Unknown Primary Carcinomas: Diagnosis and Management

April 2, 2005
Jean-pierre M. Ayoub, MD

Kevin P. Hubbard, DO

Renato Lenzi, MD, FACP

Unknown primary carcinomas are a significant health problem, constituting 3% to 10% of all tumors diagnosed in the United States each year [1,2]. While the majority of patients with metastatic carcinoma of unknown primary origin have short survival times and disease resistant to treatment, recent findings suggest that certain subsets of patients have tumors that are responsive to chemotherapy. Others can be successfully treated with regional therapy.

EpidemiologyBiologyDiagnostic EvaluationClinical EvaluationHistopathologic Evaluation and Use of Serum Tumor MarkersClinical Presentation and Treatment RecommendationsChemotherapyConclusionReferences

Unknown primary carcinomas are a significant health problem, constituting 3% to 10% of all tumors diagnosed in the United States each year [1,2]. While the majority of patients with metastatic carcinoma of unknown primary origin have short survival times and disease resistant to treatment, recent findings suggest that certain subsets of patients have tumors that are responsive to chemotherapy. Others can be successfully treated with regional therapy.

In this chapter, concepts and guidelines regarding a directed evaluation of patients with unknown primary carcinomas are presented, and the clinical course and treatment of disease presenting at each major site are reviewed.

A practical definition of unknown primary carcinoma is a biopsy-proven carcinoma with no identifiable source as determined by medical history and physical examination, complete blood cell count, chemistries, urinalysis, chest x-ray, computed tomography (CT) scan of the abdomen, CT scan or sonogram of the pelvis, and mammogram (in women). Abnormalities detected in the initial testing should be pursued if possibly indicative of a primary tumor. In patients presenting with metastatic carcinomas that have no obvious source, the subsequent search for a primary tumor is unrewarding in approximately 75% of the cases, even after the most thorough evaluation [1,3].

This fact has caused much consternation for both patients and physicians. The treatment of malignancy is traditionally based on the identification of the origin of the tumor, and treatment is chosen and initiated based upon the natural history and the most specific therapies available for a certain type of tumor. Without knowledge of the primary site, the oncologist is often hesitant to recommend therapy.

In addition, the physician may believe that he or she has somehow failed to serve the patient adequately by being unable to identify a primary malignancy. Further, in reviewing the literature, the oncologist is presented with an assortment of disparate clinical trials and retrospective studies that often arrive at contradictory conclusions and, in many cases, exclusively stress the overall poor prognosis of this disease.


Unknown primary carcinomas affect males and females equally. The median age at presentation is 56 to 60 years, and 10% of patients with an unknown primary tumor have a history of an antecedent cancer [4]. Postmortem examination will identify the occult primary tumor in 60% to 80% of cases [5,6]. In an autopsy series on patients with unknown primary tumors reported by Nystrom, the most commonly identified primary sites were pancreas (20%) and lung (18%)[7]. The poor prognosis of these malignancies reflects the overall poor prognosis of unknown primary tumors as a group. Although breast and prostate cancers represent the most common cancers in women and men, these accounted for only 4% and 2%, respectively, of the primary sites found in a recent series [8].

About 60% of the patients who present with an unknown primary carcinoma will be diagnosed with metastatic adenocarcinoma. An additional 30% of patients will have undifferentiated or poorly differentiated carcinomas. The remaining patients will have one of a variety of carcinomas that include squamous-cell carcinomas, neuroendocrine tumors, and malignancies that upon more detailed study will be found to be sarcomas, lymphomas, germ-cell tumors, melanomas, and unclassifiable undifferentiated malignant neoplasms.


In 20% to 40% of cases of unknown primary tumors, the primary site will not be identified even after autopsy [5]. Unknown primary carcinomas therefore exhibit unique biologic behavior, with the development of clinically devious metastatic lesions in the absence of a detectable primary tumor. Certain cancers such as melanoma and renal cancer are known to exhibit spontaneous regression of the primary tumor in association with the host's immunologic response. It is clear that many of these tumors have acquired the capacity to metastasize even though a primary lesion is not clinically evident.

The exact mechanisms of these phenomena are unclear, but models in which genetic alterations result in early metastasis have been proposed for other tumors [9] and may be applicable to unknown primary carcinomas as well. The notion that a genetic alteration is responsible is supported by data that demonstrated a partial or complete loss of the short arm of chromosome 1 in 12 of 13 patients with unknown primary tumors [10]. This chromosomal aberration has been reported by others as being characteristic of the karyotype most often associated with advanced malignancy and confirms the concept that structural changes of chromosome 1 exist in most forms of advanced cancer [11]. Multiple other karyotypic abnormalities have been demonstrated in unknown primary malignancies.

Molecular studies have shown overexpression of c-myc, ras, and c-erbB-2 in unknown primary tumors [12]. Although these are highly metastatic advanced tumors that would have been expected to have a high incidence of p53 mutations, the frequency of p53 mutations in a series of such tumor biopsy samples was found to be lower than anticipated (26%) [13].

Diagnostic Evaluation

Much has been written about the proper approach to diagnostic evaluation of unknown primary tumors. Some authors advocate the use of a minimal diagnostic analysis, its scope limited to separating treatable disease from untreatable disease [14]. Others have supported a more aggressive approach wherein a complete assessment of the extent of the disease as well as detection of the primary tumor site is attempted.

We believe that a pragmatic approach is best. Extensive evaluation of all patients presenting with metastases of unknown primary carcinoma is an expensive and wasteful extreme (particularly in this era of heightened consciousness over the costs of health care). One study reported the average cost of investigation of a patient with an unknown primary tumor to be $17,973 [15]. In that study, patient survival time reflected the natural history of unknown primary tumors: Mean survival time was 8.1 months, with only 18% of patients surviving 1 year. However, a strictly minimalist approach may result in the oversight of treatable (and potentially curable) neoplasms.

An important determinant of the appropriate extent of evaluation for a given patient with an unknown primary carcinoma is whether the data obtained by a diagnostic test will influence treatment decisions. If the initial data point to a treatable or potentially curable cancer (eg, a germ-cell tumor or lymphoma), then further investigation should proceed until a precise clinical diagnosis can be made. Clinical data support the belief that patients who have unknown primary tumors that are later proved to have originated from a given site or from unique histology have an overall prognosis similar to that of patients who present with a known primary tumor [16]. The recommended general approach is thus one of directed evaluation based upon the clinical presentation, the initial pathologic findings, and the characteristics of the individual patient.

Clinical Evaluation

In each case, a thorough medical history should be obtained, and a physical examination including a digital prostate exam in men and a pelvic exam in women should be performed. Determination of the patient's performance status, nutrition, and the presence or absence of concomitant medical illnesses and malignancy-related complications (such as paraneoplastic syndromes, painful metastases, or spinal-cord compression) that may affect patient care is required.

Laboratory tests should include routine biochemical and hematologic surveys. Initial radiographic studies should include chest x-ray films, CT scans of the abdomen, and a CT scan or sonogram of the pelvis. The role of CT scanning of the abdomen or pelvis is documented in patients with unknown carcinoma; detection of the site of origin in 30% to 35% of these patients has been reported [17,18]. Not surprisingly, the tumors most often identified on CT scans are those that arise from the pancreas, kidney, hepatobiliary tract, and ovary. CT scans of the chest should be performed for patients who have abnormalities evident on their chest radiographs. A cytologic analysis of the sputum should be performed in any patient with a radiographically documented chest abnormality.

All women with unknown primary carcinoma should undergo mammography and a careful pelvic examination. In cases of suspicious findings on a breast examination and a negative mammogram, patients should have a breast sonogram and a biopsy as indicated. Imaging or endoscopy of the upper and lower gastrointestinal (GI) tract is indicated for patients with abdominal complaints, ascites, liver metastases, or other findings in the initial workup that point to a possible GI primary tumor.

Patients with upper or mid-cervical adenopathy should receive a thorough evaluation including endoscopy by a head and neck surgeon as well as CT scanning of the head and neck region. Lower cervical or supraclavicular adenopathy suggests a primary tumor arising from below the clavicle, most commonly from the lung. Patients who have a pathologic diagnosis of papillary carcinoma in the neck or chest region should undergo imaging studies of the thyroid. Patients with inguinal lymphadenopathy may have a detectable primary site in the perineal or anorectal area, and anoscopy and colposcopy should be performed [1]. Evaluation by a urologist may reveal a primary carcinoma of the distal urinary tract.

Histopathologic Evaluation and Use of Serum Tumor Markers

All pathologic material obtained at biopsy should be evaluated by an experienced pathologist who is familiar with the special diagnostic problems of unknown primary carcinomas. The pathologist should also be informed of the patient's pertinent history and clinical findings and may be able to recommend further analysis based upon this information.

Light microscopy will show that about 60% of patients with unknown primary tumors have adenocarcinoma and that another 5% have squamous-cell carcinoma. In the remaining patients, light microscopy is much less specific and identifies the tumor as a poorly differentiated neoplasm, a poorly differentiated carcinoma, or a poorly differentiated adenocarcinoma [1].

Subsequent evaluation of this group of poorly differentiated lesions by means of special immunohistochemical techniques is warranted because some of these tumors are curable or very responsive to treatment. Many immunohistochemical reagents are at the disposal of the pathologist, making the histologic classification of the tumor easier. Especially useful are the antibodies to common leukocyte antigens present in lymphoma and the antibodies to the prostate-specific antigen (PSA) present in most prostate cancers [19,20].

The use of cytogenetic analysis in the diagnosis of unknown primary tumors is still limited but holds promise. Specific chromosomal abnormalities have been identified in several types of lymphoma (8,14 translocation in small non-cleaved-cell non-Hodgkin's lymphoma) [21], germ-cell tumors i(12p) [22], and neuroendocrine tumors t(11,22)[23]. Because these tumors require specific therapies, identification of the specific tumors and directed therapy is preferable to empirically treating all patients who have poorly differentiated neoplasms with a generic cisplatin (Platinol)-based regimen of chemotherapy [24].

Recently, Motzer et al have used cytogenetic techniques to identify a treatable subset of patients with poorly differentiated carcinomas. Response to cisplatin-based therapy correlated with the finding of i(12p) in tumors by either molecular or cytogenetic studies [25]. Because i(12p) is a chromosomal marker that characterizes germ-cell tumors, cytogenetic studies will provide another way to identify this treatable subset of poorly differentiated carcinomas.

The role of tumor markers in the evaluation of patients with unknown primary tumors is unclear. Most tumor markers are nonspecific, and they may be most useful in monitoring a patient's response to therapy [26]. Tumor markers do not appear to be predictive of response to chemotherapy [27], but their directed use seems valuable in complementing the overall evaluation of patients with unknown primary cancers [28,29].

Men who present with adenocarcinoma should have PSA and prostatic acid phosphatase levels measured, even in the absence of bony metastases. When tumor histology is undifferentiated, beta-human chorionic gonadotropin and alpha-fetoprotein levels should be measured, especially if the clinical presentation suggests an extragonadal germ-cell tumor [29]. In patients with hepatic tumors, alpha-fetoprotein levels should also be measured. The adenocarcinoma markers (CEA, CA125, CA15-3, and CA19-9) are often elevated in patients with unknown primary tumors and cannot be reliably used to identify a specific primary site.

Using the diagnostic approach outlined, it is possible to detect a primary neoplasm in 20% to 30% of patients presenting with unknown primary malignancies. In a review of primary tumors identified by this approach, investigators at M.D. Anderson Cancer Center found that the most common sites of origin for epithelial histologies were the lung (15%), the pancreas (13%), the colon/rectum (6%), the kidney (5%), and the breast (4%). Melanomas, sarcomas, and lymphomas each made up 6% to 8% of the total. The remaining cases were primary tumors of the stomach (4%), ovary (3%), liver (3%), esophagus (3%), mesothelioma (2%), prostate (2%), and a variety of other histologies (19%)[8].

Clinical Presentation and Treatment Recommendations

Not surprisingly, patients with unknown primary carcinomas present with many of the same symptoms as patients with advanced malignancies of known origin. In one review, clinical presentation included general deterioration (73%), digestive symptoms (58%), liver enlargement (58%), abdominal pain (56%), respiratory symptoms (45%), ascites (26%), and node enlargement (16%)[30].

Approximately 50% of patients who have unknown primary tumors present with multiple sites of involvement. In those with a dominant (or sole) site, the most common sites are the liver (25%), bone (22%), lung (20%), lymph nodes (15%), pleural space (10%), and brain (5%).

The clinical course of patients with unknown primary carcinomas varies widely. Although as a group, their median survival time is poor, there are certain prognostic variables that suggest longer survival, such as involvement of lymph nodes only, pathologic diagnoses of squamous carcinoma and neuroendocrine carcinoma, and disease limited to one organ site. On the other hand, variables suggestive of a poor prognosis include male sex, pathologic diagnosis of adenocarcinoma, and metastatic involvement of the liver, lung, bone, pleura, or brain [31].

For the majority of patients with unknown primary carcinomas, chemotherapy has not been shown to favorably affect survival. A small number of patients within this heterogeneous group, however, are candidates for curative treatment.

Clinicopathologic Subgroups

Extragonadal Germ-Cell Syndrome: As a group, patients who have undifferentiated or poorly differentiated carcinoma, are less than 50 years old, and present with rapidly growing midline tumors involving the lymph nodes, mediastinum, or retroperitoneum have been found to have tumors that are very responsive to chemotherapy, particularly to platinum-containing regimens. It is believed that these individuals have poorly differentiated extragonadal germ-cell tumors. Investigators have referred to this disease entity as “extragonadal germ-cell syndrome” or “germ-cell equivalents. [32-34]” These patients have response rates to chemotherapy of 35% to 50%, and those who achieve a complete response often enjoy a durable remission.

Disease Limited to the Lymph Nodes: Another important clinical subset is made up of patients who present with lymph-node involvement with no other apparent metastases. These patients usually do better than those who show evidence of visceral involvement. The favorable prognosis may be related to a low tumor burden at the time of diagnosis rather than to intrinsic sensitivity to treatment of the tumor.

Cervical and Supraclavicular Adenopathies: Patients presenting with involvement of the mid- or high-cervical nodes are believed to have an occult primary tumor in the head or neck. Both squamous-cell carcinomas and adenocarcinomas are encountered, but patients who present with adenocarcinoma appear to have a poorer prognosis.

Supraclavicular adenopathy is another fairly common presentation, affecting 15% of the unknown primary tumor patients in the M.D. Anderson Cancer Center series [31], and represents either an undiagnosed tumor in the head and neck region or, more often, far-advanced malignant disease from a distant site. The majority of these tumors are squamous-cell or undifferentiated carcinomas. Patients with disease in the supraclavicular nodes have a worse prognosis than do patients with disease in other lymph-node-bearing areas. Carcinoma affecting lymph nodes on the right side most commonly arises from occult primary tumors of the lung and breast. When disease affects the lymph nodes on the left side, spread from intra-abdominal malignancies by way of the thoracic duct (Virchow's node) is an additional possibility.

Most individuals with squamous-cell carcinoma or adenocarcinoma involving only the cervical or supraclavicular lymph nodes should receive regional therapy and careful follow-up. Patients whose disease is limited to the mid- or high-cervical nodes benefit from local and regional therapy. The 3-year survival rate after radical neck dissection and/or radical cervical irradiation ranges from 35% to 60% [1]. Within this group, patients with N1 disease (according to the tumor, nodes, and metastasis, or TNM, staging system) have the better prognosis; patients with N3 disease, regardless of the local treatment modality used (surgery, radiotherapy, or both), fail to achieve complete remission in 65% of cases.

Recently, investigators have tried to improve the outcome of patients with N3 disease by adding systemic chemotherapy to local treatment modalities. In one study, induction chemotherapy with cisplatin and fluorouracil was followed by radiation therapy. The rate of complete remission improved to 60% in patients with N3 disease, and although the study sample was small, these complete remissions appeared durable (median, 39 months)[35]. Patients with low cervical or supraclavicular nodes have a much worse survival rate.

Axillary adenopathy is an additional disease subset with a favorable prognosis. Women who present with adenocarcinoma in the axillary lymph nodes should undergo mammography because the most likely primary is breast cancer. Lesions identified by mammography should be biopsied. Suspicious areas not well imaged by mammography should be evaluated by breast ultrasonography; biopsy can be performed using ultrasound guidance.

These patients are best treated as if they had stage II breast cancer, and they may enjoy a significant survival benefit. Modified radical mastectomy has been recommended in these patients even when physical examination and mammography studies fail to identify a primary breast cancer. A report by Ellerbroek et al documented actuarial disease-free survival rates of 71% at 5 years and 65% at 10 years [36]. Survival rates were higher in patients who received systemic chemotherapy plus radiation therapy. Local control was also enhanced by irradiating the affected breast and axilla.

The prognosis is not as favorable in men who present with axillary adenopathy only. In these patients, a reasonable therapeutic approach is to proceed with axillary node dissection in an attempt to remove all known disease and to add local radiation, if needed. Patients are then observed without further therapy. If the disease is rapidly growing or bulky, systemic chemotherapy is administered before surgery or radiation therapy.

Inguinal Adenopathy: A few patients with unknown primary tumors present with inguinal adenopathy. Undifferentiated (anaplastic) carcinoma is identified in at least half of these cases. Some of these anaplastic “carcinomas” appear to be melanomas with no obvious primary skin lesion. The remaining patients have adenocarcinomas arising from the skin, genitourinary tract, anus, or pelvis. A detailed investigation for primary lesions in these areas is warranted. In patients with adenocarcinomas and poorly differentiated carcinomas that are strictly confined to the groin nodes, a superficial groin dissection should be performed. Further treatment is often withheld in these cases, while the patients are monitored for recurrent disease [4]. Chemotherapy may be offered in the face of systemic disease, preferably in the context of a clinical trial.

Melanoma involving the lymph nodes is a fairly common occurrence. Most patients present with clinical stage II disease and have a 5-year survival rate of 27% to 33% [4]. Authorities have recommended radical lymph-node dissection for melanomas presenting as unknown primary tumors, because a 5-year survival rate of 40% to 50% was reported for patients who had been aggressively treated [37].

Malignant Ascites: Patients with malignant ascites usually belong to one of two subsets, each with a very different natural history of disease. The first group consists of patients with mucin-producing adenocarcinoma, who may present with ascitic fluid that contains signet-ring cells. These patients often have multiple peritoneal implants. This group has a poor prognosis and responds poorly to currently available treatment regimens.

The second group is made up of women with papillary adenocarcinoma in the peritoneal fluid; this disease is often associated with pelvic adenopathy or masses. These patients may have elevated CA125 levels but do not have detectable ovarian cancer. Some investigators consider these patients to have true unknown primary ovarian tumors or primary serous carcinomas of the peritoneum [38,39]. This disease has not been documented in males. Disease management should be the same as that for women with ovarian carcinoma. Response rates ranging from 32% to 39% have been reported in patients who undergo cytoreductive surgery followed by cisplatin-based chemotherapy [1].

Neuroendocrine Tumors: Patients with poorly differentiated neuroendocrine tumors may present with patterns of metastases strongly resembling those of extragonadal germ-cell tumors. It may be difficult to separate this subset of neoplasms from the germ-cell group. However, immunohistochemical stains for neuron-specific enolase and chromogranin A (markers of neuroendocrine differentiation)[40] and for alpha-fetoprotein, beta-human chorionic gonadotropin, and placental alkaline phosphatase (markers supporting a diagnosis of germ-cell tumors) may aid in differentiation. Neuroendocrine tumors can also be diagnosed by electron microscopy, which can reveal the neurosecretory granules characteristic of such tumors [41].

Although the nature and course of this clinical entity is still being defined, the clinical data suggest that this type of anaplastic tumor is responsive to cisplatin-based chemotherapy, even when the primary site is unknown. Achievement of a reasonable duration of response is also supported by clinical data [42,43].

Pulmonary Metastases and Pleural Effusions: Individuals with multiple pulmonary metastases constitute the largest group of patients with unknown primary tumors, and lung cancer is the most frequent primary diagnosis made in this group. The diagnosis is commonly based on the results of chest radiography, CT scans, and either sputum cytology or bronchoscopy. It is very unusual for patients with pulmonary metastases to be candidates for surgical resection; these patients most often receive systemic chemotherapy. With the exception of some young patients whose tumors fit the criteria of germ-cell equivalents, these patients usually do very poorly.

Malignant pleural effusions are relatively common, affecting about 10% of patients with unknown primary tumors. Most of these patients have adenocarcinomas, which may be difficult to differentiate from mesotheliomas. Therapy for this group of patients is conservative. After initial diagnostic thoracentesis, patients are monitored for fluid reaccumulation. If the effusion reaccumulates quickly, pleurodesis may be attempted to slow the rate of fluid reaccumulation. When the effusion reaccumulates over a longer period or if new sites of disease develop, systemic chemotherapy should be administered.

Bony Metastases: When metastasis to bone is detected, men should be evaluated for prostate cancer and women for breast cancer. Exclusion of these diagnoses leaves a group of patients with both a poor prognosis and a potentially painful condition. Patients with a single bony metastasis should be given local treatment with surgery and/or radiation and then monitored. Patients with multiple-site disease and those whose tumors progress after radiation therapy should be offered a trial of chemotherapy. Many experimental agents are currently available in ongoing clinical trials. Therapy with bone-seeking radioisotopes (eg, strontium 89) may be useful in the treatment of disseminated painful bone metastases [44].

Hepatic Metastases: Patients with hepatic metastases constitute 20% to 30% of individuals with unknown primary tumors. The most important diagnostic considerations in this class are to distinguish primary liver tumors from cancers that have metastasized to the liver and to identify the subset of patients with neoplasms of a more indolent nature (eg, neuroendocrine tumors). A careful pathologic review of liver biopsy specimens is therefore essential. Once these histologies are excluded, the most common diagnoses are adenocarcinoma and undifferentiated carcinoma. Initial therapy for these patients is systemic chemotherapy; hepatic intra-arterial therapy is an option for patients who do not respond to intravenous chemotherapy. Surgery should be considered an option for patients with resectable disease.

Brain Metastases: The important factor in treating patients who have brain lesions is to distinguish individuals with metastatic disease from those with primary brain tumors. Once this distinction has been made, patients with single metastatic lesions should be considered for surgery and those with multiple lesions should receive radiotherapy. Following treatment, patients should be monitored for recurrence or the appearance of a primary tumor, which in one report occurred most often in the lung [45].

Other Sites

Single Sites Discovered Incidentally on Resection: Unknown primary tumors are notorious for unusual isolated presentations. Such lesions may appear on the skin, in single isolated lymph nodes removed during surgery for benign conditions, and at other, even more unusual sites. Patients should be examined for primary tumors and other sites of metastasis as described earlier. If no primary tumor and no additional sites of metastases are found, complete removal of the lesion must be ensured; this often requires additional excision with wider margins. The patient should then be monitored without therapy, regardless of the tumor histology involved. Many such patients enjoy a prolonged survival time. Patients with isolated skin lesions may have an undifferentiated primary integumentary tumor with potential for cure after adequate local surgical treatment.

Multiple-Site Visceral Disease: Developing a strategy to care for patients with unknown primary carcinoma with metastases involving the viscera has proven exceedingly difficult. As noted previously, some subsets of patients in this category have disease that is responsive to therapy (eg, those with germ-cell tumors and their equivalents, women with papillary abdominal carcinomatosis, and patients with neuroendocrine tumors). In these patient subsets, the durations of remission and survival are considerable if complete remission is achieved with appropriate therapy [46]. Such patients should be treated aggressively with platinum-based chemotherapy regimens and may have overall response rates as high as 50% and complete response rates ranging from 20% to 35% [32,33].


Data from chemotherapy trials enrolling patients with unknown primary carcinomas have historically been difficult to interpret. Many early studies were done prior to the era of controlled clinical trials, and the methods used in interpreting the results of these studies have been questioned. Additionally, combination regimens using newer chemotherapeutic agents have consistently demonstrated greater benefit than did older single-agent therapies.

Several difficulties arise when survival and response rates reported in different chemotherapy trials are compared. For example, histologic criteria for patient selection often varied from study to study. Moreover, in older studies, immunohistochemical methods were not used in the evaluation of pathologic specimens.

Patients who present in one of the subsets discussed previously should be managed as described in the appropriate section. For example, women with axillary adenopathy should receive local therapies in addition to chemotherapy with a combination regimen typically used for breast cancer. Such a protocol might be the FAC regimen (fluorouracil, doxorubicin [Adriamycin, Rubex], and cyclophosphamide [Cytoxan, Neosar])[47]. However, some generalizations can be made regarding the useof chemotherapy for patients with unknown primary tumors:

  • In patients with widespread metastases and poor performance status, systemic chemotherapy is likely to be more harmful than beneficial, and only supportive therapy is usually indicated.
  • Whenever possible, patients with unknown primary carcinoma who do not belong in the previously defined subgroups should be treated within the context of a clinical trial.
  • Patients with moderately differentiated or well-differentiated adenocarcinomas who do not fit into one of the subsets previously discussed and who require chemotherapy have response rates of about 20% to 30%, and their median survival time ranges from 7 to 11 months when they are treated with combinations of doxorubicin and mitomycin (Mutamycin), with or without additional agents [48]. More recently, a cisplatin-based regimen has generated similar response rates [49].
  • Patients with undifferentiated or poorly differentiated adenocarcinomas other than those fitting in the extragonadal germ-cell or neuroendocrine tumor groups have traditionally been given a trial of a cisplatin-based regimen. Although most of these patients have a poor prognosis, Hainsworth and Greco have identified 5% to 15% as potentially curable by cisplatin-based chemotherapy [50]. A recent study, however, found that patients with a diagnosis of poorly differentiated carcinoma or poorly differentiated adenocarcinoma had similar survival rates whether they were treated with a combination of vincristine, doxorubicin, and cyclophosphamide or with a cisplatin-based regimen [51].
  • Patients with squamous-cell carcinomas who require chemotherapy are also often effectively treated using a cisplatin-based regimen. Investigators have discussed the merits of a combination of fluorouracil and cisplatin in squamous-cell carcinomas of the head and neck [35].
  • Patients with the extragonadal germ-cell syndrome or germ-cell equivalent and women with papillary abdominal carcinomas should be treated with a cisplatin-based regimen.
  • Two recent studies examined the effectiveness of oral etoposide (VePesid) in unknown primary tumors. The drug was shown to have limited or no activity [52,53].


It is clearly inappropriate to think of all patients with unknown primary neoplasms as having an untreatable disease and a poor outlook. Significant benefit may be achieved by administering regional or specific systemic therapies, and many of these patients can expect a prolonged survival. All patients should undergo a directed diagnostic evaluation for the primary tumor.

Some subsets of patients defined by clinical criteria (eg, axillary nodes with adenocarcinoma in women or peritoneal carcinomatosis with a papillary or serous histology), histologic criteria (eg, neuroendocrine, small-cell, or germ-cell tumor), or a combination of clinical and histologic criteria (eg, undifferentiated tumors with a midline presentation in patients less than 50 years old) may benefit significantly from aggressive treatment with platinum-containing regimens.

For the majority of patients who present with advanced unknown primary tumors, however, the prognosis remains poor and no treatment of established efficacy is available. For those patients, enrollment into a clinical trial of an experimental therapeutic approach appears appropriate. To effect significant improvement in the treatment and survival rate of patients with unknown primary malignancies, new insights into the biology of the disease and the development of new agents active against adenocarcinomas are critical.



1. Hainsworth JD, Greco FA: Treatment of patients with cancer of an unknown primary site. N Engl J Med 329:257–263, 1993.

2. Wingo P, Tong T, Bolden S: Cancer statistics, 1995. CA Cancer J Clin 45:8–30, 1995.

3. Raber MN, Abbruzzese JL, Frost P: Unknown primary tumors. Curr Opin Oncol 4:3–9, 1992.

4. Casciato DA, Tabbarah HJ: Metastases of unknown origin, in Haskell CM (ed): Cancer Treatment, 3rd Ed, pp 798–814. Philadelphia, WB Saunders, 1990.

5. Daugaard G: Unknown primary tumours. Cancer Treat Rev 20:119–147, 1994.

6. Maiche AG: Cancer of unknown primary. Am J Clin Oncol 16:26–29, 1993.

7. Nystrom JS, Weiner JM, Heffelfinger-Juttner J, et al: Metastatic and histologic presentations in unknown primary cancer. Semin Oncol 4:53–58, 1977.

8. Abbruzzese JL, Abbruzzese MC, Lenzi R, et al: Analysis of a diagnostic strategy for patients with suspected tumors of unknown origin. J Clin Oncol 1995 (in press).

9. Fearon ER, Vogelstein B: A genetic model for colorectal tumorigenesis. Cell 61:759–767, 1990.

10. Bell CW, Pathak S, Frost P: Unknown primary tumors: Establishment of cell lines, identification of chromosomal abnormalities, and implications for a second type of tumor progression. Cancer Res 49:4311–4315, 1989.

11. Atkin NB: Chromosome 1 aberrations in cancer. Cancer Genet Cytogenet 21:279–285, 1986.

12. Pavlidis N, Briassoulis E, Bai M, et al: The expression of c-myc, ras, and c-erbB-2 in patients with carcinoma of unknown primary. Proc Am Soc Clin Oncol 13:1374, 1994.

13. Bar-Eli M, Abbruzzese JL, Lee-Jackson D, et al: p53 Gene mutation spectrum in human unknown primary tumors. Anticancer Res 13:1619–1624, 1993.

14. Stewart JF, Tattersall MHN, Woods RL, et al: Unknown primary adenocarcinoma. Br Med J 1:1530–1533, 1979.

15. Schapira DV, Jarret AR: Cost of diagnosis and survival of patients with unknown primary cancer. Proc Am Soc Clin Oncol 13:481,1994.

16. Horning SJ, Carrier EK, Rouse RV, et al: Lymphomas presenting as histologically unclassified neoplasms: Characteristics and response to treatment. J Clin Oncol 7:1281–1287, 1989.

17. Darsell PR, Sheedy PF II, O'Connell MJ: Computed tomography in search of cancer of unknown origin. JAMA 9:427–433, 1982.

18. McMillan JH, Levine L, Stephens RH: Computed tomography in the evaluation of metastatic adenocarcinoma of unknown origin. Radiology 143:143–146, 1982.

19. Warnke RA, Gatter CK, Falini B, et al: Diagnosis of human lymphoma with monoclonal antileukocyte antibodies. N Engl J Med 309:1275–1281, 1983.

20. Allhoff EP, Proppe KH, Chapman CM, et al: Evaluation of prostate specific acid phosphatase and prostate specific antigen in identification of prostate cancer. J Urol 129:315–318, 1983.

21. Arnold A, Cossman J, Bakhshi A, et al: Immunoglobulin gene rearrangements as unique clonal markers in human lymphoid neoplasms. N Engl J Med 309:1593–1599, 1983.

22. Bosl GJ, Dmitrovsky E, Reuter V, et al: i(12p): A specific karyotypic abnormality in germ cell tumors. Proc Am Soc Clin Oncol 8:131, 1989.

23. Whang-Peng J, Triche TJ, Knutsen T, et al: Chromosome translocations in peripheral neuroepithelioma. N Engl J Med 311: 584–585, 1984.

24. Ilson DH, Motzer RJ, Rodriguez E, et al: Genetic analysis in the diagnosis of neoplasms of unknown primary tumor site. Semin Oncol 20:229–237, 1993.

25. Motzer RJ, Rodriguez E, Reuter VE, et al: Molecular and cytogenetic studies in the diagnosis of patients with poorly differentiated carcinomas of unknown primary site. J Clin Oncol 13:274–282, 1995.

26. Bates SE: Clinical applications of serum tumor markers. Ann Intern Med 115:623–638, 1991.

27. Pavlidis N, Kalef-Ezra J, Briassoulis E, et al: Evaluation of six tumor markers in patients with carcinoma of unknown primary. Med Pediatr Oncol 22:162–167,1994.

28. Shahangian S, Fritsche HA: Serum tumor markers as diagnostic aids in patients with unknown primary tumors. Cancer Bulletin 41:152–156, 1989.

29. Abbruzzese JL, Raber MN, Frost P: The role of CA125 in patients with unknown primary tumors. Proc Am Soc Clin Oncol 10:39, 1991.

30. Mayordomo JI, Guerra JM, Guijarro C, et al: Neoplasms of unknown primary site: A clinicopathological study of autopsied patients. Tumori 79:321–324, 1993.

31. Abbruzzese JL, Abbruzzese MC, Hess KR, et al: Unknown primary carcinoma: Natural history and prognostic factors in 657 consecutive patients. J Clin Oncol 12:1272–1280, 1994.

32. Richardson RC, Schoumacher RA, Fer MF, et al: The unrecognized extragonadal germ cell syndrome. Ann Intern Med 94:181–186, 1981.

33. Greco FA, Vaughn WK, Hainsworth JD: Advanced poorly differentiated carcinoma of unknown primary site: Recognition of a treatable syndrome. Ann Intern Med 104:533–547, 1986.

34. Van der Gaast A, Verweij J, Henzen-Logmans SC, et al: Carcinoma of unknown primary: Identification of a treatable subset? Ann Oncol 1:119–122, 1990.

35. Jeremic B, Zivic DJ, Matovic M, et al: Cisplatin and 5-fluorouracil as induction chemotherapy followed by radiation therapy in metastatic squamous cell carcinoma of an unknown primary tumor localized to the neck. A phase II study. J Chemother 5:262–265, 1993.

36. Ellerbroek N, Holmes F, Singletary E, et al: Treatment of patients with isolated axillary nodal metastases from an occult primary carcinoma consistent with breast origin. Cancer 66:1461–1467, 1990.

37. Chang P, Knapper WH: Metastatic melanoma of unknown primary. Cancer 49:1106–1111, 1982.

38. Gershenson DM, Silva EG: Serous ovarian tumors of low malignant potential with peritoneal implants. Cancer 65:578–585, 1990.

39. Strand CM, Grosh WW, Baxter J, et al: Peritoneal carcinomatosis of unknown primary site in women: A distinguishing subset of adenocarcinomas. Ann Intern Med 111:213–217, 1989.

40. Heitz PU: Neuroendocrine tumor markers. Curr Rev Pathol 23:279–306, 1987.

41. Garrow GC, Greco FA, Hainsworth JD: Poorly differentiated neuroendocrine carcinoma of unknown primary tumor site. Semin Oncol 20:287–291, 1993.

42. Moertel CG, Kvols LK, O'Connell MJ, et al: Treatment of neuroendocrine carcinomas with combined etoposide and cisplatin: Evidence of major therapeutic activity in the anaplastic variants of these neoplasms. Cancer 68:227–233, 1991.

43. Hainsworth JD, Greco FA: Poorly differentiated carcinoma and germ cell tumors. Hematol Oncol Clin North Am 5:1223–1231, 1991.

44. Porter AT, McEwan AJ, Powe JE, et al: Results of a randomized phase III trial to evaluate the efficacy of strontium-89 adjuvant to local field external beam irradiation in the management of endocrine resistant metastatic prostate cancer. Int J Radiat Oncol Biol Phys 25:805–813, 1993.

45. Moser RP, Johnson ML, Yung WKA: Clinical presentations of unknown primary tumors: IV. Metastatic brain cancer in patients with no known primary site. Cancer Bulletin 41:173–177, 1989.

46. Hainsworth JD, Dial TW, Greco FA: Curative combination chemotherapy for patients with advanced poorly differentiated carcinoma of unknown primary site. Am J Clin Oncol 11:138–145, 1988.

47. Legha SS, Buzdar AU, Smith TL, et al: Complete remissions in metastatic breast cancer treated with combination drug therapy. Ann Intern Med 91:847–852, 1979.

48. Sporn, JR, Greenberg BR: Empiric chemotherapy in patients with carcinoma of unknown primary site. Am J Med 88:49–55, 1990.

49. Lenzi R, Raber MN, Frost P, et al: Phase II study of cisplatin, 5-FU, and folinic acid in patients with carcinoma of unknown primary origin. Eur J Cancer 29A:1634, 1993.

50. Hainsworth JD, Greco FA: Poorly differentiated carcinoma and poorly differentiated adenocarcinoma of unknown primary tumor site. Semin Oncol 20:279–286, 1993.

51. de Campos ES, Menasce LP, Radford J, et al: Metastatic carcinoma of uncertain primary site: A retrospective review of 57 patients treated with vincristine, doxorubicin, cyclophosphamide (VAC) or VAC alternating with cisplatin and etoposide (VAC/PE). Cancer 73:470–475, 1994.

52. van der Gaast A, Henzen-Logmans SC, Planting AS, et al: Phase II study of oral administration of etoposide for patients with well- and moderately-differentiated adenocarcinomas of unknown primary site. Ann Oncol 4:789–790, 1993.

53. Akerley W, Thomas A, Miller M, et al: Phase II trial of oral etoposide for carcinoma of unknown primary. Proc Am Soc Clin Oncol 13:1386, 1994.