Recent results indicate that low-dose computed tomography (LD-CT) screening reduces lung cancer mortality in high risk subjects. However, high false positive rates, costs and potential harm highlight the need for complementary biomarkers. Led by Dr Ugo Pastorino, a group of researchers from Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy, retrospectively evaluated a non-invasive plasma miRNA signature classifier in prospectively collected samples from smokers within the randomised Multicentre Italian Lung Detection (MILD) trial. Their findings indicate that microRNA signature classifier has predictive, diagnostic and prognostic value and its combined use with LD-CT may improve screening performance. The results were presented in a proffered papers session at the 4th European Lung Cancer Conference (26-29 March 2014, Geneva, Switzerland).
miRNAs are tissue- and disease-specific molecules, actively released by cells in the circulatory system. Circulating miRNAs are rather stable and easily detectable in body fluids suggesting the possibility for potential use as a new promising class of biomarkers.
The study authors have previously reported in the Proceedings of the National Academy of Sciences journal that miRNA profiling in plasma samples of disease-free smokers enrolled in two independent spiral-CT screening trials, resulted in the generation of miRNA signatures with strong predictive, diagnostic, and prognostic potential.
miRNA profiling in plasma
In the current study, the diagnostic performance of a non-invasive plasma miRNA signature classifier was retrospectively evaluated in samples prospectively collected from smokers included in the MILD trial. There were a total of 939 subjects included in the trial, from those 69 patients with lung cancer and 870 disease-free individuals. The LD-CT arm comprised 652 subjects and there were 287 subjects in the observation arm.
Plasma samples were analysed by PCR-based assay. Diagnostic performance of miRNA signature classifier was evaluated in a blinded validation study using pre-specified risk groups.
The diagnostic performance of miRNA signature classifier for lung cancer detection was 87% for sensitivity and 81% for specificity across both arms, and 88% and 80% respectively in the LD-CT arm.
For all subjects, miRNA signature classifier had a negative predictive value of 99% and 99.86% for detection and death-by-disease respectively. LD-CT had sensitivity of 79% and specificity of 81% with a false positive rate of 19.4%.
Diagnostic performance of miRNA signature classifier was confirmed by time dependency analysis.
Combination of both miRNA signature classifier and LD-CT resulted in a five-fold reduction of LD-CT false positive rate to 3.7%. miRNA signature classifier risk groups were significantly associated with survival (p < 0.0001).
The authors concluded that their large validation study indicates predictive, diagnostic and prognostic value of miRNA signature classifier. It could reduce the false positive rate of LD-CT and improve the efficacy of lung cancer screening by detecting for up a tumour up to two years before it can be found by an LD-CT scan. It determines the likelihood of a patient developing lung cancer with 87% sensitivity 81% specificity and identifies high-risk patients, revealing both the presence and aggressiveness of the disease, and those at risk of developing it. Furthermore, the false positive rate of only 4% (when used together with LD-CT) compares favourably with 96,4% in LD-CT scans alone.
Dr Johan Vansteenkiste, who discussed the study results, said that NLST trial brought level I evidence from three annual rounds of low-dose CT screening in current or former smokers, 55-74 years old individuals with 30 pack/year history and showed a 20% decrease in lung cancer-specific mortality, but it requires screening 320 subjects to prevent one lung cancer death.
According to Dr Vansteenkiste, opportunities from miRNA screening are modelling and testing with more refined imaging features (in term of 2D techniques regarding shape, margins, density of nodule and in term of 3D techniques regarding growth pattern of nodule), validation in other cohorts, and prospective demonstration of lower false positive findings, decrease in numbers needed to screen, and further lung cancer mortality reduction.
The authors published their findings online on 13 January 2014 in the Journal of Clinical Oncology.
Abstract 24O: Clinical utility of a plasma-based microRNA signature classifier within computed tomography lung cancer screening.
All authors have declared no conflicts of interest.
The European Lung Cancer Conference (ELCC) is organised by the European Society for Medical Oncology (ESMO) and the International Association for the Study of Lung Cancer (IASLC). During the four-day programme, attendees benefit from educational and scientific updates provided by thoracic oncology specialists on different multidisciplinary topics important for research and clinical practice in the field of lung cancer.