Background: head and neck cancer patients experience increased risk for depression and compromised quality of life. Pre-emptively identifying patients at risk for depression during head and neck cancer therapy can help establish targeted interventions.
Methods: Data analyses from a randomized trial examining prophylaxis with antidepressant escitalopram oxalate, in 125 patients diagnosed with new or recurrent stage II-IV head and neck epidermoid carcinoma, was aimed at identifying predictors of depression development. Depression was measured using the Quick Inventory of Depressive Symptomatology-Self Rated (QIDS-SR). We analyzed data from the baseline visit to determine if any of the variables were associated with development of moderate or greater depression (primary endpoint of the study).
Data were stratified by treatment intervention to determine the predictive cutoff values for development of depression in both the escitalopram and placebo groups using QIDS-SR. Continuous variables were analyzed using t-tests and categorical variables were analyzed using Fisher’s exact test or Chi-square test. Multivariable receiver operating characteristic (ROC) analyses were used to identify the optimum cutoff points for depression development. All tests were two-sided with p-value set at 0.05 for statistical significance.
Results: The mean baseline QIDS-SR scores of those who developed moderate or greater depression over the study period were significantly higher than the baseline scores of those individuals who did not develop depression over the study period (6.2 vs. 4.2, respectively). No subject who developed depression had a baseline score of less than 2, and the majority of the depressed group had baseline scores of 4 or greater. Independent of randomization to treatment arm, for the overall patient cohort, a baseline score of 2 or higher correctly identified 100% of patients who eventually developed moderate or severe depression during the course of their treatment for head and neck carcinoma. Negative predictive value (NPV) at this level is 100 percent, while the positive predictive value (PPV) is 19 percent.
Development of moderate or severe depression was also associated with type of initial treatment (radiation). In the placebo group, 12 of 31 patients (39%) who received radiotherapy as part of their initial definitive therapy, developed depression during the study period, compared to only 4 out of 34 (12%) who did not receive radiation as part of their initial treatment strategy (p=0.020). In the absence of other confounders, particularly use of the trial drug escitalopram, these findings suggest association between development of depression and use of radiotherapy in these patients.
Conclusions: Baseline symptoms (QIDS-SR) and initial radiation-based therapy may predict depression in patients receiving treatment for head and neck cancer. QIDS-SR baseline scores of 2 or greater served well to identify patients who may benefit the most from pharmacologic prophylaxis of depression.