Blog

Introduction

Cervical cancer (CC) is one of the most common malignant tumors affecting women worldwide, with human papillomavirus (HPV) infection recognized as its necessary but insufficient cause. Persistent infection with high-risk HPV (HR-HPV), particularly types such as HPV16 and HPV18, plays a pivotal role in the development of cervical intraepithelial neoplasia (CIN) and CC.¹ According to the World Health Organization, over 600,000 new CC cases and more than 340,000 deaths occur annually, with the highest burden concentrated in low- and middle-income countries (LMICs), particularly in sub-Saharan Africa, Southeast Asia, and Latin America.² These global disparities are influenced not only by differences in screening and vaccination coverage but also by variations in vaginal microbiota composition and host immune responses across populations.³ However, only a small proportion of HPV infections progress to malignancy, suggesting that host immunity, viral genotype, and the cervicovaginal microenvironment critically influence the natural history of HPV infection.⁴ As HPV vaccine uptake and screening programs advance in high-income countries, understanding the microbiological and immunological drivers of HPV progression in diverse global populations has become increasingly urgent for eliminating CC as a public health problem.

The vaginal microecosystem is a complex, dynamic environment composed of microbial communities, immune factors, hormones, and metabolites. In healthy reproductive-age women, this microenvironment is usually dominated by Lactobacillus species such as L. crispatus, L. gasseri, L. iners, and L. jensenii, which help maintain acidic pH, produce hydrogen peroxide (H₂O₂), and compete against pathogenic microbes.⁵ Disruption of this balance, known as vaginal dysbiosis, is characterized by a depletion of Lactobacillus and a concurrent increase in anaerobic bacteria such as Gardnerella, Atopobium, and Prevotella. Over the past decade, increasing global evidence has indicated that disturbances in vaginal microecology and associated immune dysregulation play critical roles in HPV susceptibility and persistence,^6,7 and progression to cervical disease.⁸ Worse still, this abnormal condition is associated with increased susceptibility to sexually transmitted infections, including HPV.^9,10 For instance, bacterial vaginosis (BV)—a common dysbiotic state—has been consistently associated with reduced HPV clearance rates and elevated CIN risk.¹¹ It is inferred that BV-associated bacteria and their metabolites, including sialidase and polyamines, may compromise mucosal barrier function and alter local immune responses, facilitating viral persistence.¹² Moreover, an imbalance in microbiota can lead to altered production of cytokines such as IL-6, IL-8, and TNF-α, which are linked to chronic inflammation and impaired viral clearance.¹³ Studies from Africa, Asia, and Europe confirm that women with non-Lactobacillus-dominated microbiota are at higher risk of HPV persistence and progression to high-grade cervical lesions.^4,14,15 These findings underscore the importance of understanding microbe-host-HPV interactions across different populations.

Hydrogen peroxide-producing Lactobacillus species, in particular, appear protective against HPV. These microbes can modulate immune responses, reduce local inflammation, and inactivate viruses via oxidative stress.¹⁶ Conversely, reduced H₂O₂ levels have been observed in HPV-infected individuals, suggesting a compromised microbial defense.¹⁷ Furthermore, recent evidence has shown that HPV infection itself may downregulate mucosal innate peptides used by Lactobacilli for growth, thereby reinforcing dysbiosis and perpetuating viral persistence—a novel mechanism of immune evasion.⁶ Furthermore, cytokines such as IL-6 and IL-12 play roles in antiviral defense and T-cell recruitment. Studies have shown that decreased IL-12 and elevated IL-6 levels in the vaginal milieu may impair effective immune responses, leading to persistent HPV infections.¹⁸ Additionally, factors such as vaginal pH, leukocyte esterase (LE) activity, clue cells, and overall cleanliness score serve as indirect markers of local immunity and microbial composition, and may provide clinical clues to HPV infection status.¹⁹ Thus, it is reasonable to pay more attention to the factors potentially modulating the abundance of Lactobacilli and immune state in vaginal microecosystem during the management of HPV infections.

Despite accumulating evidence regarding the HPV infection and dysbiosis of vaginal microbiome, most prior studies are either cross-sectional or based on small sample sizes. Moreover, few have integrated microbial, immunological, and HPV genotyping data longitudinally. There remains a need for real-world, retrospective analyses that explore the interaction between vaginal microecological characteristics and HPV persistence and clearance over time. The clarification of these relationships is essential for developing predictive models, identifying high-risk patients, and optimizing early interventions.

Thus, in the current study, we performed a two-year retrospective study involving 312 women, which aimed to investigate the correlation between vaginal microecological status and HPV infection comprehensively, with a focus on identifying microbial and biochemical predictors of viral persistence or clearance. By integrating pH, cleanliness grade, enzyme activity (H₂O₂, LE, sialidase), and HPV genotyping data, the current study provides new insights into how host-microbe-viral interactions influence the natural course of HPV infection. Our findings may provide additional information regarding the risk stratification of CC, and guide targeted strategies to enhance HPV clearance in additional to the previous publications.

Methods

Participants

The current study is a retrospectively observational study, which conducted at the Fifth Hospital of Xiamen based on records collected between January 2023 and December 2024. Eligible participants were women aged 20 to 65 years who underwent comprehensive gynecological screening, including HPV genotyping (performed using the Hybribio 21 HPV Genotyping Assay; DNA extracted with Qiagen QIAcube and PCR run on ABI 7500 Real-Time PCR System), vaginal microecology assessment, cytological evaluation, and biochemical analysis of vaginal secretions. Inclusion criteria were: (1) availability of baseline and follow-up HPV testing, (2) standardized and complete vaginal microecological testing, and (3) complete demographic and clinical records. Exclusion criteria included prior treatment for cervical neoplasia, history of cervical surgery, immunosuppression (eg, HIV infection, corticosteroid use), pregnancy during the study period, antibiotic or probiotic use within four weeks before sampling, and incomplete data. Ethical approval was granted by the Fifth Hospital of Xiamen, and the study adhered to the Declaration of Helsinki guidelines. All the patients have signed an informed consent regarding the use of clinicopathological information prior to study commencement.

Clinical and Laboratory Data Collection

Demographic information (age, parity, smoking status, contraception method), clinical examination findings, and results of HPV genotyping, vaginal microecology, cytology, and inflammatory marker detection were retrieved from the hospital information system by trained researchers. Data were cross-verified independently to ensure accuracy.

HPV Detection and Genotyping

Cervical samples were collected using sterile cytobrushes and stored in liquid-based cytology medium. HPV DNA extraction was performed using a commercial DNA extraction kit following standardized protocols. HPV genotyping was conducted using a multiplex PCR assay (ABI 7500 Real-Time PCR System, USA), which is capable of identifying 14 high-risk types (including HPV 16, 18, 31, 33, 45, 52, and 58) and 5 low-risk types (HPV 6, 11, 42, 43, and 44). Where available, viral load was inferred from Ct values, with Ct <30 indicating high viral burden, 30–35 moderate, and >35 low viral burden. Infection outcomes were categorized as clearance, persistence, or progression, depending on genotype detection across serial timepoints.

Vaginal Microecology Evaluation

Vaginal secretions were sampled before pelvic examination using sterile cotton-tipped applicators. Samples were immediately evaluated for multiple parameters. pH was measured using narrow-range indicator strips (precision 0.1 pH units), and values ≥4.5 were considered elevated, indicating potential dysbiosis. Cleanliness was graded via a phase-contrast microscopy (Leica DM750, Germany). (400× magnification) into four grades: Grade I (predominant Lactobacillus, rare leukocytes), Grade II (moderate Lactobacillus with occasional mixed flora), Grade III (mixed flora, moderate leukocytes), and Grade IV (predominance of pathogenic flora, heavy leukocytes).

Gram-stained smears were subjected to Nugent scoring: (1) Lactobacillus morphotype (large Gram-positive rods), (2) Gardnerella morphotype (small Gram-variable rods), and (3) Mobiluncus morphotype (curved Gram-negative rods). Nugent scores were stratified into normal (0–3), intermediate (4–6), and bacterial vaginosis (7–10). Furthermore, samples were subclassified into “low-intermediate” (4–5) and “high-intermediate” (6) groups to capture finer gradations.

Clue cells, Candida spp., and Trichomonas vaginalis were identified by a phase-contrast microscopy (Leica DM750, Germany). The proportion of clue cells was recorded (% of epithelial cells affected), with >20% considered significant. Budding yeast and pseudohyphae indicated fungal infection; motile flagellated organisms were diagnostic of trichomoniasis.

Hydrogen peroxide (H₂O₂) production was assessed semi-quantitatively by colorimetric testing and graded as strong, weak, or negative using a microplate reader (Thermo Fisher Scientific, USA). Weak or negative production suggested impaired Lactobacillus function. Lactobacillus subtype identification (where available) was performed by culture and biochemical profiling, differentiating between L. crispatus, L. iners, L. gasseri, and L. jensenii species. L. crispatus predominance was considered indicative of optimal microecology, while L. iners predominance was classified as intermediate.

Biofilm formation on epithelial cells was evaluated qualitatively via Gram staining and categorized as absent, sparse, moderate, or dense biofilm layer, particularly noting Gardnerella or Atopobium dominance.

Leukocyte esterase (LE) activity was measured by urinary dipstick on vaginal secretions: LE positivity (≥1+) indicated local inflammatory activation. Vaginal epithelial integrity was evaluated by the presence of parabasal cells, indicating atrophic change.

Sialidase enzyme activity was detected using a colorimetric rapid detection kit under a microplate reader (Thermo Fisher Scientific, USA), indicating enzymatic disruption of epithelial barriers associated with BV pathogens.

Community state types (CSTs) were determined based on dominant flora patterns: CST I (L. crispatus-dominant), CST II (L. gasseri-dominant), CST III (L. iners-dominant), CST IV (diverse anaerobes), and CST V (L. jensenii-dominant).

Definitions

Normal vaginal microecology was defined as pH <4.5, Nugent score 0–3, H₂O₂ strong positive, dominant Lactobacillus spp. (preferably L. crispatus or L. jensenii), absence of biofilm and clue cells, and negative LE and sialidase tests. Dysbiosis was defined as pH ≥4.5, Nugent score ≥4, decreased or absent H₂O₂, dominance of anaerobes (eg, Gardnerella, Atopobium, Prevotella), positive LE or sialidase, and evidence of biofilm formation. HPV persistence was defined as detection of the same HPV genotype across two consecutive tests ≥12 months apart.

Cytological Examination

Cervical cytology specimens were collected via the ThinPrep^® method, fixed in PreservCyt^® solution, and processed within 7 days. Slides were reviewed by two independent cytopathologists according to the 2014 Bethesda System, classifying samples as negative for intraepithelial lesion or malignancy (NILM), atypical squamous cells (ASC-US, ASC-H), low-grade squamous intraepithelial lesion (LSIL), high-grade squamous intraepithelial lesion (HSIL), or atypical glandular cells (AGC). All disagreements were resolved by consensus conference.

Biochemical and Inflammatory Marker Detection

For a subset of patients, cervicovaginal secretions were analyzed for soluble immune and inflammatory markers. Concentrations of interleukin-6 (IL-6) (H007-1-1, Nanjing Jiancheng Bioengineering Institute, China), interleukin-12 (IL-12) (H010-1-2, Nanjing Jiancheng Bioengineering Institute, China), tumor necrosis factor-alpha (TNF-α) (H052-1-2, Nanjing Jiancheng Bioengineering Institute, China), and secretory immunoglobulin A (sIgA) (H108-2-1, Nanjing Jiancheng Bioengineering Institute, China) were measured using high-sensitivity enzyme-linked immunosorbent assay (ELISA) kits following standard protocols. Systemic inflammation was assessed by measuring C-reactive protein (CRP) levels in serum samples where available (E023-1-1, Nanjing Jiancheng Bioengineering Institute, China). Additionally, oxidative stress markers, including malondialdehyde (MDA) levels, were measured in vaginal fluids using thiobarbituric acid reactive substances (TBARS) assays (A003-1-2, Nanjing Jiancheng Bioengineering Institute, China).

Statistical Analysis

All statistical analyses were conducted using SPSS version 26.0 (IBM Corp., Armonk, NY, USA) and R software version 4.2.1. Continuous variables were assessed for normality using the Kolmogorov–Smirnov test. Normally distributed variables were presented as mean ± standard deviation (SD) and compared using independent-sample t-tests; non-normally distributed variables were expressed as median (interquartile range, IQR) and compared using the Mann–Whitney U-test. Categorical variables were compared using the Chi-square test or Fisher’s exact test, as appropriate. Multivariate logistic regression was performed to identify independent predictors of HPV persistence and clearance, adjusting for potential confounders. Odds ratios (ORs) and 95% confidence intervals (CIs) were reported. ROC curve analysis was performed to assess the predictive power of key parameters, and the area under the curve (AUC) was calculated. Kaplan–Meier survival curves were generated to compare HPV clearance rates between different vaginal microecology groups, and statistical significance was assessed using the Log rank test. A two-sided P value <0.05 was considered statistically significant.

Results

Baseline Characteristics

A total of 312 women were included in the final analysis cohort. As shown in Table 1, the mean age of the participants was 38.7 ± 9.6 years. Most women were multiparous (68.6%), and the majority used condoms as their primary contraceptive method (42.9%), while 10.9% reported smoking habits. Baseline HPV positivity was observed in 210 cases (67.3%), and elevated vaginal pH (≥4.5) was detected in 59.0% of the population (Table 1). The bacterial vaginosis, defined by a Nugent score ≥7, was present in 28.2% of all the case (Table 1). Half of the women exhibited positive hydrogen peroxide production (50.0%), while leukocyte esterase and sialidase activities were positive in 35.9% and 23.7%, respectively (Table 1). L. crispatus was the predominant strain in 39.1% of cases, and CST I (L. crispatus-dominant) accounted for 35.3% of the vaginal community structures. Biofilm formation was noted in 33.3% of cases. Other parameters including cytokine levels, oxidative stress markers, and inflammatory indicators are detailed shown in Table 1.

Analysis of Vaginal Microecological Status of the Study Population

Vaginal microecological parameters significantly differed between HPV-negative and HPV-positive groups. Women with HPV infection exhibited a significantly higher rate of elevated vaginal pH (76.2% vs 23.5%, P<0.001), increased prevalence of bacterial vaginosis (37.1% vs 9.8%, P<0.001), and reduced hydrogen peroxide production (41.0% vs 68.6%, P<0.001) compared to HPV-negative women (Table 2). Similarly, leukocyte esterase positivity and sialidase activity were significantly more common among HPV-positive participants (Table 2). Furthermore, CST IV (diverse anaerobic flora) predominated in HPV-positive individuals (34.3% vs 13.7%, P<0.001), and biofilm formation was substantially more frequent (41.9% vs 15.7%, P<0.001) (Table 2).

Table 2 Vaginal Microecological Characteristics Stratified by Human Papillomavirus (HPV) Infection Status

Analysis of Immune and Inflammatory Markers of the Study Population

Among the 210 HPV-positive women, 132 (62.9%) cleared the infection during the follow-up, 66 (31.4%) exhibited persistent infection, and 12 (5.7%) progressed to higher-grade cervical lesions (Table 3). Cytokine profiling revealed that women with persistent or progressive HPV infections had significantly higher IL-6 levels compared to those who cleared the infection (Persistence: 10.5 pg/mL vs Clearance: 6.8 pg/mL; Progression: 14.2 pg/mL, P<0.001) (Table 3). Conversely, IL-12 levels were significantly lower in persistence and progression groups compared to clearance (Persistence: 10.2 pg/mL vs Clearance: 14.3 pg/mL, P<0.001) (Table 3). TNF-α levels and oxidative stress marker MDA were also elevated in persistent and progression groups (Table 3). Additionally, CRP levels were higher and sIgA levels lower among women who failed to clear HPV (Table 3).

Table 3 Comparison of Immune and Inflammatory Markers Between Human Papillomavirus (HPV) Clearance, Persistence, and Progression Groups

Multivariate Analysis for Predictors of HPV Persistence

Multivariate logistic regression identified several independent risk factors associated with HPV persistence (Table 4). Elevated vaginal pH (OR=2.61, 95% CI: 1.52–4.50, P<0.001), bacterial vaginosis (Nugent score ≥7) (OR=3.23, 95% CI: 1.75–5.96, P<0.001), negative hydrogen peroxide production (OR=2.87, 95% CI: 1.54–5.35, P=0.001), and positive sialidase activity (OR=3.18, 95% CI: 1.67–6.04, P<0.001) were all significantly associated with persistence. Furthermore, CST IV (compared to CST I) (OR=2.91, 95% CI: 1.47–5.75, P=0.002), biofilm presence (OR=2.45, 95% CI: 1.32–4.57, P=0.005), and elevated IL-6 levels (>9 pg/mL) (OR=3.05, 95% CI: 1.62–5.75, P=0.001) also independently predicted persistent infection (Table 4).

Table 4 Multivariate Logistic Regression Analysis for Predictors of Human Papillomavirus (HPV) Persistence

Predictive Performance of Microecological and Inflammatory Markers

ROC analysis based on multivariate analysis demonstrated that IL-6 levels provided the highest predictive value for HPV clearance among single markers (AUC = 0.789, 95% CI: 0.721–0.856), followed closely by Nugent score (AUC = 0.778) and vaginal pH (AUC = 0.752) (Table 5). Hydrogen peroxide production and sialidase activity also showed moderate predictive accuracy. Importantly, a combined predictive model incorporating microecological and immune markers yielded the highest AUC value (0.842, 95% CI: 0.780–0.904), significantly improving sensitivity and specificity (Table 5).

Table 5 Receiver Operator Curve Analysis for Predictive Markers of Human Papillomavirus (HPV) Clearance

Further, HPV clearance rates over time differed significantly between women with normal vaginal microecology and those with dysbiosis. At 24 months, 90.1% of women with normal microecology cleared HPV compared to only 66.2% of those with dysbiosis (P<0.001) (Table 6).

Table 6 Human Papillomavirus (HPV) Clearance Rates at Different Follow-up Points Based on Vaginal Microecology Status

Discussion

The current study retrospectively explored the association between vaginal microecological status and HPV infection outcomes among 312 women over a two-year period. The findings revealed that disturbances in the vaginal microbiota, notably elevated pH, BV, decreased H₂O₂ production, presence of biofilms, and shifts towards diverse anaerobic CST IV, were strongly associated with HPV persistence and progression. Furthermore, inflammatory markers, particularly elevated IL-6 levels and oxidative stress markers such as MDA, emerged as significant predictors of persistent HPV infection.

Dysbiosis, characterized by increased vaginal pH and BV, predisposed to HPV persistence aligns with earlier reports emphasizing the critical role of a Lactobacillus-dominant environment in maintaining mucosal immunity.^4,8,9 Previous longitudinal studies have similarly demonstrated that women with a Lactobacillus-depleted microbiota, especially those dominated by Gardnerella, Atopobium, or Prevotella, are at higher risk for HPV acquisition and lower clearance rates.^20,21 In particular, CST IV, characterized by anaerobic dominance, was significantly associated with persistent infection in our cohort, corroborating previous observations.²² H₂O₂-producing Lactobacillus species, notably L. crispatus, have been shown to provide a protective effect against HPV persistence through direct antiviral activities and maintenance of low vaginal pH.²³ Our finding that reduced H₂O₂ production independently predicted persistence supports this mechanistic model. Furthermore, sialidase activity, a marker of BV-associated bacteria, emerged as a strong predictor, consistent with its role in disrupting epithelial barriers and modulating local immune responses unfavorably.²⁴ The immune microenvironment appeared equally crucial. Elevated IL-6 and reduced IL-12 levels among women with persistent or progressive HPV infection reflect a shift towards a pro-inflammatory, but ineffective antiviral state.^12,25,26 IL-6 is known to promote chronic inflammation and immune evasion by HPV, while IL-12 is essential for promoting cytotoxic T-cell responses necessary for viral clearance.²⁷ Similar cytokine profiles have been reported in earlier cross-sectional studies and CC precursor lesion studies.^28–30 These findings directly address our primary research question: whether specific vaginal microecological features and immune markers are predictive of HPV persistence or clearance. The results validate our hypothesis that a dysregulated vaginal environment characterized by Lactobacillus depletion, elevated pH, enzymatic imbalances, and altered cytokine profiles are critical contributors to persistent HPV infection. By systematically integrating microbiota composition, biochemical markers, and immune parameters, the study moves beyond simple associations to propose a more comprehensive model of HPV pathogenesis.

The ROC analysis demonstrated that while individual microecological markers such as pH, Nugent score, and H₂O₂ status had moderate predictive value, the combination of microecological and immune parameters significantly improved the prediction of HPV clearance. This finding highlights the necessity of an integrated approach in risk stratification for women with HPV infection, an aspect underappreciated in many previous reports.³¹ The presence of biofilm, frequently associated with Gardnerella vaginalis and Atopobium vaginae, was also independently associated with persistence. Biofilms can shield pathogenic bacteria from host immunity and antibiotics, contributing to chronic infections and sustained mucosal inflammation.³² Recent advanced microscopy studies have confirmed the existence of polymicrobial biofilms in women with BV and HPV co-infection, supporting our findings.³³ Our findings are consistent with previously international studies. A large European cohort study by Mitra et al found that a Lactobacillus-depleted, diverse vaginal microbiota significantly correlated with CIN2+ development across multiple populations.⁸ Similarly, Brotman et al in the United States demonstrated that changes in vaginal microbiota precede HPV detection, underscoring causality.²¹ In Sub-Saharan Africa, Happel and et al observed that women with high-diversity vaginal communities were more likely to harbor persistent high-risk HPV genotypes.³⁴ These global findings corroborate our results and underscore the widespread relevance of vaginal microbiome profiling in HPV management. Hence, the significance of our study extends beyond national boundaries and offers implications for diverse populations.

Clinically, our results emphasize the importance of considering vaginal microecological assessments as part of HPV management strategies. Women with evidence of dysbiosis or elevated inflammatory markers may benefit from closer surveillance or interventions aimed at restoring healthy vaginal microbiota. Probiotic therapies, particularly those using L. crispatus strains, have shown promise in small trials for BV treatment and may warrant exploration in the context of HPV infection.³⁵ However, several limitations must be acknowledged. First, the current study was retrospective and single-centered, which may introduce selection bias. Second, although we assessed a wide range of microecological and immune parameters, other factors such as sexual behavior, co-infections with other sexually transmitted infections, and host genetic susceptibility were not analyzed. Third, cytokine measurements were performed on cervicovaginal lavage rather than tissue samples, which may not fully reflect tissue-level immune dynamics. Future studies should focus on longitudinal interventional designs to determine whether correcting vaginal dysbiosis or modulating immune responses can actively promote HPV clearance. Advanced sequencing methods such as shotgun metagenomics may also provide deeper insights into the microbial community functional profiles associated with HPV outcomes.³⁶ Furthermore, integration of systemic markers such as plasma cytokine levels and host immunogenetic profiling could refine predictive models.

Conclusions

In conclusion, our retrospective study uniquely integrated microbial enzymatic markers, vaginal ecological indicators, and HPV genotyping to explore the interplay between vaginal microenvironmental factors and HPV infection risk. The novelty of this work lies in its real-world clinical dataset, comprehensive evaluation of non-invasive biochemical parameters, and combined predictive value of vaginal microecology and immune markers, which were rarely analyzed together in prior studies. Our findings highlights that specific vaginal microecological disturbances are associated with HPV positivity, underscoring the diagnostic and preventive value of assessing host–microbe–virus interactions in clinical practice. However, our study is limited by its retrospective design, single-center sample, and absence of metagenomic or immunological profiling, which restricts causal inference and generalizability. Unmeasured behavioral or demographic factors may also confound the observed associations. Even though for these limitations, interventional studies targeting the vaginal microbiome may offer new strategies for enhancing HPV clearance and preventing cervical disease progression. Future research should incorporate prospective, multi-center cohorts and apply integrated multi-omics approaches to deepen understanding of host-microbiome-HPV interactions.

Data Sharing Statement

The data will be provided by the corresponding author on reasonable request.

Statement of Ethics

All the investigation of the current study was performed under the approval of the ethic committee of the Fifth Hospital of Xiamen as well as the Declaration of Helsinki.

Author Contributions

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Funding

This study is supported by Guiding Project of Medical and Health Care in Xiamen City (No. 3502Z20209233).

Disclosure

The authors disclose no conflicts of interest in this work.

References

1. Walboomers JM, Jacobs MV, Manos MM, et al. Human papillomavirus is a necessary cause of invasive cervical cancer worldwide. J Pathol. 1999;189(1):12–19. doi:10.1002/(sici)1096-9896(199909)189:1

2. Organization WH. Cervical cancer. 2022.;

3. Serrano B, Brotons M, Bosch FX, Bruni L. Epidemiology and burden of HPV-related disease. Best Pract Res Clin Obstet Gynaecol. 2018;47:14–26. doi:10.1016/j.bpobgyn.2017.08.006

4. Usyk M, Zolnik CP, Castle PE, et al. Cervicovaginal microbiome and natural history of HPV in a longitudinal study. PLoS Pathog. 2020;16(3):e1008376. doi:10.1371/journal.ppat.1008376

5. Chen X, Lu Y, Chen T, Li R. The female vaginal microbiome in health and bacterial vaginosis. Front Cell Infect Microbiol. 2021;11:631972. doi:10.3389/fcimb.2021.631972

6. Kudela E, Liskova A, Samec M, et al. The interplay between the vaginal microbiome and innate immunity in the focus of predictive, preventive, and personalized medical approach to combat HPV-induced cervical cancer. Epma J. 2021;12(2):199–220. doi:10.1007/s13167-021-00244-3

7. Norenhag J, Du J, Olovsson M, Verstraelen H, Engstrand L, Brusselaers N. The vaginal microbiota, human papillomavirus and cervical dysplasia: a systematic review and network meta-analysis. Bjog. 2020;127(2):171–180. doi:10.1111/1471-0528.15854

8. Mitra A, MacIntyre DA, Lee YS, et al. Cervical intraepithelial neoplasia disease progression is associated with increased vaginal microbiome diversity. Sci Rep. 5:16865. doi:10.1038/srep16865

9. Ma B, Forney LJ, Ravel J. Vaginal microbiome: rethinking health and disease. Annu Rev Microbiol. 2012;66:371–389. doi:10.1146/annurev-micro-092611-150157

10. Saraf VS, Sheikh SA, Ahmad A, Gillevet PM, Bokhari H, Javed S. Vaginal microbiome: normalcy vs dysbiosis. Arch Microbiol. 2021;203(7):3793–3802. doi:10.1007/s00203-021-02414-3

11. Li Y, Wu X. Vaginal microbiome distinction in women with HPV+, cervical intraepithelial neoplasia, and cervical cancer, a retrospective study. Front Cell Infect Microbiol. 2024;14:1483544. doi:10.3389/fcimb.2024.1483544

12. Lebeau A, Bruyere D, Roncarati P, et al. HPV infection alters vaginal microbiome through down-regulating host mucosal innate peptides used by Lactobacilli as amino acid sources. Nat Commun. 13(1):1076. doi:10.1038/s41467-022-28724-8

13. Sims TT, Colbert LE, Klopp AH. The role of the cervicovaginal and gut microbiome in cervical intraepithelial neoplasia and cervical cancer. J Immunother Precis Oncol. 2021;4(2):72–78. doi:10.36401/jipo-20-17

14. Chen Y, Qiu X, Wang W, et al. Human papillomavirus infection and cervical intraepithelial neoplasia progression are associated with increased vaginal microbiome diversity in a Chinese cohort. BMC Infect Dis. 20(1):629. doi:10.1186/s12879-020-05324-9

15. Mitra A, MacIntyre DA, Marchesi JR, Lee YS, Bennett PR, Kyrgiou M. The vaginal microbiota, human papillomavirus infection and cervical intraepithelial neoplasia: what do we know and where are we going next? Microbiome. 4(1):58. doi:10.1186/s40168-016-0203-0

16. Mahajan G, Doherty E, To T, et al. Vaginal microbiome-host interactions modeled in a human vagina-on-a-chip. Microbiome. 10(1):201. doi:10.1186/s40168-022-01400-1

17. Zhang W, Yin Y, Jiang Y, et al. Relationship between vaginal and oral microbiome in patients of human papillomavirus (HPV) infection and cervical cancer. J Transl Med. 22(1):396. doi:10.1186/s12967-024-05124-8

18. Li WQ. Correlation Between Female Vaginal Microecological Environment, Local cytokines IL-6 and IL-12, and HPV Virus Infection. Shenyang Medical College; 2024.

19. Correlation between vaginal microecology and high-risk HPV infection Hebei North University. 2023.

20. Kyrgiou M, Mitra A, Moscicki AB. Does the vaginal microbiota play a role in the development of cervical cancer? Transl Res. 2017;179:168–182. doi:10.1016/j.trsl.2016.07.004

21. Brotman RM, Shardell MD, Gajer P, et al. Interplay between the temporal dynamics of the vaginal microbiota and human papillomavirus detection. J Infect Dis. 210(11):1723–1733. doi:10.1093/infdis/jiu330

22. Gao W, Weng J, Gao Y, Chen X. Comparison of the vaginal microbiota diversity of women with and without human papillomavirus infection: a cross-sectional study. BMC Infect Dis. 13:271. doi:10.1186/1471-2334-13-271

23. O’Hanlon DE, Moench TR, Cone RA. Vaginal pH and microbicidal lactic acid when lactobacilli dominate the microbiota. PLoS One. 2013;8(11):e80074. doi:10.1371/journal.pone.0080074

24. Muzny CA, Schwebke JR. Biofilms: an underappreciated mechanism of treatment failure and recurrence in vaginal infections. Clin Infect Dis. 61(4):601–606. doi:10.1093/cid/civ353

25. Shiri Aghbash P, Rasizadeh R, Sadri Nahand J, Bannazadeh Baghi H. The role of immune cells and inflammasomes in modulating cytokine responses in HPV-related cervical cancer. Int Immunopharmacol. 145:113625. doi:10.1016/j.intimp.2024.113625

26. Hu B, Wang R, Wu D, et al. A promising new model: establishment of patient-derived organoid models covering HPV-related cervical pre-cancerous lesions and their cancers. Adv Sci (Weinh). 2024;11(12):e2302340. doi:10.1002/advs.202302340

27. Hildesheim A, Wang SS. Host and viral genetics and risk of cervical cancer: a review. Virus Res. 2002;89(2):229–240. doi:10.1016/s0168-1702(02)00191-0

28. Borgogna JC, Shardell MD, Santori EK, et al. The vaginal metabolome and microbiota of cervical HPV-positive and HPV-negative women: a cross-sectional analysis. Bjog. 2020;127(2):182–192. doi:10.1111/1471-0528.15981

29. Brotman RM, Shardell MD, Gajer P, et al. Association between the vaginal microbiota, menopause status, and signs of vulvovaginal atrophy. Menopause. 2018;25(11):1321–1330. doi:10.1097/gme.0000000000001236

30. Gillet E, Meys JF, Verstraelen H, et al. Bacterial vaginosis is associated with uterine cervical human papillomavirus infection: a meta-analysis. BMC Infect Dis. 11:10. doi:10.1186/1471-2334-11-10

31. Lee JE, Lee S, Lee H, et al. Association of the vaginal microbiota with human papillomavirus infection in a Korean twin cohort. PLoS One. 2013;8(5):e63514. doi:10.1371/journal.pone.0063514

32. Pagar R, Deshkar S, Mahore J, et al. The microbial revolution: unveiling the benefits of vaginal probiotics and prebiotics. Microbiol Res. 2024;286:127787. doi:10.1016/j.micres.2024.127787

33. Castro J, Machado D, Cerca N. Unveiling the role of Gardnerella vaginalis in polymicrobial Bacterial Vaginosis biofilms: the impact of other vaginal pathogens living as neighbors. Isme J. 2019;13(5):1306–1317. doi:10.1038/s41396-018-0337-0

34. Happel AU, Balle C, Havyarimana E, et al. Cervicovaginal human papillomavirus genomes, microbiota composition and cytokine concentrations in South African Adolescents. Viruses. 15(3). doi:10.3390/v15030758

35. Mastromarino P, Vitali B, Mosca L. Bacterial vaginosis: a review on clinical trials with probiotics. New Microbiol. 2013;36(3):229–238.

36. France MT, Ma B, Gajer P, et al. VALENCIA: a nearest centroid classification method for vaginal microbial communities based on composition. Microbiome. 8(1):166. doi:10.1186/s40168-020-00934-6

Introduction

Lung cancer remains one of the most lethal malignancies worldwide, with a disproportionate impact on both developed and developing nations. According to GLOBOCAN 2020 data, lung cancer constitutes 11.4% of all new cancer cases and 18.0% of cancer-related deaths globally, surpassing breast and colorectal cancers in mortality burden. Lung cancer continues to be a major global health challenge, with its pathogenesis driven by a dynamic interplay of modifiable and non-modifiable risk factors. Cigarette smoking remains the predominant contributor, responsible for nearly 80–90% of cases, as highlighted by recent meta-analyses confirming the dose-dependent relationship between smoking duration and adenocarcinoma risk. Lung cancer stands as the most frequently diagnosed type of cancer and is also the primary cause of cancer-related mortality among males. Non-small cell lung cancer (NSCLC) constitutes roughly 80% of all lung cancer cases.¹ Regrettably, the vast majority of patients afflicted with NSCLC are diagnosed with the carcinoma at a stage where they already have locally advanced or metastatic disease^2,3.Despite advances in therapeutic strategies, the 5-year survival rate for lung cancer patients remains dismal, largely due to late-stage diagnosis exacerbated by socioeconomic disparities in screening access.⁴ Current diagnostic approaches primarily rely on imaging modalities and tumor biomarker assays, yet both exhibit significant limitations that hinder early detection and accurate diagnosis.

Computed tomography (CT) and other imaging technologies form the basis of lung cancer screening. However, high-resolution CT has not yet achieved optimal specificity in detecting small lung nodules, leading to frequent false positives and unnecessary invasive surgeries.^5,6 Additionally, research indicates that radiomics is poised to become an advanced AI-driven imaging characterization technique capable of directly predicting the response to immunotherapy for various solid tumors, a method that has been extensively studied in NSCLC.^7,8 Tumor biomarker assays, including carcinoembryonic antigen (CEA), cytokeratin 19 fragment (CYFRA 21–1), and neuron-specific enolase (NSE), are non-invasive alternatives. However, their clinical utility is hampered by low sensitivity (40–60%) and specificity (75–85%) in early-stage disease, as well as cross-reactivity with benign conditions For instance, CYFRA 21–1 exhibits poor performance in squamous cell carcinoma detection, while NSE is unreliable for small-cell lung cancer due to interference from hemolysis. Such constraints highlight the urgent demand for novel biomarkers with higher diagnostic accuracy.

The discovery of new biomarkers is critical to revolutionizing lung cancer diagnostics. Glycodelin-derived spliceosome-mediated RNA debris (GDSMD), a recently identified splice variant of the glycodelin family, has emerged as a promising candidate. Gasdermin D (GSDMD), a key executioner of pyroptosis, has emerged as a pivotal molecule in cancer biology. As a substrate for inflammatory caspases (caspase-1/4/5/11), GSDMD is cleaved into N-terminal (GSDMD-N) and C-terminal (GSDMD-C) fragments. The N-terminal oligomerizes to form plasma membrane pores, facilitating the release of pro-inflammatory cytokines (eg, IL-1β, IL-18) and promoting immunogenic cell death, while the C-terminal fragment suppresses apoptosis.^9,10 In lung cancer, aberrant GSDMD expression correlates with tumor microenvironment remodeling and chemotherapy resistance, yet its dual role—either activating anti-tumor immunity or driving pro-tumorigenic inflammation—remains poorly understood.¹¹ The canonical pathway of gasdermin D (GSDMD)-induced pyroptosis is primarily initiated by NLRP1 inflammasomes. Exogenous pathogens and endogenous damage signals are recognized by other inflammasome sensors, including NLRC4, NLRP3, and AIM2. The activation of these inflammasomes underscores the critical link between pyroptosis and the tumor immune microenvironment. Studies have demonstrated that NLRP1 inflammasomes can be activated by dipeptidyl peptidase (DPP). Once activated, NLRP1 promotes the secretion of pro-inflammatory cytokines and enhances Th1 cell-mediated immune responses. This process not only amplifies the therapeutic efficacy of anti-PD1 antibodies in tumor treatment but also reduces the infiltration of CD8+ T cells at metastatic sites, suggesting a dual role in modulating anti-tumor immunity.¹² Despite its potential, the diagnostic relevance of GDSMD in human lung cancer remains unexplored, and its mechanistic role in disease progression is poorly understood.

Studies have shown the biological function of GSDMD in non-small cell lung cancer (NSCLC). GSDMD protein levels are significantly upregulated in NSCLC. High GSDMD expression is associated with aggressive features, including larger tumors and more advanced lymph node metastasis (TNM) staging. High GSDMD expression in lung adenocarcinoma (LUAD) indicates poor prognosis, and GSDMD knockout inhibits tumor growth in vitro and in vivo. In GSDMD-deficient tumor cells, both endogenous and exogenous activation of the pyroptotic (NLRP3/caspase-1) signaling pathway induces another type of programmed cell death (apoptosis). GSDMD depletion activates caspase-3 and PARP cleavage and promotes cancer cell death through the intrinsic mitochondrial apoptosis pathway.

This research focuses on assessing the effectiveness of GSDMD in the diagnosis of lung cancer and delving deeply into its biological functions. We are committed to bridging the critical gaps in non – invasive diagnosis. Meanwhile, we aim to conduct an in – depth exploration of the predictive value of the combined diagnosis of GSDMD with other markers. If GSDMD can be successfully validated, it will lay the foundation for its integration into the clinical workflow. Ultimately, this will lead to an increase in the early detection rate of lung cancer and an improvement in patient prognosis.

Materials and Methods

Study Population

In the lung cancer group, 114 lung cancer patients who visited the Thoracic Surgery Department of Hebei General Hospital from January 2024 to November 2024 were collected. There were 63 males and 51 females, with an average age of (61.14 ± 8.30) years. The benign group consisted of 87 patients with pulmonary nodules hospitalized during the same period, including 40 males and 47 females, with an average age of (59.31 ± 11.64) years. Retrospective analysis was carried out with 100 healthy subjects in the same period as the control group, who were matched with the lung cancer and pulmonary nodule patients in terms of gender and age.

Inclusion and Exclusion Criteria

Inclusion criteria: Patients hospitalized for the first time for lung cancer treatment; untreated lung cancer patients; lung cancer patients diagnosed by surgery, lung biopsy, or fiberoptic bronchoscopy biopsy; lung cancer patients with good compliance and complete case records. Exclusion criteria: Patients with incomplete medical records and basic information; patients with other tumors or systemic diseases.This study was approved by the Ethics Committee of Hebei General Hospital [Approval Number: 2024(288)]. Informed consent was obtained from all participants.

Collection of Clinical Data

Data were collected on patient gender, age, GSDMD, carcinoembryonic antigen (CEA), neuron-specific enolase (NSE), cytokeratin 19 fragment (CYFRA21-1), squamous epithelial cell carcinoma antigen (SCC), gastrin-releasing peptide precursor (ProGRP), human ependynein 4 (HE4), interleukin 6 (IL-6), and C-reactive protein (CRP) were collected.

Sample Collection

All subjects were required to fast for at least 12 hours. The next morning, 5 mL of non-anticoagulated venous blood was collected on an empty stomach and allowed to clot at room temperature for 2 hours. After centrifugation (1000×g, 4°C, 15 min), the serum was collected and stored in a refrigerator at −80°C until use.

The plasma concentration of Gasdermin D (GSDMD) was quantified through a chemiluminescence assay provided by Beijing Meide Taikang Biotechnology Co., Ltd. The operations were strictly in accordance with the instructions of the reagent kits.

The intra-assay and inter-assay variability of GSDMD measurements were <8% and <15%, respectively.CRP was measured by immunoturbidimetric assay using a Beckman AU5800 automated biochemical analyzer and its associated reagents.

Tumor markers and IL-6 were measured by Roche 602 electrochemiluminescence analyzer, and the measurements were strictly carried out according to the reagent instructions and the operation procedures of the instrument.

Statistical Analysis

Data analysis and graphing were performed using SPSS 25.0 software and GraphPad Prism 8.0. Measurement data that follow a normal distribution are presented as the mean±standard deviation (X±S), whereas non – normally distributed data are expressed as the median (Q1,Q3). Given the emphasis of this study on multiple comparisons, a variance homogeneity test was carried out.For data exhibiting homogeneous variances, one – way ANOVA was applied, and this was succeeded by multiple comparisons using the Least Significant Difference (LSD) method. When variances were non – homogeneous, non – parametric tests were utilized. Specifically, the Kruskal–Wallis test was employed for multiple – group comparisons, and the Bonferroni correction was used for pairwise comparisons.Qualitative data were analyzed via chi – square tests. Additionally, Spearman’s rank correlation analysis was conducted to evaluate the relationship between plasma GSDMD levels and clinical indicators.Additionally, multiple logistic regression analysis was conducted to identify risk factors for the onset of lung cancer (LC). The cutoff value was determined using the Youden index.The diagnostic value of plasma GSDMD for LC was evaluated using the area under the receiver operating characteristic curve (AUC). We performed collinearity diagnostics on the study variables and found no evidence of multicollinearity. See Supplementary Material. All statistical tests employed a two – tailed approach, with the significance level set at a stringent P < 0.05.

Results

Comparison of Clinical Indicators Among Newly Diagnosed Lung Cancer, Pulmonary Nodule, and Healthy Controls

There were no statistically significant differences in age and gender among the three groups of patients. Nevertheless, substantial disparities emerged in other measurements across the three groups, with a statistically significant difference (P < 0.05, Tables 1). In the lung cancer group, the levels of GSDMD, CYFRA21 – 1, SCC, and CRP were the highest, followed by those in the lung nodule group, and were the lowest in the healthy control group (P < 0.05; see (Figure 1a–d). The levels of NSE and ProGRP were significantly higher in both the lung cancer group and the lung nodule group than in the healthy control group (P < 0.05). Nevertheless, there was no significant difference between the lung cancer group and the lung nodule group (Figure 1e and f). The levels of CEA, HE4, and IL – 6 were significantly higher in the lung cancer group compared to the lung nodule group (P < 0.05). However, no significant differences were observed between the lung nodule group and the healthy control group (Figure 1g–i).

Figure 1 Plasm levels of GD(a), CEA(b), NSE(c), CYFRA21-1(d), SCC(e), ProGRP (f), HE4(g), IL6(h) and CRP(i) in Healthy Control group, Pulmonary Nodule group and Lung Cancer group. Abbreviations: GD, Gasdermin D;CEA, carcinoembryonic antigen;NSE, neuron-specific enolase; CYFRA21-1, cytokeratin 19 fragment; SCC, squamous epithelial cell carcinoma antigen; ProGRP, gastrin-releasing peptide precursor; HE4, human ependynein 4; IL-6, interleukin 6; CRP, C-reactive protein. Note: **p < 0.01, ***p<0.001, ns: There was no statistical difference between the two groups.

Correlation Analysis Between Plasma GSDMD and Markers of Inflammation and Tumor Markers

Spearman’s analysis indicated correlations between GSDMD and various markers: GSDMD and CEA (r = 0.329, P < 0.001,Figure 2A), GSDMD and NSE (r = 0.266, P < 0.001, Figure 2B), GSDMD and CYFRA21 – 1 (r = 0.477, P < 0.001,Figure 2C), GSDMD and SCC (r = 0.648, P < 0.001, Figure 2D), GSDMD and ProGRP (r = 0.379, P < 0.001, Figure 2E), GSDMD and HE4 (r = 0.468, P < 0.001, Figure 2F), GSDMD and IL – 6 (r = 0.616, P < 0.001, Figure 2G), as well as GSDMD and CRP (r = 0.226, P < 0.001, Figure 2H) (Table 2).

Table 2 Correlation Between Serum GSDMD and Other Indicators (n=301)

Figure 2 Analysis of the Correlation between plasma GSDMD and Markers of Tumor and Inflammation (A) scatterplot of plasma GSDMD levels associated with CEA;(B) scatterplot of plasma GSDMD levels associated with NSE;(C)catterplot of plasma GSDMD levels associated with CYFRA21-1;(D)catterplot of plasma GSDMD levels associated with SCC;(E)catterplot of plasma GSDMD levels associated with ProGRP;(F)catterplot of plasma GSDMD levels associated with HE4;(G)catterplot of plasma GSDMD levels associated with IL-6;(H)catterplot of plasma GSDMD levels associated with CRP.

An Analytical Exploration of the Factors Affecting the Progression of Newly – Diagnosed Lung Cancer

Binary logistic regression analysis was carried out, taking the occurrence or non – occurrence of lung cancer as the dependent variable, while GSDMD, CEA, NSE, CYFRA21 – 1, SCC, ProGRP, HE4, IL – 6, and CRP were regarded as independent variables.The findings from univariate logistic regression revealed that all the observed indicators were independent influencing factors. The results of multivariate logistic regression showed that serum GSDMD, CEA, SCC, HE4, and IL – 6 were independent risk factors for lung cancer (P < 0.05) (Table 3).

Table 3 Logistic Regression Analysis of Serum GSDMD and Other Indicators in Lung Cancer

The Predictive Significance of Plasma – Based GSDMD in the Realm of Lung Cancer

The area under the ROC curve (AUC) of plasma GSDMD, registering at 0.860, outstripped those of CEA (0.801) and SCC (0.843). Even though it fell short of HE4’s AUC (0.901), the sensitivity of GSDMD soared to 95.6%, substantially higher than HE4’s 83.3%. This indicates that plasma GSDMD also boasts a relatively high diagnostic value among the individual biomarkers for lung cancer.

At a plasma GSDMD cut – off value of 39.87 pg/mL, the diagnostic sensitivity hit 95.6%, and the specificity stood at 72.2% (Table 4 and Figure 3).

Table 4 ROC Curve Analysis of the Predictive Value of Serum GSDMD for the Occurrence of First Diagnosed Lung Cancer

Figure 3 Predictive value of plasma GSDMD for Lung cancer ROC curve.

Discussion

This study meticulously explored the role of serum Gasdermin D (GSDMD) in lung cancer, offering profound insights into its potential as a biomarker and its implications within the disease’s intricate pathophysiology.

GSDMD is a pivotal protein in the pyroptosis pathway, a distinct form of programmed cell death that differs from apoptosis. Structurally, it is composed of an N – terminal effector domain and a C – terminal inhibitory domain. Under normal physiological conditions, the C – terminal domain acts as a safeguard, maintaining the N – terminal domain in an inactive state. However, upon activation by specific caspases, a significant transformation occurs. In the canonical inflammasome pathway, caspase – 1 plays a crucial role in cleaving GSDMD. In the non – canonical pathway, caspases like caspase – 3/7 are responsible for this cleavage event. Once cleaved, the N – terminal fragment of GSDMD undergoes a series of remarkable changes. It translocates to the cell membrane, where it oligomerizes, forming pores that are approximately 10–14 nm in diameter. These pores are not just simple openings; they are gateways for the release of pro – inflammatory cytokines, such as interleukin – 1β (IL – 1β) and interleukin – 18 (IL – 18). The release of these cytokines triggers a cascade of immune responses, leading to an inflammatory reaction that can have far – reaching consequences.¹³

In the context of lung cancer, the function of GSDMD is a double – edged sword. On one hand, the induction of pyroptosis can be seen as a defense mechanism against cancer cells. By promoting cell death, GSDMD – mediated pyroptosis has the potential to halt the uncontrolled growth and spread of lung cancer cells. This is particularly important as lung cancer cells often exhibit abnormal growth patterns and resistance to traditional apoptosis. Pyroptosis offers an alternative route to eliminate these cancer cells, potentially reducing tumor burden. For example, there is emerging research evidence indicating that pyroptosis has the potential to suppress tumor growth and reverse drug resistance in cancer cells.¹⁴ On the other hand, the release of pro – inflammatory cytokines as a result of GSDMD activation can also create an environment that is conducive to tumorigenesis. Chronic inflammation has long been recognized as a key driver in the development and progression of cancer. In the lung, this inflammatory microenvironment can promote angiogenesis, the formation of new blood vessels that supply nutrients to the growing tumor. It can also stimulate tumor cell proliferation, enabling cancer cells to multiply at an accelerated rate. Additionally, the inflammatory cytokines can facilitate metastasis, allowing cancer cells to break away from the primary tumor and spread to other parts of the body.¹⁵ Therefore, understanding the delicate balance between the cell – killing and inflammatory – promoting functions of GSDMD in lung cancer is of utmost importance.

Our study, as vividly presented in Table 1, clearly showcases significant differences in serum GSDMD levels among the lung cancer group (LC), pulmonary nodule group (PN), and healthy control group (HC). The LC group had the highest GSDMD levels, with a median value of 94.05 pg/mL (interquartile range: 65.21–124.21 pg/mL), followed by the PN group with a median of 51.67 pg/mL (interquartile range: 30.25–96.69 pg/mL), while the HC group had the lowest levels, with a median of 5.06 pg/mL (interquartile range: 2.98–7.17 pg/mL). This distinct difference in levels indicates that GSDMD could potentially serve as a valuable biomarker for differentiating between lung cancer patients, those with pulmonary nodules (which may be pre – cancerous or benign), and healthy individuals.Similar trends were observed for other well – known biomarkers such as CEA, NSE, and CYFRA21 – 1. For instance, CEA levels were significantly higher in the LC group compared to the PN and HC groups. This parallel behavior among multiple biomarkers further validates the reliability of GSDMD as a potential biomarker in this context. By measuring GSDMD levels in serum, clinicians may be able to identify individuals at a higher risk of lung cancer or detect the disease at an earlier stage. In a clinical setting, this could lead to more timely interventions and potentially better patient outcomes.

Table 2 demonstrates significant positive correlations between serum GSDMD and various other biomarkers. The strong correlation with IL – 6, with a correlation coefficient (r) of 0.616 and a P – value < 0.000, suggests that GSDMD may be intricately involved in the inflammatory processes associated with lung cancer. Inflammatory cytokines play a crucial role in tumor development, and the correlation with IL – 6 implies that GSDMD could be part of a complex network of molecules driving the disease progression. IL – 6 is known to activate various signaling pathways, including the STAT3 pathway, which can promote tumor cell survival, proliferation, and invasion.¹⁶ The fact that GSDMD is correlated with IL – 6 suggests that they may act in concert to influence the development and progression of lung cancer.The correlations with tumor – associated antigens like CEA and SCC also suggest that GSDMD may share common regulatory pathways with these established biomarkers. CEA is a widely used biomarker for lung cancer, and its correlation with GSDMD indicates that there may be underlying molecular mechanisms that link the two. Similarly, SCC, which is often elevated in squamous cell carcinoma of the lung, shows a strong positive correlation with GSDMD. These correlations further highlight the potential of GSDMD as a valuable biomarker for lung cancer diagnosis and prognosis. By understanding these relationships, researchers may be able to develop more comprehensive biomarker panels that can provide a more accurate assessment of a patient’s disease status. The logistic regression analysis in Table 3 reveals that GSDMD is an independent predictor of lung cancer. In the univariate analysis, all investigated variables, including GSDMD, CEA, NSE, CYFRA21 – 1, SCC, ProGRP, HE4, IL – 6, and CRP, were significantly associated with lung cancer. However, in the multivariate analysis, which takes into account the potential confounding effects of other variables, GSDMD, along with CEA, SCC, HE4, and IL – 6, remained significantly associated with the risk of lung cancer.GSDMD had an odds ratio (OR) of 1.011 (95% confidence interval: 1.002–1.019) with a P – value of 0.012 in the multivariate analysis. This indicates that for every unit increase in GSDMD levels, the odds of having lung cancer increase by 1.011 times, independent of the other variables in the model. The independent predictive value of GSDMD suggests that it could be used in combination with other biomarkers to improve the accuracy of lung cancer risk assessment. In current clinical practice, relying on a single biomarker may not be sufficient to accurately predict the risk of lung cancer. By incorporating GSDMD into a panel of biomarkers, clinicians can potentially obtain a more comprehensive and accurate picture of a patient’s risk. This is particularly important as early detection of lung cancer is crucial for improving patient survival rates. Lung cancer is often diagnosed at an advanced stage, when treatment options are limited. Having reliable predictors like GSDMD can aid in more targeted screening and early intervention strategies, potentially saving lives.

The ROC curve analysis in Table 4 shows that serum GSDMD has a good predictive value for lung cancer, with an area under the curve (AUC) of 0.860. A value of 0.860 indicates that GSDMD can effectively distinguish between lung cancer patients and healthy controls.When combined with other biomarkers (CEA, SCC, and HE4), the predictive accuracy significantly improves, achieving an AUC of 0.959. This is in line with previous research demonstrating that multi – biomarker panels are more effective in predicting lung cancer than single biomarkers.¹⁷ The combination of these biomarkers provides a more comprehensive assessment of a patient’s disease status, taking into account different aspects of tumor biology.The high sensitivity and specificity values associated with GSDMD and the biomarker combination further support its potential use in clinical practice for the early detection of lung cancer. GSDMD had a sensitivity of 95.6% and a specificity of 72.2% at a critical value of 39.87 pg/mL. The combination of GSDMD, CEA, SCC, and HE4 had a sensitivity of 93.0% and a specificity of 82.9%. These values suggest that the biomarkers can accurately identify a large proportion of true positive cases while minimizing the number of false positive cases.

Our findings are consistent with several previous studies. High expression of GSDMD is associated with aggressive features, including larger tumor size and later tumor-node-metastasis (TNM) staging. Additionally, high expression of GSDMD in lung cancer predicts poor prognosis. Analysis revealed a correlation between GSDMD and EGFR/Akt signaling. GSDMD knockout weakened tumor proliferation by promoting NSCLC cell apoptosis and inhibiting EGFR/Akt signaling transduction. In summary, GSDMD is an independent prognostic biomarker for LUAD.¹⁸ Emerging evidence has established a correlative relationship between GSDMD and inflammatory cytokine dysregulation in pulmonary pathologies. In asthmatic patients, immunohistochemical analyses reveal markedly elevated expression of the activated N-terminal fragment of GSDMD (N-GSDMD) within airway epithelial compartments, a phenomenon observed to colocalize with heightened concentrations of the pro-inflammatory interleukins IL-1β and IL-18.¹⁹ Furthermore, mechanistic investigations in chronic obstructive pulmonary disease (COPD) models demonstrate that GSDMD-mediated pyroptosis contributes to disease pathogenesis via the NLRP3/caspase-1/GSDMD signaling axis, thereby amplifying inflammatory cascades through programmed cell death mechanisms.This consistency across studies further validates the relationship between GSDMD and inflammation in the context of lung diseases, including lung cancer.²⁰ In addition, the use of logistic regression and ROC curve analysis to assess the significance and predictive value of the biomarkers in our study is in line with the standard practice of biomarker studies and adds credibility to the findings. Regarding cost, GSDMD detection (eg, ELISA-based) shows advantages with lower equipment investment and per-test costs, compared to LDCT.For turnaround time, GSDMD assays (4–12 hours for serum/tissue) are faster than LDCT (2–4 days, limited by scheduling and image review).In comparison with LDCT, GSDMD serves as a complementary molecular marker: while LDCT excels in anatomical visualization, GSDMD aids in distinguishing benign/malignant lesions and monitoring disease activity, with no radiation risk.

While this study provides valuable insights into the role of GSDMD in lung cancer, there are several limitations that should be acknowledged. First, the study was conducted in a single center with a relatively small sample size. Larger, multi-center studies are needed to validate these findings and to explore the potential of GSDMD as a diagnostic and therapeutic target in lung cancer. Second, the study did not investigate the molecular mechanisms underlying the elevated levels of GSDMD in lung cancer. Future research should focus on elucidating the role of GSDMD in lung cancer pathogenesis, particularly its involvement in pyroptosis and inflammation.

In conclusion, this study provides strong evidence for the research value of GSDMD in lung cancer. Its unique structure and function place it at the intersection of cell death and inflammation, making it a potential target for therapeutic intervention. In addition, as a biomarker, whether used alone or in combination with other markers, it offers great hope for distinguishing between benign and malignant lung diseases and improving the diagnosis and prognosis of lung cancer.However, further research is needed to fully understand the complex mechanisms by which GSDMD influences lung cancer development and progression, and to translate these findings into clinical applications. Future studies could focus on exploring the upstream and downstream regulators of GSDMD in lung cancer, as well as developing targeted therapies that can modulate its function to benefit patients.

Data Sharing Statement

All data are available in the main text. Upon reasonable request, the corresponding author can provide anonymized clinical data supporting the results of this study.

Ethics Statement

The study was conducted in accordance with the Declaration of Helsinki, received approval from the Ethics Committee of Hebei General Hospital [Approval Number:2024(288)], and all participants provided informed consent.

Author Contributions

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Funding

This work was supported by Central Government Guides Local Funds for Science and Technology Development (246Z7750G) provided by hebei province science and technology department.

Disclosure

The authors report no conflicts of interest in this work.

References

1. Sung H, Ferlay J, Siegel RL, et al. Global Cancer Statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 Cancers in 185 Countries. CA. Cancer J Clin. 2021;71(3):209–249. doi:10.3322/caac.21660

2. Wu YL, Planchard D, Lu S, et al. Pan-Asian adapted clinical practice guidelines for the management of patients with metastatic non-small cell lung cancer. Lancet Oncol. 2022;23(3):e123–e135. doi:10.1016/S1470-2045(21)00694-3

3. Arbour KC, Riely GJ. Systemic therapy for locally advanced and metastatic non-small cell lung cancer: a review. JAMA. 2019;322(8):764–774. doi:10.1001/jama.2019.11058

4. Malhotra J, Malvezzi M, Negri E, et al. Risk factors for lung cancer worldwide. Eur Respir J. 2016;48(3):889–902. doi:10.1183/13993003.00359-2016

5. Takashima S, Sone S, Li F, et al. Small solitary pulmonary nodules (< or =1 cm) detected at population-based CT screening for lung cancer: reliable high-resolution CT features of benign lesions. AJR Am J Roentgenol. 2003;180(4):955–964. doi:10.2214/ajr.180.4.1800955

6. Toyoda Y, Nakayama T, Kusunoki Y, et al. Sensitivity and specificity of lung cancer screening using chest low-dose computed tomography. Br J Cancer. 2008;98(10):1602–1607. doi:10.1038/sj.bjc.6604351

7. Cao P, Jia X, Wang X, et al. Deep learning radiomics for the prediction of epidermal growth factor receptor mutation status based on MRI in brain metastasis from lung adenocarcinoma patients. BMC Cancer. 2025;25(1):443. doi:10.1186/s12885-025-13823-8

8. Wen Q, Qiu L, Qiu C, et al. Artificial intelligence in predicting efficacy and toxicity of Immunotherapy: applications, challenges, and future directions. Cancer Lett. doi:10.1016/j.canlet.2025.217881

9. Jia C, Chen H, Zhang J, et al. Role of pyroptosis in cardiovascular diseases. Int Immunopharmacol. 2019;67:311–318. doi:10.1016/j.intimp.2018.12.028

10. He WT, Wan H, Hu L, et al. Gasdermin D is an executor of pyroptosis and required for interleukin-1β secretion. Cell Res. 2015;25(12):1285–1298. doi:10.1038/cr.2015.139

11. Dai Z, Liu WC, Chen XY, et al. Gasdermin D-mediated pyroptosis: mechanisms, diseases, and inhibitors. Front Immunol. 2023;14:1178662. doi:10.3389/fimmu.2023.1178662

12. Li R, Zeng X, Yang M, et al. Antidiabetic DPP-4 inhibitors reprogram tumor microenvironment that facilitates murine breast cancer metastasis through interaction with cancer cells via a ROS-NF-кB-NLRP3 axis. Front Oncol. 2021;11:728047. doi:10.3389/fonc.2021.728047

13. Shi J, Zhao Y, Wang K, et al. Cleavage of GSDMD by inflammatory caspases determines pyroptotic cell death. Nature. 2015;526(7575):660–665. doi:10.1038/nature15514

14. Fang Y, Tian S, Pan Y, et al. Pyroptosis: a new frontier in cancer. Biomed Pharmacother. 2020;121:109595. doi:10.1016/j.biopha.2019.109595

15. Balkwill F, Mantovani A. Inflammation and cancer: back to Virchow? Lancet. 2001;357(9255):539–545. doi:10.1016/S0140-6736(00)04046-0

16. Johnson DE, Ra O, Grandis JR. Targeting the IL-6/JAK/STAT3 signalling axis in cancer. Nat Rev Clin Oncol. 2018;15(4):234–248. doi:10.1038/nrclinonc.2018.8

17. Zang R, Li Y, Jin R, et al. Enhancement of diagnostic performance in lung cancers by combining CEA and CA125 with autoantibodies detection. Oncoimmunology. 2019;8(10):e1625689. doi:10.1080/2162402X.2019.1625689

18. Gao J, Qiu X, Xi G, et al. Downregulation of GSDMD attenuates tumor proliferation via the intrinsic mitochondrial apoptotic pathway and inhibition of EGFR/Akt signaling and predicts a good prognosis in non‑small cell lung cancer. Oncol Rep. 2018;40(4):1971–1984. doi:10.3892/or.2018.6634

19. Wu J, Wang P, Xie X, et al. Gasdermin D silencing alleviates airway inflammation and remodeling in an ovalbumin-induced asthmatic mouse model. Cell Death Dis. 2024;15(6):400. doi:10.1038/s41419-024-06777-5

20. Traughber CA, Deshpande GM, Neupane K, et al. Myeloid-cell-specific role of Gasdermin D in promoting lung cancer progression in mice. iScience. 2023;26(2):106076. doi:10.1016/j.isci.2023.106076