Search for a command to run...
Immune checkpoint inhibitors (ICIs) have revolutionized the treatment landscape for non-small-cell lung cancer (NSCLC), but predictive biomarkers for response remain limited. Quantification of Programmed Cell Death Ligand 1 (PD-L1) expression using a phosphor-integrated dot (PID) score has been shown to predict ICI efficacy in NSCLC and other cancers (1). However, PD-L1 expression is not always a reliable predictor, particularly in patients with low PD-L1 levels, highlighting the need for alternative biomarkers.Akkermansia muciniphila (Akk) plays multiple roles and has beneficial effects on systemic metabolism, immunity, the intestinal barrier, and tumor progression (2)(3)(4). A Study has shown that intestinal Akk is overrepresented in cancer patients with progression-free survival (PFS) of more than 3 months or in ICI responders (4). A recent cohort study also demonstrated that intestinal Akk was associated with ICI efficacy in NSCLC patients, particularly in those with low PD-L1 expression (2). Furthermore, in a lung cancer animal model, it was observed that intestinal Akk could enter the bloodstream and subsequently colonize lung cancer tissue (5).Based on these findings, we hypothesized that tumor-associated Akk may be related to prognosis or ICI efficacy in NSCLC patients, independent of PD-L1 expression. In this study, we investigated the clinical significance of tumor Akk expression as assessed by immunohistochemistry (IHC) in NSCLC patients treated with ICIs.Material and methodsThis study was conducted in accordance with the guidelines of the Declaration of Helsinki and was approved by the Ethics Committees of Showa Medical University School of Medicine (approval number: 2772) and Fukushima Medical University (approval number: 2019-262). Informed consent was obtained from all patients involved in the study.This study enrolled 60 NSCLC patients with metastatic or recurrent cancer who were treated with ICIs. It was a multicenter retrospective cohort study, and patients were diagnosed and treated at Showa Medical University Hospital and Fukushima Medical University Hospital from December 2015 to December 2022. All patients received treatment regimens, including ICIs as shown in Table 1, which were administered according to clinical practice.Each patient's treatment response was evaluated using computed tomography scans for imaging assessments. Treatment efficacy was evaluated according to the Response Evaluation Criteria in Solid Tumors version 1.1 (6). Overall survival (OS) was defined as the time from the start of the first administration of treatment to the date of death from any cause or the last follow-up. Progression-free survival (PFS) was defined as the time from the start of treatment to the first documented instance of disease progression, death from any cause, or the last follow-up, whichever occurred first. The cut-off date for follow-up was set as December 2022.The "median PFS" and "median OS" from phase III pivotal clinical trials were used to uniformly evaluate the treatment efficacy of patient populations. The patients were divided into two groups (responders and non-responders) based on their treatment response. We then performed an analysis to compare Akk and PD-L1 expression in each group.All tumor tissue specimens used to evaluate Akk expression were obtained before each patient received ICI treatment. The immunohistochemistry (IHC) staining procedure using DAB, and the method of evaluating Akk expression, were performed according to standard clinical protocols. 60 Formalin-fixed, paraffin-embedded (FFPE) tissue samples obtained by biopsy or resection were prepared for analysis. FFPE cut to a thickness of 4 μm were deparaffinized and stained using an automated immuno-stainer BOND-III (Leica Biosystems, Germany) using the manufacturer's IHC protocol. Antigen retrieval was performed using the BOND Enzyme Pretreatment Kit (Leica Biosystems, AR9551) at 37°C for 10 min, followed by incubation with Akkermansia muciniphila primary antibody (1:500, Sigma-Aldrich, SAB4200870, Germany) at room temperature for 15 min. DAB detection was performed using the BOND Polymer Refine Detection (Leica Biosystems, DS9800). Three independent pathologists independently evaluated all 60 immuno-stained slides. We defined Akk positivity as clear intracellular staining of Akk in multiple fields of view. Appropriate positive controls were included in each IHC run. The same tumor tissue specimens were used to evaluate PD-L1 expression. The method for PD-L1 PID scoring was conducted as previously reported (1).We also performed immunofluorescence staining of CD3 and CD68 on specimens with low PD-L1 expression (n=30). We sliced 4-μm sections from paraffin blocks and placed them on glass slides, which were then deparaffinized and rehydrated. ). An All-in-One Fluorescence Microscope (BZ-X810, KEYENCE, OSAKA, Japan) was used to assess the positivity according to double immunostaining, as previously reported (7).Total RNA from tumor tissues was isolated by RNeasy Micro kit (Qiagen, Netherlands). The RNA integrity score was calculated with the TapeStation High Sensitivity RNA Kit (Agilent, CA, USA) in a 2200 TapeStation (Agilent, CA, USA). RNA-Seq libraries were prepared with the SMART-Seq® Stranded Kit (# 634444, Takara Bio, Japan). The libraries were sequenced on the NovaSeq 6000 system (Illumina) as paired-end 150 base reads.RNA libraries were sequenced on an Illumina NovaSeq 6000 platform, generating 2 × 150 bp pairedend reads. Read alignment was performed using STAR (version 2.7.10a) with default parameters, mapping to the human genome (GRCh38) and transcriptome (GENCODE version 40) as reference datasets. Gene expression levels were quantified as fragments per kilobase of exon per million reads mapped (FPKM) using StringTie (version 2.2.1).To visualize the effects of Akk positivity on FPKM expression levels between Akk positive and Akk negative in all specimens (n=6 per group), PD-L1 low specimens (n=3 per group) and PD-L1 high specimens (n=3 per group).For RNA-sequencing data in all specimens, PD-L1 low specimens and PD-L1 high specimens, gene annotation enrichment analysis was performed for KEGG pathway analysis, using the functional annotation tool in DAVID Bioinformatics Resources 2021 (https://davidbioinformatics.nih.gov/home.jsp).Statistical analyses were performed, and figures were generated using GraphPad Prism 8.4.3 software (GraphPad Software Inc., San Diego, CA, USA) or JMP software (SAS institute., NC, USA). Spearman's correlation coefficient was used to analyze the associations between variables. Un-paired t test was used to compare values between two groups. Statistical significance was defined as a p-value <0.05.For survival analyses, survival durations (PFS and OS) were assessed using the Kaplan-Meier method and Cox proportional hazard model. All tests were two-sided. When comparing two groups using the log-rank test, p-values <0.05 were considered statistically significant.Representative images of Akk-negative and Akk-positive tumors are shown in Figure 1 We examined the correlation between tumor Akk and PD-L1 expression using the PID method. Tumor Akk positivity was not significantly associated with the PD-L1 PID score (Figure 2, P=0.082, unpaired t-test).Kaplan-Meier survival analyses with log-rank tests were performed to compare progression-free survival (PFS) and overall survival (OS) between the Akk-negative and Akk-positive groups. In the overall patient population (n=60), there was no significant difference in PFS or OS between the two groups (Figure 3, Hazard ratio = 1.18 (0.67-2.07) or 1.25 (0.64-2.43)). Similarly, in patients with high PD-L1 expression, Akk status was not associated with either PFS or OS (Figure 4, Hazard ratio = 0.78 (0.34-1.79) or 0.79 (0.27-2.32)). However, among patients with low PD-L1 expression, Akkpositive patients had worse PFS compared to Akk-negative patients (Figure 5, Hazard ratio = 2.31 (1.01-5.32), P = 0.0487), although there was no significant difference in OS (Hazard ratio = 1.76 (0.74-4.19)).Univariable or multivariable Cox regression analysis was performed to adjust for potential confounders (sex, age, PD-L1 expression, or ICI regimen). There was no significant difference in PFS or OS between the two groups (Table 2).In summary, these results suggest that tumor Akk expression may serve as a predictive marker for ICI efficacy in NSCLC patients with low PD-L1 expression, but not in those with high PD-L1 expression.RNA-sequencing was performed to understand how tumor Akk positivity effected the gene expressions in all specimens (n=6 per group), PD-L1 low ones (n=3 per group) and PD-L1 high ones (n=3 per group). All RNA-sequencing results are provided in Supplementary Table S1. A total of 7,854 genes were expressed in one of the replicates in all groups. In Akk positive samples, compared to Akk negative samples, the groups of genes showing the top 100 enhanced or suppressed expression were defined as upregulated or downregulated genes, respectively. For these genes, Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis was performed using the Database for Annotation, Visualization, and Integrated Discovery (DAVID software). In all samples or PD-L1 low ones, Akk positivity upregulated pathways linked to amyotrophic lateral sclerosis and oxidative phosphorylation, while downregulating pathways linked to Ribosome biogenesis in eukaryotes, Ribosome, and spliceosome (Figure 6A). On the other hand, in PD-L1 low ones, Akk positivity suppressed these pathways other than a spliceosome pathway (Supplementary TableS2). ). On the other hand, in PD-L1 high ones, Akk positivity suppressed genes related to these pathways other than spliceosome.Because Akk induces homeostatic immune responses (3), we evaluated Tumor-Infiltrating Immune Cells, such as CD3+ T cells or CD68+ macrophages, in specimens with PD-L1 low expression (Figure 7A,n=30). The number of CD3+ T cells or CD68+ macrophages between Akk-negative (n=12) and Akk-positive tumors (n=18) were not different (Figure 7B, P=0.64 or 0.76, unpaired ttest).Here, we report the results of a retrospective, multicenter study on NSCLC patients treated with ICIs. Tumor Akk expression was associated with poor prognosis or non-response to ICIs in patients with low PD-L1 expression, but not in those with high PD-L1 expression. This result contrasts with findings regarding intestinal Akk, where ICI efficacy and prognosis were better in patients with a low abundance of intestinal Akk compared to those with high abundance or absence. This tendency was particularly observed in patients with low PD-L1 expression, but not in those with high PD-L1 expression (2). Recently, in a lung cancer animal model, it has been shown that intestinal Akk can enter the bloodstream and subsequently colonize lung cancer tissue (5). It is possible that the effect of tumor Akk on cancer immune response differs from that of intestinal Akk. Recent studies have highlighted the complexity of the tumor microbiome and its impact on cancer therapy. Nejman et al. demonstrated that various tumor types harbor distinct intracellular bacteria, which can modulate immune responses and influence treatment outcomes (8). In addition to Akk, other genera such as Bifidobacterium and Ruminococcus have been implicated in modulating ICI efficacy (9). The functional consequences of these bacteria within the tumor microenvironment remain to be fully elucidated.Within the tumor microenvironment, Akk may interact with immune cells, modulate metabolic pathways, or influence the local immune milieu. Our transcriptomic analysis revealed that Akk positivity in PD-L1-low tumors was associated with upregulation of oxidative phosphorylation and neurodegenerative disease pathways, and downregulation of spliceosome-related pathways, suggesting a potential impact on tumor cell metabolism and immune regulation. Low intestinal Akk expression correlates with increased oxidative stress and inflammatory responses (10). Although multiple models have shown that Akk in intestinal bacteria may contribute to the pathogenesis of ALS via oxidative stress alleviation and activation of the PI3K/Akt pathway (11) (12), there is currently no clear evidence regarding its association with the spliceosome. Alternative splicing of PD-1 or PD-L1 induces the production of soluble PD-1 or PD-L1, which suppresses the tumor immune response (13). However, it remains unclear how alternative splicing affects other immune checkpoints. Although this is the first report of transcriptome analysis of tumor Akk, it is necessary to elucidate the interaction between Akk and tumor microenvironment constituent cells and its relationship with spliceosomes in the future.Recent findings indicate that the majority of intratumoral microbiota are encapsulated within tumorinfiltrating macrophages ( 14). It has also shown that Akk secretes threonyl-tRNA synthetase (AmTARS) riggers M2 macrophage polarization and orchestrates the production of antiinflammatory IL-10 (15). Furthermore, in vitro studies on the interaction between Akk and macrophages revealed that repeated exposure to Akk induced the upregulation of certain immune checkpoints, such as the Siglec family, but not PD-L1 (16). This resulted in increased bacterial intracellular survival and reduced inflammation. In this study, these immune checkpoints could not be detected well by RNA-sequencing. Therefore, Akk derived proteins or metabolites could enhance the expressions of immune checkpoint factors other than PD-L1. Further molecular studies are needed to elucidate this pathogenesis. This is the first study to detect proteins derived from tumor microbes of NSCLC patients, using a anti Akk antibody. Until now, methods for studying tumor microbes have focused on targeting DNA or RNA (14). However, detecting DNA or RNA in FFPE samples can sometimes be challenging compared to detecting proteins. This new method could improve our understanding of the tumor microbiome in various cancers.The limitations of this study are as follows: First, it was a retrospective analysis and primarily exploratory, examining the utility of tumor Akk expression through immunohistochemistry (IHC). Second, unlike previous studies on tumor microbiota that employed RNA in situ analysis (10), we did not conduct such an analysis here. Third, interactions between tumor mutational burden and tumor Akk expression were not considered. Fourth, RNA sequencing results varied widely because the quality of FFPE-derived RNA varied widely from sample to sample. Additionally, this study did not investigate the functional and clinical relevance of Akk commensals, such as Bifidobacterium adolescentis (2). Nevertheless, our results suggest that tumor microbiota profiling, particularly targeting Akk, may provide additional prognostic information, especially for NSCLC patients with low PD-L1 expression. Combining tumor microbiome analysis with established biomarkers such as PD-L1, tumor mutation burden, and immune cell infiltration could improve patient stratification and guide therapeutic decision-making. Further studies are warranted to validate these findings and explore the therapeutic potential of modulating the tumor microbiome.In conclusion, our study suggests that tumor-associated Akk may serve as a negative predictive biomarker for ICI efficacy in NSCLC patients with low PD-L1 expression. Tumor microbiota profiling could represent a promising approach to refine patient selection and optimize immunotherapy strategies.The present study was supported by Konica Minolta, Inc. (Tokyo, Japan). The QUIK software used for data analysis in the present study was also supplied by this company. The authors declare that they have no competing interests. The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.1211 1312 Tables Table 1. Clinical, pathological and molecular characteristics of non-small lung cancer cases according to tumor Akkermansia muciniphila (Akk) expression.