The Predictive Value of Changes in the Absolute Counts of Peripheral Lymphocyte Subsets for Progression and Prognosis in Breast Cancer Patients

Background As the number and proportion of lymphocyte subsets are an important indicator of the immune function, an in depth understanding of the immune function of patients with malignant tumor has important clinical values for the treatment, prognosis, and evaluation of the disease. This retrospective study was to evaluate the clinical value of the absolute counts of lymphocyte subsets as potential blood biomarkers for progression and prognosis in breast cancer patients. Methods A total of 237 BC patients and 55 age-matched female normal healthy donors were included in this study. Flow cytometry was used to determine the absolute counts and the percentages of CD3+, CD4+, CD8+, B, and NK cells. The receiver operating characteristic curve (ROC) was used to evaluate the accuracy of absolute count of lymphocyte subsets in the curative efficacy assessment. The clinicopathological parameters influencing the disease progression were determined by Cox proportional hazards regression. Progression-free survival (PFS) was estimated using the Kaplan–Meier method with the log-rank test. Results: Compared with the healthy donors, the absolute counts of lymphocyte subsets in patients decreased significantly. ROC analysis showed that the area under the curve of the CD4+ absolute count was 90% (95% confidence interval 0.859–0.940), and the sensitivity and specificity were 80.9% and 85.3%, respectively. The analysis of Cox regression showed that the cutoff value of the CD4+ absolute count ≥451 cells/μL might be a favorable prognostic factor. Multivariate analysis of prognostic factors of PFS showed that the CD4+ and CD8+ absolute count were independent factors for predicting PFS. Conclusions The remarkably impaired absolute counts of the CD3+, CD4+, CD8+, B, and NK cells in patients with breast cancer can be used as potential susceptible biomarkers to evaluate the patient's immune status. The higher level of CD4+ and CD8+ absolute counts probably contributed to the longer PFS and favorable outcome of BC patients.


Introduction
Breast cancer (BC) has become one of the main causes of cancer death in women worldwide. According to the latest statistics in 2020, with an estimated 2.3 million new BC cases and 685,000 deaths caused by BC [1], the number of BC patients in 185 countries and regions has overtaken that of lung cancer patients, and BC becomes the most commonly diagnosed cancer. In recent years, the incidence rate of BC has been increasing, and the age of its victims has been getting younger [2]. At present, the main treatments of BC are surgery, chemotherapy, endocrine, and other comprehensive therapies. For patients with early and mid-term BC, postoperative adjuvant chemotherapy can improve the local control rate and reduce the recurrence rate of local and regional lymph nodes [3].
It has been found that the recurrence and metastasis of tumors are closely related to the decline of the immune function, which in turn leads to the tumor immune escape [4]. e classical lymphocyte subsets include CD3 + , CD4 + , CD8 + , B, and natural killer (NK) cells, which are essential for regulating the immunity and specific killing of tumor cells [5]. Although their compositions are simple, their roles in immune responses are diversified, including both innate and adaptive immunity, and cellular and humoral immunity as well.
e CD4 + can activate other lymphocyte subsets by releasing cytokines and suppress tumor development by directly killing tumor cells that express adequate levels of major histocompatibility complex (MHC) class II molecules [6]. e CD8 + cells, a kind of cytotoxic cells, can recognize tumor cells that express MHC class I molecules presented by antigen-presenting cells and produce IFN-c, perforin, and granzyme B for targeting and killing tumor cells [7]. B cells play an important role in the humoral immunity, which not only can present antigens to T cells and participate in the immune response of T cells but also can recognize different antigen epitopes, secrete immunoglobulin, and participate in the humoral immune response [8]. NK cells, as indispensable effector cells in the innate immune system, work as the first line of defense for the host immune defense against cancer and pathogens [9]. erefore, analyzing peripheral lymphocyte subsets is a significant and convenient way to assess the immune function, which includes the percentages of lymphocytes (PL) and the absolute count of lymphocytes (ACL). In particular, the ratio of the CD4 + to CD8 + may indicate the balance of the immune system, which is maintained by the proportion and the number of peripheral blood lymphocytes [10]. Growing evidence has revealed the absolute count of peripheral blood lymphocyte subsets is positively correlated with the prognosis of cancer patients [11][12][13].
Traditionally, the ACL was measured by the dual-platform technology, which combined the percentage of lymphocyte subsets assessed by flow cytometry with the absolute lymphocyte count assessed by an automatic hematology analyzer. However, this method produced obvious discrepancies among different laboratories [14][15][16][17]. Herein, we adopted a single-platform technology to carry out the detection entirely on flow cytometer. is method significantly improved the accuracy of the analysis. Previous studies showed that lymphopenia, or low peripheral blood lymphocyte count, can predict higher mortality and increased risk of recurrence after primary surgery and neoadjuvant therapy, whereas higher absolute lymphocyte counts can predict lower mortality from early-stage triplenegative BC [18,19]. Nevertheless, these studies only analyzed the total number of lymphocytes but did not further analyze the influence of the ACL on the immune function. As a result, the correlations between prognosis and the ACL in BC patients remain unclear.
is study aimed at evaluating the predictive value of PL and ACL for BC progression and prognosis, and analyzing the importance and susceptibility of ACL used as biomarkers of immune impairment. e endocrine therapy for premenopausal BC patients with positive estrogen/progesterone receptor (ER/PR) was dominated by tamoxifen, whereas the endocrine therapy for the postmenopausal patients with positive ER/PR was dominated by letrozole.

Efficacy Evaluation.
Based on RECIST (version 1.1) [20], the efficacy evaluation was divided into complete response (CR), partial response (PR), stable disease (SD), and progress disease (PD). CR + PR was considered to be the response (R) group, whereas SD + PD was considered the nonresponse (NR) group.

Sample
Collection. Two milliliters of fresh whole blood drawn from the NCs and BC patients were stored in the EDTA-anticoagulant blood collecting tubes.

Cellular Staining and Analyzing.
Whole blood of the 292 participants was collected and assayed by flow cytometry following the BD operating instruction. In brief, for each sample, two trucount tubes were labeled with letters A and B to distinguish them from each other. 20 μL of BD Multitest CD3 + /CD4 + /CD8 + /CD45 + and CD3 -/CD16 + CD56 + / CD45 + /CD19 + reagents was added to the bottom of tubes A and B, respectively. en, 50 μL of well-mixed whole blood was added to the bottom of every tube. Finally, 450 μL of lysing solution was added into every tube, mixed, and incubated for 15 min in dark at room temperature for analysis.

Statistical Analysis.
e normality of statistical data was evaluated by the Shapiro-Wilk test, using two independent sample t-tests to analyze the differences in PL and ACL between BC patients and NCs. e differences among three or more groups of continuous numerical variables with normal distribution were analyzed by one-way ANOVA (the Bonferroni test or the Tamhane test). Progression-free survival (PFS) was defined as the time from the date of enrollment to disease progression, recurrence, or death. e follow-up deadline was 31 January 2021. e receiver operating characteristic curve (ROC) was used to evaluate the accuracy of ACL in the curative efficacy assessment. Categorical variables were analyzed using the chi-squared test. Survival curves were constructed using the Kaplan-Meier method. Univariate and multivariate Cox proportional hazards regression analyses were performed to assess the relationship between clinicopathological parameters and  PFS. Variables with a P value < 0.05 by a univariable analysis were entered for multivariable analysis. Hazard ratio (HR) was reported with 95% confidence interval (CI). All tests were 2-tailed and P < 0.05 was considered statistically significant. e data were analyzed by SPSS 25.0 (IBM Corporation, USA). Figures were prepared using GraphPad Prism software (version 9.0) (San Diego, USA).

Comparison of ACL between BC Patients and NCs.
To investigate the changes in the ACL of BC patients, we compared the ACL of BC patients and NCs. e results showed that there was no difference in the percentages of ACL between the two groups (P > 0.05) (Figure 1(a) (Figure 1(b)). ese results suggested that it was ACL rather than PL that decreased in BC patients. To our knowledge, PL represents the proportion or composition of each subset, revealing the development and differentiation of lymphocytes, whereas the ACL demonstrates the exact number of peripheral lymphocyte subsets, revealing the proliferation of lymphocyte progenitor cells. ese results indicated that the proliferating ability of lymphocytes was impaired noticeably in BC patients.
In addition, the PL in BC patients at stages I-II and III-IV, compared with that in NCs, showed no significant difference (P > 0.05) (Figures 1(c) and 1(e)), whereas significant differences between the two groups were observed in their ACL (P < 0.001) (Figures 1(d) and 1(f )). ese results suggested that a significant decrease in the ACL was characteristic of immune impairment or a sign of immunodepletion in BC patients.

ACL in BC Patients at Different Clinical Stages.
e relationship between ACL and severity of BC patients was further investigated. With the progression of the disease, the AC of CD3 + , CD4 + , CD8 + , B, and NK cells showed varying degrees of decline (Figure 2). e CD3 + AC in patients at stage IV was lower than that in patients at stages I and II (P < 0.001), whereas BC patients at stage III were not significantly different from that in BC patients at stages I and II (Figure 2(a)). e decline of CD4 + AC was particularly significant in BC patients. Compared with that in BC patients at stages I, II, and III, the CD4 + AC in BC patients at stage IV decreased most strikingly (P < 0.001), which was followed by stage III (P < 0.01), stage II (P < 0.01), and stage I (P < 0.001) (Figure 2(b)). Compared with BC patients at stage I, the CD8 + AC decreased in BC patients at stage II (P < 0.05) and stage III-IV (P < 0.01) (Figure 2(c)). Compared with BC patients at stage I, the B AC decreased in BC patients at stage IV (P < 0.01) (Figure 2(d)). In addition, the difference in AC of NK cells was only found in BC patients at stages II and IV (P < 0.001) (Figure 2(e)). ese results suggested that the decrease of CD4 + and CD8 + AC was closely related to different clinical stages; that is, CD4 + and CD8 + AC declined with the exacerbation of BC.

Differences in ACL between the Response and Nonresponse
Groups. To further elucidate the relationship between ACL and the curative effect, we divided the BC patients into two groups according to curative effect, namely, response (R) group and nonresponse (NR) group. e results revealed that AC of CD3 + , CD4 + , CD8 + , B, and NK cells in the R group was significantly higher than that in the NR group ( Figure 3). In brief, the results indicate that the higher the ACL, the better the efficacy.

e Accuracy of ACL in the Efficacy Evaluation.
To further clarify the accuracy of ACL in determining efficacy, ROC curve analysis was conducted. e ROC curve showed that CD4 + AC had the highest prediction accuracy. When the Youden index reached a maximum of 0.662, the area under the curve of CD4 + AC was 0.90 (95% CI 0.859-0.940), the optimal cutoff value of CD4 + AC was 468.5 cells/μL, and the sensitivity and specificity were 80.9% and 85.3%, respectively. e results suggested that CD4 + AC can be used as an index to predict the curative effect ( Figure 4).

Correlation between CD4 + AC and the Clinicopathological Parameters of BC.
To further analyze the correlation between CD4 + AC and clinicopathological parameters, we used the median of CD4 + AC (451 cells/μL) as the cutoff value and divided 237 BC patients into low level (<451 cells/μL) (118 cases) and high level (<451 cells/μL) (119 cases. e results showed that family history, previous medical history, pathological category, clinical stages, lymphatic metastasis, distant metastases, vessel carcinoma embolus, and treatments were significantly correlated with the CD4 + level (P < 0.05), whereas age, age of menarche, menopause, tumor size, and differentiation degree were not correlated with the CD4 + level (P > 0.05) ( Table 2).

Discussion
e study showed that it was convenient and accurate to apply the single-platform flow cytometry to analyze the PL and ACL in BC patients. Currently, the PL is most commonly used to evaluate the immune function of tumor patients in clinical practice, whereas the detection of ACL is relatively insufficient, which may lead to a neglect of serious impairment of the immune function of cancer patients. e percentage of lymphocyte subsets represents the composition and proportion of lymphocytes, and reflects the Contrast Media & Molecular Imaging development and differentiation ability of lymphocytes, whereas the absolute counts of lymphocyte subsets represent the number of lymphocytes and reflects the proliferation ability of lymphocytes [21]. At present, there are some controversies in evaluating the immune status and prognosis of tumor patients by the percentage of lymphocyte subsets. Studies have found [22] that the percentage of CD3 + , CD4 + , CD4 + /CD8 + , and NK cells in peripheral blood of NSCLC patients with stage IV was negatively correlated with the number of circulating tumor cells. A study [23] showed that NSCLC patients had higher baseline percentages of CD3 + , CD4 + , and CD8 + cell, while lower percentages of NK cell had longer OS. In contrast, another study [24] found that neither the percentage of CD8 + , CD19 + , and CD56 + lymphocyte subsets nor the CD4 + /CD8 + ratio in peripheral blood was significant predictors or prognostic factors in patients with metastatic breast cancer. e percentage of lymphocyte subsets has some uncertainties in evaluating the immune status and prognosis of patients. erefore, in recent years, more and more researchers have used the absolute count of lymphocyte subsets as an important immune indicator for the prognosis and efficacy of tumor patients. e immune system, as a line of defense of the human body against the invasion of pathogens and malignant tumors, consists of cellular immunity and humoral immunity. Studies have shown that cellular immunity plays a vital role in the process of antitumor immunity [25,26]. e CD3 + T, as a major in the process of cellular immunity, mainly consists of CD4 + T and CD8 + T. e CD4 + T, playing an immune-modulating role, can assist the B cells to produce antibodies and in the meantime secrete cytokines to enhance the CD8 + T to kill tumors [27]. CD4 + helps to initiate a gene expression program of CD8 + via multiple molecular mechanisms to enhance the function of CD8 + CTLs and overcome the obstacle of antitumor immunity [28]. us, the CD4 + T can target tumor cells in various ways, either Contrast Media & Molecular Imaging directly by eliminating tumor cells through the cytolytic mechanism or indirectly by modulating the tumor microenvironment [29]. e increase in the number of CD4 + T indicates that the immune response is improved and the antitumor activity is reinforced [30]. e CD8 + T can be classified into cytotoxic T lymphocytes (CTLs) and inhibitory cells. CTLs, as a preferred tool to target tumors, can detect extracellular antigens presented by MHC class I molecules expressed by all tumor cell types [31]. B cells, whose surface marker is CD19 + , are mainly involved in humoral immunity, and thus, its expression level can be used to evaluate the strength of humoral immunity [32]. NK cells, expressing a specific marker of CD16 + or CD56 + , are a group of cells with special properties and do not have antigenrecognition receptors on the surface [33]. NK cells are the immune cells first activated in the antitumor immunity and important cytotoxic cells in the innate immunity, which can be activated and then perform their effector functions without antigenic stimulation [34]. e decreased number of NK cells indicates that the immune system becomes less capable of monitoring, killing, and clearing tumor cells [35,36]. erefore, the detection of the ACL in tumor patients is of great significance for understanding the changes of patients' condition and prognosis [37,38].
Our study suggested that ACL can be served as a potential biomarker of peripheral blood immune impairment in BC patients to monitor the immune function and to predict the prognosis and therapeutic effects. It is the first of its kind to report the clinical value of peripheral ACL in BC patients for prognosis, despite that some studies claimed that the total number of peripheral blood lymphocytes was associated with the prognosis and therapeutic effects [14,15,19,39,40]. Our study has made the following findings. Firstly, there was no significant difference in the percentages of CD3 + , CD4 + , CD8 + , B, and NK cells between NCs and BC patients. However, the AC of CD3 + , CD4 + , CD8 + , B, and NK cells decreased markedly in BC patients compared to that in NCs. is finding sheds some light on the necessity and importance of detecting ACL in BC patients in addition to the percentage. is may help us to understand the immunologic injury of BC patients, analyze the clinical condition, and predict the curative effect [11]. Secondly, CD4 + had a close negative correlation to clinical stages and thus decreased with the development and deterioration of BC. Compared with that in BC patients at stages I and II, the CD4 + AC in BC patients at stages III and IV decreased more significantly. Compared with BC patients at stage I, the CD8 + AC decreased in BC patients at stages II, III, and IV, too. Interestingly, ACL were positively correlated with curative effect. In patients in the response group, AC of CD3 + , CD4 + , CD8 + , B, and NK cells were significantly higher than that in patients in the nonresponse group. Finally, our study reveals that BC patients with CD4 + AC ≥451 cells/μL or CD8 + AC ≥324 cells/L  Figure 5: e forest plots of factors affecting the progression-free survival. OR > 1 indicates that the variable is considered a risk factor. OR < 1 represents that the variable is considered a protective factor. e bolded value represents that the P value was significant. demonstrated a longer PFS than BC patients with CD4 + AC <451 cells/μL or CD8 + AC <324 cells/μL, which is consistent with our previous study [41]. is indicates that more attention should be paid to CD4 + and CD8 + AC in the antitumor immunity. e increased number of CD4 + and CD8 + can signify a good prognosis in BC patients, which is consistent with previous studies [42,43]. CD4 + T can promote the activation of CD8 + CTLs, boost the effector and memory functions of CTLs, and reduce the immunosuppression of CTLs, which helps T cells amplify their response to tumor-associated antigens without generating an autoimmune response [44]. Additionally, the specific binding of antigen to CD4 + can enable the dendritic cells to optimize antigen presentation and deliver cytokines and costimulation signals specific to CD8 + , eventually promoting their cloning, amplification, and differentiation into effector or memory T cells [45]. In brief, the low ACL in BC patients indicates that the immune function of patients is impaired. e patient's impaired immune system could not stop the tumor from progressing [46]. us, it is essential to enhance the immune function of BC patients and improve their antitumor ability during the treatment.
Consequently, as the ACL not only reflects the immune status of the body but also can be used to predict disease prognosis and the curative effect, more attention should be paid to the changes of the ACL in clinical practice to provide more referential values to predict the patient's condition and prognosis in clinical treatment.
Our study has some limitations. Firstly, this study is a retrospective study with a relatively small sample size, so some important prognostic factors were not included in the analysis, which may cause some information bias. Secondly, all data included in the study were from a single center.
Multicenter studies with large sample sizes are needed to generalize our findings.

Conclusions
is study shows that it was ACL, rather than PL, that was severely impaired in BC patients, which was correlated to the progression of the tumor, PFS, and efficacy in clinical treatment. us, it is necessary and pivotal to detect ACL in clinical practice. In a nutshell, our study suggests that ACL can be used as a potential peripheral blood biomarker for monitor immune impairment and predicting the prognosis and therapeutic effect of BC patients.

Data Availability
e datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Ethical Approval
All enrolled participants signed the informed consent. e present study involving human samples was approved by the Ethics Committee of the First Teaching Hospital, Tianjin University of Traditional Chinese Medicine (Tianjin, China) (TYLL2017 [K] 002, 25 December 2017), and was conducted in accordance with the Declaration of Helsinki. e clinical trial was registered at Chinese Clinic Trial Registry (ChiCTR-IOR-17014139).

Conflicts of Interest
e authors declare that they have no conflict of interest.

Authors' Contributions
In this work, Jianchun Yu conceived and designed the study. Yingjie Jia, Yongtie Guo, Huayu Chen, and Xu Liu were responsible for data acquisition and interpretation. Guan Zhang and Yunhe Liu performed the statistical analysis. Aqing Liu and Ying Xia drafted the manuscript. Wentao Li, Songshan Ye, Zhijie-ruo Zhao, and Yanjie Yang critically revised the manuscript for important intellectual content. All authors read and approved the final manuscript.