Combination Chemotherapy for Unknown Primary

Abstract & Commentary

By William B. Ershler, MD, Editor, INOVA Fairfax Hospital Cancer Center, Fairfax, VA; Director, Institute for Advanced Studies in Aging, Washington, DC.

Synopsis: Carcinoma of unknown primary remains a management problem without an established approach demonstrated to prolong survival. In a phase II study, the combination of carboplatin, gemcitabine, and capecitabine was shown to be fairly well tolerated and, for certain subsets, effective in producing transient tumor regression.

Source: Schneider BJ, et al. Cancer. 2007;110:770-775.

Optimal treatment for patients who present with metastatic cancer of unknown primary remains unsettled. Such presentation is not unusual, accounting for approximately 2% of all newly diagnosed cancer.1 Although difficult to establish in individual cases, autopsy series indicate the lung, pancreas and hepatobiliary tree are the most common sites of original disease.2 To date, no treatment regimen has been clearly demonstrated to provide significant prolongation of survival, particularly for those with moderately or well differentiated adenocarcinomas involving the liver.3

Schneider and colleagues report a phase II trial of carboplatin, gemcitabine and capecitabine in 33 patients with unknown primary. Treatment consisted of 21 day cycles. Carboplatin dose was targeted to an area under the curve (AUC) of 5 (mg/ml x min) and administered as a 30 minute intravenous infusion on day 1, just after the gemcitabine dose. The gemcitabine dose is 1000 mg/m2 and given over 30 minutes on days 1 and 8. Capecitabine was taken orally twice daily for 14 days at a total daily dose of 1600 mg/m2. Treatment continued for up to 8 cycles with demonstration of objective response or disease stability.

Over a 4.5 year period (August 2001 through January 2006) 33 patients were enrolled. The median age at study entry was 58 years and no patients had received prior chemotherapy or radiation for this condition. A median of 5 cycles (range 1-8) was administered to each patient. In general the treatment was well tolerated. The most common grade 3-4 toxicities were neutropenia (67%) and thrombocytopenia (48%). All 9 patients with an ECOG performance score of 2 had grade 3/4 hematologic toxicity compared to 10 of 24 patients who were ECOG PS 0-1.

Thirteen of the 33 patients had a partial response to treatment and the median response duration was 3.9 months (1.4-11.2 months). By using intent-to-treat analysis, the median time to treatment failure was 4.5 months (95% confidence interval [CI], 2.8-5.7). Median progression free survival time was 6.2 months (95% CI, 5.4-8.0 months), and at 6 months over half (54.5%; 95% CI, 36.3%-69.6%) were alive and progression free. Median survival was 7.6 months (95% CI, 6.3-14.1 months), and the 1-year and 2-year survival rates were 35.6% (95% CI, 19.7-51.8%) and 14.2% (95% CI, 4.6%-29.1%), respectively.

Serum tumor markers (CEA and CA19-9) were evaluated in a subset. Patients with a > 50% drop in a marker survived longer than those with elevated markers that did not drop (median survival, 19.4 months vs 7.1 month, P = 0.03).


Over the past two decades there has been a significant reduction in the absolute numbers of those diagnosed "unknown primaries." This, no doubt, is the result of better imaging and bolder pathologists. Yet, despite this, there remains less than acceptable treatment success and no single regimen stands ahead of the rest. Among many reasons for this is the likely great heterogeneity in reported series. Perhaps with the rapidly advancing molecular technology, there will be more precise identification of tumor type, greater homogeneity in clinical trials and more specific and directed therapies. Yet, even if we become capable of defining lung, pancreas or biliary tree primary, there will not be much satisfaction until treatments for these tumors yields greater success.

That stated, oncologists faced with an individual with metastatic cancer of unknown primary should attempt to make an educated prediction whether the tumor arose from above or below the diaphragm. If lung primary is suspected, a platinum/taxane regimen might offer greater chance for treatment response. However, if the suspected primary is below, the triplet offered in this report seems as effective as any. For patients with good performance status, it is fairly well tolerated and for those with liver involvement, a group known to have particularly poor prognosis, a response rate of 40% is remarkable.


1. Jemal A, et al. Cancer statistics, 2006. CA Cancer J Clin. 2006;56(2):106-130.

2. Mayordomo JI, et al. Tumori. 1993;79:321-324.

3. Fizazi K. Ann Oncol. 2006;17 Suppl 10:x177-80.