兽研Abstract
Background/Aim
Regulatory T cells (Tregs) are immunosuppressive lymphocytes that maintain immune homeostasis but may also facilitate tumor immune evasion. Their role in canine cutaneous lymphoma (CCL) remains insufficiently understood. We hypothesized that dogs with CCL would exhibit increased Treg levels in both peripheral blood and tumor tissue, thereby contributing to local and systemic immune modulation. Our objective was to evaluate Treg levels in blood and skin of dogs with CCL compared with healthy controls.
Materials and Methods
Twenty-four dogs were enrolled: 12 with T-cell cutaneous lymphoma confirmed by histopathology and immunohistochemistry and 12 clinically healthy controls, recruited from a veterinary teaching hospital population. In this prospective study, circulating Tregs were quantified by flow cytometry, and FoxP3+ cells in skin were evaluated by immunohistochemistry. Statistical analyses included t-tests, non-parametric tests, and Kaplan–Meier survival analysis.
Results
Dogs with CCL had significantly higher circulating Tregs than controls (10.84±1.18% vs. 5.67±0.89%, p=0.0021). FoxP3+ cells were detected exclusively in neoplastic skin (mean, 8.27±5.71%), with no labeling in healthy skin (p=0.0007). No significant differences were observed between epitheliotropic and non-epitheliotropic subtypes. Treg levels and the Treg/CD8+ ratio were not associated with survival outcomes (p>0.20).
Conclusion
Dogs with CCL demonstrated increased Tregs in both peripheral blood and intratumorally, suggesting a role of these cells in CCL-associated immunomodulation. Although Treg levels were not prognostic in this cohort, their accumulation highlights potential therapeutic targets in veterinary oncology.
Keywords: Immune evasion, regulatory T cells, tumor microenvironment, veterinary oncology
Introduction
Canine cutaneous lymphoma (CCL) is an uncommon neoplastic disorder, representing approximately 3% to 8% of all lymphomas diagnosed in dogs (1, 2). Despite its lymphoid origin, CCL constitutes only about 1% of all cutaneous tumors in this species (3, 4) and accounts for merely 0.02% of routine veterinary consultations in canine practice. Histologically, CCL is classified as either epitheliotropic or non-epitheliotropic. The epitheliotropic subtype most frequently arises from T lymphocytes, with neoplastic cells confined to the epidermis and demonstrating tropism for hair follicles and sweat glands (5). Based on criteria established in human medicine, the epitheliotropic form can manifest in three distinct variants: mycosis fungoides, pagetoid reticulosis, and Sézary syndrome (6). Non-epitheliotropic lymphoma, on the other hand, is characterized by infiltration of neoplastic lymphocytes in the mid to deep dermis and hypodermis, sparing the epidermis and papillary dermis. These lesions may appear as dermal or subcutaneous plaques or nodules, either solitary or multiple (7).
Regulatory T cells (Tregs) are a distinct subset of T lymphocytes with immunosuppressive properties. Under physiological conditions, they act to prevent autoimmune and allergic responses by suppressing other immune cells (8). Furthermore, they have been extensively studied for their role in regulating immune responses in various infectious and neoplastic diseases, as they can also inhibit immune activity against pathogens and tumor cells (9). The evaluation of Tregs in lymphoma has gained increasing relevance due to their pivotal role in modulating the antitumor immune response and influencing disease progression and therapeutic outcomes (10).
Several studies have demonstrated that Tregs, typically characterized by the CD4+FoxP3+ phenotype, are elevated in dogs with different anatomical forms of lymphoma, indicating their potential role in tumor immune evasion. O’Neill et al. (11) reported that dogs with multicentric lymphoma had significantly higher percentages of circulating Tregs and an increased Treg/CD8+ T-cell ratio compared with healthy controls, suggesting that lymphomas may promote immunosuppressive microenvironments. Similar findings were observed by Vascellari et al. (12), who demonstrated an increased number of Tregs in both peripheral blood and lymphoid tissues of dogs with high-grade B-cell lymphoma. Furthermore, studies have shown that chemotherapy can modulate Treg levels. Post-treatment evaluations revealed a significant decrease in circulating Tregs following doxorubicin-based protocols, and dogs with lower post-chemotherapy Treg counts exhibited prolonged survival (13). These findings support the hypothesis that Tregs not only participate in the immunopathogenesis of canine lymphoma but may also serve as biomarkers for treatment response and prognosis, reinforcing their relevance as potential therapeutic targets in veterinary oncology.
To the best of the authors’ knowledge, no published studies have specifically evaluated Tregs in canine cutaneous lymphoma. Therefore, the present study aimed to assess Tregs in both peripheral blood and tumor tissue of dogs with cutaneous lymphoma using flow cytometry and immunohistochemistry, with the goal of elucidating their role in tumor progression and prognosis.
Materials and Methods
Experimental groups and patient enrollment. The study population consisted of 24 dogs (both males and females), divided into two groups: a control group (CG) comprising 12 clinically healthy dogs, and a lymphoma group (CLG) composed of 12 dogs diagnosed with T-cell immunophenotyped cutaneous lymphoma.
Only dogs diagnosed with cutaneous lymphoma were included. These dogs underwent a thorough physical examination, diagnostic imaging (thoracic radiography and abdominal ultrasonography), and hematologic and serum biochemical analyses, including liver and kidney function tests, as part of the routine pre-treatment assessment. Animals presenting with skin nodules or lesions clinically suspected of CCL were initially subjected to cytological examination via fine-needle aspiration of the affected area(s). Following cytological confirmation of lymphoma, an incisional biopsy of the lesion was performed for histopathological evaluation. In dogs from the control group, skin biopsies were obtained, with owner consent, from dogs undergoing castration.
Dogs with lymphomas in other anatomical sites, as identified by diagnostic imaging, and dogs with recurrent disease (neoplastic or infectious) were excluded from the study.
Sample collection. Peripheral blood samples were collected from both the CG and the CLG via jugular venipuncture under aseptic conditions, using a vacuum collection system and 10-ml heparinized tubes. Samples were obtained prior to treatment initiation in the CLG, following histopathological confirmation of the neoplasm, and were processed for subsequent flow cytometric analysis. After diagnosis, 10 of the 12 dogs in the CLG received chemotherapy. The treatment protocol consisted of oral lomustine administered at doses ranging from 70 to 90 mg/m2 every 21 days, in combination with oral prednisone at 1 mg/kg/day, tapered according to the clinical course of each patient. The remaining two animals did not undergo treatment because of owner decision.
Formalin-fixed (10% buffered, pH 7.6) cutaneous lymphoma tissue samples and controls were transferred to 70% ethanol after 24 h and processed using standard histological techniques. Histopathological grading and classification into epitheliotropic or non-epitheliotropic subtypes were performed according to previous reports (14). Immunohistochemical analyses were conducted for immunophenotyping and for evaluation of regulatory T-cell expression, following established protocols in the literature (15).
Immunohistochemistry evaluation and assessment. Immunohistochemical staining was performed using the horseradish peroxidase (HRP) polymer method (Dako K4068, Glostrup, Denmark), with 3,3′-diaminobenzidine tetrachloride (DAB; Dako K3468-1) and the LSAB 2 System-HRP (DakoCytomation, Code K0672, Carpinteria, CA, USA). Briefly, tissue sections (4 μm) were mounted on positively charged slides, incubated at 57°C for 24 h, deparaffinized in xylene, and rehydrated through a graded ethanol series. Antigen retrieval was carried out by heat-induced epitope retrieval using a citrate buffer solution (pH 6.0) in a pressure cooker (Pascal®; Dako) for approximately 30 at 128°C. After cooling to room temperature, slides were rinsed in deionized water and incubated twice for 10 min in 3% hydrogen peroxide to block endogenous peroxidase activity, followed by another 5 min rinse in deionized water. Nonspecific protein binding was blocked using a commercial serum-free protein block (Dako X0909).
Primary antibodies were applied at standardized dilutions and incubated overnight at 4°C (18 h) for anti-CD3 and anti-CD79α, and for 2 hours at room temperature for anti-FoxP3. Mouse monoclonal anti-human CD3 (clone F7.2.38, M7254; Dako; dilution 1:100) and mouse monoclonal anti-human CD79α (clone HM57, M7051; Dako; dilution 1:50) were widely used and validated for cross-reactivity with canine tissues (16, 17). For Treg analysis, a mouse anti-human FoxP3 antibody (clone eBio7979, dilution 1:100, eBioscience, San Diego, CA, USA), previously validated for cross-reactivity with canine tissue, was employed (18). Slides were subsequently washed in Tris-buffered saline (TBS). Detection was performed using the EnVision polymer (Dako) for CD3 and CD79α, and the LSAB system for FoxP3. Slides were washed three times in TBS, followed by chromogen development with DAB (DakoCytomation) and counterstained with Harris hematoxylin. Finally, slides were dehydrated through graded ethanol and xylene and coverslipped with mounting resin.
Lymph nodes from healthy dogs were used as positive controls. For negative controls, the primary antibody was replaced with mouse immunoglobulin at the same concentration as each primary antibody. CD3 and CD79α staining was used to determine the immunophenotype of cutaneous lymphomas, with CD3-positive/CD79α-negative samples classified as T-cell lymphomas. Only these were included in the study, as B-cell lymphomas were underrepresented.
FoxP3 staining was evaluated under light microscopy at 400× magnification. Each tissue sample represented a single biological replicate obtained from an individual dog. Immunohistochemical staining was performed once for each sample, and no technical replicates were generated. Quantification was based on the evaluation of five representative microscopic fields per section, with a total of 500 cells (both positive and negative) counted per sample. Mean values were used to calculate the percentage of regulatory T cells in both cutaneous lymphoma and healthy skin samples.
Flow cytometry analysis. Flow cytometry was performed according to Munhoz et al. (13). Peripheral blood mononuclear cells were isolated from 15mL of venous blood collected in heparinized tubes. Samples were diluted 1:1 (v/v) in sterile phosphate-buffered saline (PBS) and layered over a Ficoll-Hypaque density gradient (1.077 g/ml; Ficoll-Paque PLUS, Amersham Biosciences, Uppsala, Sweden). After centrifugation at 720 × g for 30 min, the mononuclear cell layer was collected, washed in PBS, and centrifuged again at 720 × g for 10 min. The leukocyte pellet was resuspended in 2 ml of ACK lysis buffer (NH4Cl 0.15 mM, KHCO3 10 mM, EDTA 0.1 mM; pH 7.3) for 5 min to lyse residual erythrocytes, followed by washing and resuspension in 1 ml of PBS. The cells were then diluted 1:20 in 0.5% trypan blue and counted using a Neubauer chamber.
For flow cytometric analysis, cells were distributed into six distinct labeling conditions. Unstained and isotype controls were prepared for both surface and intracellular staining. For surface marker analysis, cells were incubated with anti-CD4 FITC, anti-CD8 PE, and PanT APC antibodies. For regulatory T-cell analysis, cells were incubated with anti-CD4 FITC, PanT APC, and anti-FoxP3 PE antibodies (Table I). Tubes designated for intracellular staining (FoxP3) underwent fixation and permeabilization using commercial buffers (eBioscience®, cat. no. 00-5223 and 00-8333) according to the manufacturer’s instructions. To minimize nonspecific binding, 5 μl of heat-inactivated autologous serum (56°C for 30 min) was added to each sample prior to both surface and intracellular staining, with incubation periods of 30 and 15 min, respectively. After staining, 100 μl of 1% paraformaldehyde in PBS was added to each tube for cell preservation. Flow cytometry was performed within 24 hours of blood collection using a FACSCanto II cytometer (BD Biosciences®, San Jose, CA, USA). Each blood sample represented a single biological replicate obtained from an individual dog. Flow cytometric analysis was performed once per sample without technical replicates. A minimum of 10,000 lymphocyte-gated events were acquired and analyzed for each specimen with FACSDiva software.
[表格内容请查看原文]
Statistical analysis. An independent-samples t-test was used for comparisons between two variables that met the assumptions of normality and homogeneity. When these assumptions were not met, the Mann–Whitney test was applied. For comparisons involving three variables, a nonparametric analysis of variance (Kruskal–Wallis) was performed, followed by pairwise comparisons using Dunn’s test. Survival curves were evaluated using the Kaplan–Meier method, with results dichotomized into low and high groups based on the median value. Correlations between techniques were assessed using Spearman’s rank correlation test. A p-value <0.05 was considered statistically significant. All statistical analyses were conducted using GraphPad Prism 5 (GraphPad Software, San Diego, CA, USA).
Overall survival for dogs diagnosed with T-cell cutaneous lymphoma was calculated from the date of histopathological diagnosis to the time of death.
Ethics. The experimental protocol was reviewed and approved by the Ethics Committee on Animal Use (approval number 10207/16). All procedures followed internationally recognized standards of best clinical practice for veterinary patient care. The animals enrolled in the study were privately owned patients attended at the Veterinary Teaching Hospital of São Paulo State University, and their owners provided written informed consent for their participation in the research.
Results
Clinicopathological data. This study included 12 dogs diagnosed with CCL. The mean age was 7.5±2.94 years, with 7 females and 5 males. Most were mixed-breed, whereas five were purebred. The mean survival time was 71.1±63.9 days, and the mean percentage of FoxP3-positive regulatory T cells was 8.27±5.71% (Table II). Based on the histological distribution of neoplastic lymphocytes within the epithelial tissue, cutaneous lymphomas were classified as either epitheliotropic (n=5; 42%) or non-epitheliotropic (n=7; 58%). All dogs presented T-cell lymphomas (n=12). The mean survival time was 71.1±63.9 days, ranging from 17 to 180 days. Analysis of survival data in relation to epitheliotropism revealed no statistically significant differences (p=0.35).
[表格内容请查看原文]
Immunohistochemistry and flow cytometry analysis. Treg cell staining was confirmed by the presence of brown nuclear staining (Figure 1). No FoxP3 immunolabeling was detected in healthy skin samples, consistent with the absence of regulatory T cells in these tissues. By contrast, CCL samples contained a mean of 8% (±5%) FoxP3-positive cells, indicating enrichment of Tregs within the tumor microenvironment. Therefore, a significantly higher number of Treg cells was observed in CCL compared with normal skin (p=0.0007).
[图：Assessment of T reg cells (FoxP3) in canine cutaneous lymphoma (CCL). (A) Nuclear FoxP3 immunoexpression, indicating the]
No significant difference was observed in the proportion of FoxP3+ cells between the epitheliotropic and non-epitheliotropic subtypes (11±7% and 7±4.9%, respectively). Nonetheless, both subtypes exhibited significantly higher FoxP3+ cell percentages compared with the control group (p=0.0011) (Figure 2A). Survival analysis was conducted by stratifying CLG dogs into high and low FoxP3 expression groups based on the median tissue expression level. Although a trend toward differential survival was noted, no statistically significant difference was detected between the subgroups (p=0.677), suggesting that FoxP3 expression alone may not be a prognostic determinant in this cohort (Figure 2B).
[图：FoxP3 immunoexpression associated with clinicopathological factors. (A) Percentage values of FoxP3 tissue expression in ]
In the flow cytometry analysis, dogs in the CLG group showed a significantly higher mean percentage of regulatory T cells (within the PanT CD4+ gate) (10.84±1.18) compared with the control group (5.67±0.89) (p=0.0021). Regarding the percentages of CD4+ and CD8+ T-cell subpopulations, no significant differences were observed between the control group (51.01±4.92 and 31.78±5.65, respectively) and the CLG group (48.26±4.54 and 30.32±4.82, respectively), with p-values of 0.68 and 0.84, respectively.
The percentage of Tregs did not differ significantly between dogs with epitheliotropic cutaneous lymphoma (9.52±1.41) and those with non-epitheliotropic cutaneous lymphoma (11.79±1.76). However, a statistically significant difference was observed between the non-epitheliotropic group and the control group (5.67±0.89), indicating a higher proportion of regulatory T cells in non-epitheliotropic cutaneous lymphoma (p=0.001).
Regarding the proportions of CD4+ and CD8+ T cells, no significant differences were found between the epitheliotropic (45.82±7.84 and 33.30±9.26, respectively) and non-epitheliotropic subtypes (50.00±5.86 and 28.19±5.51, respectively), nor when compared with the control group (51.01±4.92 and 31.78±5.65, respectively), with p-values of 0.58 and 0.79. Group means and standard errors of the mean (SEM) are shown in Figure 3.
[图：Evaluation of T-lymphocyte subpopulations in the peripheral blood of healthy dogs (control group, CG) and dogs with epit]
When assessing survival distribution among dogs with cutaneous lymphoma and either low or high percentages of circulating Tregs, no statistically significant difference was observed (p=0.20). Similarly, survival analysis based on the Treg/CD8+ ratio showed no significant difference (p=0.25). Representative survival curves for Treg percentage and Treg/CD8+ ratio are shown in Figure 4.
[图：Survival curves of dogs with cutaneous lymphoma according to low and high percentages of circulating T regulatory (T reg]
Spearman’s correlation test revealed no significant correlation between the percentage of Tregs in peripheral blood and the percentage of Tregs within the neoplastic tissue of dogs with cutaneous lymphoma (p=0.83). For this analysis, the percentage of Tregs (within the PanT CD4+ gate) obtained by flow cytometry was compared with the percentage of FoxP3-positive cells determined by immunohistochemistry.
Discussion
Canine cutaneous lymphoma is a rare neoplasm, most commonly diagnosed within the field of dermatology and considered uncommon among all tumor types (3). In the present study, the mean age of affected dogs was 7.5 years, closely matching values reported in the literature, and no sex predisposition was observed (3, 19, 20). Most of the dogs in this cohort were mixed-breed, which limits direct comparison with the literature because of regional differences in breed distribution.
Chemotherapy with lomustine resulted in a low median survival time of 71 days. Nonetheless, this outcome is consistent with previous reports involving T-cell immunophenotypes (3, 21) and supports the continued use of lomustine as a treatment option for canine cutaneous lymphoma.
Ongoing advances in cancer research have focused on identifying tumor microenvironmental mechanisms that facilitate neoplastic cell proliferation, dissemination, and immune evasion. Targeting these pathways may provide new therapeutic strategies to alter disease progression. Within this context, the immune system plays a central role, particularly Tregs, which are increasingly studied for their involvement in suppressing antitumor immune responses and are considered promising targets for novel immunotherapies (8, 22).
Among the various markers proposed for identifying Tregs, FoxP3 remains the most widely used in both human and veterinary research, across oncology and other disciplines (23-28). In the present study, significantly higher FoxP3 expression was observed by immunohistochemistry in the neoplastic tissue of dogs with cutaneous lymphoma. In contrast, healthy skin samples lacked detectable FoxP3 immunostaining, suggesting a lower frequency of FoxP3-positive immune cells in normal skin. These findings indicate an increased accumulation of FoxP3-positive cells within the tumor microenvironment.
Although FoxP3 is widely regarded as the master transcription factor of Tregs, its expression is not entirely restricted to this population. Activated conventional T lymphocytes may transiently express FoxP3, and FoxP3-positive cells have also been identified in other tumor-associated immune populations. Therefore, immunohistochemical detection of FoxP3 should be interpreted cautiously, as it may not exclusively represent suppressive Tregs. Recent studies have highlighted the biological heterogeneity of FoxP3-expressing cells and emphasized that the prognostic and immunological significance of FoxP3-positive infiltrates may depend on both their cellular composition and functional state within the tumor microenvironment (29-31). Nevertheless, increased FoxP3-positive infiltrates are generally associated with immunosuppressive conditions and remain a useful surrogate marker for evaluating regulatory immune responses in neoplastic tissues.
Kim et al. (32) and Mucha et al. (22) also used immunohistochemistry to investigate Tregs in the tumor microenvironment of canine mammary carcinomas and reported high levels of immunosuppressive Tregs. Similarly, Pinheiro et al. (28) identified increased FoxP3+ cells in lymph nodes of dogs with multicentric B-cell lymphoma. These studies reported negative prognostic associations, including high Treg levels in high-grade tumors and shorter disease-free and overall survival in lymphomas.
However, no significant differences were observed between FoxP3+ cell levels in the tissue and histopathological subtype or survival outcomes in our study. This may be attributed to the limited number of animals included. Moreover, canine cutaneous lymphoma is often aggressive and lacks clearly established prognostic factors, apart from immunophenotype. By selecting only T-cell cases, the present study achieved greater group homogeneity, which may have reduced intergroup variability. Another possible explanation is that, regardless of histological subtype, neoplastic lymphocytes may activate similar signaling pathways that promote Treg differentiation or recruitment within the tumor microenvironment.
The mechanisms leading to increased Tregs in the tumor microenvironment remain poorly understood. It has been suggested that local Treg proliferation may occur through the conversion of naïve CD4+ T cells upon exposure to cytokines such as IL-10 and TGF-β secreted by the tumor, or via selective migration mediated by tumor-derived chemotactic signals (8, 33-36). Tominaga et al. (36) demonstrated that dogs with melanoma exhibited increased Tregs associated with elevated expression of TGF-β, COX-2, and prostaglandin E2, which may promote both local proliferation and in situ conversion of CD4+ cells into Tregs.
In peripheral blood, dogs with cutaneous lymphoma showed more than a 90% increase in Treg frequency compared with control dogs. Mucha et al. (22) similarly reported increased circulating Tregs in bitches with mammary tumors at various clinical stages. Comparable findings have been reported across different tumor types (11, 13, 24, 25, 37), supporting the validity of our results. Anai et al. (37) found that dogs with multicentric lymphoma had higher Treg percentages than healthy controls, and Treg levels decreased markedly during chemotherapy with a 19-week CHOP protocol.
In the present study, circulating Tregs and tissue FoxP3-positive cell populations were elevated compared with controls. It is possible that proliferation and/or differentiation of Tregs within the tumor contributed to their accumulation in secondary lymphoid organs, such as lymph nodes (which were not assessed here), thereby increasing their concentration in peripheral blood. However, when comparing tissue and blood Treg values within each individual, no correlation was observed, likely because of the small sample size and the use of different methodologies (flow cytometry vs. immunohistochemistry).
Tregs are known to suppress the proliferation and activity of B cells, natural killer cells, and CD4+ and CD8+ T cells through various mechanisms, including cell-to-cell contact and secretion of inhibitory cytokines such as IL-10 and TGF-β (22, 33). CD4+ and CD8+ T cells play essential roles in antitumor immunity and are commonly found in human and murine tumors (38). In this study, no significant differences were observed in CD4+ and CD8+ T-cell percentages between healthy dogs and those with cutaneous lymphoma, which is consistent with previous findings in dogs with various cancers (13, 22, 37, 39).
Conversely, O’Neill et al. (11) reported reduced CD8+ T cells in dogs with cancer and decreased CD4+ cells in those with carcinomas. Biller et al. (40) also reported lower CD8+ counts in dogs with osteosarcoma, although CD4+ levels remained unchanged. Horiuchi et al. (27, 41) observed decreased CD4+ and CD8+ cells in dogs with metastatic tumors and oral melanoma, suggesting that metastasis may be associated with impaired type I immunity.
It is important to note that this study assessed only the numerical frequency of lymphocyte populations, without evaluating their functional status. Therefore, equal frequencies between groups may not reflect similar immunological activity. Tregs may inhibit the function of other immune cells without altering their numbers
Currently, no drugs are available that selectively target Tregs in dogs (42). However, some therapeutic agents are being investigated for their ability to reduce Treg numbers while simultaneously exerting cytotoxic effects on tumor cells. Our findings suggest that further studies are warranted to evaluate such therapies in dogs with cutaneous lymphoma, ideally through longitudinal assessments of Treg levels and clinical outcomes. Given the rarity of this tumor, the limited treatment options, and the often poor prognosis, immunomodulatory approaches may represent a promising strategy to improve therapeutic success in affected dogs.
Conclusion
The tumor microenvironment in canine lymphoma exhibits an increased presence of Tregs, and dogs with cutaneous lymphoma have a higher percentage of circulating Tregs compared with healthy animals. No differences were observed in the percentages of CD4+ and CD8+ T cells. Moreover, no association was found between Treg levels or the Treg/CD8+ ratio and overall survival in affected dogs.
Conflicts of Interest
The Authors declare that they have no conflicts of interest relevant to this study.
Authors’ Contributions
Lívia Maria Souza Semolin: Writing – review & editing, writing – original draft, visualization, project administration, methodology, investigation, formal analysis, data curation, conceptualization; Letícia Abrahão Anai: writing – review & editing, writing – original draft, visualization, project administration, methodology, investigation, formal analysis, data curation, conceptualization; Paulo César Jark: visualization, methodology, investigation, validation, writing – review & editing; Thiago Demarchi Munhoz: methodology, investigation, data curation; Giovanna Gabrielle Cruvinel: methodology, investigation, data curation; Rafael Costa Bittencourt: methodology, investigation, data curation; Márcia Ferreira da Rosa Sobreira: methodology, investigation, data curation; Annelise Carla Camplesi: methodology, investigation, data curation; Juliana Rodrigues Leitão: validation, writing – review & editing; Aureo Evangelista Santana: validation, writing – review & editing; Carlos Eduardo Fonseca Alves: writing – review & editing, visualization, validation, supervision, project administration, conceptualization.
Acknowledgements
The Authors acknowledge financial support from the National Council for Scientific and Technological Development (CNPq), Brazil.
Funding
The study was funded by National Council for Scientific and Technological Development (CNPq), grant number: # 308368/2025-8.
Artificial Intelligence (AI) Disclosure
During the preparation of this manuscript, a large language model (ChatGPT, OpenAI) was used solely for language editing and stylistic improvements in select paragraphs. No sections involving the generation, analysis, or interpretation of research data were produced by generative AI. All scientific content was created and verified by the authors. Furthermore, no figures or visual data were generated or modified using generative AI or machine learning–based image enhancement tools.
References
[内容请查看原文]