The study population comprised of patients with NPC who were initially diagnosed and underwent CRT at Meizhou People’s Hospital from January 2018 to March 2021. The determination of lymphocyte subpopulations was a routine examination conducted on a voluntary basis, as informed by the doctors for nearly all lymphoma and other tumor patients. This study focused primarily on the peripheral immune profiles in patients with non-metastatic NPC. Immune data was collected retrospectively at three different time points: prior to any treatment, at the end of CRT, and one month after CRT. The inclusion criteria for this study were as follows: (1) newly diagnosed, pathologically confirmed nasopharyngeal undifferentiated non-keratinizing carcinoma; (2) age 18-75 years, with no distant metastasis; (3) receiving CRT (induction chemotherapy followed by concurrent chemoradiotherapy; concurrent chemoradiotherapy; induction chemotherapy followed by radical radiotherapy); (4) complete TBNK (T cells, B cells, NK cells) subsets and follow-up data; (5) no local radiotherapy boost for residual lesions; (6) ECOG (Eastern Cooperative Oncology Group) performance status score of 0-2; (7) normal heart, kidney, and liver functions. Exclusion criteria were as follows: (1) presence of other types of malignancies; (2) receiving hormone therapy or other treatments (targeted therapy or surgery); (3) presence of additional chronic diseases or known/suspected active autoimmune diseases. This study was approved by the Medical Ethics Committee of the Meizhou People’s Hospital (Huangtang Hospital) (the ethical approval number: 2024-C-43). The Committee waived the requirement for informed consent as it was a retrospective study. All data were collected in accordance with approved guidelines.

The collected clinicopathological data included age at diagnosis, gender, smoking history, tumor TNM stage, induction chemotherapy regimen, concurrent chemotherapy regimen, use of nimotuzumab, radiotherapy dose, comprehensive treatment plan, and clinical efficacy evaluation. Laboratory data for analysis consisted of the percentage and absolute count of peripheral blood lymphocytes at three time points: pre-treatment, at the end of CRT, and one month after CRT. This encompassed T cells characterized by CD3 expression (CD3⁺ T cells), B cells defined by CD19 expression (CD19⁺ B cells), and T lymphocyte subsets identified by the presence of CD4 (CD3⁺CD4⁺ T cells) and CD8 (CD3⁺CD8⁺ T cells). The combination of CD56 and CD16 facilitated the identification of natural killer (NK) lymphocyte populations. CD45 was used to identify leukocyte antigens in peripheral blood. The clinical efficacy of CRT was evaluated using magnetic resonance imaging (MRI). According to the Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1, clinical efficacy was classified into complete response (CR), partial response (PR), stable disease (SD), and progressive disease (PD).

According to the protocol, the staining steps for lymphocyte subsets were as follows: Peripheral blood lymphocytes were analyzed using six-color flow cytometry (FACSCanto II, BD Biosciences, San Jose, CA, USA). Fifty microliters of EDTA-K2 anticoagulated venous blood were added to the bottom of the absolute counting tube using a reverse pipetting method. Then, 20 uL of fluorescein-labeled monoclonal antibodies (6C-TBNK Lymphocyte Test Kit, BD Biosciences, San Jose, CA, USA) were added to the bottom of the absolute counting tube. The mixture was gently vortexed and incubated in the dark for 15 minutes for staining. Next, 450 uL of lysing solution was added to the tube, gently vortexed, and allowed to react for 15 minutes before the samples were analyzed using flow cytometry. All instruments and reagents were purchased from BD Biosciences (USA). In accordance with the flow cytometry platform of our hospital, the normal reference ranges were defined as follows: CD3⁺ T cells (48.0–82.0%), CD4⁺ CD8⁺ T cells (13.0–42.0%), CD3⁺ CD4⁺ T cells (21.0–46.0%), CD4/CD8 (0.71-2.78), CD19⁺ B cells (4.0–18.0%), CD16⁺CD56⁺ NK cells (7.0–41.0%), CD3⁺ T cells absolute count (668-2474/uL), CD3⁺CD8⁺ T cells absolute count (161-1159/uL), CD3⁺CD4⁺ T cells absolute count (292-1309/uL), CD19⁺ B cells absolute count (79-562/uL), CD16⁺CD56⁺ NK cells absolute count (142-1270/uL).

All statistical data analyses in this study were conducted using R software Version 4.2.1 (https://www.R-project.org). Descriptive statistics were employed to assess the baseline characteristics of patients. Results were presented as the median of the interquartile range for the percentage or absolute count of positive lymphocytes. Differences in the percentages or absolute counts of immune cells pre- and post-CRT were detected using the Wilcoxon test or the Kruskal-Wallis test for related samples. The alteration in the proportion of circulating lymphocytes pre- and post-CRT was defined as Δ lymphocyte = post-CRT lymphocyte proportion/baseline lymphocyte proportion. The relationship between Δ lymphocyte and clinical characteristics was estimated using the Wilcoxon rank sum test. A P-value of less than 0.05 was considered statistically significant.

A total of 91 patients met the inclusion criteria and were comprehensively evaluated. All patients completed CRT. Among the 91 non-metastatic NPC patients, 73 (80.22%) were male, and 18 (19.78%) had a history of smoking. The median age of the patients was 49 years (range 42-54 years). According to the 8th edition of the AJCC staging system, there were 70 patients with stage IVA, 19 with stage III, and 2 with stage II. The primary induction chemotherapy regimen consisted of docetaxel combined with platinum, which was used in 85 patients (93.41%), followed by paclitaxel combined with platinum in 4 patients (4.39%). There were three treatment modalities: 85 patients (93.41%) received induction chemotherapy followed by concurrent CRT, and 5 patients (5.49%) received induction chemotherapy followed by radical radiotherapy. The main concurrent chemotherapy drug used during radiotherapy was cisplatin, which was administered to 68 patients (74.73%). Nimotuzumab was utilized in 76 patients (83.52%) during radiotherapy. All patients underwent intensity-modulated radiotherapy (IMRT). The doses were as follows: primary tumor (PGTV) received 70-74 Gy/30-33 fractions, metastatic lymph nodes (PGTVnd) received 70 Gy/30-33 fractions, subclinical lesions and high-risk areas (PCTV1) received 60 Gy/30-33 fractions, and low-risk areas (PCTV2) received 50-52 Gy/28 fractions. At the end of CRT, 26 patients (28.57%) achieved CR. The baseline characteristics of patients are shown in Table 1 .

The expression of CD3, CD4, CD8, CD56, and CD19 on peripheral blood cells was detected by flow cytometry and the composition of lymphocyte subsets before treatment, post-completion of CRT, and one month following CRT was retrospectively analyzed ( Supplementary Figure S1 ). Figure 1 and Table 2 depicted the variations in circulating lymphocyte subsets before treatment, post-completion of CRT, and one month following CRT. In comparison to the baseline, the proportion of CD3⁺ T cells exhibited a significant increase at the end of CRT (p<0.01), followed by a marked decrease one month later (p<0.01) ( Figure 1A ). The proportion of CD3⁺CD4⁺ T cells, the CD4⁺/CD8⁺ ratio, and the proportion of CD19⁺ B cells continued to decline at baseline, at the end of CRT, and one month after CRT (p<0.001) ( Figures 1B, D, E ); whereas, the proportions of CD3⁺CD8⁺ T cells and CD16⁺CD56⁺ NK cells continued to rise (p<0.001) ( Figures 1C, F ). The absolute counts of circulating lymphocyte subsets demonstrated a trend of initial decrease followed by an increase at baseline, at the end of CRT, and one month after CRT. Compared to baseline, the absolute counts of CD3⁺ T cells, CD3⁺CD4⁺ T cells, CD3⁺CD8⁺ T cells, CD45⁺, CD19⁺ B cells, and CD16⁺CD56⁺ NK cells sharply decreased at the end of CRT (p<0.001) but rapidly increased one month after CRT (p<0.001) ( Figures 1G-L ).

Furthermore, the median percentages of CD3⁺ T cells, CD3⁺CD8⁺ T cells, and CD16⁺CD56⁺ NK cells were elevated or declined within the normal ranges before and after CRT ( Table 2 ). At baseline, the proportion of CD19⁺ B cells in 89/91 (97.80%) patients was within normal range, but at the end of CRT and one month after CRT, the percentages of CD19⁺ B cells of most patients (50/91, 54.95%; 67/91, 73.63%) were decreased, with the median further declining. The median percentage of CD3⁺CD4⁺ T cells was normal at both baseline and the end of CRT, but, one month after CRT, the proportion of CD3⁺CD4⁺ T cells in 52/91 (57.14%) patients was lower than the normal range. Similarly, we observed that the CD4⁺/CD8⁺ ratio in 60/91 (65.93%) patients was lower than normal one month after CRT. At baseline, the absolute counts of CD3⁺ T cells in 82/91 (90.11%) patients were normal; whereas, the absolute counts of CD3⁺ T cells of most patients (91/91, 100%; 65/91, 71.43%) were decreased at the end of CRT and one month after CRT. The median absolute count of CD3⁺CD8⁺ T cells was within the reference range at baseline and one month after CRT, but at the end of CRT, the CD3⁺CD8⁺ T cells absolute count was declined in 72/91 patients (79.12%). The median absolute count of CD3⁺CD4⁺ T cells was within the normal range at baseline, but at the end of CRT and one month after CRT, the CD3⁺CD4⁺ T cells absolute counts were decreased in most patients (91/91, 100%; 84/91, 92.31%). Similarly, at the end of CRT and one month after CRT, the CD19⁺ B cells absolute counts were decreased in 91/91 (100%) and 86/91 patients (94.51%), respectively. At the end of CRT, the CD16⁺CD56⁺ NK cells absolute counts were declined in 80/91 (87.91%) patients, whereas the median count was normal at baseline and one month after CRT. These results indicate that CRT triggers complex changes in the composition of circulating lymphocytes in the peripheral blood of non-metastatic NPC.

Notably, CD8⁺ T cells primarily differentiate into cytotoxic T cells, which directly eradicate tumor cells. Thus, we further focused on the alterations in CD8⁺ T cells after CRT. The proportions of CD8⁺ T cells in 34/91 (37.36%) patients after CRT were higher than 42% (normal reference range, 13.0-42.0%), but at baseline, the proportions of CD8⁺ T cells in only 5/91 (5.49%) patients was over 42% ( Figure 2A ). Also, we found that in these 34 patients, the proportions of CD3⁺ T cells were all less than 48% (normal reference range, 48.0-82.0%) after CRT, and the CD4⁺/CD8⁺ ratio was lower than 0.71 (normal reference range, 0.71-2.78) ( Figures 2B-C ). It is well-known that a low CD4⁺/CD8⁺ ratio can result from a diminished proportion of CD4⁺ T cells and/or an augmented proportion of CD8⁺ T cells. After CRT, among the 45 patients with elevated proportion of CD8⁺ T cells, the proportions of CD4⁺ T cells in 26 patients were less than 21%, and the proportions of CD4⁺ T cells in 8 patients were over 46% (normal reference range, 21.0-46.0%) ( Figures 2D-F ). These results reveal that the overall immune function in non-metastatic NPC remains normal after CRT, and the increased percentages of CD8⁺ T cells might be attributed to the immune system response in generating more cytotoxic T cells following CRT.

The clinical characteristics and treatment modalities of patients may influence the changes in the composition of circulating lymphocyte subsets before and after CRT. As shown in Table 3 , the results suggested that changes in circulating CD3⁺ T cells and CD16⁺CD56⁺ NK cells were associated with the N stage. Patients with stage N2-3 existed a significantly lower proportion of CD3⁺ T cells after CRT compared to patients with stage N0-1 (N0-1 vs N2-3, p = 0.03). The proportion of CD16⁺CD56⁺ NK cells was closely higher in patients with stage N2-3 after CRT than in those with stage N0-1 (N0-1 vs N2-3, p = 0.02). Patients with stage IVA displayed a dramatically higher proportion of CD3⁺CD8⁺ T cells after CRT compared with patients with stage II-III (II-III vs IVA, p = 0.02). Furthermore, patients who received DP induction chemotherapy had a markedly lower proportion of CD16⁺CD56⁺ NK cells after CRT than in patients who received other induction chemotherapy regimens (DP vs. others, p = 0.02). However, there was no significant correlation between changes in the composition of circulating lymphocyte subsets and factors such as gender, T stage, use of nimotuzumab, concurrent chemotherapy regimen, and treatment modality. Figure 3 further implicated the association between changes in CD3⁺ T cells, CD16⁺CD56⁺ NK cells, and CD3⁺CD8⁺ T cells with TNM stage and induction chemotherapy regimen.

The impact of dynamically measured immune characteristics on the clinical response was further evaluated in patients with non-metastatic NPC. As shown in Figure 4 , the dynamic trends of circulating lymphocyte subsets at baseline, at the end of CRT, and one month after CRT were largely consistent between the CR and PR groups, but there were significant differences at specific time points. Figures 5 , 6 separately illustrated the dynamic changes in the percentages and absolute counts of circulating lymphocyte subsets in the CR and PR groups at baseline, at the end of CRT, and one month after CRT. At the end of CRT, the proportion of CD3⁺CD4⁺ T cells was distinctly higher in the CR group than that in the PR group (p<0.05, Figure 4B ). The baseline proportion of CD19⁺ B cells was also significantly higher in the CR group (p<0.05, Figure 4E ). Conversely, at the end of CRT, the percentages of CD16⁺CD56⁺ NK cells were markedly lower in the CR group compared to the PR group (p<0.05, Figure 4F ). Furthermore, the baseline absolute counts of CD3⁺CD4⁺ T cells, CD45⁺, and CD19⁺ B cells were tightly higher in the CR group than that in the PR group (p<0.05, Figure 4H ; p<0.01, Figure 4J ; p<0.001, Figure 4K ). At the end of CRT, the absolute count of CD3⁺CD4⁺ T cells remained remarkedly higher in the CR group than that in the PR group (p<0.01, Figure 4H ). Notably, there were no statistical differences in the percentages and absolute counts of lymphocyte subsets between the CR and PR groups one month after CRT. These results ascertained that the dynamic changes of circulating lymphocyte subsets between the CR and PR groups were profoundly impressive, particularly at the end of CRT.