This page is an English translation of the Korean original. If there is any discrepancy, the Korean version prevails.
| Classification codes | C80 (KCD-8 / ICD-10) |
| Special Copayment Reduction | V193 · 5 years from the cancer patient registration date · 5% copayment |
| Synonyms | Cancer of unknown primary, carcinoma of unknown primary, CUP (Cancer of Unknown Primary), malignant tumor of unknown primary, Carcinoma of unknown primary (CUP), Occult primary malignancy |
| Adult onset | Very high (median age at onset about 65 years; also occurs in the 40s and younger) |
Contents
| A. | Classification codes | B. | Synonyms |
| 1. | History | 2. | Causes |
| 3. | Genetics | 4. | General epidemiology |
| 5. | Pathophysiology | 6. | Symptoms |
| 7. | Signs | 8. | Screening |
| 9. | Diagnosis | 10. | Treatment |
| 11. | Prognosis | 12. | Prevention |
| 13. | References | ||
| Code | Name |
|---|---|
| C80 | Malignant neoplasm without specification of site (cancer of unknown primary) |
| C80.0 | Malignant neoplasm of unknown primary site |
| C80.1 | Malignant neoplasm, unspecified, primary site specified |
| C80.2 | Malignant neoplasm, unspecified (related insurance and contract terms) |
Cancer of unknown primary (CUP) is defined as a histologically confirmed metastatic malignancy whose primary site cannot be identified even after standard diagnostic evaluation. Failure to identify the primary site does not simply mean that tests were not performed, but refers to a situation in which the primary site cannot be determined even after completing a systematic and comprehensive diagnostic evaluation.
Cancer of unknown primary (CUP: Cancer of Unknown Primary), carcinoma of unknown primary, occult primary malignancy, metastatic cancer of unknown site, malignant tumor of unknown primary. Clinically, it is also called "metastatic cancer of unknown origin" or "tumor of unknown primary." The ESMO (European Society for Medical Oncology), NCCN (National Comprehensive Cancer Network) and Korean Association for Clinical Oncology guidelines treat CUP as a separate clinical entity.
The concept of cancer of unknown primary began to be systematized in the early 20th century with advances in pathology. Initially it was regarded simply as "metastatic cancer whose primary site was not found," but the introduction of immunohistochemical techniques in the mid-20th century made it possible to estimate the tissue of origin to some extent.
In the 1970s–80s, empirical chemotherapy was the main treatment for CUP. Cisplatin-based combination chemotherapy was effective in some CUP subtypes, but the overall response rate was low and the prognosis poor. In the 1990s, advances in gene expression profiling (GEP) opened new horizons for CUP research. Molecular tests predicting the tissue of origin (TOO) were developed, and the strategy of predicting a specific tissue of origin and then applying targeted therapy for that primary cancer began to be tried.
In the 2000s, GEP-based classifiers such as the Pathwork Tissue of Origin Test were developed, making it possible to predict 15 tumor types from expression patterns of 87 genes. Since the 2010s, as clinical application of next-generation sequencing (NGS) expanded, actionable mutations such as EGFR mutations, BRAF V600E, ALK fusions and HER2 amplification have been found in CUP as well, creating opportunities for targeted therapy. In the 2020s, liquid biopsy and ctDNA methylation analysis are drawing attention as technologies used to estimate the primary tissue of origin. The ESMO 2023 clinical practice guideline and the NCCN 2024 guideline both recommend NGS panel testing and TOO molecular testing as central to the CUP diagnostic algorithm. Immune checkpoint inhibitors (pembrolizumab, MSI-H indication) and NTRK inhibitors (larotrectinib, entrectinib) are applicable to CUP as treatment based on site-agnostic molecular targets (tumor-agnostic therapy).
The causes of cancer of unknown primary are basically the same carcinogenic factors as for other solid cancers. However, CUP's unique feature—that the primary tumor is extremely small, has regressed spontaneously, or is undetectable with current technology—sets it apart in terms of mechanism of development.
Reasons the primary site cannot be identified:
Carcinogenic risk factors (general): smoking, heavy drinking, obesity, carcinogen exposure, immunosuppression, viral infection (EBV, HPV) and other general carcinogenic risk factors act in the same way in CUP. In the histologic distribution of CUP, adenocarcinoma is most common, followed by squamous cell carcinoma, poorly differentiated carcinoma and neuroendocrine tumors.
CUP is not a single molecular entity but a heterogeneous group derived from various primary cancers. However, CUP shows several common molecular features.
Characteristic molecular profile of CUP: in genomic studies, CUP has a higher rate of TP53 mutations than ordinary solid cancers (about 60%), and KRAS/NRAS mutations, CDKN2A/B loss and MYC amplification are frequent. Interestingly, CUP often does not fully match the gene profiles of standard solid cancers, leading some to argue that CUP is itself an independent molecular entity.
Molecular tests predicting tissue of origin:
Immunohistochemistry (IHC)-based estimation of origin:
Cancer of unknown primary (CUP) accounts for about 3-5% of all malignancies and is about the 8th most common cancer worldwide by incidence. It is of great clinical importance because, with its poor prognosis, it is about the 5th leading cause of cancer death.
Global epidemiology: about 220,000-270,000 people worldwide are diagnosed with CUP each year. The median age at onset is about 65 years, and the difference in incidence between men and women is small. By histologic type, adenocarcinoma is most common (about 60%), followed by squamous cell carcinoma (about 5%), undifferentiated carcinoma (about 25%) and neuroendocrine tumors (about 2%).
Korean epidemiology: in national cancer registry data, about 1,000-2,000 people are diagnosed with CUP each year, accounting for about 1-2% of all cancers. CUP appears to have a lower registration rate in Korea than in other developed countries, which may reflect differences in diagnostic systems and registration methods. With population aging and the growing number of cancer survivors, along with an increase in patients with complex metastatic cancer, the proportion of CUP diagnoses is also expected to change.
Common sites of metastasis (at CUP detection): lymph nodes (20-30%), liver (20-30%), lung (10-15%), bone (5-10%), brain (5%), peritoneum (10-15%). Clinical subtypes are distinguished by the predominant metastatic pattern, with differences in prognosis and treatment strategy.
The pathophysiology of CUP is explained by two main hypotheses.
Hypothesis 1: Early Metastasis Hypothesis. In CUP, the gene expression program related to metastasis is activated from the time the primary tumor is very small and early in formation. In the usual model of cancer progression, the tumor gradually acquires metastatic capacity as it grows, but in CUP this process occurs at a much earlier stage. In molecular studies, genomic differences between primary and metastatic tumors in CUP are small, providing evidence that the metastatic phenotype is strongly expressed from the earliest cells.
Hypothesis 2: Primary Tumor Regression Hypothesis. The primary tumor disappears through the host immune response or regresses spontaneously through ischemic necrosis caused by excessive angiogenesis. Spontaneous regression in melanoma is a representative example, and some cases of CUP are explained by this mechanism.
Common molecular mechanisms: CUP is characterized by overexpression of metastasis-related genes (MTA1, S100A4, SPARC, etc.), activation of epithelial-mesenchymal transition (EMT)-related genes, and activation of apoptosis-resistance mechanisms. Immunosuppression in the tumor microenvironment (PD-L1 expression, increased regulatory T cells) also contributes to immune evasion in CUP. Changes in DNA methylation patterns silence tissue-specific genes and suppress expression of differentiation markers, making estimation of the primary site difficult.
The symptoms of CUP vary greatly depending on the site of metastasis and are characterized by the absence of symptoms specific to a primary site.
Common symptoms: unexplained weight loss (the most common and important early symptom), generalized fatigue, anorexia, fever (malignant fever) and night sweats. Patients often present with these nonspecific symptoms, and metastatic cancer is found during workup.
Symptoms by site of metastasis:
Specific clinical features of undifferentiated carcinoma: undifferentiated carcinoma arising in the mediastinal or retroperitoneal lymph nodes in young patients suggests a germ cell tumor, and a trial of BEP (bleomycin+etoposide+cisplatin) chemotherapy along with AFP/βhCG testing is needed.
Clinical signs: the main physical examination findings that raise suspicion of CUP are unexplained lymphadenopathy (especially supraclavicular lymphadenopathy—a palpable Virchow node suggests abdominal/thoracic malignancy), hepatomegaly (irregular margin, tenderness), ascites and pleural effusion. Palpation of the supraclavicular lymph node (Virchow's node) is an important sign of metastatic cancer of unknown primary. Skin metastasis (Sister Mary Joseph's nodule around the umbilicus indicates intra-abdominal malignancy) and shifting dullness from peritoneal metastasis.
Laboratory findings:
Prognostic indicators: ECOG performance status, serum LDH, number of metastatic sites, metastatic pattern (favorable vs. unfavorable type) and histologic type are the main prognostic factors.
There is no general population screening for CUP. For patients suspected of having CUP, it is important to perform a minimum standard diagnostic evaluation while avoiding excessive unnecessary testing.
Initial evaluation checklist for patients with suspected CUP:
CUP diagnosis proceeds in three steps: (1) confirmation of a metastatic malignancy, (2) systematic search for the primary site, and (3) treatment-oriented classification (distinguishing favorable vs. unfavorable subtypes).
Obtaining sufficient tissue is essential. Whenever possible, sufficient tissue is obtained by excisional biopsy or core needle biopsy. Fine-needle aspiration (FNA) may be insufficient for marker analysis.
Basic IHC panel: CK (pan-cytokeratin), CK7, CK20, EMA, S-100, CD45 (LCA), HMB45/Melan-A (melanoma), PLAP (germ cell), CD56/Synaptophysin/Chromogranin (neuroendocrine), TTF-1, CDX2, PSA, PAX8, GATA3, ER, PR, HER2, WT1, Napsin A.
Chest-abdomen-pelvis CT is the baseline. PET-CT provides additional information when CT does not identify the primary site. Breast MRI (women), prostate MRI (men with elevated PSA), brain MRI (neurologic symptoms), GI endoscopy (if needed) and bronchoscopy (respiratory symptoms) are added depending on the situation.
CUP treatment is divided into (1) specific treatment according to clinical subtype, (2) empirical chemotherapy, and (3) targeted therapy based on molecular markers.
| Clinical subtype | Presumed primary cancer | Treatment |
|---|---|---|
| Axillary lymph node metastasis in women (adenocarcinoma) | Breast cancer | Breast cancer treatment protocol (surgery, radiotherapy, hormone therapy) |
| Peritoneal carcinomatosis (women, adenocarcinoma) | Primary peritoneal cancer/ovarian cancer | Platinum-based chemotherapy (carboplatin+paclitaxel) |
| Squamous cell carcinoma in cervical lymph nodes | Head and neck cancer | Head and neck cancer treatment (surgery+radiotherapy, cetuximab) |
| Mediastinal/retroperitoneal undifferentiated carcinoma (young men) | Germ cell tumor | BEP chemotherapy |
| Isolated bone metastasis (prostate-specific IHC) | Prostate cancer | Androgen deprivation therapy |
| CUP with neuroendocrine differentiation | Neuroendocrine tumor | Streptozocin+etoposide, or everolimus/octreotide |
Platinum-based combination regimens are used as standard first-line chemotherapy. Carboplatin+paclitaxel (or docetaxel) is the most widely used empirical regimen, with an overall response rate of 20-40% and median survival of 8-12 months. FOLFOX and FOLFIRI can also be used. Gemcitabine+cisplatin is considered in CUP of presumed lung or biliary tract origin.
When an actionable alteration is found by NGS panel:
This strategy treats the primary cancer corresponding to the tissue of origin predicted by molecular testing. Randomized controlled trials (CUPS, GEFCAPI, etc.) have not consistently demonstrated a survival advantage over empirical chemotherapy, but there is benefit in certain favorable subtypes, and the ESMO guideline supports this approach.
The prognosis of CUP is generally poor. Overall median survival is 6-12 months, the 1-year survival rate 30-40%, and the 5-year survival rate only 5-10%. However, favorable subtypes show much better outcomes.
| Clinical subtype | Median survival | Notes |
|---|---|---|
| Squamous cell carcinoma in cervical lymph nodes (presumed head and neck primary) | 36-60 months | Most favorable prognosis |
| Axillary lymph node adenocarcinoma in women (presumed breast cancer) | 30-48 months | Responds to breast cancer treatment |
| Peritoneal carcinomatosis (women, presumed ovarian cancer) | 18-30 months | Responds to platinum-based treatment |
| Undifferentiated carcinoma (mediastinal, men) | A small minority, but curable | Responds to BEP chemotherapy |
| Unfavorable type (multiple organ metastases) | 4-8 months | Shorter with poor performance status |
Prognostic factors: ECOG PS 0-1, a single or few metastatic sites, no liver metastasis, normal LDH and a favorable subtype are associated with good prognosis. Improved prognosis is expected when an alteration actionable by molecular marker-based targeted therapy is found.
There is no prevention strategy specific to CUP itself. General cancer prevention principles—smoking cessation, reduced drinking, a healthy diet, maintaining a healthy weight, regular exercise, reducing exposure to carcinogens, and HPV and hepatitis B vaccination—lower the risk of primary cancers and may thereby indirectly reduce the risk of CUP.
Importance of early diagnosis: it is important to see a specialist early when there is unexplained weight loss, persistent fatigue, new lymphadenopathy or unexplained pain. When suspicious symptoms occur, early detection of treatable subtypes through a rapid and systematic diagnostic evaluation can improve prognosis.
Managing hereditary cancer risk: in those with a hereditary cancer predisposition such as BRCA1/2 or Lynch syndrome, specialist counseling and targeted screening help detect primary cancers early, and CUP arising in these individuals is also more likely to have alterations treatable with molecular therapy.
Multidisciplinary approach: CUP requires a multidisciplinary team approach involving medical oncology, surgery, radiation oncology, pathology and radiology. Applying a standardized diagnostic algorithm and systematically using NGS-based molecular profiling are key strategies for optimal treatment of patients with CUP.