In 2001, the Institute of Medicine (IOM) (part of the US National Academy of Sciences) reviewed the scientific basis for tobacco harm reduction. They introduced the term ‘potential reduced-exposure products’ (PREPs) for modified products which gave a substantial reduction in one or more tobacco or smoke toxicants and could reasonably be expected to reduce the risk of one or more specific diseases or other adverse health effects . The IOM encouraged the development of biomarkers of exposure and effect to assist the evaluation of candidate PREPs.
Not long after the IOM report  became publicly available, the Life Sciences Research Office (LSRO) report on “Evaluation of Potential Reduced-Risk Tobacco Products” was published . In agreement with the IOM report, the LSRO report states that assessment of the biological response to a novel product in comparison to a conventional product can only be achieved in clinical studies using biomarkers.
While there are no biomarkers of effect which are qualified to the extent that they can predict the risk of disease, it is generally accepted that such measurements, together with information on smoke chemistry and biomarkers of exposure, will contribute to the assessment of the toxicity of PREPs relative to conventional cigarettes [6,7].
We believe that a panel of biomarkers of biological effect which are related to the major smoking-related diseases will enable us to assess the biological response to reduced exposure to cigarette smoke constituents when consumers smoking conventional products switch to reduced toxicant prototypes (RTPs). Hence, we have developed a biomarker programme to identify, develop and qualify biomarker candidates with potential utility for RTP assessment.
Biomarkers of effect are biological indicators of the body’s response to exposure. They indicate early sub-clinical changes, which if sustained, may go on to have pathological consequences.
For the purposes of candidate RTP assessment, a biomarker of effect would need to be a robust measure in response to cigarette smoking and minimally affected by inter-individual variability. For those biomarkers of effect that are reversible following cessation of cigarette smoke exposure, there will also be a reasonable expectation that they will be reversible if a smoker switches from a conventional control product to a RTP. Ideally, our candidate biomarkers would be related to a disease-specific endpoint, but this requirement may not necessarily be achievable in the context of biomarkers of effect for smoking-related diseases. In addition, the timeframe needed to see a change in the biomarker behaviour has to be appropriate for pre-market testing (e.g. months, weeks, days). Use of the biomarker should also be practical in terms of availability of measurement methodology, analytical reproducibility, sensitivity, specificity and cost. Finally, biomarkers of effect being investigated should preferably be widely accepted by the general scientific community.
We are identifying suitable biomarkers of effect through literature reviews, scientific engagement, in-house research and external collaborations. Our biomarker research is supported by a number of new tools including databases such as BiomarkerCenter (Thomson Reuters), enabling us to mine relevant regulatory information in the context of biomarker qualification. In addition, software applications such as MetaCore (GeneGo) allow us to explore our experimental or clinical study data in combination with publicly available literature to identify novel biomarker candidates with potential utility for RTP assessment.
To date, a series of biomarkers of effect have been identified that range from biomolecules found in tissue or bodily fluids to physiological measurements such as lung function tests and arterial imaging. The table below contains a list of candidate biomarkers of effect associated with cardiovascular disease (CVD), chronic obstructive pulmonary disease (COPD) and cancer, which could be used in clinical studies to assess the biological response to a RTP in comparison to a conventional control product.
|Neutrophils / Macrophages||Inflammation||Sputum / Blood|
|IL-1, 1beta, 6, 16||Inflammation||Sputum / Blood|
|GRo-alpha||Inflammation||Sputum / Blood|
|MCP-1||Inflammation||Sputum / Blood|
|TNF-alpha||Inflammation||Sputum / Blood|
|MMP-1, 9||Extracellular matrix degradation||Sputum / Blood|
|8-epi-PGF2-alpha||Oxidative stress||Sputum / Blood / Urine|
|Superoxide dismutase||Oxidative stress||Blood|
|8OHdG||DNA damage / repair||Urine|
|cis-Thymidine glyco||DNA damage / repair||Urine|
|ICAM-1, VCAM-1||Cardiovascular disease||Blood|
|Endothelial progenitor cells||Cardiovascular disease||Blood|
The smoking machine measurements of smoke chemical yields are limited in their ability to accurately represent the variability in human smoking behaviour. Therefore, there is a need for other systems to estimate human smoke exposure and assess the potential toxicity of cigarette smoke . As one of a number of approaches to address this topic, studies of biomarkers were recommended by the Institute of Medicine  and by the World Health Organisation . However, the biomarkers that could be used for such a purpose were not identified by either organisation.
We therefore conducted a number of proof-of-concept studies to establish which biomarkers would be suitable for the future assessment of biological response(s) to novel product in comparison to conventional product. The primary aim of these studies was to obtain baseline data on biomarkers in healthy smoker study groups that were clearly separated based upon acute smoking exposure in comparison to never and former smokers.
One such study, carried out in 2006, examined biomarkers previously suggested in the literature to be significantly different between smokers and non-smokers [6, 7, 8]. We identified some potentially useful biomarkers for future clinical research including:
Although we confirmed that these biomarkers were significantly different between smokers and non-smokers, we also observed considerable inter-individual variability. From these observations it is apparent that further work will be required before these and other biomarkers can be used to predict the risk of disease in smokers or to assess the potential of RTPs for risk reduction.
Once biomarkers have been identified by the methods described above, additional clinical studies will be necessary to enable us to better understand the behaviour of these biomarkers in defined groups of healthy smokers and to provide data to assist in the assessment of whether the biomarkers of choice can reliably predict a reduction in disease risk.
One of our studies, whose clinical phase has been successfully completed, is investigating the differences in biomarker levels in healthy twins discordant for smoking, i.e. one sibling smokes whereas the other does not.
Another study nearing its completion is aimed at the evaluation of biomarker behaviour in the context of smoking cessation. Not only are we testing the biomarkers that proved to be informative in our proof-of-concept studies, but we are also employing transcriptomics (the analysis of gene expression) to elucidate potential pathways involved in the systemic biological response to cigarette smoke and identify biomarker candidates (gene expression signature) that could be applied in future studies to assess the biological response(s) to an RTP in comparison to a conventional tobacco product.
We hope to publish the results of both of these studies upon completion.