We thank Trivanovic et al for sharing their institutional experience of 83 patients with carcinoma of unknown primary, in which performance status (PS) was a more powerful prognostic factor than the presence of liver metastasis.. We and others investigators have previously demonstrated that PS was a powerful adverse clinical prognostic factor1–3 in the setting of carcinoma of unknown primary site. In our univariate analysis of 370 patients, short survival was related more strongly to PS (P < .0001) than to the presence of liver metastasis (P = .001).1 However, the presence of liver metastasis and low serum albumin levels were the most powerful adverse prognostic factors on multivariate analysis, which led us to develop and publish our new prognostic model. This new prognostic model outperforms the previous prognostic model based on PS and serum lactate dehydrogenase (LDH) levels.3 This improvement in prognostic accuracy is because of the high rate of elevated LDH in patients who are classified as good-risk in our model. Larger prospective studies, including both clinical and biologic parameters, are warranted now to validate our prognostic model.
Although we agree with Trivanovic et al that PS may used to guide chemotherapy treatment decisions, we previously reported that factors other than PS carry greater weight in the decision to use chemotherapy.2