In recent years, the global incidence of thyroid cancer has escalated rapidly. Over the past three decades, the incidence of thyroid cancer has increased by approximately 300%, with PTC being the primary subtype contributing to this rise (1). This increase is largely attributable to advancements in imaging technologies, which have led to an upsurge in the detection of thyroid nodules. Consequently, there has been a significant rise in the number of patients undergoing nonessential investigations and surgeries. Notably, a substantial proportion of these patients, who are diagnosed with incidentally detected thyroid nodules, ultimately undergo surgery. Yet, the majority of these cases are found to be benign upon final pathology (2).

The BRAF mutation in thyroid tissue is mainly caused by the T→A conversion at 1799 nucleotide of exon 15, that is, the mutation at codon V600E. The BRAF^V600E oncoprotein is known to enhance kinase activity, which subsequently activates the MAPK (MEK1/2 and ERK1/2) signaling pathway (3–6). However, the role of the BRAF^V600E mutation as an independent predictor of thyroid cancer remains inconclusive in clinical practice.

The overall immune activity in thyroid cancer is elevated, enabling various immune cells to exert both pro- and anti-tumor effects through distinct pathways, thereby shaping a conducive immune landscape. Tumorigenesis recruits peripheral immune cells, such as B cells and T cells, into the tumor microenvironment, facilitating their interaction with tumor cells. Consequently, monitoring changes in circulating immune cells can provide indirect insights into the tumor microenvironment. Given the lack of quantitative indicators for distinguishing benign from malignant thyroid nodules, peripheral blood cytology testing holds significant clinical potential in aiding the diagnosis of papillary thyroid carcinoma.

Ultrasonography serves as the primary modality for thyroid nodule evaluation, providing real-time assessment of morphological features (size, margins, echogenicity) and vascular patterns. When integrated with the C-TIRADS classification system, it enables reliable differentiation of benign and malignant lesions and guides clinical decision-making regarding fine-needle aspiration biopsy or surgical intervention (7). Its non-invasive nature and diagnostic accuracy establish ultrasonography as the first choice for initial screening, diagnostic evaluation, and longitudinal monitoring of thyroid nodules (8).

This study systematically evaluates associations among BRAF^V600E mutation status, peripheral blood lymphocyte subsets, and ultrasonographic features with PTC risk in thyroid nodules classified as C-TIRADS categories 3+. With PTC as the primary endpoint, key predictive biomarkers were identified and incorporated into a validated nomogram for clinical risk stratification. We anticipate that these findings will help reduce unnecessary surgeries for benign thyroid nodules while providing a foundation for personalized treatment strategies. Specifically, BRAF^V600E mutation testing may guide targeted therapy selection, including BRAF/MEK inhibitors such as dabrafenib/trametinib combination therapy, while lymphocyte subset profiling could provide immunological insights for potential PD-1/PD-L1 checkpoint inhibitor applications in refractory cases.

Building upon prior research, this study innovatively investigated the diagnostic value of lymphocyte subsets in differentiating benign from malignant thyroid nodules. Our analysis revealed that NK cell counts serve as a protective factor against malignancy in thyroid nodules classified as C-TIRADS category 3 or higher. We developed a novel nomogram model incorporating five key variables (BRAF^V600E genotype, NK cell counts, NK cell percentage, CD4+ T cell percentage, and age) to predict malignancy risk in these nodules. Validation demonstrated that the model exhibits excellent discriminative ability, strong calibration, and superior predictive performance compared to the traditional C-TIRADS. These findings suggest significant potential for clinical translation in thyroid nodule risk stratification.

The Bethesda System for Reporting Thyroid Cytopathology (TBSRTC) is a standardized classification system for thyroid FNA cytology, demonstrating high sensitivity and specificity in diagnosing thyroid cancer (9). Previous studies have confirmed the diagnostic accuracy of TBSRTC categories II to VI (10). Accordingly, this study enrolled 210 patients with thyroid nodules classified as C-TIRADS category 3 or above. Based on FNAB results, nodules categorized as Bethesda II were assigned to the benign group, while those classified as Bethesda V or higher and subsequently confirmed as PTC through postoperative histopathological examination comprised the PTC group. The study objectives were to identify potential risk factors associated with PTC development and establish a risk prediction model for discriminating between benign thyroid nodules and PTC, thereby facilitating more accurate clinical diagnosis and treatment decision-making.

This study demonstrated that the BRAF^V600E mutation serves as an independent risk factor for PTC. As a major oncogenic driver, the BRAF^V600E mutation is present in approximately 60% of PTC cases (11). Located on chromosome 7q34, the BRAF gene encodes a threonine/serine protein kinase belonging to the RAF family. As a critical component of the MAPK signaling pathway, BRAF regulates cell growth, proliferation, and apoptosis. Dysregulation of this pathway due to BRAF mutations can contribute to tumorigenesis (12).

Our analysis identified age as an independent protective factor against PTC, corroborating existing epidemiological evidence (13). The observed inverse association may be attributed to age-related biological mechanisms including telomere attrition, stem cell depletion, impaired macroautophagy, and cellular senescence - processes known to mediate environment-dependent tumor suppression and potentially attenuate carcinogenesis risk (14, 15).

The tumor microenvironment (TME) comprises the cellular and molecular milieu supporting tumorigenesis and progression. This complex ecosystem includes not only tumor cells themselves, It also includes innate immune cells (such as macrophages, mast cells, granulocytes, myeloid-derived suppressor cells, dendritic cells, natural killer cells, etc.), adaptive immune cells (such as T lymphocytes, B lymphocytes), interstitial fibroblasts, extracellular matrix, vascular and lymphatic networks, and inflammatory factors and growth produced by them and tumor cells through autocrine or paracrine Factors, chemokines and other non-cellular components. These components engage in dynamic crosstalk that collectively governs tumor progression and metastatic dissemination.

Specific cellular or molecular features in TME have been shown to be independent prognostic factors. Accumulating evidence indicates that lymphocytic infiltration is associated with favorable outcomes across various cancer types (16–19). During thyroid cancer progression, patients exhibit systemic immune activation accompanied by a marked expansion of pro-tumorigenic immune cell populations (20). Notably, PTC with robust pericarcinoma lymphocytic infiltration exhibits an improved prognosis compared to PTC lacking such infiltration (21). This phenomenon may arise from lymphocyte-derived cytokine-mediated tumor suppression or cancer cell evasion of immune surveillance via mechanisms such as major histocompatibility complex (MHC) modulation. Peripheral blood immune cells serve as both systemic immune indicators and active participants in tumor immunomodulation. These cells migrate to tumor sites where they influence local immune responses through multiple mechanisms including chemotaxis, cell-cell signaling, and systemic immune modulation, ultimately affecting tumor progression and therapeutic outcomes (22). Quantitative monitoring of circulating lymphocyte populations may offer valuable insights into cancer development and prognostic evaluation. Given these observations, we hypothesized that peripheral blood lymphocyte subset alterations might influence PTC development. NK cells, critical effectors of innate immunity, contribute substantially to antitumor responses and cancer immunosurveillance (23, 24). Depletion of NK cells has been associated with elevated tumor susceptibility (25, 26). Our findings demonstrate that NK cells serve as protective factors in PTC. Specifically, a 1-cell/μL increment in NK cell counts was associated with a 0.5% reduction in PTC risk. This finding is consistent with previous reports in lung adenocarcinoma (27), colorectal cancer (28), gastric cancer (29), and melanoma (30). Higher tumor-infiltrating NK cell abundance was significantly associated with better clinical outcomes in different types of malignancies. This may be related to the fact that NK cells achieve tumor control by recognizing and killing tumor cells and promoting adaptive T-cell immune responses (31–33).

Our analysis of thyroid nodules classified as C-TIRADS category 3 or higher revealed a distinct pattern in CD4+ T-cell distribution. While absolute CD4+ T-cell counts showed no significant difference between benign lesions and PTC, the relative percentages of CD4+ T cells were markedly elevated in PTC patients compared to their benign counterparts. Notably, this parameter persisted as an independent predictive factor in our LASSO-optimized risk stratification model. These findings contrast with previous reports (34), suggesting complicated thyroid-specific immunomodulatory mechanisms. An investigation in oligometastatic non-small cell lung cancer (NSCLC) has demonstrated significantly elevated peripheral CD4+ T-cell levels in patients with brain metastases compared to healthy controls (35). These findings suggest CD4+ T cells may contribute to anti-tumor immune responses, potentially through specific mechanisms. Substantial interindividual variability exists in CD4+ T cell measurements, with absolute counts being particularly susceptible to confounding factors including active infections, pharmacologic interventions, chronic inflammatory conditions, leukocytosis, and post-splenectomy status (36). Notably, while these factors may significantly elevate absolute CD4+ counts, they generally induce only marginal fluctuations in CD4+ percentages. Consequently, the CD4+ percentages appear to represent a more stable immunological indicator.

Mantovani et al. demonstrated that both innate and acquired immune responses can either promote cancer initiation and tumor progression or exert anticancer effects (37). However, current evidence on the diagnostic value of peripheral blood lymphocyte subsets in PTC remains limited and inconsistent. Some studies report no significant differences in the percentages of CD8+ T cells, CD4+ T cells, and NK cells in peripheral blood between PTC patients and those with benign thyroid nodules (38). In contrast, other research has identified a higher proportion of T cells, CD3+ T cells, CD4+ T cells, CD8+ T cells, and regulatory T cells (Tregs) in the peripheral blood of PTC patients compared to benign cases (34), correlating with PTC aggressiveness (39). Investigations focusing on lymphocyte subset infiltration within thyroid tissue have consistently linked CD8+ T cells, CD4+ T cells, and B cells to PTC (38, 40–45). Notably, our findings align with prior reports (45) indicating the minimal influence of peripheral blood T cells and B cells on PTC development. The observed discrepancies may reflect methodological variations across studies, particularly regarding BRAF^V600E mutation status adjustment, which was adequately controlled in our analysis. Moreover, we employed high-resolution flow cytometry to precisely quantify both absolute lymphocyte counts and subset percentages simultaneously, a methodological advancement over conventional morphological analysis that significantly improves measurement accuracy. The compartmentalized distribution of lymphocyte subsets - with distinct peripheral blood profiles contrasting sharply with tissue infiltration patterns - suggests a model where T and B lymphocytes undergo primary activation within the thyroid tumor microenvironment before systemic dissemination, potentially accompanied by either clonal expansion or functional exhaustion. Consequently, while intratumoral T cells exhibit pronounced activation in PTC, peripheral blood alterations may remain subclinical (38, 46).

In this study, we employed LASSO regression to identify significant predictors for PTC risk among thyroid nodules classified as C-TIRADS category 3 or above. To facilitate clinical implementation, we constructed a nomogram that visually represents the model parameters, enabling intuitive and individualized risk stratification for PTC. The nomogram demonstrated favorable discrimination, accuracy, and clinical utility. Comparative analysis with the conventional C-TIRADS system revealed superior performance, offering enhanced clinical value for individualized risk assessment of thyroid nodules.

This study has several limitations that warrant consideration. Firstly, this study employed peripheral blood analysis to characterize lymphocyte subsets while recognizing that circulating immune profiles may not accurately represent the local thyroid tissue microenvironment, particularly regarding tumor-infiltrating lymphocytes (TILs) and tissue-resident NK cells. The biological interpretation of our findings requires consideration of potential discrepancies in immune cell migration and phenotype between peripheral circulation and thyroid tissue. Furthermore, peripheral lymphocyte proportions may be influenced by systemic inflammatory status and hormonal fluctuations. Future investigations should incorporate single-cell RNA sequencing of FNA specimens or multiplex immunofluorescence analysis to directly evaluate spatial lymphocyte distribution and functional states within thyroid nodules, thereby validating the tissue-level applicability of our model.

Secondly, the relatively small sample size from a single institution restricted the more comprehensive variable analysis to prevent overfitting, and although LASSO regression was used for variable selection, potential residual confounders may remain. Importantly, post-hoc power analysis confirmed our study maintained 92% statistical power to detect moderate effect sizes, ensuring robust reliability of the reported associations. Additionally, the scarcity of Bethesda VI nodules necessitated combining categories V and VI for statistically reliable analysis. These constraints highlight the need for future multi-center studies with larger cohorts to enable stratified analyses and external validation. In subsequent research, we plan to collect comprehensive clinical data (including peripheral blood lymphocyte subsets, C-TIRADS classification, BRAFV600E genotype, and histopathological type) from a minimum of 500 thyroid nodule patients across two provincial tertiary hospitals over a 24-month period. This multicenter validation study will assess the model’s generalizability across diverse populations and the consistency of the NK cell counts-PTC association in different demographic and clinical subgroups.

Thirdly, the cross-sectional nature of our study design precludes causal inference regarding the relationship between identified factors and PTC development. Longitudinal studies would be required to establish temporal relationships and causality.

Finally, while FNAB represents our primary diagnostic modality, its accuracy is inherently influenced by several technical and biological factors including nodule characteristics (size, location), sampling technique, and cytological preparation quality. To maximize diagnostic reliability, we implemented rigorous quality control measures: all procedures were performed by experienced senior clinicians (chief or deputy chief physicians), with strict adherence to the Bethesda System for Reporting Thyroid Cytopathology. This standardized approach helped mitigate potential variability, though the intrinsic limitations of cytological interpretation remain an acknowledged constraint of this study.

This study developed and validated a clinically useful nomogram incorporating BRAF^V600E genotype and lymphocyte subsets, which demonstrated superior diagnostic accuracy in differentiating benign lesions from PTC among C-TIRADS category 3 and higher nodules. The model may help refine risk stratification and reduce unnecessary invasive interventions in indeterminate cases.