The Frequency of DNA Mismatch Repair Deficiency Is Very Low in Surgically Resected Lung Carcinoma

Introduction DNA mismatch repair (MMR) deficiency leads to changes in the length of nucleotide repeat sequences of tumor DNA. In that situation, DNA replicational errors occur and accumulate during DNA replication. As a result, this mechanism frequently affects the coding regions of oncogenes and tumor suppressor genes and causes carcinogenesis. Recently, DNA MMR deficiency has been recognized as a predictive biomarker for immunotherapy. The aim of this study is to examine the frequency of DNA MMR deficiency and clinicopathological characteristics in surgically resected lung carcinoma (LC) and their correlation. Methods A total of 1153 LCs were examined. Tissue microarrays were constructed. The status of MMR deficiency was evaluated by immunohistochemical analysis of MMR protein expression (hMLH1, hMSH2, hMSH6, and hPMS2). Microsatellite instability analysis, BRAF mutation, and MLH1 methylation analysis were performed for cases that showed MMR deficiency. Results Only 2 of the 1153 cases (0.17%) showed a loss of hMLH1/hPMS2 protein expression. They also had high levels of microsatellite instability (MSI-H), had neither MLH1 promoter methylation nor BRAF mutation, and were male smokers. Histopathologically, one was a squamous cell carcinoma, and the other was combined small cell carcinoma with squamous cell carcinoma. Regarding PD-L1 protein expression, one had high expression, and the other had none. Conclusion The frequency of MMR deficiency was very low in LC. However, our two cases were non-adenocarcinoma and differed from previous studies. Because of its very low frequency, MMR deficiency is not a practical biomarker to predict the effect of immune checkpoint inhibitors in LC.


INTRODUCTION
Lung carcinoma (LC) is the leading cause of cancer-related death in developed countries (1)(2)(3). Although multidisciplinary therapy has improved the outcome for LC patients, most of them are diagnosed at an advanced stage, and the 5-year survival rate for LC is still about 18% (1)(2)(3). Recently, immunotherapy using immune checkpoint-blocking antibodies targeting programmed death 1/ programmed death ligand 1 (PD-1/PD-L1) has improved the outcomes for patients with a variety of malignant tumors (4,5). PD-1 is expressed on the surface of T cells, while PD-L1 and PD-L2 are expressed on the surface of various cells, including many malignant cells. PD-1 negatively controls its activity by way of interaction with its ligands PD-L1 and PD-L2. This interaction decreases the T-cell activity, resulting in tumor cell avoidance of the immune system (4,6). Pembrolizumab is one of a humanized monoclonal antibodies against PD-1 that has antitumor activity in advanced non-small cell lung carcinoma. Its therapeutic effect is closely related to PD-L1 expression in cancer cells, and PD-L1 protein expression has been suggested to be a predictive biomarker of the response to immunotherapy (7). Therefore, the status of PD-L1 protein expression has been suggested as a predictive biomarker for immunotherapy.
Meanwhile, DNA mismatch repair (MMR) deficiency has also been recognized as a predictive biomarker for immunotherapy (8)(9)(10). DNA MMR deficiency leads to changes in the length of nucleotide repeat sequences of tumor DNA, and a phenomenon called microsatellite instability (MSI) occurs (11). In that situation, DNA replicational errors occur that accumulate during the DNA replication cycle. As a result, this mechanism frequently affects the coding regions of oncogenes and tumor suppressor genes, causing carcinogenesis (12). Le et al. have reported that MMR deficiency predicts the response of various malignant tumors to PD-1 blockade (9). Therefore, it is important to investigate the frequency of DNA MMR deficiency in LC. However, there are few publications regarding the relationship between DNA MMR deficiency and LC, and their relationship is thus inadequately known.
In the present study, we studied the frequency of DNA MMR deficiency and clinicopathological characteristics in surgically resected LC and examined their correlation.

Patients
A retrospective review of a prospectively maintained surgical database was performed to identify patients who underwent primary LC resection with curative intent from 2008 to 2018 (patients from 2008 to 2014, Yamagata Prefectural Central Hospital; patients from 2015 to 2018, Iwate Medical University). Histopathologic diagnoses had been made in accordance with the eighth edition of the TNM Classification of the Union for International Cancer Control and the 2015 WHO classification (13,14). Patients were excluded from the current evaluation if they underwent neoadjuvant therapy, underwent incomplete resection, had multiple primary lung cancers, or had incomplete follow-up data. In the end, a total of 1153 patients with primary LC were included. This study was approved by the Institutional Review Board of both Iwate Medical University (reference number: MH2020-163) and Yamagata Prefectural Central Hospital (reference number: 15) and was conducted according to the principles of the Declaration of Helsinki. Written informed consent was waived because this was a retrospective study, the patient data remained anonymous, and an opt-out approach was used.

Tissue Samples and Tissue Microarray Preparation
A total of 1153 formalin-fixed, paraffin-embedded samples from consecutive resected LC collected from 2008 to 2018 were used for the preparation of tissue microarrays (TMAs). Briefly, we marked one representative tumor area and arrayed a cylindrical 3-mm tissue core from the corresponding paraffin blocks into a recipient block using a tissue arrayer (KIN-2, Azumaya, Japan). Hematoxylin and eosin (HE) staining was used to evaluate tumor cells in each TMA specimen.

Evaluation of MMR Protein Expression
MMR protein expression was examined by immunohistochemistry (IHC). The TMA blocks were sliced to 4-mm thickness, deparaffinized, and stained for anti-hMLH1 (clone: M1), anti-hMSH2 (clone: G219-1129), anti-hMSH6 (clone: SP93), and anti hPMS2 (clone: A16-4) (Ventana Medical Systems, Inc., Tucson, AZ) on a Benchmark XT system with an automated staining protocol. Tumors showing a total absence of nuclear staining, with the adjacent normal tissue showing the presence of nuclear staining, were regarded as having lost MMR protein expression. Lymphoid cells and stromal cells served as the internal positive staining controls. Two certified pathologists (NY and NU or MO) used a multi-headed microscope to evaluate the slides together and discussed until the consensus was acquired.

Evaluation of PD-L1 Protein Expression and p53 Protein Expression
PD-L1 protein expression was examined by IHC. The TMA blocks were sliced at 4-mm thickness, deparaffinized, and stained for PD-L1 using the PD-L1 IHC 22C3 pharmDx assay (Agilent Technologies, Inc., Santa Clara, CA, USA) and the Autostainer Link 48 using an automated staining protocol. Two certified pathologists (NY and NU or MO) evaluated the slides. If the cell membrane of the tumor cells was stained, the cells were considered positive for PD-L1 protein expression. The tumor proportion score (TPS) of PD-L1 in tumor cells was defined as the percentage of PD-L1 positive tumor cells among all tumor cells in the TMA tumor sections and was estimated in increments of 5%, except for the 1% value. We classified PD-L1 expression into three patterns based on the TPS as follows: no expression if the TPS was <1%, low expression if the TPS was 1-49%, and high expression if the TPS was ≥50%. We also performed immunohistochemical staining using antibodies against p53 (clone:DO-7, Roche) in the TMAs. The immunohistochemical result was evaluated as follows: staining of p53 protein expression in ≥10% of tumor cells were defined as p53 positivity.

BRAF Mutation and MLH1 Methylation Analyses
Mutations in BRAF (V600E) genes were examined using a PyroMark Q24 pyrosequencer, as described previously (18,19). Each reaction contained 1 × PCR buffer, 1.5 mM MgCl 2 , 0.2 mM each dNTP, 5 pmol forward primer, 5 pmol reverse primer (biotinylated), 0.8 U HotStarTaq DNA polymerase (Qiagen), 10 ng template DNA, and dH 2 O to a final volume of 25 mL. Cycling conditions were as follows: 95°C for 15 min; 38 cycles of 95°C for 20 s, 53°C for 30 s, and 72°C for 20 s; and a final extension at 72°C for 5 min, with holding at 8°C. Following amplification, 10 mL biotinylated PCR product was immobilized on streptavidin-coated sepharose beads (Streptavidin Sepharose High Performance; GE Healthcare Bio-Sciences Corp., Piscataway, NJ, USA) and washed in 70% EtOH. The purified biotinylated PCR products were loaded into the PyroMark Q24 (Qiagen) using PyroMark Gold reagents (Qiagen) containing 0.3 lM of the sequencing primer and annealing buffer. Regarding MLH1 methylation, pyrosequencing for methylation of MLH1 was performed using a PyroMark Q24 instrument (Qiagen, Valencia, CA, USA), as described in a previous study (20,21). Briefly, the PCR product (25 mL) was bound to streptavidin Sepharose HP (GE Healthcare, Brøndby, Denmark), purified, washed, denatured in 0.2 M NaOH, and washed again. Before pyrosequencing, 0.3 mM sequencing primer was annealed to the purified single-stranded PCR product by heating to 80°C for 2 min. The primers were designed with the Pyromark Assay Design Software package (Qiagen NV) and included 3 or 4 CpG sites to analyze the methylation of the promoter. The cutoff value for positive methylation was set as 30% of the tumor cells (20).

The Frequency of DNA MMR Deficiency and Its Correlation With Clinicopathological and Molecular Characteristics
Loss of MMR protein expression was only observed in 2 of 1153 cases (0.17%). Both of them showed loss of hMLH1 and hPMS2 protein expression (Figure 1). They also had MSI-H ( Figure 2). They had neither MLH1 promoter methylation nor BRAF mutation. Briefly, both were males and smokers. One was pathological stage I and the other was stage II. One is alive and the other was died in 6 months after surgical resection because of recurrence. One had a history of gastric carcinoma, and the other had a history of colon carcinoma. Neither had any remarkable family history and not adapted to criteria about Lynch syndrome (22,23). Histopathologically, one was a squamous cell carcinoma, and the other was a combined small cell carcinoma with squamous cell carcinoma. The latter showed loss of hMLH1/ hPMS2 protein expression together with the presence of MSI-H only in the squamous cell carcinoma area. We also examined immunohistochemistry for our cases. We used four antibodies as follows: TTF-1 (8G7G3/1, Dako, Glostrup, Denmark), Napsin A (IP64, Nichirei, Tokyo, Japan), p40 (BC28, Biocare Medical, Pacheco, CA, USA) and CK5/6 (D5/16 B4, Dako, Glostrup, Denmark). p40 and CK 5/6 were both positive (Figure 1), on the other hand, TTF-1 and Napsin A were both negative. One had high expression of PD-L1 protein while the other had none, and both showed overexpression of p53 protein.  (26). The frequency of MMR deficiency was from 0.3% to 0.8%. Our frequency of 0.17% was a little lower than the 0.3% to 0.8% of these previous studies. Differences in the method of detection may underlie the small differences among these studies. Results from these previous studies are shown together with our results in Table 2. Note that all MSI-H patients for whom smoking history was reported were smokers. MMR-deficient tumors have been reported to show a good response to pembrolizumab in some types of carcinoma that were reported to have a higher nonsynonymous mutation burden (10,27,28). Tumor mutation burden is used as a biomarker to predict a good response to immunotherapy because tumors with a high mutation burden present neoepitopes that can behave as neoantigens (29). Tobacco smoke carcinogens induce DNA damage, p53 mutation, and nonsynonymous mutation burden (27,30,31), and the molecular signature with smoking is significantly correlated  with nonsynonymous mutation burden (25,27). Higher nonsynonymous mutation burden in tumors is associated with improved objective response, durable clinical benefit, and progression-free survival after treatment with immune checkpoint inhibitor (32). Thus, we consider that the tumor of our MMR-deficient LC patient showed mutational signatures associated with both smoking and DNA MMR deficiency. As to histopathological subtypes, our cases were not adenocarcinomas. Only adenocarcinomas were examined in the two previous studies (24,25), while another study did not report histological subtypes (26). MSI is frequently observed in endometrial carcinoma, gastric and colorectal carcinoma, and small intestine malignancies, and most of these are considered to be adenocarcinoma (26). On the other hand, MSI was not detected in head and neck squamous carcinoma and esophageal carcinoma, which are considered to be squamous cell carcinomas (26). Therefore, we think our cases were rare.

DISCUSSION
In sporadic colon and gastric carcinoma, BRAF mutation and the epigenetic inactivation of MLH1 expression by promoter methylation result in MMR deficiency (17,33). Our cases showed loss of hMLH1/hPMS2 protein expression and MSI-H, and therefore we examined BRAF mutation and MLH1 promoter  This study has some limitations. First, we used TMAs rather than whole tissue sections, possibly missing changes in consequence of heterogeneous immunohistochemical staining. Secondly, we assessed the deficiency of MMR protein by using IHC prior to MSI analysis, whereas previous studies did MSI analysis in advance of determining the deficiency of MMR protein using IHC. Thirdly, we could not examine somatic MLH1 gene mutation.

CONCLUSION
The frequency of MMR deficiency was very low in LC. Interestingly, our cases were not adenocarcinomas and were thus different from those of previous studies. Because its frequency is so very low, MMR deficiency does not appear to be a practical biomarker to predict the effect of immune checkpoint inhibitors in LC.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.

ETHICS STATEMENT
This study was approved by the Institutional Review Board of both Iwate Medical University (MH2020-163) and Yamagata Prefectural Central Hospital (15) and was conducted according to the principles of the Declaration of Helsinki. Written informed consent was waived because this was a retrospective study, the patient data remained anonymous, and an opt-out approach was used.

AUTHOR CONTRIBUTIONS
Conception and writing of the manuscript: NaY and TS. Collection of the clinical data: NaY, SS, ME, MM, and HS. Pathological diagnosis and immunohistochemical analyses: NaY, S-yO, NoY, RS, MO, and NU. Collection of the samples for the molecular analyses: NaY. All authors contributed to the article and approved the submitted version.