A 65-year-old woman with a history of resected breast cancer 10 years ago and resected lung cancer 3 years ago presented to hospital services feeling unwell. CT imaging demonstrated a new 2.8 cm right upper-lobe lung mass with hilar and mediastinal adenopathy, as well as multiple liver masses. She underwent ultrasound-guided biopsy of one of the liver lesions (shown below). A broad panel of immunohistochemical stains was performed.
Positive stains: CAM5.2, AE1/3, CK7, POU2F3, Ki-67: 75%
Negative stains: TTF1 (clone SPT24 as well as 8G7G3/1), p40, Chromogranin, Synaptophysin, GATA3, GCDFP1, ER, PAX8, WT1, CDX2, HepPar1, Arginase
The correct answer is ...
Small cell carcinoma.
The tumor shows an organoid arrangement of abnormal cells with increased N:C ratio, nuclear molding, and finely granular chromatin. There is abundant mitotic activity, as well as both punctate single cell and geographic tumor necrosis. Even without stains, the morphologic features demonstrate a high-grade, poorly differentiated neoplasm.
Historically, in the setting of a lung mass with hilar adenopathy, a diagnosis of small cell carcinoma (SCLC) could be made on morphology alone.1 However, given the history of previous malignancy in this patient, additional studies are prudent.
The extensive immunohistochemical work-up in this patient yields some interesting results. Whilst diffuse keratin expression confirms epithelial origin, additional markers are largely negative. Stains for mammary origin (GCDFP1, ER, and GATA3) are non-reactive. Additional less likely considerations, such as squamous cell, ovarian, hepatocellular, and gastrointestinal carcinoma, are effectively ruled out by negative staining in the remainder of the panel.
Most SCLC (about 80%) show strong nuclear positivity for TTF1, and show at least focal expression of classic neuroendocrine markers such as synaptophysin, chromogranin, and CD56 (the latter not performed in this case). However, up to 20% of SCLC may be negative for both synaptophysin and chromogranin, and even with the introduction of INSM1, they may be completely negative for neuroendocrine expression.2
Recent studies exploring the molecular landscape of SCLC has revealed insight into why some tumors lack neuroendocrine expression and defined a distinct subtype of SLCL with a “tuft cell” phenotype.3 These are characterized by the expression of POU2F3, a recently identified marker expressed in 10%-12% of SCLC. Interestingly, POU2F3-positive cases show low or negative expression of standard neuroendocrine markers. Future studies into the now well-defined molecular subtypes of SCLC, including the tuft cell type, may yield important prognostic and treatment implications for patients.
It is important to note that the use of POU2F3 requires the appropriate clinical and histopathologic context, as it is not specific for SCLC. This marker has been documented to stain large cell neuroendocrine carcinoma and squamous cell carcinomas.4
Ariel Sandhu, M.B., B.Ch.
Fellow, Pulmonary Pathology
Melanie Bois, M.D.
Consultant, Anatomic Pathology
Associate Professor of Laboratory Medicine and Pathology
Mayo Clinic College of Medicine and Science