Abstract
Developing prognostic tools is a fundamental component in the management of patients with advanced cancer. The clinical course of head and neck cancer varies widely, and no prognostic tools are available for head and neck cancer patients receiving palliative care. Therefore, there is a need to establish a predictive prognostic tool for head and neck cancer patients under palliative care.
We enrolled 47 patients with pathologically proven head and neck cancer receiving palliative care at Kagawa university hospital from April 2013 to March 2019 to investigate the feasibility of using the palliative prognostic index (PPI) and Glasgow prognostic score (GPS) as predictors of the life expectancy of these patients. The patient demographic data and the indices needed to calculate the PPI and GPS were collected retrospectively from the electronic medical records.
The median age of the patients was 68 y.o. (range, 54-93 y.o.), and 40 patients were male. Patients were categorized into subgroups by the PPI (18 cases in Group A, 10 cases in Group B and 19 cases in Group C) and GPS (1 case with Score 0, 10 cases with Score 1 and 36 cases with Score 2). There were significant differences in the survival among patients categorized into Group AB and Group C according to the PPI (median survival 65 days (IQR, 34-133 days), 20 days (IQR, 11-30 days) in the patients categorized into Group AB, and C, respectively). There were also significant differences in the survival among the patients with scores of 0-1 and 2 on the GPS (median survival 105 days (IQR, 57-152 days), 26 days (IQR, 17-47 days) in patients with score 0-1 and 2). The sensitivity, specificity, positive predictive value, and negative predictive value for 14-day survival were 88%, 69%, 37%, 96%, respectively, for PPI, and 100%, 28%, 22%, and 100%, respectively, for GPS. The corresponding results for 30-day survival were 67%, 81%, 74%, and 75%, respectively, for PPI, and 95%, 39%, 56%, and 91%, respectively, for GPS.
This is the first report to evaluate the feasibility of using PPI and GPS as prognostic tools to predict the survival of head and neck squamous cell cancer patients under palliative care. Our results suggest that both PPI and GPS have the potential to be useful as prognostic tools to predict the life expectancy in patients with head and neck cancer receiving palliative care.
