Performance status as prognostic factor in phase III trials of first-line chemotherapy of unresectable or metastatic pancreatic cancer: A trial-level meta-analysis

For patients with unresectable or metastatic pancreatic adenocarcinoma (mPDAC), there are no standardized prognostic and predictive factors beyond performance status (PS). A poor PS, as defined by Eastern Cooperative Oncology Group (ECOG) score of 2 or more, has been related with a detrimental effect of chemotherapy. Therefore, even more trials enrolled patients with good PS. The current analysis aims to evaluate the results of PS as a prognostic factor in phase III trials of patients with mPDAC receiving first-line chemotherapy. A literature search on two databases, from 2000 to 2019, and a further selection of clinical trials were performed by predefined criteria. Twelve phase III studies have been included in the analysis: the trials, enrolling 5619 patients, confirmed the worse prognosis of patients with higher ECOG PS scores (hazard ratio [HR] = 1.45; 95% confidence interval [CI], 1.21-1.74; p-value < 0.001), and a similar trend was evident for patients with an ECOG PS 1 versus 0 (HR = 1.61; 95% CI, 1.43-1.80; p-value < 0.001) in six studies, enrolling 2799 patients. Heterogeneity of trials was high, with I-2 = 91%. Some possible moderators have been suggested, such as the number of drugs in the chemotherapy regimen and the male gender. In conclusion, a low ECOG PS score appears to be related with a longer survival even in trials that excluded patients with an ECOG PS 2 score, but the meta-analyses reported high heterogeneity and publication bias.