Establishing exposure-response markers is imperative for the assessment of candidate modified risk tobacco products (MRTPs) against conventional combustible cigarettes (CC). Biomarkers derived from the primary site, such as the airway, require invasive sampling, whereas blood offers a non-invasive alternative for the general population. Various diseases and exposures, including cigarette smoke, have been shown to alter the molecular profile of the blood. To identify exposure-response relationships, we have conducted a whole genome Affymetrix microarray analyses from blood derived from a clinical study comparing smokers switching from CC to candidate MRTP THS 2.1, and from smokers who continued smoking their own CC for 5 consecutive days. The open-label, randomized, controlled, 2-arm parallel group recruited 42 healthy smokers of both genders, aged between 23 and 65 years and was conducted according to Good Clinical Practices (GCP) and is registered on ClinicalTrials.gov, number NCT01780714. By analyzing the whole blood transcriptome of the study subjects and leveraging previously identified genes that exhibited a response to smoking in blood, we derived a classification tool using statistical learning methods. The classifier consisted of eight genes and differentiated users of THS 2.1 from smokers of CC with a specificity and sensitivity of 0.81 and 0.74, respectively. Several of these genes have been previously reported to be differentially expressed in the blood of smokers as compared to never smokers. In conclusion, our systems pharmacology approach showed that the impact of switching to THS 2.1 can be detected in 5 days in whole blood transcriptome, thus providing a sensitive and non-invasive tool to assess exposure-response. These data alone do not substantiate reduced risk, and a range of further studies, including clinical studies, are being conducted on MRTPs.