The Role of DNA and RNA From Extracellular Vescicles Obtained by Bronchial Lavage in NGS Analyses for Advaced Stage NSCLC Patients

In patients with advanced non-small cell lung cancer (NSCLC), who often undergo biopsies, there is a growing need to clearly define the clinical role of liquid biopsies. Lung tumors can only be visualized directly during bronchoscopy when located in central positions, sometimes making sampling difficult when they are located in more peripheral regions of the lung. With the development of drugs with specific therapeutic targets, the clinical need for re-biopsy or even repeated biopsies is increasing; these biopsies are necessary to identify the mechanisms of drug resistance in target lesions. Liquid biopsy of blood was introduced to overcome the difficulties in obtaining a quality solid tumor tissue sample through direct biopsy of tumor lesions, but it remains a secondary diagnostic technique compared to tissue biopsies due to its low sensitivity. Liquid biopsy of blood is a technique used to detect circulating deoxyribonucleic acid (DNA) derived from tumor cells (free DNA fragments in the blood). Furthermore, since blood is a complex body fluid consisting of many components, there is a need to identify a different biological resource from which to perform the liquid biopsy. In addition, the characteristics of 'naked' DNA ('Cell-Free DNA' - cfDNA) extracted from plasma after the removal of red blood cells have never been adequately defined, as this DNA may have originated from floating DNA, protein-DNA complexes and/or from within extracellular vesicles (EV). EVs, nano-sized vesicles secreted by almost all types of cells, transport bioactive molecules such as proteins, glycans, lipids, metabolites, and nucleic acids (RNA and DNA), enclosed in a double lipid layer, and act as essential mediators in cell-cell communication. EVs are an ideal biomarker for cancer, as the content of EVs originating from cancer cells reflects the molecular and genetic composition of the parent cells and is secreted in higher quantities than EVs from normal cells.

Researchers identified tumor-derived EVs carrying double-stranded DNA (dsDNA) and EV-DNA representing the entire genome, even reflecting the mutational status of the parental tumor cells. Several studies have also shown that, although EV-DNA exists in short and long sizes, most are 10 kb or longer, and have demonstrated that tumor-related mutations can be comprehensively detected using EV DNA. In addition, two types of DNA were identified, internally protected and external surface DNA, which could be used as a basis for the development of diagnostic methods.

Long-stranded, concentrated EV-DNA is easy to amplify, making it a suitable candidate for NGS analysis. It has been suggested that NGS analysis using EVs isolated from pleural effusion and urine could effectively replace NGS on tissue in cases where there is a shortage of tissue. The encapsulation of DNA within the double lipid layer of EVs appears to improve DNA stability by protecting it from the external environment. The amount of cfDNA released by tumors is normally low, representing only a fraction of the tumor's genomic heterogeneity released by dying tumor cells. In contrast, the process of isolating EVs from fluids increases the final concentration of EV-DNA obtained, and in addition, EVs are secreted by both viable and dying tumor cells. In a study comparing the use of EV-DNA and plasma cfDNA from NSCLC patients in liquid biopsies, EV-DNA improved the detection sensitivity of EGFR exon 19 deletion, L858R, and T790M mutations. This study also highlighted the diagnostic and prognostic capabilities of EV-DNA by demonstrating that mutations identifiable by liquid biopsy on EV-DNA included the specific acquired resistance mutation T790M. Furthermore, the specificity and sensitivity of liquid biopsy were further increased with the use of EV-DNA from bronchoalveolar lavage (BAL) fluid. EVs are widely distributed in various body fluids, making them easier to sample using less invasive methods than cancer cells. More importantly, specificity and sensitivity can increase significantly when EV-DNA is used for liquid biopsy compared to cfDNA. For these reasons, liquid biopsy using EV-DNA could be a clinically useful means of isolating cancer-specific DNA.

Lung cancer cells produce and release EVs in abundance in the tumor microenvironment (TME), which is easily accessible via bronchial wash (BW) or bronchoalveolar lavage (BAL). These procedures are mainly performed for diagnostic purposes, to obtain microbiological or cytological information from the main airways down to the subsegmental bronchi in the case of bronchial lavage, or from the lower respiratory tract and alveoli in the case of bronchoalveolar lavage. The results of liquid biopsy on EVs isolated from BAL and BW in patients with NSCLC have been shown to be superior to those of the same biopsy performed on plasma cfDNA. Recent studies have shown that EVs successfully isolated from BAL in patients with lung cancer contain abundant amounts of dsDNA, and that liquid biopsy for genotyping of the epidermal growth factor receptor (EGFR) using BAL is tissue-specific and extremely sensitive compared to the use of cfDNA. Furthermore, sensitivity increases proportionally with disease stage progression, reaching 100% in samples from stage IV patients. These results seem to suggest that EGFR genotyping from BAL-derived EVs is a highly promising liquid biopsy method, and particularly efficient for patients requiring repeated biopsies during disease progression. Furthermore, it is noteworthy that EGFR testing of BAL-derived EVs in stage I lung cancer showed a relatively high predictive value of 85.7%, despite a low sensitivity of 35.3%. BAL is a minimally invasive procedure associated with a low complication rate; however, it is technically more complex than bronchoalveolar lavage. The superiority of liquid biopsy was confirmed even on EV isolated from BW in patients with NSCLC compared to plasma cfDNA biopsy, achieving a diagnostic accuracy of 94.5% for EGFR genotyping.

In the diagnostic field, clinicians often find it difficult to manage slow-growing ground glass nodules (GGNs), especially in cases of low-intermediate risk of the lesion being tumorous. Three-quarters of resected GGNs tested positive for EGFR, KRAS, ALK, or HER2 mutations. Therefore, considering the high positive predictive value of EV genotyping on BAL and BW, the detection of a mutation through this sampling could be useful in directing treatment towards surgical resection rather than careful follow-up. The low sensitivity of the analysis could be improved by using more sensitive techniques such as next-generation sequencing and droplet digital PCR.