Early detection of lung cancer using artificial intelligence-enhanced optical nanosensing of chromatin alterations in field carcinogenesis

Sep 24, 2026

Uploaded by Putri

Supranucleosomal chromatin structure regulates gene expression, and dysregulation is linked to carcinogenesis. Building on field carcinogenesis evidence that buccal mucosa carries lung-cancer molecular signatures, the study tests whether chromatin structural changes can predict lung cancer, while addressing limits of diffraction-limited optical microscopy for nanoscale chromatin domains. The authors develop csPWS (chromatin-sensitive partial wave spectroscopic microscopy) and apply AI to csPWS nanocytology of 179 patients across two sites, distinguishing Stage-I lung cancer from cancer-free controls. Results show diagnostic separation with site-specific AUC performance, motivating a minimally invasive screening approach.

LanguageEnglish

CategoryHealthcare

FormatPDF

Pages13

Reading Time33 min

AccessFree to download

Introduction

Field carcinogenesis and chromatin changes

Screening need and limitations of existing tests

Methods

csPWS optical nanosensing approach

AI-enhanced diagnostic pipeline

Study design and patient cohort

Results

Stage-I detection performance at two clinical sites

Discussion

Interpretation of biomarker limits

Prospects for minimally invasive screening

[www. nature.com/scientificreports](www. nature.com/scientificreports) OPEN Early detection of lung cancer using artificial intelligence‑enhanced optical nanosensing of chromatin alterations in field carcinogenesis Ali Daneshkhah1,5, Sravya Prabhala1,5, Parvathi Viswanathan2, Hariharan Subramanian1,2, Jianan Lin2, Andrew S Chang1, Ankit Bharat4, Hemant Kumar Roy3 & Vadim Backman1* Supranucleosomal chromatin structure, including chromatin domain conformation, is involved in the regulation of gene expression and its dysregulation has been associated with carcinogenesis. Prior studies have shown that cells in the buccal mucosa carry a molecular signature of lung cancer among the cigarette‑smoking population, the phenomenon known as field carcinogenesis or field of injury. Thus, we hypothesized that chromatin structural changes in buccal mucosa can be predictive of lung cancer. However, the small size of the chromatin chain (approximately 20 nm) folded into chromatin packing domains, themselves typically below 300 nm in diameter, preclude the detection of alterations in intradomain chromatin conformation using diffraction‑limited optical microscopy. In this study, we developed an optical spectroscopic statistical nanosensing technique to detect chromatin packing domain changes in buccal mucosa as a lung cancer biomarker: chromatin‑sensitive partial wave spectroscopic microscopy (csPWS). Artificial intelligence (AI) was applied to csPWS measurements of chromatin alterations to enhance diagnostic performance. Our AI‑enhanced buccal csPWS nanocytology of 179 patients at two clinical sites distinguished Stage‑I lung cancer versus cancer‑free controls with an area under the ROC curve (AUC) of 0.92 ± 0.06 for Site 1 (in‑state location) and 0.82 ± 0.11 for Site 2 (out‑of‑state location) . Cancer screening tests should, ideally, identify cancer before symptoms have appeared and while the tumor is small in order to effectively increase the chance of treatment and reduce mortality. Lung cancer is the leading cause of cancer deaths across races and genders in the U.S. with an overall 5-year survival rate of 22.9% which is notably lower than colorectal (65.1%), breast (90.6%), and prostate cancers (96.8%)1. However, if lung cancer is detected at an early stage it is highly curable through surgical resection. The 5-year survival rate for late-stage (distant) non-small lung cancer (NSLC) is less than 8% but improves to 64% if detected at a localized stage, and reaches 80% if detected at Stage-IA2. Low-dose computed tomography (LDCT) has been established as the gold standard for lung cancer screening and is associated with a 20% decrease in mortality among patients screened with the technique. Accessibility, cost, stigma, and lack of adherence to LDCT guidelines are among the major challenges limiting its impact, as only about 5% ofthe LDCT-eligible population undergoes screening3, resulting in 55% of lung cancer cases being detected at an advanced stage where the survival rate is below 8%4. We therefore propose a minimally invasive, accessible, sensitive, and accurate screening test with high sensitivity (Se) to early-stage lung cancer. Screening methods other than LDCT such as chest X-rays and sputum cytology have proven unsatisfactory when evaluated in large-scale clinical screening settings5. New methods based on standard protein biomarkers used for the detection of cancer do not provide sufficient sensitivity and specificity (Sp)6. Recently, there has been significant interest in the development of protocols that rely on tumor secretions in the blood, such as liquid biopsy. Tests being developed by companies including Grail, Freenome, Guardant, Delfi, and Thrive identify cancer by analyzing circulating tumor DNA (ctDNA) or tumor-derived circulating free DNA (cfDNA) properties such as gene mutations, methylation, and fragmentation7–11. Although initial results have shown promise in the detection of various cancers, including lu

FAQs

Why focus on chromatin structural changes in buccal mucosa for lung cancer detection?

Because field carcinogenesis suggests buccal mucosa contains a lung-cancer molecular signature, the study hypothesizes that chromatin structural changes may be predictive of lung cancer.

What is csPWS and what problem does it address?

csPWS (chromatin-sensitive partial wave spectroscopic microscopy) is designed to detect chromatin packing domain changes that diffraction-limited optical microscopy cannot reliably resolve at nanoscale dimensions.

How does AI improve diagnostic performance in this study?

Artificial intelligence is applied to csPWS measurements of chromatin alterations to enhance diagnostic performance, enabling Stage-I lung cancer separation from cancer-free controls in the patient cohort.

Sep 24, 2026

Uploaded by Putri