Integrated imaging and circulating biomarkers for the assessment of drug-induced hepatotoxicity

Abstract

Drug-induced liver injury (DILI) is a major burden for health care systems and a principal reason for attrition during drug development. Prediction of clinical DILI remains difficult and diagnosis is usually based on exclusion using blood-based biomarkers for cell injury. Currently used chemistry parameters lack sensitivity, specificity and prognostic capability. More importantly, they provide little insight into the underlying mechanistic basis of DILI hindering stratification and appropriate therapeutic intervention. A panel of circulating biomarkers that can identify liver injury in patients with DILI with enhanced sensitivity and specificity over currently used tests and translational to animal models has been previously described. Serum microRNA-122 (miR-122) has shown liver specificity and higher sensitivity than alanine aminotransferase (ALT) and High mobility group box 1 (HMGB1) is an informative and early serum indicator of cell death and inflammatory processes in Paracetamol (APAP)-induced liver injury (APAP-ILI). The mechanistic nature of these biomarkers allows for a deeper understanding of the basis to the toxicological events in DILI scenarios. However, this biomarkers panel is made up of analytes that reflect cell death mode or inflammation processes and lacks novel mechanisms for determining liver function. Multispectral Optoacoustic Tomography (MSOT) is an imaging modality that facilitates the whole-body imaging of toxicological processes in organs that can be observed in vivo, real-time and with high resolution. MSOT imaging operates in the near-infrared (NIR) spectral window and high detection specificity is achieved by multispectral reconstruction which allows accurate monitoring of biodistribution of relevant molecules. It has been used preclinically to assess kidney function in Adriamycin-induced nephropathy mouse model where clearance of the near infrared dye IRDye 800CW carboxylate was assessed. Indocyanine green (ICG) is a fluorescent dye that can also be detected in the system by MSOT imaging. It is removed from circulation exclusively by the liver parenchyma and is clinically used in medical diagnostics as a marker for liver function. Here for the first time, we demonstrate that ICG clearance from blood can be monitored using real-time and non-invasive MSOT imaging. Established (ALT and total bilirubin; TBIL) and novel (miR-122 and HMGB1) serum biomarkers were analysed and histopathological examination for liver injury was assessed in APAP-ILI mouse model. Data showed a significant increase of ICG half-life at 3 hours (h) post-APAP followed by a gradual decrease at 24 h time-point and a second peak of ICG clearance delay at 48 h with full recovery at 96 h post- APAP. The strong correlation between parameters demonstrated the utility of MSOT imaging for the assessment and monitoring of liver function in DILI mouse model. Furthermore, a cohort of mice received APAP antidote, N-Acetyl-Cysteine (NAC), for the evaluation of MSOT imaging utility in reporting the efficacy of therapeutic intervention. We were able to monitor recovery after NAC treatment using non-invasive imaging. This work was then extrapolated to a chronic DILI model where mice received carbon tetrachloride (CCl4) twice per week for up to 12 weeks (w) and had the same parameters than above measured at different timepoints throughout the timecourse. MSOT imaging delivered interesting and novel data regarding liver function in this model. Finally, a cohort of mice received a potential cell therapy and MSOT imaging was used to monitor efficacy of this novel therapy as well as tracking the fate of these cells in the animal. This time MSOT proved the multimodal utility in drug development. Furthermore, the lack of understanding of the mechanisms of release in blood of biomarkers and its relationship with tissue injury was addressed. We showed for the first time the tissue-periphery distribution of miR-122 and HMGB1 in APAP-ILI. Thorough histopathological examination and quantification of miR-122 in situ hybridisation (ISH) and HMGB1 immunohistochemistry (IH) and liver injury severity assessment was performed. This data was then correlated with serum biomarkers levels. Serum miR-122 and HMGB1 elevation directly correlated to high expression of miR-122 and HMGB1 translocation in hepatocytes undergoing cell stress. In summary, the data in this thesis provides a novel approach for the diagnosis of DILI using a novel photoacoustic imaging technology which allows the non-invasive and real-time monitoring of liver function in combination with novel panel of circulating biomarkers for liver injury. Altogether proved to be more sensitive, specific and importantly more informative of the underlying mechanistic basis of DILI.