Translating buccal nanocytology for lung cancer screening into clinical practice

The overarching goal of this Academic-Industrial Partnership is to develop an at-home, self-administered lung cancer screening test based on the analysis of chromatin architecture in cells swabbed from the buccal (cheek) mucosa, which can identify patients at risk for early-stage lung cancer who may benefit from definitive follow up evaluation such as CT. Lung cancer mortality depends on stage at diagnosis. The combination of recently developed immunotherapy, chemoradiation, and robotic surgery has led to the near curation of cancers detected at early stages such as IA. However, most patients are still diagnosed at late stages. Current screening guidelines recommend all patients with smoking history exceeding 20 pack-years to receive LDCT. Due to underreporting, noncompliance, and the rapidly increasing rates of lung cancer in non-smoking subjects as well as quit- and second-hand smokers, most lung cancer deaths now occur in patients who would not qualify for LDCT. There is an urgent need for an easy-to-administer screening test capable of detecting early-stage lung cancer. We propose a new approach that leverages lung field carcinogenesis and chromatin conformation alterations as the biomarker source and type, respectively. An advantage of the approach is its practicality (i.e., analysis of swabbed buccal cells) and the sensitivity to early-stage cancer regardless of tumor size. We have found that nanoscale alterations in chromatin domains play a critical role in fostering the transcriptional plasticity of precancerous cells in field carcinogenesis and have developed a statistical optical spectroscopic nanosensing technology, chromatin-sensitive partial wave spectroscopic (csPWS) microscopy, to measure chromatin domain conformation with sensitivity down to the size of the 20nm chromatin chain. In the proposed project we will build upon these results and finalize the development of csPWS. We will optimize chromatin conformation biomarkers by combining the quantification of chromatin conformation using csPWS within domains defined by histone marks and transcriptional activity imaged using spectroscopic single molecule localization microscopy with deep learning methods to capture the complexity of the chromatin conformation-transcriptional plasticity association. Chromatin scanning transmission electron tomography will be used to identify chromatin conformation features as biomarker candidates. Finite-difference time-domain computational electrodynamics simulations will provide the link between the chromatin features and csPWS measurements. Molecular theory simulations will be used to confirm the etiological association between the chromatin domain conformation biomarkers and proneoplastic transcriptional patterns. The prediction rule will be validated in an independent patient dataset. The endpoint will be detection of stage I lung cancer across population, including LDCT-eligible/ineligible smokers and nonsmokers. We will ensure that the prediction rule is robust with regards to tumor histology, demographics, and risk factors. Upon the completion of this project, the lung cancer screening test based on the optical analysis of buccal chromatin conformation will be ready for a definitive clinical trial.