Introduction

Chronic obstructive pulmonary disease (COPD) is a highly prevalent inflammatory disorder of the airways, representing a major global health concern and ranking among the leading causes of morbidity and mortality1. Characterized by irreversible airflow limitation resulting from structural damage to both the airways and alveoli, COPD is frequently punctuated by exacerbations that accelerate the decline in pulmonary function and contribute to disease2. Alongside this deterioration, reduced physical activity—driven by progressive respiratory limitation and systemic comorbidities—is commonly observed in COPD3,4 and is strongly associated with increased risk of exacerbations and mortality5,6.

Beyond the mechanical and inflammatory components traditionally implicated in COPD, accumulating evidence suggests a central role for immune dysregulation in its pathogenesis7. In particular, chronic inflammation appears to compromise immune homeostasis through the upregulation of inhibitory immune checkpoints and the dysfunction of T-cell responses8. Among these, the programmed cell death protein 1 (PD-1) and its ligand PD-L1 have been identified as critical regulators of T-cell activation, antigen-specific responses, and immunosenescence9,10. In brief, PD-1, an inhibitory receptor expressed on activated T cells, binds to its ligand PD-L1, expressed on antigen-presenting cells such as monocytes, macrophages and dendritic cells, to suppress T cell responses and prevent tissue damage and autoimmunity. However, their aberrant expression has been associated not only with increased susceptibility to COPD exacerbations, but also with a higher risk of tumorigenesis, positioning the PD-1/PD-L1 axis as a potential therapeutic target in this population11,12,13.

Importantly, although PD-1 and PD-L1 are primarily membrane-bound proteins, circulating soluble forms (sPD-1 and sPD-L1) can also be detected in plasma, likely generated through alternative splicing or proteolytic shedding. These soluble isoforms are increasingly recognized as immunomodulatory molecules, capable of interfering with the PD-1/PD-L1 pathway and serving as biomarkers of disease activity and prognosis in several chronic and malignant conditions14,15. Nevertheless, their role in COPD and their relationship with surface expression or T-cell function remain poorly defined.

Interestingly, physical activity has been shown to exert systemic anti-inflammatory effects and to counteract immunosenescence, thereby enhancing immune surveillance16. In oncological settings, regular physical activity has also been found to synergize with immune checkpoint blockade therapies, suggesting that lifestyle interventions may influence the expression and function of these regulatory pathways17. However, the relationship between physical activity and immune checkpoint modulation in COPD remains poorly understood.

In this context, the present study aims to evaluate the impact of physical activity on PD-1/PD-L1 expression in patients with COPD and to explore its association with clinical outcomes, particularly exacerbation frequency. We further examine the effects of a structured physical activity program on the regulation of this immune checkpoint axis. We hypothesize that increased physical activity may attenuate PD-1/PD-L1 expression and improve T-cell function, potentially contributing to a more favorable disease trajectory and offering a novel adjunctive strategy in the comprehensive management of COPD.

Methods

All methods were performed in accordance with the relevant guidelines and regulations.

Study subjects

COPD patients were consecutively recruited from the Pulmonology Department at Hospital Universitario La Paz (Madrid, Spain). Inclusion criteria were: age > 35 years; previous clinical diagnosis of COPD; a post-bronchodilator FEV1/FVC ratio below the lower limit of normal; post-bronchodilator FEV1 < 70% of the predicted value; current or former smoking history ≥ 10 pack-years; and stable pharmacologic treatment aligned with current guidelines, with no changes in the preceding 8 weeks. Detailed exclusion criteria are provided in the Online Supplement Material. The study was approved by the institutional ethics committee of the Hospital Universitario La Paz, Madrid, Spain (PI-2816) and all participants provided written informed consent. Research involving human research participants has been performed in accordance with the Declaration of Helsinki.

Intervention

Participants enrolled in a 6-month individualized physical activity promotion program specifically designed for patients with COPD. The program was implemented by physicians and a study nurse from a specialized COPD outpatient clinic, ensuring continuity of care and a high degree of disease-specific expertise. Each participant attended an initial educational and motivational counseling session lasting approximately 30 min, followed by monthly reinforcement sessions of similar duration. Educational content addressed COPD self-management, including recognition and prompt management of exacerbations, correct inhaler technique, adherence to prescribed pharmacological therapy, smoking cessation support where appropriate, nutritional advice to optimize physical conditioning, and strategies to overcome common barriers to maintaining regular activity. Motivational techniques included the establishment of progressive, individualized goals, the use of self-reward strategies, and problem-solving approaches for periods of reduced adherence. Printed educational materials summarizing the key points were provided to each participant.

Daily self-monitoring of physical activity was encouraged through the use of a pedometer (ONWALK 100, SMT, Kedah, Malaysia), and participants were instructed to maintain a daily activity log. A personalized walking plan was developed for each participant based on their baseline physical activity, as recorded during the initial week of monitoring. The progression targets were individually adjusted at each monthly follow-up visit according to the pedometer log from the previous month. Specifically: (i) participants averaging fewer than 6,000 steps/day were encouraged to exceed this threshold; (ii) those achieving between 6,000 and 9,000 steps/day were advised to increase their daily steps to above 9,000; (iii) participants already achieving between 9,000 and 10,000 steps/day were encouraged to reach the 10,000-step target; and (iv) individuals consistently exceeding 10,000 steps/day were advised to maintain this level of activity. This stepwise, goal-oriented approach was designed to ensure realistic, achievable increases in physical activity.

Measurements

At baseline, data were collected on demographics, smoking history, dyspnea (mMRC scale), comorbidities, prior-year COPD exacerbations, and current treatment. Pulmonary function was assessed via spirometry, body plethysmography, and diffusion capacity for carbon monoxide (DLCO). Respiratory and peripheral muscle strength were measured using maximal inspiratory pressure (PImax) and handgrip dynamometry, respectively. Exercise tolerance was evaluated with the six-minute walk test (6MWT). A full description of study procedures is available in Supplementary Material.

Daily physical activity was objectively quantified before and after the intervention using a validated multisensor armband (SenseWear, BodyMedia Inc., Pittsburgh, PA, USA), worn continuously (except during hygiene activities) for 7 consecutive days18. The device integrates a biaxial accelerometer and physiological sensors to estimate step count, energy expenditure, and metabolic equivalents19,20. Physical activity level (PAL) was calculated as total daily energy expenditure divided by resting (sleep) energy expenditure21. PAL categories were defined as: ≥1.65 (moderately active), 1.35–1.65 (predominantly sedentary), and < 1.35 (very sedentary)18,20,22. A minimum of 4 valid days (including ≥ 1 weekend day, with ≥ 22 h of wear time/day) was required for data inclusion.

Blood collection and cell isolation

Peripheral blood (18 mL) was collected by venipuncture into EDTA tubes at baseline and post-intervention. Immediately after blood sample collection, peripheral blood mononuclear cells (PBMCs) were isolated via Ficoll-Paque Plus (Amersham Biosciences, UK) density gradient centrifugation and processed for immunophenotyping of T cells and monocytes using multicolor flow cytometry. Plasma aliquots were stored at − 80 °C for subsequent soluble immune checkpoint analysis. PBMCs were preserved for flow cytometry, mRNA extraction and proliferation assays.

Quantification of soluble immune checkpoints

Soluble PD-1 and PD-L1 levels in plasma were measured using a bead-based multiplex immunoassay (BioLegend, San Diego, CA; #740961). Bead acquisition was performed on a FACSCalibur flow cytometer (BD Biosciences, Franklin Lakes, NJ), and data were analyzed using LEGENDplex™ Data Analysis Software v8 (BioLegend).

Flow cytometry and T cell proliferation assays

PBMCs were processed for immunophenotyping of T cells and monocytes using multicolor flow cytometry. Cells were stained with fluorescently labeled antibodies targeting surface markers including CD14, CD3, CD4, CD8, PD-1, and PD-L1, and data were acquired using a FACSCalibur flow cytometer and cytometry results were analyzed using FlowJo™ v10.8 Software (BD Life Sciences) (Figure S1).

T cell proliferation capacity was assessed using the CellTrace CFSE assay. PBMCs were labeled with CFSE, cultured under basal conditions or stimulated with pokeweed mitogen, and incubated with or without the PD-1 blocking antibody nivolumab. After 96 h, cells were stained with CD4 or CD8 markers, and CFSE dilution was analyzed by flow cytometry to quantify proliferative responses.

A more detailed description of flow cytometry protocols, antibody panels, stimulation conditions, and data analysis is available in the Supplementary Material.

Statistical methods

Based on preliminary data from our group, we assumed a standard deviation of 15.5 pg/mL for plasma PD-1 concentrations. Under these conditions, the minimum clinically relevant difference was defined as 5 pg/mL. For a two-sided paired comparison (pre–post), with a type I error (α) of 0.05 and a type II error (β) of 0.20 (power 80%), the required sample size was calculated to be 76 evaluable patients. Considering an anticipated dropout rate of 15%, the total sample size was set at 91 patients.

Descriptive results are presented as mean ± standard deviation, median (interquartile range), or frequency (%), as appropriate. Between-group comparisons were assessed using the χ² test for categorical variables, one-way ANOVA for normally distributed continuous variables, and the Kruskal–Wallis test with Dunn’s post hoc correction for non-normally distributed continuous variables. Paired comparisons employed the Wilcoxon signed-rank test or two-way ANOVA with Sidak’s post hoc test. Comparisons of the evaluated biomarkers among the three patient groups according to physical activity level were adjusted for sex, age, BMI, smoking status, and baseline FEV₁ (% predicted) using univariate analysis of variance within the general linear model framework. Correlations were analyzed using Spearman’s rank correlation coefficient (ρ). Statistical significance was set at an adjusted p-value < 0.05. Analyses were conducted with GraphPad Prism v8 (GraphPad Software) and SPSS v20 (IBM). Correlograms were generated using RStudio with the “corrplot” package, displaying only statistically significant Spearman correlations. Additional methodological details, including sample sizes and replicate numbers, are provided in the corresponding figure legends.

Results

Baseline characteristics

A total of 91 patients with clinically stable COPD were recruited. Their baseline characteristics are summarized in Table 1. Based on physical activity level (PAL), 21 patients were classified as moderately active, 42 as moderately sedentary, and 28 as severely sedentary. Severely sedentary patients had lower FEV₁/FVC ratios, higher functional residual capacity, and reduced exercise tolerance—reflected in shorter distances in the six-minute walk test—compared with the other subgroups. No significant differences were found across groups regarding anthropometric variables, comorbidities, or treatment. As expected, the severely sedentary group displayed markedly lower levels of active energy expenditure, step count, and total physical activity time.

Table 1 General characteristics of the study groups*.
Full size table

Plasma PD-1/PD-L1 levels and physical activity

Plasma concentrations of soluble PD-1 and PD-L1 were significantly elevated in patients with severe sedentarism compared to those classified as moderately active (Fig. 1A–B). These differences remained after adjustment for sex, age, body mass index, smoking status, and baseline FEV₁ (Table S3). Both markers showed a significant inverse correlation with physical activity indices, including daily step count and energy expenditure (Fig. 1C–D). Neither PD-1 nor PD-L1 levels were associated with spirometric values or clinical features. However, plasma PD-1 was significantly higher in patients with ≥ 1 moderate-to-severe exacerbation in the previous year (Fig. 1E), while PD-L1 levels were not significantly different (Fig. 1F).

Fig. 1
figure 1

Association between physical activity and plasma levels of soluble PD-1 and PD-L1 in COPD patients. (A) Plasma PD-1 and (B) PD-L1 concentrations in COPD patients stratified by physical activity level (PAL): moderate activity (MA, n = 21), moderate sedentarism (MS, n = 41), and severe sedentarism (SS, n = 27). (C–D) Spearman’s correlation between plasma levels of PD-1 (C) or PD-L1 (D) and objective physical activity parameters (n = 89). Spearman’s rho (ρ) and p-values are shown. (E–F) Plasma concentrations of PD-1 (E) and PD-L1 (F) in patients with no exacerbations in the previous year (n = 64) versus those with ≥ 1 moderate-to-severe exacerbation (n = 25). Data are presented as medians with 95% confidence intervals. Differences are analyzed by Kruskal–Wallis test with Dunn’s post hoc correction; adjusted p-values are indicated.

Full size image

PD-1/PD-L1 expression in peripheral blood cells

Flow cytometry revealed increased PD-1 expression in CD4⁺ T lymphocytes from patients with moderate and severe sedentary behavior, relative to those who were moderately active (Fig. 2A). This pattern was not observed in CD8⁺ T cells. PD-L1 expression in CD14⁺ monocytes was significantly higher in the severely sedentary group (Fig. 2B) (Table S3).

Fig. 2
figure 2

Expression of PD-1/PD-L1 and T-cell proliferative capacity according to physical activity levels in COPD patients. Patients were stratified by physical activity level (PAL) into three groups: moderate activity (MA; PAL > 1.65, yellow), moderate sedentary (MS; PAL 1.35–1.65, orange), and severe sedentary (SS; PAL < 1.35, red). (A) Frequency of PD-1–expressing CD4⁺ T cells (left) and CD8⁺ T cells (right) in MA (n = 20), MS (n = 40), and SS (n = 25) groups. (B) Frequency of PD-L1–expressing CD14⁺ monocytes in the same patient groups. (C) Proliferation of CD4⁺ (left) and CD8⁺ (right) T cells from peripheral blood mononuclear cells (PBMCs) after 96 h stimulation with 2 µg/mL pokeweed mitogen in MA (n = 19), MS (n = 36), and SS (n = 20) patients. (D) Correlation matrix between CD4⁺ and CD8⁺ T-cell proliferation and clinical, physical activity, and immune parameters (n = 75). Only significant Spearman’s correlations (P < 0.05) are shown. (E) T-cell proliferation (CD4⁺, left; CD8⁺, right) following 96 h culture under basal conditions or with PD-1 blockade (Nivolumab) in MA (n = 9), MS (n = 12), and SS (n = 10) groups. Paired data are shown. Data are presented as medians with 95% confidence intervals. Differences are analyzed using Kruskal–Wallis tests with Dunn’s post hoc correction or Wilcoxon matched-pairs signed-rank test, as appropriate. Adjusted P values are shown; n.s., not significant.

Full size image

The proliferative response of CD4⁺ T cells to pokeweed mitogen was significantly greater in moderately active patients than in those who were sedentary (Fig. 2C) and data from correlation slightly positively correlated with physical activity and inversely with PD-1 and PD-L1 expression (Fig. 2D). CD8⁺ T-cell proliferation did not differ across groups (Table S3). To further explore the immune mechanism underlying these effects, we performed an exploratory ex vivo assay using the PD-1 antibody (nivolumab) to assess T-cell proliferation. When PBMCs were cultured in the presence of the PD-1-blocking antibody nivolumab, CD4⁺ T cells from sedentary patients showed restored proliferative capacity (Fig. 2E), suggesting PD-1 signaling in the observed dysfunction. CD8⁺ T-cell proliferation increased with PD-1 blockade across all activity levels.

Effect of physical activity promotion on immune checkpoints and T-cell function

Of the 91 patients initially recruited, 59 completed the 6-month intervention (Table S1). A total of 25 withdrew voluntarily (due to relocation, consent withdrawal, or loss to follow-up during the COVID-19 pandemic), and 7 were excluded due to exacerbations requiring prolonged corticosteroid therapy. A sensitivity analysis revealed no differences between patients who completed the study and those who discontinued (Table S2). In the 59 patients who completed the intervention and were analyzed, adherence to the program was optimal: all participants attended every scheduled counseling session, with 14% of sessions requiring rescheduling but subsequently completed. Compliance with daily physical activity recording was also high, with a mean of 6 ± 5 days without pedometer data over the 6-month period, corresponding to > 90% completion of the activity log in all cases.

Post-intervention, significant increases were observed in physical activity parameters, including daily step count and total activity time (Table 2). Overall, 36 patients (61%) improved their PAL, while 23 (39%) showed no change or a decline.

Table 2 Physical activity characteristics pre- and post-intervention.
Full size table

Patients with increased PAL exhibited a significant reduction in plasma PD-1 levels after the intervention (Fig. 3A); no changes were observed in PD-L1 levels (Fig. 3B). Flow cytometric analysis showed that the proportion of PD-1⁺ CD4⁺ and CD8⁺ T cells decreased significantly in those who improved their PAL (Fig. 3C–D). Notably, PD-L1 expression in CD14⁺ monocytes decreased across all patients, regardless of PAL change (Fig. 3E).

Fig. 3
figure 3

Modulation of PD-1/PD-L1 axis expression following a physical activity intervention in COPD patients. Patients were categorized based on the change in physical activity level (PAL) after intervention: decreased activity (ΔPAL ≤ 0, n = 23) and increased activity (ΔPAL > 0, n = 36). (A) Plasma concentrations of soluble PD-1 and (B) PD-L1 before and after the intervention. (C–E) Frequency of PD-1⁺ CD4⁺ T cells (C), PD-1⁺ CD8⁺ T cells (D), and PD-L1⁺ CD14⁺ monocytes (E) pre- and post-intervention in patients who decreased (n = 23) or increased (n = 32) their PAL. Data are presented as medians with 95% confidence intervals. Comparisons were performed using two-way ANOVA with Sidak’s multiple comparisons test. Adjusted and interaction P values are reported; n.s., not significant.

Full size image

Functionally, CD4⁺ T-cell proliferation improved significantly only among patients who increased their PAL (Fig. 4A), whereas no changes were observed in CD8⁺ T-cell proliferation (Fig. 4B). The ex vivo proliferation with PD-1 antibody data showed an enhanced CD4+ T-cell proliferation after physical activity promotion and PD-1 inhibition (Fig. 4C). In contrast, CD8⁺ T-cell proliferation improved with nivolumab independently of changes in PAL (Fig. 4D). In summary, these results and the ex vivo assays indicate that physical activity enhancement in COPD patients is associated with decreased PD-1 expression and improved CD4⁺ T-cell function, suggesting a modulatory role of activity level on immune checkpoint dynamics. Together, these findings highlight a potential synergistic role for lifestyle interventions and immune modulation strategies in restoring adaptive immune function in COPD.

Fig. 4
figure 4

T-cell proliferative responses before and after a physical activity intervention in COPD patients. T-lymphocyte proliferation was assessed in peripheral blood mononuclear cells (PBMCs) collected pre- (orange) and post-intervention (green). (A–B) Proliferative CD4⁺ (A) and CD8⁺ (B) T-cell responses after 96 h stimulation with 2 µg/mL pokeweed mitogen in patients who decreased (ΔPAL ≤ 0, n = 17) or increased (ΔPAL > 0, n = 19) their physical activity level. (C–D) CD4⁺ (C) and CD8⁺ (D) T-cell proliferation pre- and post-intervention, cultured for 96 h under basal conditions or in the presence of a PD-1 blocking antibody (Nivolumab) (n = 21). Data are presented as medians with 95% confidence intervals. Statistical analysis was performed using two-way ANOVA with Sidak’s multiple comparisons test. Adjusted and interaction P values are shown; n.s., not significant.

Full size image

Discussion

Regular physical activity is widely recommended for patients with COPD due to its association with reduced risk of exacerbations, hospitalizations, and mortality23. However, the immunological mechanisms underlying these benefits remain incompletely understood. This study provides new evidence highlighting the beneficial effects of physical activity on immune function in patients with COPD. Specifically, physical activity was associated with reduced expression of the PD-1/PD-L1 immune checkpoint pathway and enhanced CD4⁺ T-cell proliferative capacity. These findings suggest that engaging in daily physical activity may help restore immune competence and improve overall immune surveillance in COPD, underscoring the importance of promoting an active lifestyle as part of disease management.

Our findings align with the concept of immune dysfunction in COPD, a disease marked by paradoxical immune cell accumulation in the airways alongside heightened susceptibility to infections24,25. The observed elevation of circulating PD-1 and PD-L1 in sedentary patients supports a systemic signature of immune exhaustion, similar to that seen in malignancies26 and fibrotic lung disease27. Notably, physical activity was sufficient to downregulate these checkpoint molecules, pointing to a potentially modifiable target through non-pharmacologic intervention. Furthermore, local inflammatory processes may interact with systemic immune responses. In particular, airway inflammation and excessive nitric oxide production in the peripheral airways are key features of COPD28 and may contribute to the systemic immune alterations observed. Considering these interactions provides a more comprehensive pathophysiological framework for understanding immune dysregulation in COPD”.

Mechanistically, PD-1 is a well-established regulator of T-cell exhaustion, dampening proliferation and cytokine production29,30,31,32. In our cohort, increased PD-1/PD-L1 expression in sedentary individuals correlated with reduced CD4⁺ T-cell proliferation, reinforcing its functional significance. Conversely, patients who improved their activity levels demonstrated enhanced CD4⁺ responses and reduced PD-1 expression—an effect not observed in CD8⁺ T-cells. This selective CD4⁺ responsiveness may reflect the immunosenescent shift commonly seen in COPD, characterized by an inverted CD4⁺/CD8⁺ ratio and accumulation of senescent CD8⁺ subsets33.

Although the present study was not designed to address mechanistic questions, it is nonetheless reasonable to speculate on the potential effects of exercise on the PD-1/PD-L1 immune checkpoint axis in patients with COPD, based on the known determinants of its basal dysregulation. Hypoxia and inflammation, both hallmarks of COPD, are capable of upregulating PD-1/PD-L1 through distinct molecular pathways. For instance, PD-L1 is a direct transcriptional target of HIF-1α, and both PD-1 and PD-L1 have been shown to increase in a hypoxia-dependent manner in chronic respiratory diseases34,35,36. In parallel, NF-κB activation driven by oxidative stress promotes their transcription37,38,39, while persistent T cell activation sustains PD-1 expression via NFAT–TOX signaling40. Regular physical activity may mitigate these processes by improving respiratory function, reducing hypoxia and systemic inflammation41,42, and enhancing the expansion of regulatory T cells, thereby limiting excessive T cell activation43.

Consequently, the restoration of CD4⁺ T-cell function following the intervention suggests that physical activity may partially reverse immunosenescence, a hypothesis supported by previous studies showing that regular exercise expands naïve and central memory CD4⁺ T-cell compartments, reduces senescence markers, and improves antiviral responses44,45,46. These effects are particularly relevant in COPD, where compromised immune function contributes to frequent exacerbations, poor vaccine responses, and increased cancer risk47.

To explore the effects of physical activity on the immune response, we employed a PD-1 antibody (nivolumab), as previously reported by other authors48. Our data showed that physical activity selectively improved CD4⁺ T-cell function. The absence of similar effects in CD8⁺ T cells may reflect the limitations of the ex vivo model rather than a true biological difference. Specifically, (1) the ex vivo assays were intended to provide insights into mechanisms that may occur in vivo; however, these controlled experimental conditions differ markedly from the complex physiological environment of patients. (2) In contrast, a physical activity intervention in vivo exerts systemic effects within the integrated milieu of the human body, involving coordinated interactions among the immune system, skeletal muscle, endocrine pathways, and metabolic networks. Although our study does not provide mechanistic data to definitively explain the selective enhancement of CD4⁺ T cell function, several hypotheses can be considered. Prior research indicates that physical activity may preferentially stimulate T helper cell subsets: in people living with HIV, higher physical activity levels are associated with selective recovery and functional improvement of CD4⁺ T cells49, and in murine models, exercise significantly increases the proliferation of CD4⁺ T cells in response to allogeneic stimulation50. Additionally, physical activity reduces chronic low-grade inflammation and improves lymphoid tissue microenvironments, changes that may preferentially preserve CD4⁺ T helper subsets, which are more dependent than CD8⁺ T cells on sustained survival signals and niche quality51. Together, these factors may explain why systemic exercise-induced immune modulation could have a more pronounced impact on CD4⁺ T cells, in contrast to the direct receptor blockade achieved with Nivolumab.

Although the design of the present study was limited to evaluating the effect of physical activity on the expression of the PD-1/PD-L1 axis, some secondary results allow further insight into which components of exercise may be more strongly implicated in the observed response. Specifically, the increase in physical activity time resulting from the intervention was inversely correlated with changes in plasma PD-1 levels (ρ=−0.275, p = 0.044), PD-1 expression in CD8⁺ T lymphocytes (ρ=−0.344, p = 0.022) and PD-L1 expression in CD14⁺ cells (ρ=−0.259, p = 0.049). In contrast, the increase in daily step count was directly correlated with CD4⁺ T-cell proliferation (ρ = 0.343, p = 0.049), while no significant associations were observed between changes in activity energy expenditure or metabolic activity and the modulation of PD-1/PD-L1 parameters. Taken together, and acknowledging the limited statistical power of our study for this secondary objective, our findings suggest that interventions targeting exercise duration, likely reflecting low-intensity activity, may exert a greater modulatory effect on PD-1/PD-L1 expression than increases in exercise intensity.

Despite these encouraging results, several limitations must be acknowledged. First, the mechanistic pathways linking physical activity to PD-1/PD-L1 modulation remain to be elucidated. Second, our analyses focused exclusively on peripheral immune cells, without assessment of airway-resident lymphocyte populations, which may differ in phenotype and function. Third, to avoid confounding effects of corticosteroids on immune outcomes, we excluded patients receiving systemic or inhaled corticosteroids—thereby limiting the generalizability of the findings to the broader COPD population. Fourth, the study experienced a substantial dropout rate between baseline and follow-up assessments. Although common in long-term behavioral interventions, attrition may introduce selection bias and limit the robustness of longitudinal conclusions. Additionally, some losses occurred due to pandemic-related restrictions, further impacting retention. Lastly, the duration of follow-up was restricted to the intervention period, precluding assessment of long-term immunological or clinical outcomes.

Conclusion

This study highlights physical activity as a potentially modifiable factor influencing immune checkpoint regulation and T-cell function in COPD. Increased physical activity was associated with reduced PD-1/PD-L1 expression and enhanced CD4+ T-cell proliferation, suggesting a potential improvement in immune competence that may help reduce susceptibility to infections and comorbidities. These findings support the integration of structured physical activity programs into comprehensive COPD management. Future studies with longer follow-up and mechanistic exploration are warranted to determine the durability and clinical significance of these effects.