Preclinical Biomarkers

Measurements of motor and sensory nerve conduction velocities and action potential amplitudes have been proposed as sensitive preclinical biomarkers of peripheral neuropathy in workers repeatedly exposed to /7-hexane (ATSDR 1999b), but this effect is specific to //-hexane (and perhaps a few other aliphatic hydrocarbons in the EC5-EC8 fraction) among petroleum hydrocarbons. 

Current clinical and preclinical biomarkers suffer from lack of target organ specificity, sensitivity issues, and poor mechanistic insight. There is therefore a lot of interest in the field of biomarker discovery to overcome these issues, for both clinical and preclinical applications. In this chapter we set out to demonstrate how in vitro techniques are an indispensable tool in the development and discovery of novel mechanistically based biomarkers. We provide examples of several novel biomarkers which have been either discovered in vitro or where such systems have been used to elucidate key mechanistic information. Many of these biomarkers are more than innocent bystanders leaked into the surrounding tissue, with most highly implicated in cell and tissue survival as well as tissue differentiation. These new biomarkers will be not only useful for current preclinical and clinical applications but also advantageous in the development of better in vitro systems in order to reduce or replace animal testing. 

Kidney injury molecule-1 outperforms traditional biomarkers of kidney injury in preclinical biomarker qualification studies. Nat. Biotechnol. 28(5) 478. 

BEST PRACTICES IN PRECLINICAL BIOMARKER SAMPLE COLLECTIONS... 

Vaidya VS, Ozer JS, Dieterle F, CoUings FB, Ramirez V, Troth S, Muniappa N, Thudium D, Gerhold D, Holder DJ, Bobadilla NA, Marrer E, Perentes E, Cordier A, Vonderscher J, Maurer G, Goering PL, Sistare FD, Bonventre JV. Kidney injury molecule-1 outperforms traditional biomarkers of kidney injury in preclinical biomarker qualification studies. Nat Biotechnol. 2010. 28(5) 478-85. 

The resulting device has demonstrated both FDG and FLT labelling at yields of 98% and 90%, respectively, in 100 s compared to typical macro-scale labelling of 65% in 45 min for FDG and 30% in 90 min for FLT. The use of acetonitrile, DMSO and HC1 have shown no degrading effect on the system. Extremely efficient labelling illustrates the effectiveness of flow-based micro-reactors for PET biomarker synthesis. Multiple biomarkers can be produced in 1-2 min, while using only micro-litres of precursor and can revolutionise the production of radiotracers. Small reaction volumes, improved yields, and the ability to synthesise small quantities of a variety of new compounds will allow preclinical and clinical evaluation of new PET agents with potential for clinical utilisation. 

Many biopharmaceutical preparations are heterogeneous and may be difficult to fully characterise. Certain fractions of a preparation may have different biological activity or kinetics than the intended product. It is important that such fractions are appropriately qualified. The proportions of these fractions may be altered when production changes are made or they may be different between similar products produced by different manufacturers. Because of their proteinaceous nature and their novel mechanisms of action, all preclinical and clinical development steps must be re-evaluated. For pharmacokinetic studies, blood concentrations should be measured by specific analytical techniques (most often ELISA), which quantify the active protein and not one of its fragments or inactive forms, such as antigen-antibody complexes. For PK-PD studies of monoclonal antibodies, relevant biomarkers are most often circu-... 

No information is available concerning the effects of 2,3-benzofuran in humans. Acute oral exposure to 2,3-benzofuran has been shown to alter levels of enzyme activity in the livers of female mice (Heine et al. 1986), but much more work would need to be done to determine whether there is a pattern of enzyme alteration specific to 2,3-benzofuran exposure. Other effects found in animals following oral exposure to 2,3-benzofuran are kidney and liver damage and kidney, lung, liver, and stomach cancer (see Section 2.2.2). Such generalized responses do not suggest the basis for any specific biomarker of clinical or preclinical effects caused by 2,3-benzofuran. 

This is not to say that cardiotoxicity is not seen with biopharmaceuticals. Cardiomyopathy is now a well-recognized complication of trastuzumab and and has been reported with bevacizumab treatment, in particular in combination with other cytotoxic cancer therapies [20]. Myocarditis and pericarditis are a well-documented complications of vaccinia immunization [21], and could also complicate use of a pox-virus vector for other therapeutics. In 1995 Genetics Institute suspended phase 2 cancer trials of Interleukin-12 for serious tox-icities including cardiac arrhythmia. However, such toxicities are best detected by incorporation of biomarkers for myocardial damage such as troponin-T into preclinical and early clinical studies, and continual ECG monitoring for arrhythmia in preclinical and early clinical studies, not by in vitro explorations of electrophysiology. 

While the use of preclinical disease models offers many advantages, such as the potential for identification of biomarkers for clinical trials and a direct estimation of the therapeutic index, there are also disadvantages to this approach. The disadvantages can include a paucity of historical/baseline data and inherent variability of the model, the technical feasibility of using a particular disease model, the onset/severity of the injury in the animals as this may be different from humans, and the fact that the model may only emulate select aspects of the human pathophysiology of the disease [43], It is important to use behavioral models that have adequate sensitivity and specificity to allow for correlation with morphological and biochemical findings, as in the area of spinal cord injury. 

Further, the imaging biomarkers identified in preclinical safety assessment studies can also be used in clinical drug safety studies, as MRI is widely available and safe to use in volunteer studies. This can be an advantage for the preclinical safety assessment function as it provides feedback on predictability of animal models to human disease using the same endpoint. Clearly, one would not run MRI on all clinical safety studies but in those cases where there is no cheaper, simpler safety biomarker available and there is doubt about the degree of risk posed in man, for example, because of species differences or because the effect size in the placebo group is expected to be very high. 

---