Integrative Analysis of Targeted Genomic Profiling and Immune Cell Infiltration in the Prognosis of Lung Adenocarcinoma

Summary

This protocol enables the integration of next-generation sequencing and multicolor immunofluorescence to profile lung adenocarcinoma. It aims to analyze the relationship between genomic heterogeneity and tumor-infiltrating immune cell density. This multidimensional approach facilitates a comprehensive evaluation of the tumor microenvironment for postoperative prognostic assessment.

Abstract

The prognosis of lung adenocarcinoma (LUAD) is traditionally evaluated via tumor-node-metastasis (TNM) staging, which does not account for the patient's immune status. This protocol integrates genomic profiling and immune microenvironment analysis to facilitate a more comprehensive postoperative prognostic evaluation. The method involves a retrospective analysis of paraffin-embedded tumor tissues using two primary techniques. First, next-generation sequencing (NGS) is performed with a customized 37-gene panel to identify mutations in driver genes and variants of uncertain significance. Second, multiplex immunofluorescence (mIF) is utilized to target markers including HLA-DR, CD68, CD163, CD206, PD-L1, and PanCK. This enables the quantification of spatial distribution and density for specific immune cell subpopulations across various tumor regions. This integrated approach enables the simultaneous assessment of genomic heterogeneity and tumor-infiltrating immune cell characteristics. The resulting data identifies specific combinations of mutational profiles—such as EGFR status—and immune cell densities. These integrated combinations enable the study of their collective impact on patient survival, offering a promising approach for the development of future lung cancer prognostic models. This protocol demonstrates a robust method for characterizing the complex biological features of the LUAD tumor microenvironment.

Introduction

LUAD is the most prevalent subtype of lung cancer, yet the average 5-year survival rate remains low at approximately 15%¹. Current prognostic evaluation primarily relies on the TNM staging system; however, this method often overlooks the patient's overall immune status, which may limit its predictive accuracy. Evidence suggests that both genomic heterogeneity and the tumor microenvironment (TME) play critical roles in LUAD tumorigenesis and disease progression^2,3.

In recent years, the transition from traditional Sanger sequencing to targeted NGS has enabled the high-sensitivity, simultaneous analysis of multiple driver genes, such as EGFR, TP53, and KRAS^4,5. While NGS provides a high-resolution map of genetic alterations, genomic data alone cannot capture the complex cellular interactions and spatial organization within the TME.

To address these complexities, mIF offers significant advantages over traditional immune profiling methods. While single-plex immunohistochemistry (IHC) is restricted to one marker per section, mIF enables the simultaneous detection of multiple protein markers while preserving the spatial architecture of the tissue. Unlike sequential IHC, which may cause tissue degradation, or flow cytometry and bulk RNA sequencing, which result in the loss of spatial localization and single-cell resolution^6,7,8, mIF combined with spatial region annotation allows for in situ, multi-parameter characterization of the immune landscape⁹.

This protocol establishes a combined workflow integrating targeted NGS, mIF, and spatial region assessment (encompassing the whole tissue, tumor boundary, and tumor core). Such a multidimensional approach is necessary to investigate the interplay between genomic alterations and immune cell density—a relationship that single-modality approaches often fail to capture. By evaluating these integrated characteristics, this protocol provides a complementary strategy for characterizing the biological features of LUAD and offers a more robust framework for postoperative prognostic stratification.

Protocol

This retrospective study was approved by the Medical Ethics Committee of Kunming Yan'an Hospital (Approval No. 2021-140-01). Informed consent was obtained from all participants.

Study design and sample collection
A cohort of 63 patients with pathologically or histologically confirmed LUAD who underwent surgical resection between 2015 and 2021 was enrolled. Exclusion criteria included the receipt of neoadjuvant chemotherapy, radiotherapy, or immunotherapy prior to surgery. Additionally, patients with concurrent malignancies, severe comorbidities, poor tissue quality, or incomplete clinical records were excluded from the analysis. Formalin-fixed paraffin-embedded (FFPE) tumor blocks were processed into 1 mm sections for deparaffinization. As illustrated in the workflow (Figure 1), these samples underwent an integrated analysis using mIF and targeted NGS. This combined approach enables the simultaneous evaluation of genomic alterations and immune microenvironment characteristics, providing a multidimensional framework for characterizing tumor biology.

Follow-up
The patient's targeted drugs, immunotherapy, and disease progression or death were tracked by telephone follow-up, and detailed information was obtained regarding patients’ survival status, disease progression or recurrence, adjuvant treatments received after surgery, and any newly diagnosed comorbidities. The follow-up time ended in July 2023, and the final observation endpoint was disease progression or death. Overall survival was defined as the time from the start of surgery to the patient's death. There were no patients lost to follow-up. The minimum follow-up time was 21 months.

NGS library preparation and sequencing
DNA extraction from FFPE tissue sections
Obtain FFPE tissue sections (3–10 µm thickness) from pathological specimens. Place sections into sterile microcentrifuge tubes. Add xylene to each tube and incubate for 5–10 min at room temperature to remove paraffin. Centrifuge at 12,000 x g for 5 min and carefully discard the supernatant. Repeat the xylene treatment 1x to ensure complete deparaffinization. Wash the pellet sequentially with 100% ethanol 2x to remove residual xylene. Air-dry the pellet at room temperature until residual ethanol has completely evaporated. Add appropriate lysis buffer and proteinase K (provided in the DNA extraction kit) and incubate at 56 °C for 1–3 h (or overnight if necessary) until the tissue is fully lysed. Perform an additional incubation at 90 °C for 1 h to reverse formalin cross-linking. Extract genomic DNA using a commercial FFPE DNA extraction kit (see Table of Materials) according to the manufacturer’s instructions. DNA quality from FFPE samples is often fragmented; therefore, DNA quality should meet minimal integrity thresholds (e.g., DIN ≥ 5) to ensure successful library preparation for targeted NGS.

DNA quantification and normalization
Quantify DNA using a fluorometric method to ensure accurate concentration measurement. Normalize DNA samples to a final concentration of approximately 5–10 ng/µL using Tris-HCl buffer (pH 8.5). Perform dilution steps gradually if necessary to ensure accuracy and stability of DNA concentration.

Library preparation and target enrichment
Prepare sequencing libraries using hybrid capture‑based targeted NGS on DNA extracted from FFPE tissue¹⁰. Perform target enrichment using a customized probe panel covering exonic regions of selected genes associated with tumor development. Assess library quality and fragment size distribution using standard methods (e.g., electrophoresis-based systems), if required.

Library pooling and sequencing
Quantify each library using a fluorometric assay. Pool libraries at equimolar concentrations according to the sequencing kit requirements. Load pooled libraries onto the Illumina sequencing platform and perform sequencing with a 150 bp pair-end sequencing pattern.

Data processing and variant detection
Raw sequencing reads were processed using standard bioinformatics pipelines, including adapter trimming, quality filtering, and alignment to the human reference genome (GRCh37/hg19). Quality control metrics were as follows: Q20 > 85%, average sequencing depth > 2500x, on-target rate > 65%, and Fold 80 base penalty ≤ 2. Quality control and alignment: Raw FASTQ data underwent quality control using FastQC and Trim Galore (or Trimmomatic) to remove adapter contamination and low-quality sequences. Cleaned reads were aligned to the GRCh37 (hg19) human reference genome using the Burrows-Wheeler Aligner (BWA). Post-alignment processing: Sorting, duplicate marking, and base quality score recalibration (BQSR) were performed using Picard and the Genome Analysis Toolkit (GATK) to minimize false-positive calls. Variant calling: Germline single-nucleotide polymorphisms (SNPs) and InDels were detected using BWA MEM (0.7.16a - r1181). Somatic SNVs and InDels were identified through a consensus approach using Mutect2 and VarScan2. Low-frequency mutations were further evaluated using the Freebayes (v 1.3.2) statistical model. Annotation and filtering: Variants were annotated using ANNOVAR, integrating data from gnomAD, ClinVar, and COSMIC. Germline variants were distinguished from somatic mutations by filtering out high-frequency population SNPs and evaluating variant allele frequencies (VAF). VAF thresholds: A VAF cutoff of ≥5% was used for clonal events (e.g., ROS1 and ALK), while a 1% threshold was applied to gene mutations (e.g., EGFR) to ensure the detection of low-frequency alterations, which was made possible by the high average sequencing depth of >500x. Germline mutations were distinguished from somatic mutations using the 1000 Genomes Project (1000G) data and a local baseline database of healthy Han Chinese individuals established in our laboratory. Pathogenicity assessment of detected somatic variants was performed according to the AMP/ASCO/CAP framework¹¹, which refines evidence grading for somatic variants through a tiered, evidence-based approach. Pathogenic/likely pathogenic (P/LP) variants are retained, with their classification grounded in Tier I/II criteria requiring A/B (FDA approval, clinical trials) or C/D (preclinical/case studies) evidence for grading. Tier III (VUS) lacks conclusive evidence, while Tier IV indicates benign variants through population frequency (e.g., gnomAD) or functional studies; these mutations are excluded from subsequent analyses. Pathogenicity assessment of detected germline variants was performed according to ACMG/AMP guidelines¹². Variants of pathogenic (P), likely pathogenic (LP), and uncertain significance (VUS) were recorded in this study. Only high-confidence variants passing quality control criteria were included in downstream analyses (QUAL ≥1; coverage depth >10 for germline mutation and coverage depth >500x for somatic mutation)

Sample preparation from FFPE tissue
Preparation of paraffin sections
Cut FFPE tissue blocks into 3 µm thick sections using a microtome. Float sections on a 40–45 °C water bath to flatten the tissue. Mount sections onto adhesive microscope slides and dry at room temperature. Place slides in a 65 °C drying oven for 1 h to ensure firm adhesion of tissue sections to the slides and to melt excess paraffin.

Deparaffinization and rehydration
Place slides in a staining rack and immerse them in clearing solution I for 5 min. Transfer slides sequentially to clearing solution II and clearing solution III, 5 min each, to completely dissolve paraffin. Transfer slides to 100% ethanol for 3 min to remove residual clearing reagent. Rehydrate tissue sections through a graded ethanol series: 95% ethanol for 3 min and 80% ethanol for 3 min. Rinse slides 2x in distilled water, allowing 3 min between washes. Xylene and ethanol used during deparaffinization were handled in a chemical fume hood due to their volatility and toxicity. All waste solutions containing organic solvents were collected separately and disposed of according to institutional hazardous waste disposal guidelines.

HE staining partition
Hematoxylin and Eosin staining
Immerse slides after deparaffinization and rehydration in hematoxylin solution for 5–10 min at room temperature. Rinse slides under running tap water for 5 min. Differentiate slides in 1% hydrochloric acid alcohol for a few seconds (3–10 s). Immediately rinse in running tap water. Blue nuclei by immersing slides in tap water or bluing solution for 5 min. Immerse slides in eosin solution for 1–3 min at room temperature. Briefly rinse in distilled water if necessary.

Dehydration and clearing
Dehydrate slides through graded ethanol: 80% ethanol for 1 min; 95% ethanol for 1 min, and 100% ethanol 2 times for 1 min. Clear slides in xylene I–II, 2–5 min each.

Mounting
Apply a drop of Neutral balsam. Place a coverslip carefully to avoid air bubbles. Allow slides to dry at room temperature before imaging.

Multiplex immunofluorescence staining of FFPE Tissue Sections
Sequential multiplex immunofluorescence staining was performed to detect multiple immune and tumor markers on the same tissue section. This approach enables in situ visualization of diverse cell populations while preserving spatial context¹³. Antigen Retrieval: Fill a staining jar with antigen retrieval buffer. Place the jar in a microwave oven and heat at high power for 3-5 min until the solution begins to boil. Carefully transfer slides after deparaffinization and rehydration into the heated retrieval buffer, ensuring the tissue sections are completely submerged. Heat slides at low microwave power for 15 min to allow efficient antigen unmasking. After heating, allow slides to cool naturally at room temperature for 30 min. Transfer slides to ice water for 5 min to stop the antigen retrieval process.

Washing and hydrophobic barrier preparation
Rinse slides 1x with distilled water to remove residual retrieval buffer. Use a hydrophobic barrier pen to draw a circle around each tissue section. Rinse slides again briefly with distilled water. Wash slides 2x with TBST buffer. After each wash, place slides horizontally for 3 min to ensure thorough removal of buffer.

Blocking
Apply blocking buffer to fully cover the tissue section (~100 µL per slide). Incubate slides in a humidified chamber at room temperature for 15 min.

Primary antibody incubation
Remove the blocking buffer by gently tapping the slide edge on absorbent paper. Primary antibodies against HLA-DR, CD68, CD163, CD206, PD-L1, and PanCK were applied at optimized working dilutions ranging from 1:100 to 1:400. Apply ~100 µL of diluted primary antibody solution to each tissue section. Incubate slides in a humidified chamber: 37 °C for 1 h, or 4 °C overnight for increased sensitivity.

Washing
Wash slides 3x with TBST buffer, 3 min each wash. Place slides horizontally after each wash to allow buffer drainage.

Secondary antibody incubation
Prepare the secondary antibody solution by diluting 1:3 in TBST buffer. Apply approximately 100 µL of secondary antibody solution to each slide. Incubate slides at 37 °C for 10 min in a humidified chamber. Wash slides 3x with TBST, 3 min each wash.

Fluorescent signal amplification
Dilute the fluorescent amplification reagent 1:100 in amplification buffer. Apply 100 µL of diluted reagent to each slide. Incubate slides at room temperature for 10 min in the dark to prevent photobleaching. Wash slides 3x with TBST, 3 min each wash.

Antigen retrieval between markers
Perform microwave antigen retrieval as described. After heating, cool slides sequentially: 5 min in warm water and 5 min in ice water.

Sequential staining of additional markers
Repeat primary and secondary antibody incubation steps for each additional antibody marker. Ensure antigen retrieval is performed after staining each marker before proceeding to the next antibody.

Nuclear counterstaining
Wash slides 1x with TBST buffer. Apply DAPI nuclear stain. Incubate at room temperature for 5 min in the dark. Wash slides 3x with distilled water. Endpoint determination: Nuclei should appear clearly stained under fluorescence microscopy.

Final marker only staining
Apply biotin solution diluted 1:200 in amplification buffer, incubate 10 min at room temperature. Wash slides 3x with TBST buffer. Perform antigen retrieval as described. Wash slides 3x with TBST buffer. Apply endogenous biotin blocking reagent, incubate 37 °C for 30 min. Apply staining dye 1:100 in TBST and incubate 30 min at room temperature in the dark. Wash slides 1x with TBST and allow to drain 3 min.

Mounting
Add a drop of Antifade Mounting Medium onto the tissue section. Carefully place a coverslip over the slide. Allow slides to dry before imaging.

Image acquisition of H&E and mIF sections
Whole-slide images (WSI) of H&E-stained and mIF FFPE tissue sections were acquired using a slide scanner. Brightfield scanning was performed for H&E sections, while fluorescence scanning was used for mIF sections. Objective: 20x magnification for overview scanning; 40x magnification for high-resolution regional imaging. Excitation and emission settings: Optimized for each fluorophore according to the manufacturer’s specifications. Exposure time and gain: Adjusted automatically by imaging software to ensure adequate signal intensity without saturation.

mIF quality control and image evaluability
To ensure the reliability of the mIF results, a rigorous QC protocol was implemented prior to tissue classification and cell quantification using the HALO digital pathology system. Artifact removal and evaluable area definition: Whole-slide images (WSIs) were meticulously reviewed to manually exclude regions containing tissue folds, staining artifacts, and non-specific background signals. This step was essential to prevent false-positive signals and ensure the accuracy of spatial analysis. Following this screening, an average of 80% (range: 50%–95%) of the tissue area across all samples was designated as the evaluable mIF region. Pathological verification and partitioning: To accurately delineate tumor and stromal compartments, mIF sections were cross-referenced with adjacent serial sections stained with H&E. Annotations were independently reviewed by two experienced pathologists, with any discrepancies resolved by consensus. Samples with a tumor tissue area of less than 30% were excluded from the study to maintain analytical stringency. Post-quantification validation: Following automated cell phenotyping and quantification, the spatial distribution and expression patterns of immune cell subpopulations were re-evaluated by pathologists. This review ensured that the digital quantitative data were consistent with the underlying histomorphological characteristics, thereby guaranteeing the reliability and accuracy of the findings.

Digital pathology analysis and region annotation
The panoramic scanned images were imported into the digital pathology image analysis platform for processing. The HALO AI Nuclei Seg-Plugin was used to segment and identify cells. Using the H&E staining as reference, the following regions of interest (ROIs) were manually annotated: Whole tumor region (WSIT); Tumor boundary (TB); Tumor core (TC). Positive signal detection was determined based on fluorescence intensity thresholds and the expected subcellular localization of each marker. This compartment-specific approach was applied to minimize background interference and improve the accuracy of immune cell identification. Immune cell density was quantified as the number of marker-positive cells per selected tumor regions. Align mIF images with corresponding H&E-derived annotations based on tissue morphology. Patients were stratified into high-density and low-density groups according to the median value of immune cell density for each marker.

Statistical analysis
SPSS 17.0 software was used for analysis. Normally distributed data were expressed as mean ± standard deviation, and non-normally distributed quantitative data were expressed as median (25%-75% quantile range). The correlation between each gene mutation type, immune cell density and clinical characteristics was analyzed by the Cox proportional hazards regression model by using python lifelines model. According to the sample size and whether the data conformed to normal distribution, Student's test or Mann-Whitney U test was used for comparison between continuous variables. Wilcoxon signed rank test was used for comparison of rank variables. Chi-squared test or Fisher's exact test was selected for analysis of unordered categorical variables according to the sample size and theoretical frequency. The Spearman test or the Pearson test was used for correlation analysis. The overall survival rate of patients was obtained by the Kaplan-Meier method, and the Log-rank test was used for testing. Survival curves were drawn using the R language based on RStudio software. P-value <0.10 was considered statistically significant.

Practical applicability and limitations
For the reliable implementation of this integrated workflow, FFPE tissue blocks must meet stringent quality standards, including an ideal tumor cellularity of ≥30% and the absence of extensive necrosis or significant tissue artifacts. The technical success of mIF is highly dependent on tissue preservation; factors such as over-fixation, excessive damage, or high background autofluorescence can result in inadequate antibody penetration and suboptimal signal intensity. Furthermore, several inherent limitations of this study should be considered. First, the analysis is based on a retrospective, single-center cohort of 63 patients, which may limit the statistical power and the broad generalizability of the findings to other populations. Second, the scope of the assessment is restricted to a fixed 37-gene NGS panel and a specific set of immune markers (HLA-DR, CD68, CD163, CD206, PD-L1, and PanCK; Supplementary Table 1). As noted in the sources, while this combined approach provides a more robust framework for prognostic stratification, larger prospective, multi-center studies are necessary to further validate these associations and refine the predictive utility of the identified biomarkers.

Results

Clinical characteristics of patients with lung adenocarcinoma
A total of 63 patients who were diagnosed with LUAD and underwent surgical treatment in our hospital from 2015 to 2021 and met the inclusion criteria were collected in this study. The clinical characteristics of the included patients are shown in Table 1. Among the enrolled patients, the median age was 60 years (47-73 years), including 30/63 male patients (46.2%) and 33/63 female patients (53.8%). There were 23/63...

Discussion

Integrated genomic and spatially resolved immunological profiling of LUAD
This study demonstrates an integrated workflow combining targeted NGS and mIF to characterize the multidimensional landscape of LUAD. Our genomic findings align with previous literature, particularly regarding ethnic variations in mutation frequencies. The observed EGFR mutation rate (approximately 40%-60%) is consistent with reported frequencies in Asian populations, which are significantly higher than those in West...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
AlphaTSA Multiplex IHC Kit	alphaxbio Technology Co., Ltd.	AXT36025011	Multiplex immunofluorescence staining
Antifade Mounting Medium	Leagene Technology Co., Ltd.	IH0252, Lot0525A17	Multiplex immunofluorescence staining
CD163	Beijing Zhongshan Golden Bridge Biotechnology Co. Ltd	ZM0428	RRID:AB_3714707;Multiplex immunofluorescence staining
CD206	Abcam, Inc	Cat# 5307-1	RRID:AB_10896526;Multiplex immunofluorescence staining
CD68	Beijing Zhongshan Golden Bridge Biotechnology Co. Ltd	ZM0060	RRID:AB_2904190;Multiplex immunofluorescence staining
customlised cancer gene panel	TWIST Co., Ltd.		NGS Library Preparation
Deparaffinization Buffe	QIAGEN Co., Ltd.	19093	DNA Extraction HE staining; Multiplex immunofluorescence staining from FFPE Tissue Sections
Dynabead M-270 streptavidin	Thermo Co., Ltd.	65602	NGS Library Preparation
Fluorometer	ThermoFisher Scientific Co., Ltd.	QUBIT 4.0	NGS Library Preparation
HALO digital pathology image analysis platform	Indica Labs, Inc., Albuquerque	version 3.6.4134	HE staining and Multiplex immunofluorescence staining analysis
Hematoxylin-Eosin Staining Kit	Wuhan elabscience Technology Co., Ltd.	E-IR-R117	HE staining
HiPure FFPE DNA Kit	Guangzhou Magen Biotechnology Co., Ltd.	D6323	DNA Extraction from FFPE Tissue Sections
HLA-DR	HUABIO Co., Ltd.	ET1610-66	RRID:AB_3069950; Multiplex immunofluorescence staining
Illumina DNA sequencer	Illumina,  Inc	550DX	Sequencing
Neutral balsam	Beijing Solarbio Science & Technology Co., Ltd.	G8590	HE staining
Panck	Beijing Zhongshan Golden Bridge Biotechnology Co. Ltd	ZM0069	RRID:AB_2941997;Multiplex immunofluorescence staining
PD-L1	R and D Systems	Catalog # AF1019	RRID:AB_354540;Multiplex immunofluorescence staining
TBST buffer	abclonla Technology Co., Ltd.	RM00013	HE staining and Multiplex immunofluorescence staining
UN-Blocker	homgen Co., Ltd.		NGS Library Preparation
VAHTS DNA Clean Beads	Vazyme Co., Ltd.	N411	NGS Library Preparation
VAHTS uniersal DNA library Prep kit for illumina	Vazyme Co., Ltd.	ND610	NGS Library Preparation
whole-slide imaging system	ZEISS Co., Ltd.	ZEISS AXIOSCAN 7 version3.3	HE staining and Multiplex immunofluorescence staining analysis
xGen Hybridization and Wash Kit	IDT Co., Ltd.	1080577	NGS Library Preparation