A New Multistate Transition Model For Effect Estimation In Randomized Trials With Treatment Switching And A Cured Subgroup

Background: Many methods, including multistate models, have been proposed in the literature to estimate the treatment effect on overall survival in randomized trials with treatment switching permit after the disease progression. Nevertheless, the cured fraction of patients has not been considered. The cured would never experience the progressive disease, but they may suffer death with a hazard comparable to that of people without the disease. With the mix of the cured subgroup, existing methods yield highly biased effect estimation and fail to reflect the truth in uncured patients. Methods: In this paper, we propose a new multistate transition model to incorporate the cure, progression, treatment switching, and death states during trials. In the proposed model, the probability of cure and the death hazard of the cured are modeled separately. For the not cured patients, the semi-competing risks model is used with the treatment effect evaluated via transitional hazards between states. The particle swarm optimization algorithm is adopted to estimate the model parameters. Results: Extensive simulation studies have been conducted to evaluate the performance of the proposed multistate model and compare it with existing treatment switching adjustment methods. Results show that in all scenarios, the treatment effect estimation of the proposed model is more accurate than that of existing treatment switching adjustment methods. Besides, the application to diffuse large B-cell lymphoma data has also illustrated the superiority of the proposed model.Conclusions: The superiority and robustness of the proposed multistate transition model qualify it to estimate the treatment effect in trials with the treatment switching permit after progression and a cured subgroup.

Improving the causal treatment effect estimation with propensity scores by the bootstrap

When observational studies are used to establish the causal effects of treatments, the estimated effect is affected by treatment selection bias. The inverse propensity score weight (IPSW) is often used to deal with such bias. However, IPSW requires strong assumptions whose misspecifications and strategies to correct the misspecifications were rarely studied. We present a bootstrap bias correction of IPSW (BC-IPSW) to improve the performance of propensity score in dealing with treatment selection bias in the presence of failure to the ignorability and overlap assumptions. The approach was motivated by a real observational study to explore the potential of anticoagulant treatment for reducing mortality in patients with end-stage renal disease. The benefit of the treatment to enhance survival was demonstrated; the suggested BC-IPSW method indicated a statistically significant reduction in mortality for patients receiving the treatment. Using extensive simulations, we show that BC-IPSW substantially reduced the bias due to the misspecification of the ignorability and overlap assumptions. Further, we showed that IPSW is still useful to account for the lack of treatment randomization, but its advantages are stringently linked to the satisfaction of ignorability, indicating that the existence of relevant though unmeasured or unused covariates can worsen the selection bias.