2.1 Patients and sputum collection
One hundred and eighteen patients who were scheduled to undergo curative resection of primary lung cancer at Kanagawa Cancer Center during 2014/9-2016/5 were enrolled in this study. Patients with pure ground glass nodule, synchronous multiple lung cancer, and uncontrollable other cancer were excluded from this study. Patients undergoing preoperative radiation therapy or chemotherapy, and bronchoscopy within one week were also excluded. All patients provided informed consent. This study was performed in accordance with relevant guideline and was approved by the Kanagawa Cancer Center institutional review board (25 Ken – 64 and 2019 Eki-14).
Each patient was given a container with YM fixative solution (50% ethyl alcohol and 2% polyethylene) for collection of sputum for 3 days before surgery . Patients were instructed to collect early morning sputum just after gargling and to shake the container approximately 20 times after each sputum collection so that the sputum mixed with the fixative. The patients were instructed to store the containers in a refrigerator.
After centrifugation, a total of two cytological specimens were prepared by rubbing the sputum sample between two glass slides. Each cytological specimen was stained with Papanicolaou stain after 95% ethyl alcohol fixation. Cytology was classified by an expert cytologist as follows: 1) Insufficient material; 2) Class I: absence of atypical or abnormal cells; 3) Class II: atypical cytology but no evidence of malignancy; 4) Class III: cytology suggestive of, but not conclusive for, malignancy; 5) Class IIIa: probably benign atypia; 6) Class IIIb: malignancy suspected; 7) Class IV: cytology strongly suggestive of malignancy; and 8) Class V: cytology conclusive for malignancy. Final diagnosis was made by a pathologist in cases of class III or higher. Sputum cytology positive (SC (+)) was defined as patients whose sputum cytology was class III or higher. Sputum cytology negative (SC (-)) was defined as patients whose sputum cytology was lower than class III, including cases with insufficient material. SNC (sputum not collected) was defined as patients who could not collect sputum because of a lack of sputum.
2.3 DNA extraction and ddPCR
The cover glass was peeled off after immersing the glass slide with the cytological specimen in xylene. After applying 50% diluted Marinol in xylene, the Marinol was cured on the extender for 30 to 60 minutes. The sample was then immersed in Milli-Q® water for approximately 15 minutes to soften the encapsulant, and the sheet-like cells were peeled off with a knife and placed in a tube. After dissolution and removal of the encapsulant, the sample was washed with alcohol and dried. DNA was extracted using the QIAamp DNA FFPE Tissue Kit (Qiagen, Inc., Valencia, CA, USA) according to the manufacturer's protocol. In all samples, the concentration of DNA was measured using Qubit® (Life Technologies, Thermo Fisher Scientific, Carlsbad, CA, USA).
ddPCR was performed using the QX200 Droplet Digital PCR System (Bio-Rad, Hercules, CA, USA) according to the manufacturer's protocol. A total of 20 μL of ddPCR reaction mixture was prepared. The volume contained 10 μL of 2× ddPCR Supermix for Probe (no dUTP) (Bio-Rad), 1 μL template DNA, 1.8 μL of forward and reverse primers (10 μM), 0.5 μL of FAM- and Hex-labeled probe, and 4.9 μL of nuclease-free water. A total of 20 μL/well of the sample solution was transferred to DG8 cartridges (Bio-Rad). After loading 70 μL/well of generator oil in the lower layer of the well, droplets were made using the Droplet Generator (Bio-Rad). After transferring 40 μL to each well of a PCR plate, the plate was sealed with a foil heat seal using a PX1 PCR Plate Sealer (Bio-Rad). PCR was performed using a C1000 Touch thermal cycler (Bio-Rad).
EGFR p.L858R c.2573T>G (Bio-Rad, Catalog #10049550, Assay ID: dHsaMDV2010019) was used as the primer/probe mix to detect the Ex21 mutation using ddPCR. The ddPCR EGFR Exon 19 Deletions Screening Kit (Bio-Rad, Catalog #12002392), which allows the quantification and screening of 15 EGFR exon 19 deletions, was used to detect Ex19 mutations. The negative template control contained reaction mixed with water, and the positive template control contained EGFR-mutated DNA (Ex21 and Ex19 mutation). QuantaSoft software (version 1.7.4) was used for the analysis. The presence or absence of EGFR mutations was determined by the threshold set based on the criterion optimized in the analysis of each sample.
2.4 Pathological findings and EGFR mutation of surgically resected lung cancer specimens
The pathological diagnosis was made by an expert pathologist (Y.T.) based on hematoxylin and eosin (H&E) staining and Alcian blue staining of the tissue sections of formalin-fixed paraffin-embedded (FFPE) specimens. Elastica van Gieson staining was used to evaluate vascular and pleural invasion. Immunostaining of thyroid transcription factor-1 was performed if necessary. Spread through air space (STAS) was defined as the spread of lung cancer cells into air spaces in the lung parenchyma beyond the edge of the main tumor [23, 24]. The existence of STAS was evaluated based on H&E stains of FFPE sections of the tumor.
FFPE sections of the resected tumor were used to extract the DNA from the samples. Eight to ten sections (5–10 μm thick) where tumor diameter was maximum were used to extract DNA. The fragment method (sensitivity <5% mutant DNA) was used to examine Ex19 mutations . The Cycleave method (sensitivity 1-5% mutant DNA) was used to examine Ex21 mutations . The concordance of EGFR mutation status detected in sputum samples and in surgically resected specimens was evaluated for SC (+) and SC (-) groups, respectively.
2.5 Statistical analyses and definitions
Continuous and categorical variables between the two groups were compared using Mann-Whitney U test and Fisher's exact test, respectively. Receiver operating characteristics (ROC) curve analysis was performed to discriminate SC (+) from SC (-) via radiological examination (CT tumor size and positron emission tomography maximum standardized uptake value [PET SUVmax]). CT tumor size was defined as the maximum tumor diameter measured using high resolution CT (level 600 Hounsfield units [HU]; width 1600 HU) of 1 to 2 mm thickness. The preoperative PET-CT scan calculated the SUVmax of the tumor lesion where fluorodeoxyglucose F 18 (18F-FDG) accumulated. Logistic regression was performed to analyze the clinicopathological characteristics of patients regarding SC (+). Significance was defined as p<0.05. Statistical analyzes were performed using EZR on R commander version 1.30 (Saitama Medical Center, Jichi Medical University, Saitama, Japan).