Choice of baseline hazards in joint modeling of longitudinal and time-to-event cancer survival data.

Longitudinal time-to-event analysis is a statistical method to analyze data where covariates are measured repeatedly. In survival studies, the risk for an event is estimated using Cox-proportional hazard model or extended Cox-model for exogenous time-dependent covariates. However, these models are inappropriate for endogenous time-dependent covariates like longitudinally measured biomarkers, Carcinoembryonic Antigen (CEA). Joint models that can simultaneously model the longitudinal covariates and time-to-event data have been proposed as an alternative. The present study highlights the importance of choosing the baseline hazards to get more accurate risk estimation. The study used colon cancer patient data to illustrate and compare four different joint models which differs based on the choice of baseline hazards [piecewise-constant Gauss-Hermite (GH), piecewise-constant pseudo-adaptive GH, Weibull Accelerated Failure time model with GH & B-spline GH]. We conducted simulation study to assess the model consistency with varying sample size (N = 100, 250, 500) and censoring (20 %, 50 %, 70 %) proportions. In colon cancer patient data, based on Akaike information criteria (AIC) and Bayesian information criteria (BIC), piecewise-constant pseudo-adaptive GH was found to be the best fitted model. Despite differences in model fit, the hazards obtained from the four models were similar. The study identified composite stage as a prognostic factor for time-to-event and the longitudinal outcome, CEA as a dynamic predictor for overall survival in colon cancer patients. Based on the simulation study Piecewise-PH-aGH was found to be the best model with least AIC and BIC values, and highest coverage probability(CP). While the Bias, and RMSE for all the models showed a competitive performance. However, Piecewise-PH-aGH has shown least bias and RMSE in most of the combinations and has taken the shortest computation time, which shows its computational efficiency. This study is the first of its kind to discuss on the choice of baseline hazards.

Sociodemographic Factors and Stage of Cancer at Diagnosis: A Population-Based Study in South India.

PURPOSE: Lower socioeconomic status is associated with inferior cancer survival in high-income countries, but whether this applies to low- and middle-income countries is not well described. Here, we use a population-based cancer registry to explore the association between educational level and stage of cancer at diagnosis in South India. METHODS: We used the Trivandrum District population-based cancer registry to identify all cases of breast and cervical cancer (women) and oral cavity (OC) and lung cancer (men) who were diagnosed from 2012 to 2014. Educational status-classified as illiterate/primary school, middle school, or secondary school or higher-was the primary exposure of interest. Primary outcome was the proportion of patients with advanced stage disease at diagnosis defined as stage III and IV (breast, cervix, or OC) or regional/metastatic (lung). RESULTS: The study population included 4,547 patients with breast (n = 2,283), cervix (n = 481), OC (n = 797), and lung (n = 986) cancer. Educational status was 22%, 19%, and 26% for illiterate/primary, middle, and secondary school or higher, respectively. Educational status was missing for 33% of patients. The proportion of all patients with advanced stage disease was 37% (breast), 39% (cervix), 67% (OC), and 88% (lung). Patients with illiterate/primary school educational status were considerably more likely to have advanced breast cancer (50% v 39% v 36%; P < .001), cervix cancer (46% v 43% v 24%; P = .002), and OC cancer (77% v 76% v 59%; P < .001) compared with patients with higher educational levels. The proportion of patients with advanced lung cancer did not vary across educational levels (89% v 84% v 88%; P = .350). CONCLUSION: A substantial proportion of patients in South India have advanced cancer at the time of diagnosis. This is particularly true among those with the lowest levels of education. Future health awareness and preventive interventions must target less-educated communities to reduce delays in seeking medical care for cancer.