Culturable and unculturable periodontal microorganisms in non-small cell lung cancer: an exploratory analysis

Abstract

The impact of the periodontitis-associated microbiota on lung cancer patients remains controversial and poorly explored. This study aims to characterize the periodontal microbiota in saliva and subgingival plaque of patients with non-small cell lung cancer (NSCLC) and healthy volunteers. Twenty-four individuals (12 with NSCLC and 12 healthy controls) were included, matched by age and stage of periodontitis. Periodontal clinical examination, saliva, and subgingival plaque sampling were processed using qPCR techniques to determine culturable and non-culturable periodontopathic microorganisms. Microbiological results revealed differences in saliva samples of NSCLC patients, with a significant decrease in the frequency of Treponema denticola (p = 0.019) and Eubacterium brachy (p = 0.045). Furthermore, a significant reduction was demonstrated in bacterial concentrations of Eubacterium brachy in plaque (p = 0.027) and saliva (p = 0.0076) in NSCLC patients. In contrast, Desulfobulbus oralis showed higher frequency (p = 0.05) and concentration (p = 0.038) in subgingival plaque in these patients. Porphyromonas gingivalis was not associated with NSCLC. Desulfobulbus oralis should be studied as a potential microbiota dysbiosis biomarker in NSCLC. These findings highlight the potential diagnostic value of periodontal microbiota in NSCLC.

Introduction

Lung cancer (LC) was the most commonly occurring cancer worldwide, with 2.5 million new cases accounting for 12.4% of the total new cases¹. In addition, LC was the leading cause of cancer death (1.8 million deaths, 18.7% of the total cancer deaths), and its re-emergence as the most common cancer is likely related to persistent tobacco use in Asia^1,2. About 60% of the world’s LC incidence is attributed to low- and middle-income countries, with the majority noted in the Asian continent². LC is classified into two main types: non-small cell carcinoma (NSCLC), which accounts for 80% of cases, and small cell carcinoma (SCLC), which comprises the remaining 20%³. The tobacco cessation efforts by the World Health Organization (WHO) have been instrumental in controlling the tobacco epidemic. The 4th edition of the WHO global report on the trends in the prevalence of tobacco use reported a global decline in tobacco use from 32.7% in 2000 to 22.3% in 2020⁴. However, some Southeast Asian and Middle Eastern countries still have a tobacco use prevalence as high as 50%⁴. The incidence of lung cancer in non-smokers is increasing globally¹. Most Asian and Latin women with lung cancer are non-smokers due to risk factors such as outdoor air pollution, environmental exposures (arsenic, asbestos, biomass exposure), genetic factors, and pulmonary conditions like pulmonary fibrosis⁵. In recent years, in addition to these traditional factors, the role of the oral microbiome as an additional risk factor for developing this disease has begun to be recognized⁶. The correlation between microbiota dysbiosis and cancer has gained extensive attention and has been widely explored⁶. Although previous findings uncovered the critical roles of microbiota in tumorigenesis and response to immunotherapy, most of them were focused on the intestinal microbiota rather than the oral and lung microbiota^7,8.

Lung cancer has been widely considered to be a complex disease caused by interactions between host and environmental factors⁹. Among diverse ecological risk factors, microbes maintain microecological balance and regulate host immune responses to multiple treatments¹⁰. Although healthy lung tissues were long considered a sterile environment, it was recently found that certain microbial species exist in the lung, impacting the balance between health and pathogenesis in this organ-microenvironment¹¹. Many of the bacteria found in the lungs come from the microaspiration of the oral cavity¹². In observational studies, there was an association between oral commensal bacteria (especially phylum Firmicutes) and LC^12,13. On the contrary, no relationship has been observed between the presence of the phylum Bacteroidetes and LC, with a significant decrease in saliva and bronchoalveolar fluid (BALF)^14–16. Despite this, a significant association has been reported between periodontitis and LC, suggesting that periodontal diseases could be related to the development of this neoplasia¹⁶. Periodontitis is an inflammatory chronic disease that responds to changes in the oral microbiome, including the Bacteroidetes and another unculturable microorganism in the sulcus or periodontal pocket, also present in the saliva^17,18. However, periodontopathic microorganisms have shown a reduction in saliva and BALF in LC individuals^14,15. This discrepancy in the microbial component between saliva and subgingival plaque could be due to a marked difference in the microenvironment between saliva and subgingival plaque in this type of cancer, which may affect the colonization and growth of microorganisms. Studies linking the oral microbiome to LC have been performed on saliva samples without considering subgingival plaque microorganisms and their potential impact on pathology and immunotherapy treatment. This study aims to characterize the culturable and unculturable target periodontal microorganisms in saliva and subgingival plaque of patients with NSCLC compared with individuals without LC.

Methods

Study design

This cross-sectional study was conducted from March 2022 to October 2024. Universidad El Bosque Ethical Committee approved it in accordance with the ethical guidelines governing health research on humans issued by the Colombian Ministry of Health (Acta No. 012–2021). All participants signed an informed consent form and adhered to the Helsinki Declaration’s international ethical guidelines for human experimentation.

Inclusion criteria

All patients admitted to the Thoracic Oncology Unit of The Luis Carlos Sarmiento Angulo Cancer Treatment and Research Center diagnosed with recurrent or metastatic NSCLC who are candidates for first-line treatment with immunotherapy alone (in case of having a PD-L1 > 50%), or chemoimmunotherapy were considered for the study. In general, all had to be over 18 years of age, with a pathologically and immunohistochemically confirmed diagnosis of a primary squamous or non-squamous lung tumor. All cases had to have tumor tissue available for genomic analysis by NGS (Illumina HiSeq system) and evaluation of the PD-L1 expression level. Likewise, patients had to have a measurable disease that allowed confirming the level of response to oncological treatment following the Response Evaluation Criteria in Solid Tumors (RECIST), version 1.1¹⁹. Patients were included with a sequential non-probabilistic sample method. They matched healthy subjects without cancer by periodontal condition (stages of periodontitis according to the 2017 World Workshop on the Classification of Periodontal and Peri-Implant Diseases and Conditions²⁰ and age (maximum difference of two years).

Exclusion criteria

Patients were excluded if they had received chronic steroid therapy (prednisolone > 10 mg or equivalent), those with uncontrolled novo brain metastasis or leptomeningeal disease, or patients with a life expectancy of less than 12 weeks according to baseline ECOG Score. In addition, patients with surgical or non-surgical periodontal therapy in the last year, those with acute periodontal conditions (periodontal abscesses or necrotizing periodontal disease) at the time of evaluation, those who had received antibiotics in the previous three months, and pregnant women were also excluded.

Periodontal clinical examination

A calibrated periodontist performed all measurements. Clinical parameters were assessed at proximal sites per tooth except for third molars and dental implants using a UNC-15 periodontal probe (HuFriedy, Leimen, Germany). Clinical parameters included pocket depth (PD) (ICC_95% 92%) and clinical attachment level (CAL) (ICC_95% 89%), expressed in millimeters (mm), plaque index (PI), and bleeding on probing (BOP), expressed as the percentage of positive sites in the mouth. Intra-examiner calibration was performed by recording duplicate measurements of PD and CAL in three patients twice during the same visit at 30-minute intervals.

Saliva collection samples

Participants had to have clean lips and cavity-free food for the saliva sample collection. In a sterile 50 mL Falcon tube containing transport medium [DNA/RNA Shield TM (Zymo Research) as a nucleic acid preservative], the patient was asked to spit for five timed minutes without touching the edge of the tube with their lips and to close it during the time they were not spitting to avoid minimal contamination with the environment. Additionally, a sterile swab was used to sweep saliva for 10 s on each surface of the oral cavity (oral mucosa of the jaws, floor of the mouth, center and sides of the tongue), then the swab was placed in the same Falcon tube and sealed. Each tube was labeled with the patient’s name and identification.

Microbiological sampling of subgingival plaque

Samples were taken from four sites by placing two consecutively inserted sterile paper points per site, left at the bottom of the sulcus/pocket for 10 s. Sampling sites were selected as follows: in periodontal health or gingivitis, samples were taken from the mesiobuccally sites of the first molars and, in their absence, from the adjacent second molars. Subgingival samples were collected from the most accessible site with the deepest periodontal pocket in each quadrant in individuals with periodontitis.

DNA extraction

Following the manufacturer’s recommendations, DNA was extracted from all samples using a Zymo BIOMICSTM™ DNA Miniprep kit. Once the DNA was obtained, the quality of the DNA was verified by determining the 260/280 and 260/230 absorbance ratios for purity verification in a Nanodrop (Nanodrop 2000). A fluorometer (Life Technologies, Invitrogen, Waltham, Massachusetts, United States) was used to quantify double-stranded DNA to confirm the quality of the DNA for performing the qPCRs¹⁸.

Quantitative PCR (qPCR)

Quantitative PCR (qPCR) was performed to detect Porphyromonas gingivalis, Tannerella forsythia, Treponema denticola, Eubacterium nodatum, Fusobacterium nucleatum, Eubacterium brachy, Eubacterium saphenum, Desulfobulbus oralis, and Filifactor alocis using specific primers and TaqMan probes. The identification of P. gingivalis was carried out according to what was reported by Boutaga et al.²¹, with a reaction mixture containing 3 mM MgCl₂, 1X buffer (GoTaq Polymerase^® from Promega, Wisconsin, United States), 0.1 mM dNTP, 0.9 µM probe, 1 µM primers and GoTaq Polymerase^® DNA G2 at 0.06 IU/µL. The protocol for T. forsythia was based on Morillo et al.²², whose master mix contained 3.5 mM MgCl₂, 0.2 mM dNTP, 2 µM primers, 0.5 µM specific probe, and GoTaq Polymerase^® DNA G2 at 0.06 IU/µL. For T. denticola, the method of Yoshida et al.²³, was followed with a reaction mixture with 2.25 mM MgCl₂, 1.5 mM dNTPs, 0.5 µM primers, 0.5 µM specific probe, and GoTaq Polymerase^® DNA at 0.125 U. Identification of E. nodatum was standardized in our laboratory, as well as the design of primers and probes with amplification conditions consisting of 3 mM MgCl₂, 0.1 mM dNTPs, 1 µM primers, 0.5 µM specific probe, and GoTaq Polymerase^® G2 DNA at 0.06 IU/µL¹⁸. Primer sequences for F. nucleatum were reported by Cross et al.²⁴. Amplification conditions were standardized with final concentrations of 2.0 mM [MgCl₂, 0.1 mM dNTPs, 0.3 µM primers, 0.3 µM specific probe, and GoTaq Polymerase^® DNA G2 at 0.06 IU/µL. E. brachy, E. saphenum, D. oralis, and F. alocis were identified following the protocol of Lafaurie et al.¹⁸ and Castillo et al.²⁵ with a reaction mixture containing 2.5 mM MgCl₂, 0.1 mM dNTPs, 0.04 µM primers, 0.03 µM specific probe and GoTaq Polymerase^® DNA G2 at 0.06 IU/µL. For amplification, 1 µL of previously quantified DNA was used. All protocols were carried out in a CFX 96 thermal cycler (BioRad^®, Hercules, CA, United States). Quantification was performed using calibration curves established for each bacterium. Positive controls were included with DNA from ATCC strains for culturable bacteria and plasmids for non-culturable bacteria. Negative controls were performed with molecular-grade ultrapure water.

Statistical analysis

Categorical data were expressed as frequencies, and associations between CP and the different microorganisms were established by the Chi-square/Fisher’s exact test. Continuous variables were tested using the Shapiro-Wilk test to define the type of distribution. Comparisons between groups were performed using the Wilcoxon rank sum test. Multiple correspondence analysis was performed to identify correlations between quantitative variables (concentration of microorganisms) among the different microorganisms in the NSCLC group, healthy patients, and combinations of both in subgingival plaque and saliva. Likewise, multiple correspondence analysis was performed to establish associations between various microorganisms in the NSCLC group and healthy patients. The analyses were performed using SPSS version 25 (SPSS Inc., Chicago, IL, USA) and STATA v14 (StataCorp LLC., USA).

Results

A baseline clinical and sociodemographic characterization

A total of 24 individuals participated in this study. Twelve patients with advanced NSCLC treated with first-line immunotherapy were included between March 2023 and October 2024. The remaining 12 subjects belonged to the control group and were healthy. Three patients were excluded from antibiotic intake due to their inability to perform the periodontal examination due to medical conditions or non-acceptance, and two patients were edentulous (Fig. 1).

Fig. 1.

Flow chart of selected population.

Table 1 shows the sociodemographic and periodontal characteristics of the NCLSC individuals and controls at baseline. All individuals with MCLCP had a diagnosis of stage IV lung adenocarcinoma and were candidates for immunotherapy as first-line treatment. The mean age for NSCLC patients and controls was 62.9 (SD +/−10.2, range 36–72) and 61.9 (SD +/- 8.3, range 38–72) years, respectively. Of all participants, six patients in each group were in periodontal stages I/II, and 6 in more advanced stages (III/IV). Although NSCLC subjects had fewer teeth than the control, there were no statistically significant differences between the groups (p = 0.27). Among NSCLC patients, 6 were men and 6 women, while 9 were women and 3 men in the control group. In the NSCLC group, all patients had a smoking history. However, 9/12 subjects were former, and only 2/12 were current smokers. In the control group, the smoking history was low; only 5 subjects were smokers, 3 were former, and only 2 were current smokers (p = 0.004). Although both groups were homogeneous in periodontitis stages, patients with NSCLC tended to accumulate more bacterial plaque (p = 0.02). They presented a greater pocket depth (PD) (p = 0.0008) and a more severe clinical attachment loss (CAL) (p = 0.015). However, there were no statistically significant differences regarding bleeding on probing (p = 0.41) (Table 1).

Table 1.

Sociodemographic aspects and periodontal clinical characterization of patients with NSCLC and control patients.

	NSCLC n = 12	CONTROL n = 12	p-value
Age (Years)
Mean SD	62.9 ± 10.2	61.9± 8.3	0.60
Gender F (%)
Male	6 (50)	9 (75)	0.40
Female	6 (50)	3 (25)
Ethnicity
Hispanic	11 (91.6)	12 (100)	0.90
Non-Hispanic White	1 (8.4)	0 (0)
Smoking habit
Never	0 (0)	7 (58)	0.004§
Current	3 (25)	2 (17)
Former	9 (75)	3 (25)
Stages of Periodontitis
I – II	6 (50)	6 (50)	1
III - IV	6 (50)	6 (50)
Number of Teeth
Median IQR	21(15–26)	24 (21–26)	0.27
Plaque Index%
Median IQR	53 (47–84)	38 (21–51)	0.02
BOP %
Median IQR	40 (30–59)	20 (0–65)	0.41
Pocket depth (mm)
Median IQR	2.42 (2.27–2.60)	1.63 (1.51–1.96)	0.0008§
CAL (mm)
Median IQR	3.16 (1.83–3.83)	1.72 (0.77–2.37)	0.015§

BOP Bleeding on probing, CAL Clinical attachment level, § statistical significance p < 0.05, NS non-significant value p > 0.05, ND Not Determined, RIQ Interquartile range (percentile 25-percetille 75).

Frequency of cultivable and non-cultivable microorganisms in subgingival plaque and saliva of patients with NSCLC and healthy patients

Figure 2A shows the frequency of Desulfobulbus oralis. This microorganism was identified in subgingival plaque samples of the NSCLC patients in 91% and 58% of healthy volunteers, respectively (p = 0.05). Treponema denticola and Filifactor alocis were comparable between groups (41% vs. 50%). However, Porphyromonas gingivalis was observed in a lower proportion concentration in both cases and controls (25% vs. 33%, respectively). Eubacterium nodatum showed a lower frequency in NSCLC patients and was absent from controls. Figure 2B shows a statistically significant reduction of T. denticola and Eubacterium brachy (p = 0.019, p = 0.045, respectively) in saliva samples from NSCLC patients compared to healthy controls. P. gingivalis (58% vs. 80%), D. oralis (25% vs. 60%), and F. alocis (25% vs. 50%) also showed lower proportions in saliva samples from cancer patients when compared between cancer and non-cancer groups without statistically significant differences. Other microorganisms, such as Eubacterium saphenum and Fusobacterium nucleatum, Eubacterium nodatum, and Tannerella forsythia, were present in most patients in both groups.

Fig. 2.

Qualitative analysis: detection of frequency of culturable and non-culturable microorganisms. A Subgingival plaque groups. B Saliva. * Statistically significant differences by groups.

Comparison of cultivable and non-cultivable microorganism concentrations of the periodontal microbiome from subgingival plaque and saliva by qPCR

As seen in Fig. 3, bacterial concentrations were similar between cases and controls for F. nucleatum, E. nodatum, T. denticola, T. forsythia, P. gingivalis, F. alocis, and E. saphenum in subgingival samples. However, E. brachy (p = 0.027) was reduced in NSCLC patients, and D. oralis overgrowth in subgingival plaque in these patients (p = 0.038). Other microorganisms did not find statistical differences. In saliva, E. brachy (p = 0.0076) and T. denticola (p = 0.033) also showed reduced concentrations in NSCLC patients (p = 0.0076).

Fig. 3.

Quantitative analysis: detecting the concentration of culturable and non-culturable microorganisms in saliva and subgingival plaque by groups. A Heatmap showing the concentration of target periodontopathic microorganisms. B Bivariate analysis compared progressive and non-progressive subjects after First-line ICI Therapy. * Statistically significant differences.

Associations and correlations between periodontopathic microorganisms from the subgingival microbiota and saliva in NSCLC and healthy subjects

Figure 4 presents a multiple correspondence analysis (MCA) of the subgingival plaque and saliva samples from patients with NSCLC and controls. The study showed more microbial interaction in NSCLC individuals than in healthy controls. While controls have poor associations between microorganisms in subgingival plaque, patients with NSCLC show strong associations between P. gingivalis and F. alocis and moderate between P. gingivalis and T. denticola or T. forsythia with F. alocis. Saliva samples in NSCLC also showed an association between F. alocis, D. oralis, T. forsythia, and T. denticola. However, P. gingivalis presents a weak association in this group. On the contrary, in voluntary health, P. gingivalis was significant in the healthy control saliva samples, which showed strong associations with other periodontopathic microorganisms such as F. alocis and D. oralis.

Fig. 4.

Multiple correspondence analysis of bacterial association in subgingival plaque and saliva in NSCLC and control group.

Spearman’s correlation was used to establish the correlations of the concentration of microorganisms in each sample (saliva or subgingival plaque) and between the two types of samples (saliva and subgingival plaque) for both groups. NSCLC subjects exhibit a more significant correlation (p < 0.05) between bacteria than healthy individuals in saliva samples. T. forsythia showed significant correlations with E. nodatum (r = 0.63) and F. nucleatum (r = 0.80), and T. forsythia with E. nodatum (r = 0.90) and F. nucleatum (0.76), F. nucleatum with E. nodatum (r = 0.90) and P. gingivalis with T. denticola (r = 0.57). Correlations between bacteria in saliva samples did not show significant results in the control group. In subgingival plaque samples, T. forsythia correlated significantly with F. nucleatum (0.73) and P. gingivalis with E. saphenum (0.70) in NSCLC samples. The microorganism correlations for the same species in saliva and subgingival plaque samples were only significant for P. gingivalis in healthy subjects (r = 0.83). Other correlations between plaque and saliva samples between different bacterial genera were more frequent in samples from subjects with cancer (Fig. 5).

Fig. 5.

Correlation between microbial species in subgingival plaque and saliva. The heatmap shows the correlation between periodontopathic target microorganisms in saliva (S) and subgingival plaque (SP).

Discussion

Our study sought to characterize periodontal microbiomes in saliva and subgingival plaque differences in the components of culturable and non-culturable microorganisms associated with periodontitis. Interestingly, a reduction in the frequency and concentration of some of these microorganisms was observed in the saliva of patients with lung cancer. T. denticola, E. brachy, F. alocis, and D. oralis have a marked reduction in saliva compared to subgingival plaque. This finding can be explained by dysbiosis, which increases commensal microorganisms. An overgrowth of Firmicutes, such as Streptococcus and Veillonella in saliva and BALF, has been associated with LC^26–28. Paradoxically, it was previously described that serum IgG antibodies to periodontal bacteria have an inversely proportional relationship with the development of LC. This finding suggests the generation of an adaptive immune response that could modulate the progression of NSCLC²⁹. In our study, P. gingivalis was low in frequency and concentration in saliva compared with other periodontopathic microorganisms in NSCLC subjects, and it did not show differences with the healthy controls similar to what was observed in other studies²⁹.

In contrast, other studies have shown a clear association between P. gingivalis in the tumor microbiome and LC proliferation^6,30,31. In this regard, Liu et al. showed that P. gingivalis in tumor tissue has an adverse prognosis affecting the tumor objective response rate (ORR) and progression-free and overall survival³². Several studies have shown that P. gingivalis causes sustained dysbiosis, favoring the induction of cellular changes through TP53 in patients with LC^33,34. Increasing P. gingivalis could generate a negative regulation that affects diverse cellular processes, including apoptotic cell death, adaptation of cellular metabolism, cell cycle arrest, cellular senescence, and DNA damage repair³⁵. Neither total nor phosphorylated MDM2 was increased in P. gingivalis-infected cells³⁶. Stress conditions, such as DNA damage, hypoxia, nutrient deprivation, or oncogene activation, lead to the alteration of MDM2 and/or direct modification of p53, resulting in the stabilization and, thus, activation of the p53 protein^32,33. The route of infection of P. gingivalis in NSCLC tumor tissue seems not to be micro-aspiration but rather hematogenous dissemination that allows the colonization of distant tissues from the oral cavity. Strikingly, P. gingivalis is the main microorganism detected in bacteremia studies of patients with periodontal disease^37,38.

Regarding subgingival plaque samples, a statistically significant increase in D. oralis was observed in the NSCLC group compared to healthy controls. This finding suggests that D. oralis could be associated with NSCLC, mediated by an alteration in the microbial ecology of the subgingival microenvironment. It is relevant to highlight that D. oralis has been described in the literature as an opportunistic microorganism with factors that can trigger an inflammatory response²⁴. Its increased prevalence could reflect dysbiosis associated with periodontal disease. D. oralis is an unculturable pathobiont associated with advanced periodontitis evaluated in different geographical regions^18,39, and it is a new putative periodontal pathogen in the study⁴⁰. This study is the first to report an association of these microorganisms in the subgingival plaque of NSCLC patients. Future studies should expand and evaluate this preliminary association.

In our study, T. denticola and E. brachy were associated with severe periodontitis and were found to be notably reduced in saliva. Therefore, we could hypothesize that the environment in patients with this type of cancer favors the proliferation of other microorganisms due to possible bacterial competition. Oral microorganism communities in saliva have also been linked to xenobiotic biodegradation, which may induce microbiome dysbiosis by increasing the number of microorganisms capable of performing this function⁴¹. However, these microorganisms have not been established so far. As observed in associations between bacterial species, NSCLC patients show more substantial associations in plaque than the control groups. Likewise, the saliva samples showed more correlations between microbial species. Ecological changes generated in saliva during NSCLC progression can influence the differences observed between plaque and saliva. However, in another study, a strong correlation exists between the subgingival microbiome and the saliva microbiome in individuals with periodontitis in a healthy population, indicating that saliva may be a sample that reflects the composition of the subgingival microbiota⁴². However, this may not occur in elderly individuals since a significant correlation between plaque and saliva was only observed for P. gingivalis in healthy individuals.

In lung cancer, the sample used in most studies is saliva, under the assumption of translocation by broncho-aspiration⁶; however, the subgingival plaque samples do not seem to correlate with the salivary samples in NSCLC subjects. Otherwise, saliva samples were quite active in the interaction between species. This activity could be advantageous for studies of oral microbiota in cancer since saliva samples harbor the same microorganisms as those isolated in subgingival plaque and are easier to obtain in these patients. Future studies should evaluate this microorganism in saliva and subgingival plaque samples as markers in NSCLC prognosis and assess the impact of periodontopathic microorganisms in immunotherapy response.

Although no significant statistical differences were found in the frequency and concentration of F. nucleatum between patients with and without cancer, this microorganism is higher in saliva and subgingival plaque. F. nucleatum has been identified as a relevant factor in studies linking oral microbiota with LC⁴³. Furthermore, previous research has shown that F. nucleatum and its products can reach the lower respiratory tract and lungs through saliva aspiration and blood circulation, causing chronic inflammation and apoptosis in the respiratory system. Its interaction with other respiratory pathogens could exert greater lung virulence in this context. While F. nucleatum has difficulty surviving in an aerobic environment, it can colonize the lungs in synergy with other respiratory pathogenic bacteria, such as Pseudomonas aeruginosa⁴⁴. Chronic, low-grade, and resultless degradation is a key factor in developing multiple diseases throughout the body, including respiratory and periodontal diseases⁴⁵. On the other hand, Chu et al. found that enriched F. nucleatum in the airways before immunotherapy treatment was associated with resistance to anti-PD-1/L1 inhibitors in NSCLC. This finding suggests that F. nucleatum may be associated with LC development and changes in the tumor stroma that favor primary resistance to immunotherapy⁴⁶. F. nucleatum also induced tumor microenvironment (TME) by T cell immunosuppression, generating a proinflammatory environment with cytokines deregulation and immune evasion, generating resistance to ICIs directed to PDL-1⁴⁷.

The differences observed between the frequency and concentration of microorganisms could be attributed to the periodontal condition of subjects with NSCLC and control subjects. However, the NSCLC patients showed lower frequency and concentration of periodontal bacteria in saliva than healthy volunteers. Because NSCLC individuals had more severe periodontal conditions, these differences cannot be attributed to their periodontal status. Another factor that may have influenced the differences between the groups is a more significant history of smoking in the NSCLC group. Most studies have reported an overgrowth of F. nucleatum in smokers⁴⁸. However, in our study, neither the frequency nor the concentration of this microorganism was similar between groups because most of the cancer patients were ex-smokers and not active smokers. D. oralis has been associated with severe periodontitis. However, it cannot be ruled out that this difference observed in the subgingival plaque of patients with NSCLC may be related to their periodontal condition. Another factor that could have influenced the results is the sample size, which was a highly selected sample and challenging to obtain due to the condition of the patients.

Conclusions

Relevant differences exist in the frequency and concentration of some periodontal bacteria in patients with NSCLC. A significant reduction of T. denticola and E. brachy periodontal bacteria occurs in saliva samples in patients with LC. D. oralis emerges as a bacterium that overgrows in subgingival plaques of cancer patients.

Strengths and limitations

This study is limited by its small sample size, which may affect the generalizability of the results and reduce the statistical power to demonstrate differences between groups. However, this is the first study to assess the difference in culturable or unculturable periodontal microorganisms between saliva and subgingival plaque. These findings could be used to study subgingival plaque as a sample that can contribute to future understanding of the association between periodontal microbiota and NSCLC. This study is not sufficiently robust or well-designed to evaluate these microorganisms as biomarkers for NSCLC.

Author contributions

Conceptualization: Cardona AF, Lafaurie GI, Chamat M, Castillo DM, Arrieta O. Medical records review and curation: Cardona AF, Rojas L, Zuluaga J, and Sanchez N, Arrieta O. Dental records review and curation : Lafaurie GI, Gómez LA, Chamat M. Microbiological and laboratory records: Castillo DM, Delgadillo NA. Formal analysis: Lafaurie GI, Diaz-Báez D, Castillo Y. Final manuscript writing and review: all authors.

Funding

This study is supported by the Universidad El Bosque, Bogotá, Colombia, by an internal grant for translational research (No. PCI-2022-11049).

Data availability

No/Not applicable (this manuscript does not report data generation or analysis).

Declarations

Consent for publication

The work described has not been published previously, and it is not under consideration for publication elsewhere. The manuscript was approved by all authors and by the responsible authorities where the work was carried out.

Competing interests

The authors declare no competing interests.