Detection of lung cancer metastasis from blood using L-MISC nanosensor: Targeting circulating metastatic cues for improved diagnosis

Abstract

Metastatic lung cancers are considered one of the most clinically significant malignancies, comprising about 40% of deaths caused by cancers. Detection of lung cancer metastasis prior to symptomatic relapse is critical for timely diagnosis and clinical management. The onset of cancer metastasis is indicated by the manifestation of tumor-shed signatures from the primary tumor in peripheral circulation. A subset of this population, characterized as the metastasis-initiating stem cells, are capable of invasion, tumor initiation, and propagation of metastasis at distant sites. In this study, we have developed a SERS-functionalised L-MISC (Lung-Metastasis Initiating Stem Cells) nanosensor to accurately capture the trace levels of metastatic signatures directly from patient blood. We investigated the signatures of cancer stem cell enriched heterogenous population of primary and metastatic lung cancer cells to establish a metastatic profile unique to lung cancer. Multivariate statistical analyses revealed statistically significant differences in the molecular profiles of healthy, primary, and metastatic cell populations. The single-cell sensitivity of L-MISC nanosensor enabled a label-free detection of MISCs with high sensitivity and specificity. By employing a robust machine learning model, our diagnostic methodology can accurately detect metastatic lung cancer from not more than 5 μl of blood. A pilot validation of our study was carried out using clinical samples for the prediction of metastatic lung cancers resulting in 100% diagnostic sensitivity. The L-MISC nanosensor is a potential tool for highly rapid, non-invasive, and accurate diagnosis of lung cancer metastasis.