Cancer of Unknown Primary

Metastatic malignancy with no identifiable primary site after standardized workup — clinicopathologic subsetting, the role of molecular tumor profiling, and the shift toward site-specific and biomarker-directed therapy

Key Points

• Cancer of unknown primary (CUP) is a biopsy-proven metastatic malignancy for which the anatomic site of origin cannot be identified after a standardized history, examination, laboratory testing, imaging, and pathologic evaluation.
• CUP accounts for approximately 1–2% of all invasive cancers; incidence has been declining with improvements in imaging (PET/CT), immunohistochemistry, and molecular diagnostics that reclassify many former CUP cases.
• Adenocarcinoma (~60%) and poorly differentiated carcinoma (~30%) are the dominant histologies; squamous cell, neuroendocrine, and undifferentiated neoplasms make up the remainder.
• Immunohistochemistry — anchored by cytokeratins CK7 and CK20 plus lineage-specific markers (TTF-1, CDX2, GATA3, PAX8, p40) — is the single most important tool for assigning a likely tissue of origin.
• The central clinical task is to separate the ~15–20% of patients who fall into favorable, treatable subsets from the majority with unfavorable disease, because the two groups have very different prognoses and management.
• Comprehensive genomic profiling frequently uncovers actionable alterations (e.g., MSI-high/dMMR, high TMB, NTRK/RET/ALK fusions, BRAF, HER2), making tissue-agnostic and biomarker-directed therapy a growing part of CUP care.
• Overall prognosis remains poor for unfavorable CUP (median OS ~9–12 months), but favorable subsets treated as their presumed primary can achieve long-term survival.

Definition & Epidemiology

Cancer of unknown primary (CUP) is defined as a histologically confirmed metastatic malignancy in which the primary tumor cannot be identified after a directed, standardized diagnostic workup. It is a diagnosis of exclusion: the primary is either too small to detect, has regressed spontaneously, or was never radiographically apparent. CUP historically accounted for 3–5% of all malignancies, but with modern PET/CT, refined immunohistochemistry, and molecular profiling, many cases previously labeled CUP are now assigned an origin, and the true incidence has fallen to roughly 1–2% of invasive…

Diagnostic Workup

The goal of the workup is efficient — not exhaustive — identification of a treatable primary or a favorable subset. Over-investigation delays therapy without improving outcomes. Standard evaluation for all patients: • Complete history and physical examination, including breast, pelvic, rectal, and genitourinary exam as appropriate. • Basic laboratory tests: CBC, comprehensive metabolic panel, urinalysis, and stool occult blood. • CT of the chest, abdomen, and pelvis with contrast. • Adequate biopsy of the most accessible lesion — core or excisional preferred over fine-needle aspiration to…

Pathology & Immunohistochemistry

Pathologic evaluation is the cornerstone of CUP management. The pathologist first classifies the tumor by light microscopy into one of the major histologic groups: well/moderately differentiated adenocarcinoma (~60%), poorly differentiated carcinoma or adenocarcinoma (~30%), squamous cell carcinoma (~5%), and undifferentiated neoplasm (~5%). Undifferentiated tumors require a broad panel to exclude lymphoma, melanoma, sarcoma, and germ-cell tumor, which have specific, effective therapies. Immunohistochemistry is then used to infer a tissue of origin. The initial cytokeratin pattern narrows…

Molecular & Gene-Expression Profiling

Two distinct molecular approaches inform CUP care, and they answer different questions. Tissue-of-origin (gene-expression) classifiers use RNA or microRNA expression signatures to predict the most likely primary. While they can assign a putative origin in the majority of cases, the pivotal randomized trial data are mixed: prospectively directing therapy by a molecular tissue-of-origin prediction has not clearly improved survival over empiric chemotherapy for unfavorable CUP. These assays are therefore adjuncts to — not replacements for — clinicopathologic judgment. Comprehensive genomic…

Favorable Subsets

Roughly 15–20% of patients with CUP belong to a favorable subset that should be treated as the corresponding known cancer, often with meaningful long-term survival: • Women with isolated axillary lymphadenopathy (adenocarcinoma): Managed as stage II–III breast cancer — axillary surgery, breast radiotherapy, and systemic therapy guided by ER/PR/HER2; occult breast MRI often localizes the primary. • Women with papillary serous peritoneal carcinomatosis (PAX8+, CA 125 elevated): Treated as advanced epithelial ovarian/primary peritoneal cancer with cytoreductive surgery and platinum–taxane…

Treatment of Unfavorable CUP & Prognosis

The majority of patients do not fit a favorable subset and have unfavorable CUP — typically disseminated adenocarcinoma or poorly differentiated carcinoma involving multiple organs (liver, lung, bone, or multiple nodal stations). For these patients: • Biomarker-directed therapy first when available: MSI-H/dMMR or high TMB → pembrolizumab; actionable fusions or mutations → the corresponding targeted agent. This is the most impactful advance in CUP care. • Empiric chemotherapy for biomarker-negative disease: A platinum-based doublet (e.g., carboplatin plus paclitaxel), or gemcitabine-based…