Tumour size, new lesions and other markers as longitudinal endpoints in oncology studies

Abstract

Cancer is a complex disease causing millions of deaths worldwide every year. To date, despite the increasing knowledge in cancer biology, risk factors and important discoveries in cancer treatments, numerous cancers cannot be treated due to a lack of effective therapies. Drug development is a long and complex process including a succession of pre-clinical and clinical phases evaluating drug efficacy and safety. In oncology clinical trials, overall survival (OS) remains the gold standard to assess treatment efficacy. However, OS necessitates the enrolment of a large number of patients in trials and a long follow-up period in order to establish drug-related clinical benefit with a significant statistical power. Therefore, regulatory agencies allow the use of surrogate endpoints for OS that offer an opportunity to assess drug-related patient benefit at an earlier stage, support go-no go decisions to the next phase and help to accelerate a new drug approval. The majority of surrogates for OS are categorical endpoints based on tumour size (TS) imaging that allow a qualitative evaluation of drug efficacy. Lately, the use of TS measurements as a longitudinal variable has been suggested to allow a quantitative evaluation of drug efficacy using modelling approaches. However, TS measurements have been observed to be subject to inter-operator variability, which could lead to misinterpretation of clinical outputs. Therefore, the first objective of this thesis was to quantify TS measurement variability in a non-small cell lung cancer Phase III clinical trial and to assess to what extent it could influence the evaluation of drug efficacy using tumour growth inhibition models. This work revealed high TS measurement variability that did not seem to affect the evaluation of drug effect at a population level using modelling approaches. In order to determine clinical response-to-treatment, it is necessary to define TS burden at baseline for each patient based on existing individual lesion measurements. Guidelines, such as the Response Evaluation Criteria in Solid Tumours (RECIST) criteria, provide a recommendation for the number of lesions to be followed. However, there is limited knowledge about the impact of target lesion selection on drug effect evaluation using tumour growth inhibition (TGI) models in malignant pleural mesothelioma patients. Therefore, the second objective of this thesis was to understand to what extent the number of lesions to be selected to follow tumour size burden over the treatment time course could influence the modelling outcomes in patients with malignant pleural mesothelioma. In this analysis, reducing the number of target lesion did not seem to compromise the determination of drug effect using TGI models. Being able to characterise drug efficacy at an early stage during oncology clinical trials is of primary necessity for patients and drug developers. Therefore, with a better understanding of the TS variable, the thesis was refocused on exploring the use of TS metrics as a surrogate for overall survival, to inform drug-related clinical benefit at an earlier stage after treatment initiation in malignant pleural mesothelioma patients. In this work, two TS-related surrogates for overall survival were identified. Chemotherapy treatments are known to have high interindividual variability in pharmacokinetics and haematological toxicity. However, therapeutic drug monitoring is rarely used in clinical practice. Thus, identifying a routinely measured biomarker that could be used to perform dose optimisation in oncology clinical practice would be of great interest. Therefore, the last objective of this thesis was to develop a modelling framework to adapt docetaxel dose at patient level using a dose-limiting toxicity biomarker, neutrophil counts, that is routinely collected in clinical practice. This analysis highlighted a neutrophil metric could be used to individualise dosing of chemotherapy with dose-limiting neutropenia.

Date of Award	1 Aug 2021
Original language	English
Awarding Institution	The University of Manchester
Supervisor	Hitesh Mistry (Supervisor), Leon Aarons (Supervisor) & Kayode Ogungbenro (Supervisor)