W e read with interest the results from the study of serum albu- min and liver metastasis in patients with carcinomas of unknown primary site by Seve et al. 1 We can present some results from our unpublished study in patients with cancer of unknown primary site in which we explored a set of possible independent prognostic factors, which include novel factors (not presented here) and some known factors like lactate dehydrogenase (LDH) levels, liver metastasis, and performance status.
Our prospective study included 83 patients who fulfilled the criteria for a diagnosis of cancer of unknown primary site. The study was done with statistic analyses using both univariate and multivariate Cox proportional-hazards regression models. Patients with and without liver metastasis survived for 113 days and 331 days, respectively, although our results barely reached the level of significance.
We hypothesized that Eastern Cooperative Oncology Group (ECOG) performance status was the more powerful prognostic factor. This was confirmed by our results. Patients who had an ECOG performance status 0 or 1 had median survival of 400 days compared with patients who had a poor performance status (ECOG 2), who had a median survival of 118 days (hazard ratio, 0.39; 95% confidence interval, 0.21-0.72; P ΒΌ .003). When they were incorporated separately into the model, serum liver metastasis, LDH levels, and albumin levels not reached the level of significance. We can conclude that, although the presence of liver metastasis at the time of diagnosis is a poor prognostic factor, performance status is a more powerful predictor and remains the cornerstone for deciding whether patients who have carcinomas of unknown primary site should receive symptomatic therapy only or should be entered onto indicated chemotherapy protocols.