Promising new cancer drugs frequently fail to demonstrate clinical benefit over approved therapies, despite reporting good efficacy in preclinical development. Existing literature highlights the need to raise experimental standards to improve reproducibility and clinical translation, as well as encouraging the selection of more relevant preclinical models. Less attention, however, has been focused on the selection of relevant, translational endpoints in preclinical development. In the clinical setting, treatments are frequently evaluated by assessing disease progression and/or survival. In contrast, in preclinical research, efficacy typically focuses on evaluating tumour-growth inhibition (TGI) following a defined treatment period. We have an opportunity to better align the preclinical metrics to those used in the clinic, thus enhancing the utility of preclinical datasets in predicting clinical outcomes.

We developed a new approach for objectively evaluating tumour regrowth in mice, assessment of durability, that more closely aligns with the current predominant clinical endpoint, progression-free survival. Mathematically, the method (INTRA - in vivo tumour regrowth analysis) is a spline-based approach for statistical detection of the day of regrowth following treatment withdrawal. We have validated the method by applying it to studies evaluating the efficacy of Tyrosine Kinase Inhibitors (TKIs) across a panel of murine patient-derived xenograft (PDX) models of non-small cell lung cancer (NSCLC).

Application of the approach to studies of Tyrosine Kinase Inhibitors targeting the Epidermal Growth Factor Receptor finds that the durability measure recapitulates multiple features of clinical outcomes that were not detected by TGI. Specifically, evaluating durability differentiates multiple generations of EGFR TKIs, closely mirroring clinical response. Additionally, durability response following Osimertinib treatment is affected by EGFR mutational status in line with patient outcomes.

We have developed a robust statistical method for evaluating durability of regression for xenograft regrowth studies. We have applied the method to several studies evaluating the efficacy of EGFR TKIs and demonstrate that preclinical outcomes recapitulate clinical findings. Our results underscore the utility of this approach in enhancing the predictive power and translational potential of preclinical development.