The patient is a 72 year old male with a history of metastatic non-small cell lung cancer (NSCLC) multiply treated with chemotherapy, check point inhibitor therapy and brain resection surgery presenting with pancytopenia (WBC:3.6 K/cumm, Hb:9.2g/dL, platelet count: 103 K/cumm) with macrocytosis.  A bone marrow examination was performed showing hypercellular marrow for age (~60% cellularity) with multilineage dysplasia (megakaryocytic and erythroid lineages) with ring sideroblasts (~18%). Blasts are not increased (~2%). On examination of the bone marrow core biopsy, there is a focal area of epithelioid cells with large pleomorphic nucleus, abundant clear to pink cytoplasm and inconspicuous nucleoli (Panel A-B). On immunohistochemical staining these cells are positive for pan-cytokeratin (AE1/AE3) (Panel C-D) and shows a dot-like cytoplasmic staining in most cells. These cells are positive for CD163 (Panel E-G) and are negative for myeloperoxidase (Panel H), lysozyme (Panel I), CD34 (Panel J) and CD117 (not shown) and other carcinoma markers TTF-1 (not shown), CAM5.2 (not shown) and Napsin A (not shown). Previous papers have suggested this dot-like cytokeratin staining is present in myeloblasts in cases of myelodysplastic syndrome (J Clin Pathol. 2001 Sep;54(9):735) and in myeloid sarcoma (Blood. 2016 Jul 21;128(3):460). However, in our case, it appears that the cytokeratin expression is not in the myeloblasts but in the macrophages. Similar findings of this aberrant cytokeratin expression in the tissue macrophages have been described in prior studies (Inflamm Res. 2008 Dec;57(12):593-600) however, has never been reported in bone marrow macrophages. These findings should be kept in mind as a diagnostic pitfall in evaluation of bone marrow in metastatic cancer patients and should be evaluated with multiple cytokeratin stains to confirm.