Background: A considerable proportion of patients (pts) with HER2-negative (HER2-) metastatic breast cancer (MBC) present with elevated serum levels of the HER2 extracellular domain (sHER2) and/or HER2-overexpressing circulating tumor cells (CTCs) during their further clinical course. These “occult” HER2-positive (HER2+) pts may be candidates for anti-HER2 therapy (Tx) albeit normally not subjected to such treatment. This retrospective study was undertaken to gain more insights into the feasibility of HER2-directed Tx in occult HER2+ MBC pts in the clinical routine. Methods: From our database, we identified 26 pts with heavily pretreated HER2- MBC (ER+, 21 pts) showing sHER2 values > 15 ng/mL (6 pts), HER2+ CTCs (6 pts), or both (14 pts) having failed 2-16 prior systemic treatments (median: 7) who did not qualify for recruitment onto a prospective clinical trial. All pts received anti-HER2 Tx with trastuzumab (H: 14 pts), lapatinib (L: 4 pts), H+L (2 pts), or H+pertuzumab (H+P: 6 pts). HER2-targeting Tx was given alone (4 pts), or in combination with endocrine agents (4 pts), cytotoxics (16 pts), or other targeted drugs (2 pts). Responses were scored according to RECIST 1.1, OS was calculated from the start of HER2-directed Ctx until death from any reason or loss to follow-up by using Kaplan-Meier statistics. Results: Anti-HER2 Tx was generally well tolerated. Median treatment duration was 16.1 wks (range 1.0-56.1 wks). In 2 pts with L and 1 pt with H+L, Tx was prematurely stopped due to toxicity (diarrhea, fatigue). 10 PR, 10 SD, 5 PD, and 1 non-evaluable (NE) pt accounted for an objective response rate (ORR) of 38.5% and a clinical benefit rate (CBR) of 76.9%. Median OS was 62.9 wks. Conclusions: Our findings indicate that anti-HER2 Tx may be a valid option in pts with heavily pretreated HER2- MBC with pathological sHER2 values and/or HER2+ CTCs in the clinical routine. Thus, results of ongoing randomized trials in this setting are eagerly awaited.