Abstract
Background/Aim: Lung adenocarcinoma and lung squamous cell carcinoma represent the most prevalent subtypes of non-small cell lung cancer eligible for surgery in the early stages. The emergence of immune checkpoint inhibitors as adjuvant therapy has shown promising potential in improving the postoperative prognosis of patients with lung cancer. Hence, a comprehensive understanding of the clinicopathological and molecular features of programmed cell death ligand-1 (PD-L1) expression in lung adenocarcinoma and squamous cell carcinoma is crucial. Patients and Methods: In this retrospective study, we conducted a comparative analysis of clinicopathological features associated with the expression of PD-L1, stratifying patients who underwent surgical resection into two distinct groups: 289 patients with lung adenocarcinoma and 66 with lung squamous cell carcinoma. Furthermore, we investigated the associations between the expression of PD-L1 and genetic alterations in well-established oncogenic driver mutations. Results: Among the cases, 52.9% exhibited negative PD-L1 expression, 32.9% had low PD-L1 expression, and 12.3% had high PD-L1 expression in adenocarcinoma, while the PD-L1 expression in squamous cell carcinoma showed a near-even distribution. Notably, male sex, smoking history, the presence of invasive pathological factors, and disease progression significantly influenced PD-L1 expression in adenocarcinoma, whereas none of these factors were associated with PD-L1 expression in squamous cell carcinoma. Additionally, the distribution of PD-L1 expression varied based on the type of specific driver gene mutation in adenocarcinoma. Conclusion: The present study revealed clinicopathological and molecular differences between lung adenocarcinoma and squamous cell carcinoma patients promoting the expression of PD-L1.
- Lung adenocarcinoma
- lung squamous cell carcinoma
- programmed cell death ligand-1
- driver gene mutation
- immune checkpoint inhibitors
Surgical resection is the primary and potentially curative treatment choice for patients with stage I-II non-small cell lung cancer (NSCLC) and for selected patients with stage IIIA disease (1). Nevertheless, the majority of mortality after complete resection is associated with the emergence of recurrence, driven by minimal residual disease and micrometastasis. These elusive tumor foci cannot be detected through conventional diagnostic methods, such as computed tomography, and may be responsible for cancer recurrence even if no clinical signs of cancer are observed (2). Therefore, the development of postoperative adjuvant therapy to reduce the risk of recurrence is anticipated. Within this context, the utilization of recently developed immune checkpoint inhibitors (ICIs) as adjuvant therapy carries the potential benefits in improving the postoperative prognosis of patients with lung cancer (3).
Immunotherapy based on ICIs represents one of the most important breakthroughs in the management of lung cancer. To evade immune surveillance, cancer cells employ a strategy binding the surface protein known as programmed cell death ligand-1 (PD-L1) to programmed death 1 (PD1) receptor found on T-lymphocytes, effectively inhibiting their cytotoxic function. Given that the PD1–PD-L1 interaction serves as a pivotal immune checkpoint implicated in immune escape throughout cancer development and progression, the presence of PD-L1 on tumor tissues is consistently correlated with positive therapeutic responses to anti-PD1 and PD-L1 inhibitors in NSCLC (4-6).
Adenocarcinoma and squamous cell carcinoma stand as the most prevalent lung cancer subtypes, collectively accounting for approximately 85% of NSCLC cases (2). These two subtypes not only differ in histological characteristics but also exhibit distinct clinicopathological and molecular profiles. The expression of PD-L1 has been reported to be up-regulated by tumor mutation burden, oncogenic signaling pathways, and inflammatory cytokines, such as interferon-γ (7, 8). Given the diverse factors influencing PD-L1 expression, it is plausible that the clinicopathological and molecular determinants driving PD-L1 expression may differ between lung adenocarcinoma and squamous cell carcinoma. However, previous studies exploring these associations have predominantly focused on patients with advanced-stage NSCLC (9, 10). Therefore, little is known about these relationships in patients with early-stage lung cancer who are eligible for surgical treatment.
Thus, in this study, we conducted a comprehensive analysis of 355 Japanese patients who underwent lung surgery for adenocarcinoma or squamous cell carcinoma. Our aim was to elucidate the differences in clinicopathological characteristics that underlie PD-L1 expression in these two distinct subtypes. Furthermore, we investigated the associations between PD-L1 expression and genetic alterations in well-established oncogenic driver mutations associated with lung cancer.
Patients and Methods
Study population. Samples from 399 Japanese patients who underwent surgical resection for lung cancer from July 2019 to December 2022 at the Respiratory Center of Matsusaka Municipal Hospital were enrolled in the study. The study cohort consisted of 289 cases of lung adenocarcinoma and 66 cases of lung squamous cell carcinoma, with anonymized data. Patients with stage 0 disease were excluded from the analysis. In the case of 19 patients with synchronous multiple lung cancer, one of the highest-grade tumors was selected for inclusion in this study’s analysis. In addition, patient medical and pathological records were reviewed to extract demographic and pathology-specific data. Pathological tumor staging was performed using the eighth edition of the American Joint Committee on Cancer Staging Manual (11). In addition, the Oncomine Dx Targeted Test (ODxTT) (Ion Torrent PGM Dx Sequencer; Thermo Fisher Scientific, Cleveland, OH, USA) was performed in 255 cases of lung adenocarcinoma and 54 cases of lung squamous cell carcinoma to determine the presence of driver gene mutations. Among adenocarcinoma cases, we investigated the association between driver gene mutations and PD-L1 expression. The study protocol received approval from the Institutional Review Board of Matsusaka Municipal Hospital (approval no. J227-230203-5-3, February 2023).
Sample processing. Samples were processed using methods previously described by our group (12, 13). In brief, small tumor tissue samples obtained through endobronchial biopsy/transbronchial biopsy, computed tomography-guided percutaneous needle biopsy, and fine-needle aspiration samples were promptly immersed in 10% neutral buffered formalin (NBF) and fixed at room temperature for 12-18 hours. In cases of surgical lung resection, the limited resection samples, including lung segmentectomy and wedge resections, were immediately placed in 10% NBF after sampling for intraoperative rapid diagnosis (IRD) and fixed at room temperature for 24-48 h. Lobectomy samples obtained between August 2019 and December 2019 were stored in a refrigerator at 4°C for less than 3 hours after sampling for IRD and then put in 10% NBF for 24-48 hours at room temperature. Starting from January 2020, a modified process was adopted for lobectomies, whereby 10 mm × 10 mm samples from tumor-rich areas were simultaneously taken for the Oncomine Dx Targeted Test (ODxTT) during IRD sampling and immediately immersed in 10% NBF. Formalin-fixed tissues were embedded in paraffin to create formalin-fixed paraffin-embedded (FFPE) blocks. LSI Medicine Laboratories (Tokyo, Japan) received 10-15 slide-mounted 5-μm sections from small biopsy samples and 5-10 slide-mounted 5-μm sections from surgical resection samples for assay by ODxTT.
Assessment of genetic testing. ODxTT was used to detect sequence variations in 46 genes on DNA and RNA isolated from FFPE specimens including for targeted therapies on driver mutations in: Epidermal growth factor (EGFR), anaplastic lymphoma kinase (ALK), ROS proto-oncogene 1, receptor tyrosine kinase (ROS1), B-Raf proto-oncogene, serine/threonine kinase (BRAF) (p.V600E), ret proto-oncogene protein (RET), Kirsten rat sarcoma viral oncogene homolog (KRAS), MET proto-oncogene (MET), Erb-B2 receptor tyrosine kinase 2 (ERBB2), and phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha (PIK3CA). LSI Medicine Laboratories performed the ODxTT based on Thermo Fisher’s Ion AmpliSeq technology. Driver gene mutations that could not be determined by ODxTT were measured as much as possible by single-gene test based on real-time polymerase chain reaction, and fluorescence in situ hybridization. These additional analyses were also performed at the central laboratory of LSI Medicine Laboratories.
PD-L1 immunohistochemical staining and scoring. FFPE tumor samples were used for staining to evaluate the PD-L1 expression. The immunostaining for PD-L1 in tumor tissues was conducted at a specialized laboratory at the LSI Medicine Corporation, utilizing the 22C3 pharmDx assay (Dako, Carpinteria, CA, USA). PD-L1 expression was quantified through the application of the tumor proportion score (TPS), which is defined as the percentage of viable tumor cells exhibiting either partial or complete membrane staining. The immunoreactivity rate within tumor tissue in laboratory-generated reports was converted to a continuous variable to assess differences between various groups. Depending on the level of PD-L1 expression, tumors were stratified into three distinct categories: Negative (<1%), low (1-49%), and high (≥50%), according to the TPS determined by assessing at least 100 viable cells.
Statistical analysis. Statistical analysis was performed using SPSS 29.0 (SPSS Inc., Chicago, IL, USA), and values of p<0.05 were considered to indicate statistical significance. Regarding the clinicopathological characteristics of lung adenocarcinoma and lung squamous cell carcinoma, we used Pearson’s chi-square test or Fisher’s exact test to compare proportions. Differences between these two groups were evaluated by the Mann–Whitney U-test and the Kruskal–Wallis exact test. Data represent the mean and standard error of the mean.
Results
Patient characteristics. During the study period, a total of 399 patients underwent lung resection at our Institution. Among these, 289 were diagnosed with lung adenocarcinoma, while 66 received a diagnosis of lung squamous cell carcinoma. In the case of 19 patients with synchronous multiple lung cancer, one of the highest-grade tumors was included in the analysis for this study. In this study, 255 out of 289 lung adenocarcinoma specimens and 54 out of 66 lung squamous cell carcinoma specimens were analyzed by ODxTT. The clinical characteristics of the study population are summarized in Table I. All patients were Japanese, and their median age was 73 years (range=36-89 years) in those with lung adenocarcinoma and 75 years (range=29-88 years) in those with lung squamous cell carcinoma. One hundred and forty (48.4%) patients in the adenocarcinoma group were male, while 57 (86.4%) patients in the squamous cell carcinoma group were male. Notably, the proportion of former or current smokers in the squamous cell carcinoma group (98.5%) was significantly higher than that in the adenocarcinoma group (57.1%).
Characteristics of patients in the lung adenocarcinoma (ADC) and lung squamous cell carcinoma (SqCC) groups.
Surgical interventions predominantly involved lobectomy for both adenocarcinoma (61.2%) and squamous cell carcinoma (69.7%). Most of the patients underwent lung resection without receiving induction therapy, accounting for 98.6% in adenocarcinoma and 95.5% in squamous cell carcinoma. In cases with documented pathological findings, the majority of tumors showed no pleural invasion (75.4% in adenocarcinoma and 71.2% in squamous cell carcinoma). Venous and lymphatic invasion were more frequent in squamous cell carcinoma compared to adenocarcinoma (53% vs. 24.9%, and 63.6% vs. 54.7%, respectively). Early-stage lung cancer, i.e., up to stage IA, accounted for 65.8% of adenocarcinoma and 46.9% of squamous cell carcinoma.
Regarding PD-L1 expression, 147 (50.9%) patients with adenocarcinoma and 18 (27.3%) patients with squamous cell carcinoma registered a negative TPS (<1%). One hundred and two (35.3%) patients with adenocarcinoma had a low TPS (1-49%), while 24 (36.4%) patients with squamous cell carcinoma expressed PD-L1 that level. Thirty-eight (13.1%) patients with adenocarcinoma and 23 (34.8%) patients with squamous cell carcinoma had a high TPS (≥50%).
Clinicopathological features associated with expression of PD-L1. The association between the expression of PD-L1 and the patients’ clinicopathological characteristics was assessed by separating lung cancer patients into two groups (Figure 1). In patients with adenocarcinoma, the expression of PD-L1 was higher in males (p<0.001), smokers (p<0.001), and in patients with invasive pathological factors (pleural invasion: p=0.001, vascular invasion: p<0.001, lymphatic invasion: p<0.001), but no differences were observed comparing between younger (<65 years) and older (≥65 years) patients (p=0.273). In addition, the expression of PD-L1 in lung adenocarcinoma was significantly higher in advanced-stage disease than in early-stage disease (p<0.001). On the other hand, no statistical significance was observed in any of the clinicopathological parameters for the squamous cell carcinoma group.
Clinicopathological comparison of tumor tissue programmed cell death ligand-1 (PD-L1) expression between lung adenocarcinoma and squamous cell carcinoma by age, sex, smoking status, pathological invasive factor, and pathological stage. ADC: Adenocarcinoma; SqCC: squamous cell carcinoma; TPS: tumor proportion score. Significantly different at: **p<0.01 and ***p<0.001.
Association between driver gene mutation and PD-L1 expression. To understand the association between the expression of PD-L1 and tumor driver gene mutations, especially those for which targeted therapeutic agents are available, we assessed 255 lung adenocarcinoma and 54 squamous cell carcinoma cases who were amenable to an analysis by ODxTT. Genetic profiling of the lung adenocarcinoma revealed that approximately 15% of the cases exhibited no discernible mutations (Figure 2A). Among the remaining adenocarcinoma cases, common EGFR mutations (Ex19del or L858R) were the most frequent (43.1%), followed by KRAS (14.7%), MET (5.4%), uncommon EGFR mutations (4.3%), ALK (2.9%), BRAF (2.9%), PIK3CA (2.5%), ERBB2 (1.8%), ROS1 (1.1%), and RET (0.7%). Single driver mutations were present in 89.8% (229/255) of cases, while multiple driver mutations were present in 10.2% (26/255) of cases. Conversely, only 22% of squamous cell carcinoma cases displayed gene mutations, with the remaining 78% lacked mutations in the genes under study (Figure 2B). Unlike adenocarcinoma, PIK3CA mutations were the most prevalent in squamous cell carcinoma (11%). The distribution pattern of TPS for PD-L1 expression according to each driver gene mutation for lung adenocarcinoma is detailed in Table II. Of note, the proportion of high PD-L1 expressions in patients with common EGFR mutations accounted for only 5% (6/119) of cases, which was considerably lower than for other driver mutations. In addition, the distribution pattern of PD-L1 expression and TPS differed for each type of driver mutation (p=0.01) (Figure 2C and D).
Venn diagram of the mutational frequencies in adenocarcinoma (ADC) (A) and squamous cell carcinoma (SqCC) (B). Patterns of programmed cell death ligand-1 (PD-L1) expression by driver mutation in adenocarcinoma showing comparison of frequency (C) and comparison of tumor proportion score (TPS) (D). ALK: Anaplastic lymphoma kinase; BRAF: B-Raf proto-oncogene, serine/threonine kinase; EGFR: epidermal growth factor; ERBB2: Erb-B2 receptor tyrosine kinase 2; KRAS: Kirsten rat sarcoma viral oncogene homolog; MET: MET proto-oncogene;PD-L1: programmed cell death ligand-1; PIK3CA: phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha; RET: ret proto-oncogene protein; ROS1: ROS proto-oncogene 1; WT: wild-type.
The pattern of programmed cell death ligand-1 (PD-L1) expression by driver mutation in lung adenocarcinoma.
Discussion
In the management of early-stage NSCLC, the primary treatment objectives are to enhance cure rates and reduce the risk of tumor recurrence. Despite undergoing surgical intervention with curative intent, a substantial proportion of patients, ranging from 30-70%, will encounter disease recurrence (14). Although adjuvant chemotherapy has contributed to the improvement of overall survival, the benefit remains modest, with a 4-5% enhancement in 5-year OS. Consequently, there has been significant anticipation regarding the potential efficacy of immunotherapy in the adjuvant setting, aimed at reducing the risk of postoperative recurrence among patients with early-stage NSCLC. To date, the phase III IMpower010 trial has recently reported the advantages of adjuvant atezolizumab (anti-PD-L1) when compared to the best supportive care following chemotherapy in patients who underwent complete resection for stage II-IIIA NSCLC (3, 15). Since this trial highlighted significant advantages in the subgroup of patients whose tumors expressed PD-L1 on 1% or more of tumor cells (3), the measurement of PD-L1 will be essential in the selection and personalization of perioperative treatments for patients undergoing lung cancer surgery.
The indications for immunotherapy in advanced-stage NSCLC differ between lung adenocarcinoma and lung squamous cell carcinoma. The current evidence indicates that lung adenocarcinoma with driver mutations, such as EGFR mutations and ALK rearrangements can be effectively treated by small-molecule tyrosine kinase inhibitors (TKIs) (16). In contrast, patients with EGFR-mutated lung adenocarcinoma have been reported to exhibit a low response to anti-PD-1 and PD-L1 inhibitors, resulting in a lack of overall survival improvement with ICIs (17). Therefore, EGFR-TKIs are considered the first-line agents for patients with advanced-stage lung adenocarcinoma with sensitizing EGFR mutations (16). On the other hand, squamous cell carcinoma rarely carries driver mutations and has not shown efficacy of TKI treatment. Therefore, ICIs are selected as the first-line agents for patients with advanced-stage lung squamous cell carcinoma. Even though the use of ICIs in an adjuvant setting is different from that in advanced-stage lung cancer, PD-L1 expression and the presence of driver gene mutations should be taken into account in perioperative treatment.
In this study, 50.9% of adenocarcinoma cases exhibited negative PD-L1 expression (TPS <1%), 35.3% had low PD-L1 expression (TPS 1-49%), and 13.1% had high PD-L1 expression (TPS ≥50%), while the distribution of PD-L1 expression was almost even in squamous cell carcinoma. The difference may be explained by the higher prevalence of EGFR mutations in adenocarcinoma and higher proportion of smokers in the squamous cell carcinoma group. Indeed, our study revealed that EGFR mutations, which have been reported to be associated with negative PD-L1 expression (18), were present in 47% of adenocarcinomas and 2% of squamous cell carcinomas. Although the distribution pattern of PD-L1 expression in adenocarcinoma in our study differs from previous reports (19, 20), this discrepancy may be due to the different prevalence of EGFR mutations between Western and East Asian populations. With regard to the smoking status, 98.5% of patients with squamous cell carcinoma were former or current smokers, in comparison to 55% of patients with adenocarcinoma. The association between smoking status and the high expression of PD-L1 has been well documented. Tobacco carcinogens have the potential to induce tumors with a higher mutational burden, resulting in the generation of more antigens and greater PD-L1 expression (7). Additionally, the inflammatory response triggered by smoking can also stimulate the production of interferon-γ, further up-regulating PD-L1 expression (7).
In our examination of the relationship between pathological factors and the expression of PD-L1, we found that the presence of invasive pathological factors and disease progression had a noticeable impact on PD-L1 expression in lung adenocarcinoma. These results align with previous studies that have shown that advanced-stage disease may be more likely to express PD-L1 in comparison to earlier-stage NSCLC (21, 22). Furthermore, metastatic tumor tissues and lymph nodes have been reported to express higher PD-L1 expression in comparison to primary lung tumors (10). Unlike lung adenocarcinoma, PD-L1 expression in lung squamous cell carcinoma was not significantly associated with these pathological factors or disease progression.
The activation of PD-L1 is known to be modulated by oncogenic signaling pathways, including ALK/signal transducer and activator of transcription 3 and extracellular-signal-regulated kinases/mitogen-activated protein kinase (9). Given the involvement of the activation and expression of PD-L1 in oncogenic signaling pathways, alterations in driver genes may also affect their activation and expression. In our study, driver gene mutations were found in 85% of adenocarcinomas but in only 22% of squamous cell carcinomas. Among lung adenocarcinoma, cases with common EGFR mutations tended to have lower PD-L1 expression levels in comparison to cases with other driver gene mutations. In addition, our study is consistent with previous reports demonstrating that cases with common EGFR mutations had lower PD-L1 expression levels in comparison to EGFR wild-type cases (7, 23). Among the driver gene mutations in lung adenocarcinoma, high TPS was most frequently observed in RET mutation (95%), followed by BRAF (40.7%), ALK (34.4%), KRAS (24.3%), and MET (23%). These groups of genetic mutations with high TPS have the potential to improve cancer prognosis through combination therapy with ICIs and targeted agents. However, in the TATTON trial, combination therapy with osimertinib (a third-generation EGFR-TKI) and durvalumab (anti-PD-L1) for patients with EGFR-mutated lung adenocarcinoma was reported to be infeasible due to increased interstitial pneumonitis (24). Although combination therapy with ICIs and targeted agents may be associated with better survival in comparison to monotherapy, the potential risk of serious side-effects from such therapy should be taken into account.
The present study is subject to several limitations that warrant acknowledgement. Firstly, this study was retrospective in nature and carried out at a single center. Therefore, the possibility of unintentional selection bias cannot be entirely ruled out. Secondly, the number of patients involved in the study was relatively limited, especially for specific genomic subgroups in the analysis of the association between driver mutations and PD-L1 expression. Thirdly, PD-L1 immunohistochemistry was performed using a single antibody in this study. Our prior investigation revealed discordance in PD-L1 expression within a sample when different antibodies were used (4). To address this concern, future studies should consider the utilization of multiple antibodies for evaluation.
Conclusion
Our study has uncovered notable differences in the distribution patterns of PD-L1 expression between lung adenocarcinoma and lung squamous cell carcinoma. Furthermore, we identified several factors influencing PD-L1 expression in adenocarcinoma, including male sex, smoking status, the presence of invasive pathological factors, and disease progression. In contrast, none of these factors exhibited a significant association with PD-L1 expression in squamous cell carcinoma. In the case of adenocarcinoma, the distribution pattern of PD-L1 expression varied based on the specific type of driver gene mutation. These results were based on real-world data, which provide useful information and new insight into the perioperative management of lung adenocarcinoma and squamous cell carcinoma.
Footnotes
Authors’ Contributions
A.I.: Conceptualization, investigation, data curation, formal analysis, and writing-original draft. K.I.: Project administration and supervision. T.T.: Investigation and data curation. Y.S.: Writing–review and editing. T.S.: Writing–review and editing. K.F.: Writing–review and editing. Y.N.: Writing–review and editing. O.T.: Writing–review and editing. H.Y.: Writing–review and editing. O.H.: Conceptualization and supervision. All Authors contributed to the critical revision of the article and approved its final version.
Conflicts of Interest
Matsusaka Municipal Hospital received research grant funding from Novartis, GlaxoSmithKline, AstraZeneca, Daiichi Snakyo, Bayer, and Boehringer Ingelheim. K. Ito has received speaker fees as honoraria Eli Lilly Japan, Chugai, AstraZeneka, MSD, Boehringer Ingelheim Japan, Ono, and Pfizer Japan. O. Taguchi received speaker fees as honoraria from AstraZeneca. O. Hataji received speaker fees as honoraria from Novartis Pharma, AstraZeneca, and Boehringer Ingelheim Japan. The remaining authors declare no conflict of interest.
Funding
This work was not supported by any funding sources.
- Received August 10, 2023.
- Revision received September 21, 2023.
- Accepted September 25, 2023.
- Copyright © 2023 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
References
In this issue
Jump to section
Related Articles
Cited By...
- Clinicopathological and Molecular Insights into Primary Retroperitoneal Squamous Cell Carcinoma: HPV Association and Discovery of Recurrent PIK3CA E545K Mutation
- MiR-140-3p Improves Sensitivity to Docetaxel by Suppressing PD-L1/ABCG2/MVP Expression in Lung Adenocarcinoma
- Prediction of Responsiveness to PD-L1/PD-1 Inhibitors Using miRNA Profiles Associated With PD-L1 Expression in Lung Adenocarcinoma and Squamous Cell Carcinoma