CI, 90.5%~95.7%) and 93.8% (95% CI, 91.7%~95.9%), respectively. No significant differences were observed between the two groups of patients in terms of IBTR (Fig. 1), DMFS (Fig. 2A), or OS (Fig. 2B) (p = 0.523, 0.461 and 0.328, respectively). Atypical hyperplasia can be further classified into mild, moderate, and severe categories which borders on ducal carcinoma in situ. Coopey et al. evaluated breast cancer events in a retrospective cohort of 2938 women with ADH, ALH, LCIS, and severe ADH, and the 10 years risk of breast cancer they estimated was 17% for women with ADH, 26 for women with severe ADH[16]. Therefore, severe atypical hyperplasia may be regarded as “higher level of risk” lesion. We further analyzed the outcomes between patients with severe atypical hyperplasia and those without atypical hyperplasia, and no significant differences were found in IBTR, DMFS, or OS between patients with severe atypical hyperplasia (n = 44) and those without atypical hyperplasia (n = 185).
An overall pCR of breast and axillary nodes was achieved in 68 patients. Patients who achieved an overall pCR had significantly better DMFS (p = 0.022) and OS (p = 0.016), but not IBTR (p = 0.365), compared with those with residual disease. Among 255 patients with residual disease after neoadjuvant chemotherapy, 37 patients received re-excision due to invasive cancer and/or in situ carcinoma at the primary margins, 63 patients received re-excision due to severe AH which is somewhat difficult to distinguish from low grade DCIS in the frozen section[17]. Among these 63 patients, 16 still had severe AH after re-resection, and 3 of these 16 patients experienced local recurrence, while 1 of 30 patients without AH at re-excision margins had local recurrence, but again no significantly difference between the two group of these 46 patients in term of local recurrence in ipsilateral breast (p = 0.059).
It has been reported previously that some clinical, pathologic, and molecular factors were associated with IBTR after BCS[4]. Therefore, a multivariate analysis was performed to assess these factors associated with IBTR, DMFS and OS in our study. There was no association between atypical hyperplasia status and IBTR in the multivariate analysis. Similarly, other clinical and pathological features, including age, tumor histologic type, initial clinic T stage, initial nodal stage, receptor status, post-NAC pathologic T stage, post-NAC pathologic nodal status or overall pCR rate, were not significantly associated with IBTR as well. On multivariate analysis, patients who achieved pCR (p = 0.037, HR 4.6, 95% CI 1.09–19.18) had better DMFS, and patients who had negative lymph nodes (p = 0.012, HR 4.8, 95% CI 1.41–16.79) after NAC had better OS.