Developing the POTOMAC Model: A Novel Prediction Model to Study the Impact of Lymphopenia Kinetics on Survival Outcomes in Head and Neck Cancer Via an Ensemble Tree-Based Machine Learning Approach.

PURPOSE: Lymphopenia is associated with poor survival outcomes in head and neck squamous cell carcinoma (HNSCC), yet there is no consensus on whether we should limit lymphopenia risks during treatment. To fully elucidate the prognostic role of baseline versus treatment-related lymphopenia, a robust analysis is necessary to investigate the relative importance of various lymphopenia metrics (LMs) in predicting survival outcomes. METHODS: In this prospective cohort study, 363 patients were eligible for analysis (patients with newly diagnosed, nonmetastatic HNSCC treated with neck radiation with or without chemotherapy in 2015-2019). Data were acquired on 28 covariates: seven baseline, five disease, seven treatment, and nine LMs, including static and time-varying features for absolute lymphocyte count (ALC), neutrophil-to-lymphocyte ratio, and immature granulocytes (IGs). IGs were included, given their hypothesized role in inhibiting lymphocyte function. Overall, there were 4.0% missing data. Median follow-up was 2.9 years. We developed a model (POTOMAC) to predict survival outcomes using a random survival forest (RSF) procedure. RSF uses an ensemble approach to reduce the risk of overfitting and provides internal validation of the model using data that are not used in model development. The ability to predict survival risk was assessed using the AUC for the predicted risk score. RESULTS: POTOMAC predicted 2-year survival with AUCs at 0.78 for overall survival (primary end point) and 0.73 for progression-free survival (secondary end point). Top modifiable risk factors included radiation dose and max ALC decrease. Top baseline risk factors included age, Charlson Comorbidity Index, Karnofsky Performance Score, and baseline IGs. Top-ranking LMs had superior prognostic performance when compared with human papillomavirus status, chemotherapy type, and dose (up to 2, 8, and 65 times higher in variable importance score). CONCLUSION: POTOMAC provides important insights into potential approaches to reduce mortality in patients with HNSCC treated by chemoradiation but needs to be validated in future studies.

Social desirability trait influences on self-reported dietary measures among diverse participants in a multicenter multiple risk factor trial.

Data collected at 4 Behavioral Change Consortium sites were used to assess social desirability bias in self-reports derived from a dietary fat screener (PFat), a dietary fruit and vegetable screener (FVS), and a 1-item question on fruit and vegetable intake. Comparisons were made with mean intakes derived from up to 3 24-h recall interviews at baseline and follow-up (at 12 mo in 3 sites, 6 mo in the fourth). A social-desirability-related underestimate in fat intake on the PFat relative to the 24HR (percentage energy as fat) was evident in women [baseline b = -0.56 (P = 0.005); follow-up b = -0.62 (P < 0.001)]. There was an overestimate in FVS-derived fruit and vegetable consumption (servings/week) in men enrolled in any intervention at follow-up (b = 0.39, P = 0.05) vs. baseline (b = 0.04, P = 0.75). The 1-item fruit and vegetable question was associated with an overestimate at baseline in men according to SD score (b = 0.14, P = 0.02), especially men with less than college education (b = 0.23, P = 0.01). Women with less than college education expressed a similar bias at follow-up (b = 0.13, P = 0.02). Differences in the magnitude of bias according to gender, type of instrument used, and randomization condition are comparable to what has been seen for other instruments and have important implications for both measuring change in studies of diet and health outcomes and for developing methods to control for such biases.