Provincial Health Services Authority (PHSA) improves the health of British Columbians by seeking province-wide solutions to specialized health care needs in collaboration with BC health authorities and other partners.
Updated: 3 August 2005
Patients with metastatic cancer of unknown primary site (primary unknown) have a histological or cytologically documented cancer, but no primary site identified after careful clinical history, full physical examination (including pelvic and rectal exam) and chest radiograph. The management of metastatic cancer when the site of origin of the malignancy is unknown requires a systematic approach.
Metastatic cancer with unknown primary site accounts for 2.5% of referrals to BC Cancer Agency. Incidence rises with age and has no clear-cut predominance by gender. The median survival is about six months and the proportion of patients alive at 1 year is 10-20%. A small proportion (about 5%) are alive at 5 years. Post-mortem examination reveals lung or pancreas primaries in about half of those patients in whom a primary tumour is ultimately identified.
Careful pathological assessment is crucial and specific tumour types should be referred to the appropriate tumour groups (i.e. sarcoma, melanoma). Male patients with midline masses (germ cell cancers) are seen by the Genitourinary Tumour Group. Squamous carcinomas of the cervical lymph nodes need review by the Head and Neck Tumour Group. However, the majority of tumours are adenocarcinomas or undifferentiated carcinomas.
Triage of patients to Radiation or Medical Oncology will depend on the symptoms needing palliation and the site(s) involved. For example, patients with brain metastases or painful bone metastases are usually treated by a radiation oncologist and patients with liver, lung or soft tissue disease are managed by a medical oncologist. A specific clinic or conference for tumours of unknown primary does not exist at this time, but difficult cases are discussed in a multi-disciplinary fashion by the appropriate specialties.
Updated: 18 August 2005
2.1 Clinico-pathologic Considerations
With the above in mind, all patients should have a careful history taken and thorough physical examination performed. The latter should include examination of all node-bearing areas, rectal examination with examination of the prostate and testes in males and a breast and pelvic examination in females.
Most patients should have a panel of relevant tumour markers performed to include CEA, ß-HCG, alphafetoprotein (AFP), prostatic specific antigen (PSA) in males, CA-125 and CA-15-3 in females (looking for gynecological and breast primary tumours respectively). All patients should have a chest X-ray as well as a complete blood count, renal and liver function tests. If breast cancer is a possibility, mammograms should be done. Abdominal and pelvic CT scan is a good screen for intra-abdominal and retroperitoneal primary tumours when these are felt to be likely and when such a finding may assist management. Bone scan with X-rays of areas of positive uptake is useful if the patient has bone pain. Fecal occult bloods should be performed for patients with symptoms or intra-abdominal adenocarcinomas. Other special studies should be done only if suggested by specific symptoms or clinical situations.
2.2 Diagnostic Pathology – Biopsy Considerations
When there is a need for pathologic assessment it is important that the physician decides upon the type of information required prior to biopsy. A variety of biopsy techniques are available including needle aspiration cytology, core biopsy or open biopsy. The choice of technique depends upon the clinical problem and differential diagnosis as well as the need for additional studies such as analysis of estrogen and progesterone receptors. Fresh, unfixed tissue is required for estrogen and progesterone receptor studies and is also preferred for analysis of DNA status, oncogene rearrangement and chromosomal studies. Fresh material and special fixatives are required for analysis of malignant lymphomas and are of particular importance in small cell, undifferentiated malignant tumours in children.
Core biopsy or open biopsy are the current standard techniques for obtaining tissue for pathological assessment. Preferably the material should be sent fresh to the laboratory with a brief outline of the clinical problem. This is particularly important when the biopsy is removing a solitary metastasis and no further tissue would be readily available. Submission of a small fragment of tissue for frozen section may be of assistance in confirming the presence of tumour. However, it is not advised to submit the tissue only for frozen section as this distorts morphology and may complicate interpretation. Fine needle aspiration cytology can also be very useful in confirming the diagnosis of malignancy as well as providing material for special stains or DNA analysis.
2.3 Tumour Markers and Pathology
While tumour markers may be performed on tissue, the immunological stains used to analyze these markers are not always reliable. In many instances the measurement of tumour markers on the patient's serum are more useful. As these markers may disappear from the patient's blood within a short time after removal of the tumour, it is important to obtain a blood sample prior to surgery in cases in which the tissue removed represents the major portion of the bulk of the patient's disease. The serum can be frozen and processed later as needed. Post-operative serum tumour markers may also be helpful, if elevated pre-op.
A careful pathologic assessment of the tumour can be helpful in narrowing the differential diagnosis. In some instances the likely site of origin can be suggested by the pathologist and may direct further diagnostic procedures. For example, the pathologic findings may point in the direction of an infradiaphragmatic primary lesion and suggest the need for an abdominal CT scan. Another example is nodal disease in the neck, which pathologically may be interpreted as suggesting nasopharyngeal carcinoma again suggesting diagnostic procedures directed to that site. New immunohistochemical stains may assist in pinpointing the primary origin: TTF-1; CSX-2; Hepar-1; GCDFP-16, etc. Newer molecular markers could be considered in some cases where molecular therapy is being contemplated: He-2neu, c-kit, EGFR.
In many instances the pathology interpretation carried out at the referring hospital is straightforward. However, particularly when the pathology has been interpreted as showing "malignancy - subtype undetermined," "carcinoma of unknown type," or "poorly differentiated or anaplastic carcinoma," then a pathology review should be carried out at the BC Cancer Agency. Advances in immunocytochemical techniques may make it possible to narrow down tumour sites quite considerably. When requesting a pathology review at BCCA, it is preferable to send a representative paraffin block from the biopsy together with the slides. Occasionally, it will be possible to reclassify poorly differentiated lesions as small cell anaplastic carcinoma, lymphoma or germ cell tumours, which would dictate a different treatment approach.
Reviewed: 3 August 2005
There is no accepted formal staging system for these patients.
Pre-treatment staging investigations are individualized for each patient and will take into account the patient's general condition, fitness for treatment, and the pathology of the biopsy material.
Reviewed: 3 August 2005
Refer a patient
Those patients who present with an apparent cancer of unknown primary, but in whom a definite diagnosis is eventually made, should receive appropriate specific treatment. Their prognosis will depend on the diagnosis and the availability of effective specific treatment and they will not be included in the rest of the discussion.
4.1 Special Situations
Some patients have clusters of findings suggesting a possible primary site and indicating a potentially fruitful approach to treatment.
4. 2 Other Primary Unknown Cancer
Treatment should be carefully selected and need not necessarily include systemic anti-neoplastics to be appropriate. The clinician should consider whether localized problems such as pain, obstruction, bleeding, cough or skin erosion might be managed with localized treatment such as irradiation or surgery. Radical radiation therapy with curative intent is rarely given except in the case of metastatic squamous cell carcinoma confined to the neck nodes or inguinal nodes. Depending on the overall condition of the patient, a course of palliative radiotherapy may vary from a single fraction to a two week course with longer courses of fractionated treatment given under special circumstances. Patients presenting with emergency conditions such as spinal cord compression should be referred immediately upon recognition of the diagnosis for assessment and treatment on an emergency basis.
If empiric chemotherapy is considered, the clinician should first decide whether it is wise for a given individual. Patients with major co-morbid cardiovascular, metabolic or other diseases which are actively symptomatic may well deteriorate more quickly under the added stress of chemotherapy. Patients with poor performance status (bed or chair-ridden most of the day) are unlikely to respond to systemic chemotherapy. Cytotoxic agents should be reserved for those patients who are sufficiently young and fit to permit a reasonable judgment that the unavoidable toxicities of such agents are worth enduring for a 20%-30% chance of a short-lived remission. No available test or assay can better identify the patient likely to profit from chemotherapy than the clinician's thoughtful appraisal.
Many chemotherapy programs have been described for the empiric treatment of patients with cancers of unknown primary site. In phase II clinical trials response rates of 0-50% have been reported. Limited phase III data is available Monotherapy with mitomycin at a dose of 15 mg/m² for two doses only given one month apart with re-assessment for evidence of response is appropriate for patients over 60 years of age or unable to tolerate cis-Platin based chemotherapy. Cis-Platin and etoposide combinations are likely appropriate for younger patients with good performance status. Clinical trials may be available. Consideration can be give to the use of a Taxane. Further treatment at a later date can be considered for the minority of patients that truly benefit. Second line chemotherapy is ineffective and unjustifiable unless an extraordinary response occurred after primary treatment.
Once therapy has been completed patients will return to their referring physicians for continued follow-up. Most patients should be seen on an intermittent basis to check for symptoms or findings of obvious progressive disease. Routine follow-up investigations would generally not be required unless the patient becomes symptomatic since the overall approach is palliative in nature.
BC Cancer Agency. All Rights Reserved.