A major goal in cancer medicine is to provide effective, affordable therapy for patients with minimal off target toxicity. Positive and negative predictive biomarkers of therapeutic response are in development with the aim to provide targeted, ‘personalised’ therapy. The thesis focuses on the integration and investigation of predictive biomarkers in lymphoid disease, with a major focus on aggressive B cell non-Hodgkin lymphoma in early phase clinical trials. The thesis highlights the conception, design and results of four trials in aggressive B cell lymphoma. Briefly, the thesis describes: a) the CXD101 trial dose escalation and expansion phase 1 trial of a novel HDAC inhibitor (CXD101) and describes the analysis of the protein biomarker HR23B in this setting; b) the largest published phase II trial in the first line setting in Richter’s syndrome - the CHOP-OR trial with a number of exploratory studies; c) the AZD1152 (aurora kinase B inhibitor) phase II trial and the analysis of c-MYC expression as a potentially predictive biomarker of response; d) the TORCH trial conception, design and set up with the aim to analyse responses of the novel dual mTORC1/2 inhibitor (AZD2014). The trial aims to analyse exploratory outcomes including PET-CT vs. CT responses, soluble tumour DNA as a predictor of relapse, and immunohistochemistry analysis of mT0RC1/2 pathway protein expression and resistance mechanisms. All chapters include published material. Five full publications have resulted to date from the work included in the thesis, alongside 7 national or international conference poster presentations. Full trial results are provided in 3 chapters with the fourth chapter focusing on conception, design and recruitment. The chapters highlight not only the results of the novel agents tested, but attempt to gain additional information about the relative importance, benefits, and challenges in interpretation of specific biomarkers integrated closely within the design of these individual trials.