Association of DOK3 and infiltrated tumor-associated macrophages with risk for the prognosis of Porphyromonas gingivalis-infected oral cancer: a 12-year data analysis of 200 patients from a tertiary teaching hospital, Urumqi, China

Oral squamous cell carcinoma (OSCC), derived from epithelia, is the leading histopathological type, accounting for approximately up to 90% of all head and neck malignancies [1]. The anatomical distribution of OSCC includes the anterior two-thirds of the tongue, lower and upper gingiva, buccal mucosa, hard palate, floor of the mouth, retromolar triangle, and vermilion mucosa [2]. OSCC is characterized by its high local invasion and easy recurrence and metastasis, as well as relatively high morbidity and mortality (31,733 new cases and 15,745 deaths in 2022 in China, and 25,210 new cases and 4,452 deaths in 2022 in the USA) estimated through GLOBOCAN (http://​globocan.​iarc.​fr/​) [3]. In addition, owing to its poor prognosis, OSCC has gradually become a serious public health issue worldwide, despite the rapid development of multidisciplinary treatment [4]. Moreover, the 5-year overall survival rate for this disease has not significantly increased [4, 5] (Supplemental Fig. 1).

Since the recognition of a causal relationship between Helicobacter pylori infection and the occurrence of gastric cancer in the 1990s [6], a greater depth of understanding has been acquired with respect to bacterial carcinogenesis. However, the effect of the microbiome on oral cancer still remains unknown. As a precondition, many structures, including the nasal cavity, oral cavity, and sinuses (e.g., frontal sinus, ethmoidal sinus, paranasal sinus, maxillary sinus and sphenoid sinus) in the oro-maxillofacial area constitute an ideal room, wherein the stable habitat of suitable salivary pH (6.5 to 7.5) and constant temperature (around 37℃) provided, whether for the growth of anaerobic bacteria or aerobes [7]. Some epidemiological studies have established chronic inflammatory diseases, such as periodontitis, as newly defined risk factors contributing to OSCC development [8‐10]. Porphyromonas gingivalis (P. gingivalis), the most dominant bacteria in periodontal lesions, is a key pathogen that mediates the local immune inflammatory response in chronic periodontitis. More importantly, P. gingivalis tends to be positively associated with orodigestive cancers and its detection in patients with oral or esophageal SCC has adverse outcomes [11‐14]. Additionally, P. gingivalis can adhere to gingival epithelial cells, interfere with the normal physiological metabolism of cells, and inhibit the cytotoxicity of programmed cell death [15]. Persistent exposure to P. gingivalis can also give rise to cell morphological changes, promote proliferative capacity with a higher S phase fraction in the cellular cycle, and facilitate cell invasion and migration [16].

It is a well-known fact that macrophage is the most plentiful and important tumor-infiltrating immune cell type, and it would be differentiated into M2-like tumor-associated macrophage (TAM) expressing CD68⁺, CD163⁺, and CD204⁺ in the local milieu of OSCC stromal spaces [17]. These TAMs have been analyzed in a broad spectrum of cancers with strong evidence demonstrating their carcinogenic function in the furtherance of metastasis and relapse. However, the potential interaction between P. gingivalis and TAMs and the effect of TAMs on the prognosis of P. gingivalis-infected OSCC remain unclear.

OSCC is the most common type of oral malignancy, which usually originates from precancerous lesions of the oral mucosa, and metastasis accounts for its poor prognosis. Although multidisciplinary and comprehensive treatment strategies have brought substantial progresses in prognostic outcomes, OSCC has a profound influence on human health and quality of life owing to its high morbidity and mortality [24]. A large amount of evidence has indicated that alcohol, tobacco products, areca nut, betel quid chewing, and genetic alterations are causative factors implicated in OSCC progression [25]. The microbiome has entered the perspective of the academic community since the commencement of the new millennium [7], and its role in the promotion of OSCC has gradually become a novel area of research. Although a definitive link between oral microflora and OSCC is yet to be established, accumulating evidence demonstrates that a variety of microbiological agents can also contribute to the progression of oral carcinogenesis in the presence of definitive risk factors such as alcoholism and smoking [7, 26]. To our knowledge, this is the first qualitative study adopting a comprehensive, theory-driven approach to investigate the the association between different expression levels of P. gingivalis and infiltrated M2-TAMs and the clinical significance impacting on the prognosis of patients with OSCC in Urumqi, China.

P. gingivalis is an important opportunistic pathogenic bacterium that can exist, survive and reproduce in the cytoplasm of infected cells [27]. In this present study, P. gingivalis immunostaining was observed in all OSCC specimens (including 61 weakly expressed and 139 strongly expressed) but was almost absent in matched adjacent non-carcinoma tissues. The vast majority of previous clinical studies have revealed that P. gingivalis is a high-risk factor for OSCC linked to a worse outcome by measuring OSCC subjects versus healthy individuals [12, 13, 28]. However, few studies have attempted to quantify different strengths, and show differences in clinical outcomes by discriminating between strong and weak immunoexpression levels. While our investigation, followed by our recent findings [13], found that strong immunoexpression of P. gingivalis was significantly associated with the presence of tobacco smoking, poor oral hygiene habits, poor periodontal condition, larger tumor size (diameter ≥ 3 cm), poor tumor differentiation, advanced T stage and clinical stage, neck lymph node metastasis, and death (all P < 0.05). Overall, a few studies have also reported that the detection of P. gingivalis in both serum and saliva is correlated with a high risk of oral cancer [12, 14, 29]. Another prospective case–control study also showed that overall oral microbiome composition was associated with risk of SCC of the head and neck, and greater abundance of periodontitis-associated bacteria was more likely to be present in the cancer cases through bacterial 16S rRNA gene sequencing [30]. These conclusions should be interpreted with caution because P. gingivalis is an opportunistic periodontopathic bacterium that can also live in the oral cavity, salivary glands, and peripheral blood circulation of healthy volunteers [31, 32]. Hence, the provenance of P. gingivalis in OSCC tissues and its abundance may explain the problem.

The products of P. gingivalis and/or its metabolic by-products significantly contribute to the development of oral carcinogenesis via their involvement in chronic inflammation. The OSCC microenvironment often resembles that of chronic inflammation induced by the dynamic interplay between tumor cells and the milieu it belongs to [33]. In addition, this proinflammatory microenvironment can increase the number of CD66b⁺ neutrophils discovered in OSCC, and these neutrophils are positively associated with poor prognosis, as reported in our previous investigation [13]. TAM constitutes the largest number of immune cells, accounting for up to 50% of solid neoplasms. Additionally, TAM, often referred to as M2-like macrophages, are capable of promoting tumor angiogenesis, immunosuppression, and metastasis in cancer progression [34]. Nevertheless the available scientific literature regarding the topic on “pathological features of TAMs related to tumor immunity and its clinical significance of P. gingivalis-infected OSCC” is nonexistent. To reasonably recognize a crucial indicator, we performed bioinformatics analysis before IHC evaluation. In our bioinformatics analysis, we identified DOK3 as an important transcript in the immune response; higher expression levels of DOK3 were positively associated with immunosuppressive M2-like TAM in OSCC (r = 0.72, P < 0.001), which is in line with the findings of Liu et al. [35] in neurogliomas. In addition, the effect of DOK3 expression on M2-TAMs infiltration significantly increased after P. gingivalis treatment (P < 0.0001). These results were further validated by IHC analysis. Other markers for M2-TAM (e.g., CD68⁺, CD163⁺, and CD204⁺) expressed in the local milieu of OSCC stromal spaces were confirmed by Petruzzi et al. [17]. Therefore a novel marker of CD206 was used to stain M2-TAMs in our immunohistochemical assessment [36]. We found that P. gingivalis immunoexpression levels were positively associated with DOK3 and CD206⁺ TAM immunoexpression levels, suggesting that P. gingivalis can affect onco-immunity in OSCC by increasing DOK3 and M2-TAM expression.

Bolz et al. [37] identified the bacterial spectra on the surface of OSCC in comparison to oral mucosa of patients with a higher risk to emerge an OSCC and a healthy control group; and they reported gram-negative anaerobes provided by biofilms on OSCC surfaces play a decisive role in the development of postoperative infections in patients with OSCC. However, currently no synthesized evidence from meta-analysis has demonstrated the prevalence rate of P. gingivalis and its association with OSCC prognosis. A few studies have reported that infection with P. gingivalis correlates with poor prognosis in patients with oral cancer [12, 29], one of which is the saliva sample type, and the densities of P. gingivalis are not stratified. In the present study, based on different staining intensities, we discovered that OSCC patients with strong immunoexpression levels of P. gingivalis had a worse prognostic outcome than those with the weak immunoexpression levels (HR = 1.674, 95%CI 1.216–4.142, P = 0.012), which enhances the existing conclusions. Similarly, detection with DOK3 (HR = 1.881, 95%CI 1.433–3.457, P = 0.042), and M2-TAM (HR = 1.649, 95%CI 0.824–3.082, P = 0.034) strong immunoexpression levels were associated with worse prognosis for OSCC patients.

Oral hygiene habits are being increasingly examined as explanatory factors for oral cancer. However, the relationship remains complex given the confounding effects of established determinants that are prevalent, as well as broader issues including a lack of access to positive oral health awareness and advanced healthcare facilities. Another key message from IHC analysis of DOK3 and M2-TAMs in P. gingivalis-infected OSCC immunomicroenvironment concluded that both strong staining of DOK3 and M2-TAMs exhibited significant associations with poor oral hygiene habits and severe dysphagia (Table 2). Likewise, a case–control study carried out by Gupta et al. [23] showed that reported habits of poor oral hygiene were significantly associated with an increased risk of oral cancer, after adjustment for other known risk factors. However, some medical co-morbidities – including hypertension and diabetes mellitus – may contribute to poor oral hygiene status [38]. In addition, whether a positive association between poor oral hygiene, tumor-infiltrated M2-TAMs and the risk of OSCC can be observed should be further investigated. Critically, tolerable oral diet without severe preoperative dysphagia is necessary to appraise oropharyngeal function, especially for the patients with floor of the mouth SCC [39]. The impact of early dysphagia should not be underestimated, even though preoperative dysphagia score exhibited no significant relation with the clinicopathological data and follow-up. By considering swallowing impairment at the primary therapy patients can profit concerning survival and comorbidity.

In conclusion, for the first time, the relevance between different immunohistochemical intensities of P. gingivalis in OSCC tissue and clinicopathological characteristics and significance in prognosis was analyzed to better understand the participation of P. gingivalis in the OSCC immune microenvironment. Based on bioinformatics analyses, DOK3 was identified as the key DEG in the TME of OSCC infected with P. gingivalis, and its effect on TAM infiltration was significantly increased after P. gingivalis treatment. Furthermore, strong expression levels of DOK3 and M2-TAM were correlated with worse prognosis in patients with OSCC. Collectively, P. gingivalis, DOK3, and M2-type TAM could be considered as three novel independent risk factors for predicting the prognosis of OSCC. However, more basic researches on the molecular mechanism of the OSCC microenvironment in P. gingivalis infection need to be conducted in the future.