Joint modeling of longitudinal outcomes and survival using latent growth modeling approach in a mesothelioma trial

Abstract

Joint modeling of longitudinal and survival data can provide more efficient and less biased estimates of treatment effects through accounting for the associations between these two data types. Sponsors of oncology clinical trials routinely and increasingly include patient-reported outcome (PRO) instruments to evaluate the effect of treatment on symptoms, functioning, and quality of life. Known publications of these trials typically do not include jointly modeled analyses and results. We formulated several joint models based on a latent growth model for longitudinal PRO data and a Cox proportional hazards model for survival data. The longitudinal and survival components were linked through either a latent growth trajectory or shared random effects. We applied these models to data from a randomized phase III oncology clinical trial in mesothelioma. We compared the results derived under different model specifications and showed that the use of joint modeling may result in improved estimates of the overall treatment effect.

Fig. 1

Path diagrams of joint models trajq_s and remq_s. Trt treatment; I intercept (b _0i); S slope (b _1i); Q quadratic coefficient (b _2i);  = η(t _k)

Fig. 2

Fitted population-level quadratic growth curve of LCSS scores in the remq_s model. The cisplatin group is in gray, and the pemetrexed/cisplatin group is in black

Fig. 3

Parameter estimates and 95 % confidence intervals for effects of covariates on random intercepts, random slopes, and TTPD in the remq_s model for ASBI5 with covariates. Confidence intervals that exclude the null value of 0 are in black