Comparison of Various Phase I Combination Therapy Designs in Oncology for Evaluation of Early Tumor Shrinkage Using Simulations

Abstract

There is still a lack of efficient designs for identifying the dose response in oncology combination therapies in early clinical trials. The concentration response relationship can be identified using the early tumor shrinkage time course, which has been shown to be a good early response marker of clinical efficacy. The performance of various designs using an exposure–tumor growth inhibition model was explored using simulations. Different combination effects of new drug M and cetuximab (reference therapy) were explored first assuming no effect of M on cetuximab (to investigate the type I error (α)), and subsequently assuming additivity or synergy between cetuximab and M. One‐arm, two‐arm, and four‐arm designs were evaluated. In the one‐arm design, 60 patients received cetuximab + M. In the two‐arm design, 30 patients received cetuximab and 30 received cetuximab + M. In the four‐arm design, in addition to cetuximab and cetuximab + M as standard doses, combination arms with lower doses of cetuximab were evaluated (15 patients/arm). Model‐based predictions or “simulated observations” of early tumor shrinkage at week 8 (ETS8) were compared between the different arms. With the same number of individuals, the one‐arm design showed better statistical power than other designs but led to strong inflation of α in case of misestimated reference for ETS8 value. The two‐arm design protected against this misestimation and, with the same total number of subjects, would provide higher statistical power than a four‐arm design. However, a four‐arm design would be helpful for exploring more doses of cetuximab in combination with M to better understand the interaction.

Relevance of primary lesion location, tumour heterogeneity and genetic mutation demonstrated through tumour growth inhibition and overall survival modelling in metastatic colorectal cancer

“…Our growth rate K gr estimate had the same order of magnitude as the 1 in that work, with both rates being very slow (3.4 Â 10 À3 and 1 Â 10 À3 wk À1 , respectively). Estimates of R Cetu parameters seem different (5.8 Â 10 À2 and 0.2 wk À1 ), but this is due to the fact that, in the Seurat et al 36 study, the development of resistance for the 3 distinct treatments was described with a shared parameter while in our model the tolerance induced by each 1 of the treatments could be characterized. Another limitation was the worse model fit achieved for the CRYSTAL study.…”

“…In a recent work a TGI model was used to perform simulations of phase I therapy designs. 36 The value of the area under the curve calculated over 2 weeks was used as the driver of cetuximab-related tumour shrinkage. Noteworthy, the K Cetu value estimated in that work was 2.5 Â 10 À4 L mg À1 wk À1 , which was similar to our K Cetu value equal to 2.3 Â 10 À4 L mg À1 wk À1 .…”

“…It was not considered to build a validation dataset using the smaller studies due to unique characteristics present in some of these studies, such as covariate values (e.g. all patients in the APEC study are Asian) or different dropout patterns.The processes of tumour cell proliferation, treatment induced tumour shrinkage, and resistance to treatment effects are represented semi-mechanistically in the model for TGI with a structure very similar to that used in other evaluations of tumour size (including a recent evaluation of cetuximab effects in phase I studies36 ) and derived from the seminal work from Claret et al20 A DO model was developed together with the TGI model to further improve model performance and to describe patients leaving the study due to disease progression. One remarkable result obtained from the modelling effort presented here is the topological and quantitative characterization of two relevant individual tumour characteristics, the KRAS mutation status and the primary lesion location.…”