Circulating the HLA-DR+ T Cell Ratio Is a Prognostic Factor for Recurrence of Patients with Hepatocellular Carcinoma after Curative Surgery

Background HLA-DR+ T cell, accounting for 1.2%–5.8% of peripheral lymphocyte, is a type of activated T lymphocyte. This retrospective study aimed to evaluate the prognostic value of HLA-DR+ T cell for progression-free survival (PFS) and overall survival (OS) in hepatocellular carcinoma (HCC) patients after curative surgery. Patients and Methods. Clinicopathological data of 192 patients who underwent curative resection for hepatocellular carcinoma in the affiliated hospital of Qingdao University between January 2013 and December 2021 were collected and analyzed. Statistical tests used in this study were the chi-square test and Fisher's exact test. The prognostic value of the HLA-DR+ T cell ratio was analyzed using univariate and multivariate Cox regression analyses. The Kaplan–Meier curves were drawn by the R programming language. Results HCC patients were divided into high (≥5.8%) and low (<5.8%) HLADR+ T cell ratio groups. Cox regression analysis indicated that a high HLA-DR+ T cell ratio was positively related to the PFS in HCC patients (P=0.003) and AFP-positive (≥20 ng/ml) HCC patients (P=0.020). HCC patients and AFP-positive HCC patients in the high HLA-DR+ T cell ratio group were prone to have a higher T cell ratio, a higher CD8+T cell ratio, and a lower B cell ratio than the low HLA-DR+ T cell ratio group. However, the HLA-DR+ T cell ratio was not a statistically significant predictor for OS in HCC patients (P=0.57) as well as PFS (P=0.088) and OS (P=0.63) in AFP-negative HCC patients. Conclusions This study confirmed that the HLA-DR+ T cell ratio was a significant predictor of PFS in HCC patients and AFP-positive HCC patients after curative surgery. This association may have guiding significance for the follow-up work of HCC patients after surgery.


Introduction
Comprising 75%-85% of primary liver cancer, hepatocellular carcinoma (HCC) is one of the most fatal cancers which presents a severe challenge to human health and life quality [1]. Despite recent advances in diagnosis and treatment, the resection rate and overall prognosis of HCC patients are still unsatisfactory. Even if the early-stage HCC patients were treated with curative surgery [2], the incidence of tumor relapse ranges from 50% to 70% [3]. Terefore, it is urgent to identify biomarkers to screen out patients with a high risk of relapse. Te tumor immune ecosystem is a key determinant and research hotspot of the progression of HCC [4]. In this study, we analyzed the relationship between HLA-DR+ T cells, one of the peripheral blood lymphocyte subsets, and the prognosis of HCC.
Te human leukocyte antigen isotype DR (HLA-DR) as an MHC class II molecule is expressed on antigenpresenting cells (APC) and activated T cells [5,6]. Numerous detailed studies have elucidated the importance of presenting antigens to T cells by professional APC expressing HLA-DR in the immune process [7,8]. HLA-DR, as a marker of T cell activation, was shown to associate with disease states, such as cancer, HIV, and autoimmune diseases [9][10][11]. However, immunosuppressive cells expressing HLA-DR, including TAM tumor-associated macrophage (TAM) and regulatory T (Treg) cells, are involved in immune escape [12,13]. Additionally, previous studies revealed HCC cells are one major source of HLA-DR in tumor tissues [14]. But in HCC, few studies reported the role of HLA-DR+ T cells in prognosis.
T cells, as the most important part of antitumor immunity, have been the focus of the research. Over these years, more and more studies revealed the signifcance of HLA-DR+ Tcells in the development and treatment of HCC. According to a recent study, mucosal-associated invariant T (MAIT) cells, as innate T cells with immunoregulatory activity, upregulated HLA-DR in peripheral blood of HCC patients but not in hepatic sinusoids and were functionally impaired [15]. Te activated but functionally impaired MAIT cells might suggest a potential role in HCC pathogenesis. In addition, the CD57+ HLA-DR+ CD8+ T cells, displaying a highly proliferative and cytotoxic phenotype, were correlated with recurrence in HCC patients after liver transplantation [16]. HLA-DR+ T cells in peripheral are increased in HCC patients with anti-CTLA4 treatment [17].
In conclusion, the role of HLA-DR+ T cells in HCC prognosis deserves further investigation.
We aimed to analyze the relationship between the preoperative HLA-DR+ T cell ratio and postoperative recurrence or death in HCC patients. We also analyzed the diferences in baseline data between the groups with different HLA-DR+ T cell ratios to investigate the potential value of HLA-DR+ T cells in HCC progression and prognosis.

Patient Selection.
Patients who received liver resection and were pathologically diagnosed as having HCC at the Department of Hepatobiliary and Pancreatic Surgery, Te Afliated Hospital of Qingdao University, between January 2013 and December 2021 were retrospectively collected. We eliminated patients with other malignancies, preoperative anti-HCC treatment, macrovascular invasion, TNM stage III, TNM stage VI, and postoperative liver transplantation for non-neoplastic liver disease, as well as patients lacking complete baseline and follow-up data, resulting in 192 patients included in the study. We acquired clinicopathological data from medical records. Tis study complied with the ethical guidelines of the Helsinki Declaration, and written informed consent was obtained from participants or their immediate families prior to surgery.

Clinicopathological
Variables. Te clinicopathological variables of HCC patients incorporated the date of surgery, gender, age, HbsAg, Child-Pugh grade, AFP (alpha-fetoprotein) levels, ALT (alanine transaminase) levels, AST (aspartate transaminase) levels, AST/ALT, GGT (gamma glutamyl transferase) levels, lymphocyte subsets, largest tumor size, tumor number, liver cirrhosis, macro-and microvascular invasion, capsular invasion, tumor satellites, and tumor diferentiation. Each subset of lymphocytes was described in terms of its ratio to total lymphocytes. For continuous variables, the upper level of normal (ULN) was used as the cut-of value. AFP positivity was defned as a serum level ≥20 ng/ml. Liver cirrhosis was diagnosed based on pathological fndings. Te curative surgery referred to complete resection of tumor with negative microscopic margins. Te 8 th AJCC cancer staging system was used to stage HCC. Partial HCC patients with microvascular invasion received prophylactic TACE (transcatheter arterial chemoembolization) within 1 month after surgery.

Postoperative Follow-Up.
All patients were followed up regularly at the outpatient clinic after discharge. AFP and image studies were performed routinely at 1 month after the operation, then at least 3 months in the frst 2 years, and at least 6 months thereafter. Dynamic contrast-enhanced computed tomography (CT) of the upper abdomen or Gd-EOB-DPTA-enhanced magnetic resonance imaging (MRI) was performed if digestive ultrasound revealed HCC recurrence and/or alpha fetoprotein (AFP) was signifcantly elevated. Further auxiliary inspection including chest CT, lower abdominal CT, pelvic CT, or positron emission tomography (PET)-CT was completed when clinically necessary. Te diagnosis of tumor recurrence was confrmed by at least two imaging examinations or liver histopathological examination. Once tumor recurrence is found, resurgical resection, local ablation, TACE, radiotherapy, systemic therapy, and supportive care can be selected according to the characteristics of tumor recurrence. Te survival information was obtained from clinical follow-up or telephone follow-up. Follow-up was up to September 1, 2021. Te starting point of follow-up was the date of surgery; the time interval from the starting point to the date of the frst recurrence, frst metastasis, death, or last follow-up was PFS; and the time interval from the starting point to the date of death or last follow-up was OS.

Statistical Analysis.
Continuous variables were categorized by ULN and described together with categorical variables as frequencies (N). Te χ 2 test or Fisher's test was performed to compare categorical variables. Using the COX proportional hazard model, we identifed independent risk factors associated with postoperative progression or death, with hazard ratios (HR) and 95% confdence intervals (CI). We plotted the Kaplan-Meier curves through the "survival" package of the R language software package. SPSS software version 22.0 (SPSS, Chicago, IL, USA) and R language software version 4.1.2 were used for statistical analysis in this study. Results were considered statistically signifcant when the P value was less than 0.05.

Patient Baseline Information.
Te baseline clinicopathological characteristics of all HCC patients were outlined in Table 1. A total of 115 (59.9%) patients were less than 60 years old, and 153 (79.7%) patients were male. Tere were 164 (85.4%) patients who had the history of viral hepatitis B, and 114 (59.4%) patients had liver cirrhosis. Tere were 187 (97.4%) patients with Child-Pugh grade A, and 5 well selected HCC patients with Child-Pugh grade B who underwent curative surgery were also included in the study according to a comprehensive review [18]. Te proportion of patients with AFP positivity (≥20 ng/ml) and a high HLA-DR+ T cell ratio (≥5.8%) were 50.0% and 52.1%, respectively. Except for 2 patients with two tumors, the rest of the patients had one tumor. 190 patients (99.0%) were TNM stage I, and 2 patients were TNM stage II. 116 patients (60.4%) were Edmondson grades I-II. Capsular invasion was noted in 70 (36.5%) patients, and microvascular invasion was noted in 67 (34.9%) patients. After curative surgery, 65 (33.9%) patients received prophylactic TACE. With a median follow-up of 48.3 months in entire study patients, postoperative tumor progression occurred in 81 (42.2%) patients and 32 (16.7%) patients died. Nevertheless, patients with a high HLA-DR+ T cell ratio were less likely to present cancer progression (35%).

Te Relationship between Clinicopathological Characteristics and the HLA-DR+ T Cell Ratio.
A total of 100 (52.1%) patients had a high HLA-DR+ T cell ratio. Te relationship between clinicopathological characteristics and the HLA-DR+ T cell ratio is shown in Table 1. Patients with a high HLA-DR+ T cell ratio had a higher T cell ratio (P � 0.005), a higher CD8+ T cell ratio (P < 0.001), and a lower B cell ratio (P � 0.005) than patients with a low HLA-DR+ T cell ratio. In addition, in Figure 1, there is Journal of Oncology 3 a positive correlation between the HLA-DR+ T cell ratio and the T cell ratio (P < 0.05) as well as the CD8+ T cell ratio, while there is a negative correlation between the HLA-DR+ T cell ratio and the B cell ratio (P < 0.05).

COX Regression Analyses between Clinicopathological
Variables Associated with PFS and OS after Curative Surgery for HCC. As indicated by the Kaplan-Meier curves in Figure 2(a), HCC patients with a high HLA-DR+ T cell ratio had better PFS (P � 0.003) than those with a low HLA-DR+ T cell ratio; unfortunately, Figure 2(b) shows that the HLA-DR+ T cell was not a prognostic biomarker for OS in HCC patients (P � 0.567). According to the results of multivariate COX regression analyses, the following features were statistically signifcant with PFS of HCC patients in Table 2: serum GGT level (P � 0.027), serum HLA-DR+ T cell ratio (P � 0.002), tumor Edmondson grade (P � 0.047), and capsular invasion (P � 0.045); and the following features were statistically signifcant with OS of HCC patients in Table 3: serum GGT level (P � 0.044), largest tumor size (P � 0.027), and microvascular invasion (P � 0.030). As a result, the HLA-DR+ Tcell ratio was a prognostic factor for PFS but not for OS of HCC patients.

COX Regression Analyses between Clinicopathological Variables Associated with PFS and OS after Curative Surgery for AFP-Positive HCC.
Te clinicopathological characteristics associated with serum AFP positivity (≥20 ng/ml) versus AFP negativity (<20 ng/ml) among HCC patients were noted in Table 4. AFP-positive HCC patients had more female patients (P � 0.048), higher proportion of HBV infection (P � 0.041), lower B cell ratio (P � 0.011), worse tumor Edmondson grade (P � 0.001), and more microvascular invasion (P � 0.049) than AFP-negative HCC patients. Figures 3(a) and 3(b) demonstrate that the HLA-DR+ T cell ratio was a prognostic factor for PFS (P � 0.020) but not for OS (P � 0.239) in AFP-positive HCC patients. However, Figures 4(a) and 4(b) depict that the HLA-DR+ T cell ratio was nonstatistically signifcant with PFS (P � 0.088) and OS (P � 0.632) of AFP-negative HCC patients. As shown in Tables 5 and 6, COX regression analyses also indicated that the HLA-DR+ T cell ratio was the only prognostic factor for PFS of AFP-positive HCC patients. Similar to the results of all patients in this study, AFPpositive patients with a high HLA-DR+ T cell ratio are more likely to have a high T cell ratio (P � 0.001), a high CD8+ T cell ratio (P � 0.020), and a low B cell ratio (P � 0.022) in Table 7, but AFP-negative HCC patients with a high HLA-DR+ T cell ratio are more likely to have a high CD8+T cell ratio (P � 0.007) in Table 8.

Discussion
With the deepening of the research on the immune environment, more and more studies have confrmed that the functional status of T cells plays a pivotal role in the occurrence and development of HCC [4,19,20]. As an important molecule for presenting antigens, HLA-DR was mostly used as a target on the surface of APC for HCCrelated research [12,21], whereas HLA-DR+ T cells are shown to be efector T cells in a variety of malignancies and serve as a predictive factor for antitumor treatment [22,23]. Based on these existing studies, we designed this project to explore the role of HLA-DR+ T cell in the anti-HCC immunity process. To the best of our knowledge, this research frst revealed that HLA-DR+ T cell was a predictive marker for recurrence of HCC after hepatectomy.
In this article, we revealed that HCC patients with a high HLA-DR+ T cell ratio were less likely to experience recurrence after curative liver resection. HLA-DR+ cytotoxic T lymphocytes (CTLs) are cytotoxic and express immune signatures of functionally activated cells, such as Granzyme B and IFN-c [24]. Meanwhile, regulatory T (Treg) cells upregulating HLA-DR expression increase in HCC and exhibit enhanced immunosuppressive activity driven by the hypoxia environment [25,26]. However, it has been recently reported that Treg cells are diferently expressed in HCC according to etiology of underlying liver cirrhosis. Treg cells play an immunosuppressive role in chronic viral liver diseases by hindering the antiviral process, while in autoimmune liver diseases, Treg cells show quantitative and functional defects so that they cannot efectively suppress self-reactive lymphocytes [27][28][29]. Besides, monitoring the genetic background (HLA) of Treg cells has potential value in assessing the extent of drug side efects during treatment with immune checkpoint inhibitors (ICIs) in patients with HCC [29]. HLA-DR is expressed not only on the surface of CTLs but also on immunosuppressive T cells so that the efect of HLA-DR on tumor immunity may be opposite. It is necessary to separately research the expression and prognostic value of HLA-DR in various T lymphocyte subsets. Besides, through the signifcance test of diference, we demonstrated that patients in the high HLA-DR+ T cell ratio group had a lower B cell ratio than patients in the low HLA-DR+ T cell ratio group. Te role of B cells in T cell activation is bidirectional. Tumor-infltrating B cells activate T cells through spatial cell-to-cell contacts to enhance antitumor activity in HCC patients [30]. On the contrary, B cell-derived GABA (c-aminobutyric acid) impairs cytotoxic T cell responses and antitumor immunity [31]. In this article, we revealed that the B cell ratio was negatively correlated with the activated (HLA-DR+) T cell ratio. In conclusion, we believed that HCC patients with a high HLA-DR+ Tcell ratio had a low B cell ratio and were less prone to tumor progression postoperatively.
AFP, as a tumor-associated antigen (TAA) of HCC [32], is considered a serum biomarker for diagnosis and a potential target for immunotherapy [33]. We, respectively, analyzed the prognostic signifcance of the HLA-DR+ T cell ratio in AFP-positive and AFP-negative HCC patients. Tere was no statistically signifcant diference in the HLA-DR+ T cell ratio between AFPpositive and AFP-negative HCC patients. But only among AFP-positive HCC patients, we found that patients in the high HLA-DR+ T cell ratio group had a lower rate of postoperative tumor progression than patients in the low HLA-DR+ T cell ratio group. It has been reported         that AFP has strong binding properties towards HLA-DR of helper T lymphocyte and T cell response against AFP contributes to the signifcantly improving survival rate [34]. As part of the normal T cell repertoire, TAA-specifc T cell responses are important for controlling HCC in diferent stages [35,36]. However, further research will be needed to confrm whether HLA-DR+ T cells are AFPspecifc T cells.

Journal of Oncology
Tis study aimed to fnd new predictors of postoperative recurrence to optimize postoperative follow-up protocols. For the frst time, HLA-DR+ T cell was used as a marker of antitumor immunity activity in the study of HCC. Simultaneously, we conducted stratifcation studies and demonstrated the specifc prognostic value of HLA-DR+ T cells for postoperative recurrence in AFP-positive HCC patients.
Moreover, apart from the HLA-DR+ T cell ratio, this study found other predictors for progression of HCC patients after curative resection, such as the GGT level, Edmondson grade, and capsular invasion.
Meanwhile, there were a number of limitations in the current research. First, this retrospective study in a single center may bring about selection bias. Consequently, a largescale, multicenter, and prospective study is needed to validate the results of this study. Second, the classifcation of continuous variables is bounded by the ULN of a single medical institution. Whether the results will difer from those of other hospitals is unknown. Tird, this study only included HBV, which is a common etiology of HCC in China, so it cannot be confrmed whether HLA-DR+ T cell have the same prognostic value in HCC caused by other    pathogenies. Last, some postoperative adjuvant treatments, such as antiviral therapy, targeted therapy, and immunotherapy, were not included in this study. We need to incorporate these factors afecting postoperative recurrence into the study for further stratifed analysis.

Conclusion
In summary, our study found that the preoperative HLA-DR+ T cell ratio served as a useful prognostic marker for PFS in HCC patients with curative resection, and the same results were obtained in AFP-positive patients. Tis is benefcial for clinicians to screen patients with poor prognoses based on clinical data in order to improve patient outcomes by developing personalized treatment and follow-up plans.

Data Availability
Te datasets analyzed during the current study are available from the corresponding author upon request.

Conflicts of Interest
Te authors declare that they have no conficts of interest. 10 Journal of Oncology