The prevalence and clinical significance of residual occult breast cancer after neoadjuvant chemotherapy: reassessing surgical pathology in cases initially described as pathological complete response

Aims Evaluation of pathological complete response (pCR) [no residual invasive carcinoma in the breast (RIC) or lymph node metastases (LNM) in surgical specimens following therapy] is typically based on evaluation of one level of haematoxylin and eosin (H&E) section. Not achieving pCR is associated with worse outcomes, and additional therapy may ensue. This study of patients with triple‐negative (TNBC) or human epidermal growth factor receptor 2 (HER2)‐positive (HER2+) breast cancer who underwent neoadjuvant therapy aims to assess whether occult residual disease (ORD) can be identified in deeper sections of tumour beds and lymph nodes in cases originally reported as pCR and whether ORD is associated with worse outcomes. Methods and results In 84 cases of pCR (2009–17) at our institution, deeper‐level recuts were assessed for ORD. Oncological and survival outcomes were compared. ORD was identified in seven of 40 TNBC (17.5%; five RIC; one LMN; one RIC and LMN) and four of 44 HER2+ (9.1%; three RIC; one LMN) cases (all residual cancer burden I). Median follow‐up was 46.7 months for TNBC (one local recurrence, four distant metastases and two deaths) and 86.8 months for HER2+ (no local recurrence, three distant metastases and two deaths). All recurrence and death events occurred in patients with pCR without ORD, with no recurrence events in patients with ORD. Conclusions In patients with TNBC and HER2+ breast cancer with pCR by standard pathological assessment, occult residual disease is not uncommon. Occult disease was not associated with worse oncological or survival outcomes, suggesting standard pathological assessment is sufficient to identify clinically meaningful disease.