Influencing Factors and Prognostic Significance of TIL Subset Abundance in Head and Neck Squamous Cell Carcinoma

doi:10.21203/rs.3.rs-99693/v1

Download PDF

Research

Influencing Factors and Prognostic Significance of TIL Subset Abundance in Head and Neck Squamous Cell Carcinoma

https://doi.org/10.21203/rs.3.rs-99693/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Objective: To evaluate the abundance of Foxp3-positive tumor-infiltrating lymphocytes (Foxp3⁺TILs), CD103⁺cells, CD8⁺TILs, and CD3⁺TILs in tissue specimens of 85 head and neck squamous cell carcinoma (HNSCC) and analyze the associations of their abundance with the clinicopathological features and prognosis.

Methods: Immunohistochemistry was performed to detect the abundance of immunocyte subsets. Both correlation analysis and multivariate logistic regression analysis were used to analyze the associations between the frequency of these cell subsets and clinicopathological features in HNSCC. Kaplan–Meier analysis and the Cox proportional-hazards model were utilized to evaluate the effects of these immunocytes on the prognosis of patients with HNSCC.

Result: Multivariate logistic regression analysis showed that an independent association of Foxp3⁺TILs abundance with preoperative chemotherapy (PC) was negative (P=0.018), and that with smoking index (P=0.001) and N stage (P=0.012) was positive. Multivariate Cox model revealed that the high abundance of Foxp3⁺TILs was independently associated with short recurrence-free survival (RFS) (P=0.015) and poor overall survival (OS) (P=0.015). The high frequency of CD103⁺cells was independently related to long RFS (P=0.001) and favorable OS (P<0.001). The population with a high abundance of CD8⁺TILs and CD103⁺immunocytes in HNSCC tissues exhibited optimal RFS and OS, while the sub-group with a low frequency of CD8⁺TILs and CD103⁺immunocytes showed poor RFS and OS. Moreover, Spearman’s correlation analysis revealed that the frequency of CD103⁺cells was positively correlated to the abundance of CD8⁺TILs in HNSCC (P<0.001).

Conclusion: Chemotherapy eliminates the frequency of Foxp3⁺TILs and improves the immune tumor microenvironment (iTME) of HNSCC. On the other hand, heavy smoking upregulates the abundance of Foxp3⁺TILs in HNSCC tissues and disrupt the balance of immune homeostasis in iTME. Additionally, both Foxp3⁺TILs and CD103⁺cells might be potential independent prognostic indicators of HNSCC.

Oncology

HNSCC

TILs

chemotherapy

smoking

prognostic indicators

The immune tumor microenvironment (iTME) consists of several factors, such as immunocytes, which have a major influence on tumor progression. Recently, a number of studies have shown that tumor-infiltrating lymphocytes (TILs) play a vital role in the therapeutic response and prognosis of patients with solid tumors [1–8]. Typically, CD3 is expressed on almost all T cells, and CD3⁺TILs are heterogeneous and divided into CD4⁺TIL and CD8⁺TIL subsets, and the prognostic value of CD3⁺TILs in different solid tumors is controversial [4, 9]. CD8⁺TILs mainly produce a cytotoxic effect on cancer cells and is favorable to prognosis in solid tumor [4, 10]. However, CD8⁺TILs also show a larger transcriptional heterogeneity among individuals [11]. In addition, the bystanders in CD8⁺TILs do not respond to cancer cells [12]. Interestingly, several studies have proposed that CD103 is a marker of antigen-educated tumor-specific CD8⁺TILs and maybe an independent and low-risk factor for predicting prognosis in solid tumors [5–8]. Previous studies have shown that forkhead box P3 (Foxp3) is a characteristic marker of regulatory T-cells (Tregs) [13, 14], and immunosuppressive Foxp3⁺Tregs are mainly CD4⁺TILs and is unfavorable to the prognosis of patients with cancer [15, 16]. Briefly, TILs are major elements in iTME and impact the prognosis of patients with solid tumors. A recent study has shown that preoperative chemotherapy (PC) improves the iTME via transforming the TILs subset landscape in breast cancer [2]. Nevertheless, the impact of the factors on the distribution of TIL subsets is yet to be investigated in depth. Therefore, immunohistochemistry (IHC) was performed to assay the abundance of TIL subsets (CD3, CD8, and Foxp3) and CD103⁺immunocytes in head and neck squamous cell carcinoma (HNSCC) and study the effect of clinicopathological factors on TIL landscape and the prognostic significance of these cell subsets in iTME.

2.1 Patients

A total of 85 patients who were histologically diagnosed with HNSCC between May 2012 and December 2013 at Tianjin Medical University Cancer Institute and Hospital (TMUCIH; Tianjin, China) were included in this study, and the focal information is described below, as reported in our previous studies [17, 18]. All the patients with HNSCC underwent radical surgery at TMUCIH. The inclusion criteria were as follows: (1) no radiotherapy before surgery, (2) pathological confirmation of the clinical diagnosis, and (3) providing sufficient samples to fulfill the requirements of experiments. The exclusion criteria were as follows: (1) patients with nasopharyngeal or oropharyngeal carcinoma, (2) patient’s age < 16 or > 80-years-old, (3) no residual cancer cells on microscopic examination, or (4) missing follow-up and deaths unrelated to tumor progression. Pathological staging was estimated according to the 7th edition of the American Joint Committee on Cancer (AJCC) staging system for HNSCC. Follow-up was conducted on all the enrolled patients with HNSCC until December 2018. The median follow-up time was 61.7 (range, 3.9–67.6; mean, 48.6) months for patients who were alive at the last follow-up. During the follow-up period, 36 (36/85) patients had events of local recurrence or distant metastasis, and 34 patients died of HNSCC progression. Recurrence-free survival (RFS) was defined as the period from the date of radical surgery to the time of local recurrence or distant metastasis. Overall survival (OS) was defined as the duration from the date of radical surgery to death from cancer cause. Smoking index (SI) was defined as the number of cigarettes smoked daily multiplied by the total number of years of smoking. The demographic and clinicopathological features of these patients are summarized in Table 1.

2.2 IHC

The method of immunohistochemistry (IHC) was similar to that reported previously[17, 18]. Briefly, the formalin-fixed, paraffin-embedded (FFPE) tissue was sliced into 4-mm tissue sections on the microscope slides (Citotest, Shenzhen, China) and preheated at 70 ℃ for 1 h. After heat drying, the sections were deparaffinized in xylene and rehydrated in the gradient of ethanol. According to preset experimental operation requirements, the citrate buffer (pH = 6.0) or ethylene diamine tetraacetic acid (pH = 9.0) was used for heat-induced epitope retrieval (HIER) at 123℃ for 3 min, and the slides were slowly or immediately cooled for antigen retrieval. Subsequently, all the slides were immersed in 3% hydrogen peroxide for 20 min to inactivate the endogenous peroxidase of the specimens, followed by incubation with bovine serum albumin (BSA) to reduce the non-specific hydrophobic binding. Then, the tissue sections were incubated with mouse monoclonal primary antibodies against Foxp3 (SP97; dilution, 1:100; Abcam, CO, USA), CD103 (EPR4166(2); dilution, 1:500; Abcam), and CD3 (F7.2.38; dilution, 1:300; Thermo Fisher Scientific, Waltham, MA, USA), and rabbit monoclonal primary antibody against CD8 (SP16; dilution, 1:300; Thermo Fisher Scientific) at 4 ℃ overnight. After washing thrice with phosphate-buffered saline (PBS), the sections were incubated with goat anti-mouse horseradish peroxidase (HRP)-conjugated secondary antibody (Maixin Technology Co., Ltd, Shenzhen, China) for 30 min at room temperature before staining with 3,3’-diaminobenzidine (DAB). The color reaction was terminated by washing with deionized water. Finally, the slides were sealed with resin and mounted after dehydration in alcohol and xylene. The specimens were examined under a light microscope.

Two pathologists independently evaluated the IHC results of all samples. A total of ten high-power fields (HPF, × 400) were randomly selected at the strong density region of DAB-chromogenic cells, and the number of positive cells was recorded. The average number of positive cells was regarded as the abundance of DAB-chromogenic immunocytes. The receiver operating characteristic (ROC) curve was constructed using the method of minimum P-value in this cohort. CD8⁺TIL ≥ 30/HPF was defined as CD8⁺TILs^high or the high abundance of CD8⁺TILs, CD8⁺TIL < 30/HPF as CD8⁺TILs^low or the low abundance of CD8⁺TILs, CD3⁺TIL ≥ 59/HPF as CD3⁺TILs^high or the high frequency of CD3⁺TILs and CD3⁺TIL < 59/HPF as CD3⁺TILs^low or the low frequency of CD3⁺TILs, Foxp3⁺TIL ≥ 8/HPF as Foxp3⁺TILs^high or the high distributional level of Foxp3⁺TILs and Foxp3⁺TIL < 8/HPF as Foxp3⁺TILs^low or the low distributional level of Foxp3⁺TILs, CD103⁺cells ≥ 21/HPF as CD103⁺cells^high or the high abundance of CD103⁺cells, and CD103⁺cells < 21/HPF as CD103⁺TILs^low or the low abundance of CD103⁺cells.

2.3 Statistical analysis

Statistical analyses were performed using SPSS 24.0 software (IBM, Armonk, NY, USA). The chi-square test and Fisher’s exact test were used to analyze the association between the abundance of immunocytes subsets (CD3⁺TIL, CD8⁺TILs, CD103⁺cells, and Foxp3⁺TILs) and clinicopathological features. The Spearman’s correlation was applied to analyze the correlation among these subsets (CD3⁺TIL, CD8⁺TILs, and CD103⁺cells). Mann-Whitney U test was utilized to analyze the correlation between smoking status and the abundance of Fxop3⁺TILs and to study the difference in the frequency of these subsets (CD3⁺TIL, CD8⁺TILs, CD103⁺cells, and Foxp3⁺TILs) between the PC and non-PC groups. Multivariate logistic regression analysis was used to screen the factors predicting the abundance of subsets (CD8⁺TILs and Foxp3⁺TILs). RFS and OS data were estimated using the Kaplan–Meier method, and the log-rank test was used to compare the differences of high vs. low abundance of subsets (CD3⁺TIL, CD8⁺TILs, CD103⁺cells, and Foxp3⁺TILs). The Cox proportional-hazards model was used for multivariate regression analysis of independent prognostic factors associated with RFS/OS. The two-tailed P < 0.05 was considered statistically significant.

3.1 Patients’ clinical and demographic characteristics

A total of 85 HNSCC cases were enrolled in this study, as reported in previous investigations; also, the clinical and demographic characteristics of the patients were presented before [17, 18]. These patients with primary HNSCC included oral cancer (n = 38), laryngeal cancer (n = 36), hypopharyngeal cancer (n = 9), as well as other cancers (n = 2). The constituent ratio of patients at pathological stages I, Ⅱ, Ⅲ, and Ⅳ was36.5%, 18.8%, 15.3%, and 29.4%, respectively. Moreover, 21 cancer tissues contained poorly differentiated clusters (PDCs). The age of the patients was 27–76 (median age, 57)-years-old, and the cohort consisted of 18 (22.2%) female cases, 49 (57.6%) with alcohol consumption history, 54 (63.5%) with the history of smoking, and 54 (63.5%), who received chemotherapy before radical surgery. In addition, the baseline characteristics of Foxp3⁺TILs abundance in patients with HNSCC are listed in Table 1, and those of other immunocyte subsets are presented in Suppl. tables

3.2 Staining and abundance of immunocyte subsets in HNSCC

The staining characteristics of CD3, CD8, CD103, and Foxp3 are illustrated in the sections of FFPE tissue from four patients (Fig. 1). The IHC of CD3 and CD8 showed yellow staining in the membrane of TILs, and the coloration of CD103 was also in the membrane of cells, which was infiltrating in carcinoma tissues; however, Fxop3 staining exhibited in the nucleus of TILs. Among 85 patients with HNSCC, 39 cases (45.9%) showed the abundance of CD3⁺TILs < 59/HPF, 46 cases (54.1%) exhibited the frequency of CD3⁺TILs ≥ 59/HPF, 26 cases (30.6%) presented the frequency of CD3⁺TILs < 30/HPF, 59 cases (69.4%) emerged the abundance of CD3⁺TILs ≥ 30/HPF, 39 cases (45.9%) showed the abundance of CD103⁺cells < 21/HPF, 46 cases (54.1%) exhibited the frequency of CD103⁺cells ≥ 21/HPF, 44 cases (51.8%) presented the frequency of Foxp3⁺TILs < 8/HPF, and 41 cases (48.2%) emerged the abundance of Foxp3⁺TILs ≥ 8/HPF. Furthermore, among 59 patients with high abundance of CD8⁺TILs, 39 cases (66.1%) exhibited CD103⁺cells^high and 20 cases (33.9%) displayed CD103⁺cells^low, while 26 patients showed low abundance of CD8⁺TILs, 7 cases (26.9%) had CD103⁺cells^high, and 19 cases (73.1%) presented CD103⁺cells^low.

3.3 Correlation between immunocyte subset abundance and clinicopathological features

In the current study, we found that a significantly high abundance of Foxp3⁺TILs was correlated to gender, site of primary cancer, and smoking history (Table 1). A significant increase was detected in the incidence of the high frequency of Foxp3⁺TILs with male gender, history of smoking (SI ≥ 1000), and primary laryngeal carcinoma, i.e., 22.2% (4/18) and 55.2% (37/67) in female cases vs. male cases (P = 0.017), 40.0% (26/65) and 75.0% (15/20) from cases with IS < 1000 vs. IS ≥ 1000 (P = 0.006), and 36.7% (18/49) and 63.9% (23/36) from cases with non-laryngeal carcinoma vs. primary laryngeal carcinoma (P = 0.013). However, an obvious decrease was noted in the incidence of a high abundance of Foxp3⁺TILs in cases with oral cancer, i.e., 63.8% (30/47) and 28.9% (11/38) from the cases with non-oral carcinoma vs. primary oral carcinoma (P = 0.001). The results of the Mann–Whitney U test revealed that PC and smoking history were significantly associated with the abundance of Foxp3⁺TILs (Fig. 2). However, no evident association was detected between the frequency of Foxp3⁺TILs and age, laryngopharynx, stages (T, N, and TNM), PDCs, or drinking. Additionally, we performed multivariate logistic regression analysis, and it was disclosed that the abundance of Foxp3⁺TILs was independently associated with PC, SI, and N stage (Table 2).

We also showed that a significantly higher frequency of CD8⁺TILs was related to the sites of primary cancer, local lymph node metastasis, and the sample mixed PDCs (Suppl. Table 2). A significant decrease was noted in the incidence of the high abundance of CD8⁺TILs with primary hypopharyngeal carcinoma, lymph node metastasis (N1–N3), and mixed PDCs, i.e., 73.7% (56/76) and 33.3% (3/9) from cases with non-hypopharyngeal carcinoma vs. primary hypopharyngeal carcinoma (P = 0.021), 79.7% (47/59) and 46.2% (12/26) from cases with N0 stage vs. N1–N3 stage (P = 0.003), and 80.3% (49/61) and 47.6% (10/21) from samples with PDCs mixed vs. PDCs absent (P = 0.004). However, an obvious increase was observed in the incidence of the high abundance of Foxp3⁺TILs with oral cancer, i.e., 59.6% (28/47) and 81.6% (31/38) from cases with non-oral carcinoma vs. primary oral carcinoma (P = 0.029). Also, no correlation was established between the frequency of CD8⁺TILs and age, gender, larynx, stages (T and TNM), PC, smoking, or drinking. Moreover, multivariate logistic regression analysis was performed, which disclosed that the abundance of CD8⁺TILs was independently associated with the primary sites of cancer (oral carcinoma) (Suppl. Table 3).

Furthermore, no correlation was found between the clinicopathological factors mentioned above and the CD3⁺TIL abundance or CD103⁺cells frequency (Suppl. Table 1, Suppl. Table 4, and Fig. 2E). Subsequently, the Spearman’s correlation analysis discovered a positive correlation was between CD3⁺TILs frequency and CD8⁺TIL abundance, CD3⁺TIL abundance and CD103⁺cell frequency, and CD8⁺TIL frequency and CD103⁺cell abundance (Fig. 2), rather than between Foxp3⁺TIL frequency and CD3⁺TILs, CD8⁺TILs, or CD103⁺cells (data not shown).

3.4 Associations of immunocytes subset abundance with prognoses in HNSCC patients

The results of Kaplan–Meier analysis and log-rank test revealed that both CD8⁺TILs^high and CD103⁺cells^high were markedly correlated with long RFS (P < 0.001 and P = 0.002, respectively) and favorable OS (all P = 0.001) rather than CD3⁺TIL^high; however, Foxp3⁺TILs^high was obviously associated with short RFS (P = 0.001) and poor OS (P = 0.004), respectively (Fig. 3). The Cox proportional-hazards regression model was utilized to identify the effects of clinicopathological prognostic factors on the survival of patients with HNSCC. Univariate Cox model analysis demonstrated that the sites of primary cancer (hypopharyngeal carcinoma), pathological stages (T, N, and TNM), degree of differentiation, and the frequency of Foxp3⁺TILs were correlated with short RFS of patients with HNSCC (all P < 0.05; Table 3). In addition, the sites of primary cancer (hypopharyngeal carcinoma), pathological stages (T, N, and TNM), and the frequency of Foxp3⁺TILs were correlated with poor OS of patients with HNSCC (all P < 0.05; Table 4) rather than the degree of differentiation. However, CD8⁺TILs^high and CD103⁺cells^high were associated with long RFS and favorable OS (all P < 0.05; Tables 3 and 4). Multivariate Cox model revealed that the high abundance of Foxp3⁺TILs was independently associated with short RFS (hazard ratio (HR) = 3.014, 95% confidence interval (CI): 1.236–7.348; P = 0.015; Table 3) and poor OS (HR = 2.802, 95% CI: 1.226–6.399; P = 0.015; Table 4). The high frequency of CD103⁺cells was independently related to long RFS (HR = 0.300, 95% CI: 0.145–0.620; P = 0.001; Table 3) and improved OS (HR = 0.253, 95% CI: 0.117–0.546; P < 0.001; Table 4) rather than CD8⁺TILs^high. Especially, the sub-population with a high abundance of CD8⁺TILs and CD103⁺immunocytes in HNSCC tissues had the best RFS and OS, while the sub-group with a low frequency of CD8⁺TILs and CD103⁺immunocytes had the worst RFS and OS (Fig. 4).

In the treatment of HNSCC, the recommendation for induction chemotherapy (IC) is commonly based on the update from RTOG 91 − 11 [19]. Several phase III trials showed significantly improved survival for patients in the 3-drug induction group (cisplatin, 5-FU plus taxane; TPF) as compared to those receiving 2 drugs (cisplatin plus 5-FU) [20–24]; this result was supported by a meta-analysis encompassing five randomized controlled trials (RCTs) [25]. Due to the adjustment of treatment strategy, after IC, some patients with HNSCC underwent radical surgery, instead of concurrent chemoradiation, which provided an advantage in understanding the transformation of iTME after chemotherapy for HNSCC.

In these patients with HNSCC, Foxp3⁺TILs was significantly eliminated in the PC sub-group as compared to the non-PC sub-population, rather than CD3⁺TILs, CD8⁺TILs, and CD103⁺cells. Docetaxel is a microtubule inhibitor and should preferentially inhibit the proliferation and function of rapidly dividing Tregs, instead of CD4⁺ effector T cells [26], in which support our finding that Foxp3⁺TILs are eliminated. In addition, several preclinical trials have shown that docetaxel also enhances CD8⁺T-cell responses and antitumor activity [27, 28], although our results showed that the frequency of CD8⁺TILs and CD103⁺cells in the sub-group that received PC was not decreased significantly than that in the non-PC sub-population. Briefly, PC may be the main reason for the transformation of the T-cell landscape, as substantiated previously [29, 30].

Smoking, drinking, chewing betel nut, and virus infections play a pivotal role in the progression and development of HNSCC. The smoking-related carcinogen, 4-(N-methyl-N-nitrosamino)-1-(3-pyridyl)-1-butanone (NNK) elevates the expression of HIF-1α and HIF-2α [31]. HIF-1α directly binds to the Foxp3 promoter and upregulates Foxp3 expression in the cutaneous T cells [32]. The result of multivariate logistic regression analysis disclosed that the abundance of Foxp3⁺TILs was independently associated with the smoking history of the patients with HNSCC. These results indicated that smoking carcinogen, such as NNK, upregulates the abundance of Foxp3TILs in HNSCC via HIF-1 signaling pathway.

Additionally, the result of multivariate Cox regression analysis showed that the high frequency of Foxp3⁺TILs is one of the independent risk factors for the survival of the patients with HNSCC. Several studies have shown that Foxp3 is a characteristic marker of Tregs [13, 14], and immunosuppressive Foxp3⁺Tregs are unfavorable to the prognosis of patients with solid tumors [15, 16]. Consequently, the result of the current study demonstrated that Foxp3⁺TIL in HNSCC might be a dominant subset of Tregs with immunosuppression.

Herein, the results of survival analysis established a correlation between CD8⁺TILs^high/CD103⁺cells^high and long RFS and optimal OS in patients with HNSCC. The Spearman’s correlation analysis demonstrated that the frequency of CD8⁺TILs was positively correlated with the abundance of CD103⁺cells, and multivariate Cox regression analyses revealed that CD103⁺cells^high was deemed as one of the independent factors for predicting favorable survival rather than CD8⁺TILs^high. Moreover, the combined analysis of the effects of CD8⁺TILs^high and CD103⁺cells^high on the survival of patients with HNSCC revealed maximal RFS and OS for the sub-population with a high abundance of CD8⁺TILs and CD103⁺immunocytes in HNSCC tissues. Several studies also showed that the frequency of CD103⁺CD8⁺TILs was associated with the level of granzyme B, which is critical to kill cancer cells, and the CD8⁺CD103⁺TILs or CD103⁺TILs are favorable to the prognosis in breast cancer, lung cancer, and laryngeal squamous cell carcinoma [5, 8, 33, 34]. These findings suggested that CD103 is a vital biomarker of cytotoxic CD8⁺TILs in solid tumors, while the frequency of CD103⁺cells may be an indicator for the prognosis of some solid tumors.

Interestingly, the abundance of CD8⁺TILs in oral carcinoma tissues was higher than that in the HNSCC originating from other sites, and the association of original site of the tumor with the frequency of CD8 + TILs was independent. Nevertheless, the independent and positive association of CD8⁺TILs^high with oral carcinoma needs to be elucidated with respect to the mechanisms underlying this phenomenon. Since the comprehensive knowledge of the mechanism is yet lacking, we speculate that the phenomenon may be important when considering the trial of immunotherapy strategies for HNSCC.

Chemotherapy significantly eliminates the frequency of Foxp3⁺TILs and improves the iTME of HNSCC. Conversely, heavy smoking upregulates the abundance of Foxp3⁺TILs in HNSCC tissues and overturn the balance of immune homeostasis in iTME. Additionally, both Foxp3⁺TILs and CD103⁺cells may be potentially independent prognostic indicators of HNSCC.

DAB: 3,3’-diaminobenzidine

Foxp3: forkhead box P3

FFPE: formalin-fixed, paraffin-embedded

HNSCC: head and neck squamous cell carcinoma

HPF: high-power fields

iTME: immune tumor microenvironment

IHC: immunohistochemistry

IC: induction chemotherapy

NNK: 4-(N-methyl-N-nitrosamino)-1-(3-pyridyl)-1-butanone

OS: overall survival

PC: preoperative chemotherapy

PDCs: poorly differentiated clusters

RFS: recurrence-free survival

SI: smoking index

TILs: tumor-infiltrating lymphocytes

Tregs: regulatory T-cells

Availability of data and materials

The datasets generated and/or analyzed during the current study are not publicly available due to them being collected within the institution but are available from the corresponding author on reasonable request.

Authors’ contributions

Xinwei Zhang and Li Zhou contributed to the conception and design of the study. Preparation of materials, performing the experiments, data collection, and data analysis were conducted by Xuezhou Wang Li Zhou, Ruifeng Dong, Baihui Li, and Wenwen Yu. The first draft of the manuscript was prepared by Xuezhou Wang. All the authors read the first version of the manuscript and provided their valuable comments, and Xinwei Zhang and Xiubao Ren revised the manuscript accordingly. All the authors read and approved the final version of the manuscript.

Ethics statement

The present study was approved by the Ethics Committee of TMUCIH, and the study was conducted in accordance with the Declaration of Helsinki. Patients or their guardians have given their written informed consent.

Consent for publication

We just extracted data and did not involve the private information of patients.

Funding

This study was financially supported by the National Natural Science Foundation of China (Grant No. 81702702 and 81702405), the Key Projects of Tianjin Health Industry (Grant No. 15KG145), Natural Science Foundation of Tianjin Province (Grant No. 19JCQNJC10800), and the Scientific Research Program of Tianjin Education Commission (Grant No. 2019KJ185).

Conflict of interest

The authors declare that there is no conflict of interest.

Acknowledgments

We sincerely appreciate the efforts of all the patients, physicians, and students during this study.

Kochi M, Iwamoto T, Niikura N, Bianchini G, Masuda S, Mizoo T, et al. Tumour-infiltrating lymphocytes (TILs)-related genomic signature predicts chemotherapy response in breast cancer. Breast Cancer Res Treat. 2018 Jan;167(1):39-47.
Goto W, Kashiwagi S, Asano Y, Takada K, Takahashi K, Hatano T, et al. Predictive value of improvement in the immune tumour microenvironment in patients with breast cancer treated with neoadjuvant chemotherapy. ESMO Open. 2018 Aug 30;3(6):e000305.
Kashiwagi S, Asano Y, Goto W, Takada K, Takahashi K, Noda S, et al. Use of Tumor-infiltrating lymphocytes (TILs) to predict the treatment response to eribulin chemotherapy in breast cancer. PLoS One. 2017 Feb 6;12(2):e0170634.
Rapp C, Dettling S, Liu F, Ull AT, Warta R, Jungk C, et al. Cytotoxic T Cells and their Activation Status are Independent Prognostic Markers in Meningiomas. Clin Cancer Res. 2019 Sep 1;25(17):5260-5270.
Wang ZQ, Milne K, Derocher H, Webb JR, Nelson BH, Watson PH. CD103 and Intratumoral Immune Response in Breast Cancer. Clin Cancer Res. 2016 Dec 15;22(24):6290-6297.
Webb JR, Milne K, Nelson BH. Location, location, location: CD103 demarcates intraepithelial, prognostically favorable CD8+ tumor-infiltrating lymphocytes in ovarian cancer. Oncoimmunology. 2014 Jan 10;3:e27668.
Mokrani M, Klibi J, Bluteau D, Bismuth G, Mami-Chouaib F. Smad and NFAT pathways cooperate to induce CD103 expression in human CD8 T lymphocytes. J Immunol. 2014 Mar 1;192(5):2471-2479.
Koh J, Kim S, Kim MY, Go H, Jeon YK, Chung DH. Prognostic implications of intratumoral CD103+ tumor-infiltrating lymphocytes in pulmonary squamous cell carcinoma. Oncotarget. 2017 Feb 21;8(8):13762-13769.
Li J, Wang J, Chen R, Bai Y, Lu X. The prognostic value of tumor-infiltrating T lymphocytes in ovarian cancer. Oncotarget. 2017 Feb 28;8(9):15621-15631.
Teng F, Meng X, Wang X, Yuan J, Liu S, Mu D, et al. Expressions of CD8+TILs, PD-L1 and Foxp3+TILs in stage I NSCLC guiding adjuvant chemotherapy decisions. Oncotarget. 2016 Sep 27;7(39):64318-64329.
Li Z, Chen G, Cai Z, Dong X, Qiu L, Xu H, et al. Genomic and transcriptional Profiling of tumor infiltrated CD8+ T cells revealed functional heterogeneity of antitumor immunity in hepatocellular carcinoma. Oncoimmunology. 2018 Nov 10;8(2):e1538436.
Simoni Y, Becht E, Fehlings M, Loh CY, Koo SL, Teng KWW, et al. Bystander CD8+ T cells are abundant and phenotypically distinct in human tumour infiltrates. Nature. 2018 May;557(7706):575-579.
Fontenot JD, Rudensky AY. A well adapted regulatory contrivance: regulatory T cell development and the forkhead family transcription factor Foxp3. Nat Immunol. 2005 Apr;6(4):331-337.
Hori S, Nomura T, Sakaguchi S. Control of regulatory T cell development by the transcription factor Foxp3. Science. 2003 Feb 14;299(5609):1057-1061.
Frydrychowicz M, Boruczkowski M, Kolecka-Bednarczyk A, Dworacki G. The Dual Role of Treg in Cancer. Scand J Immunol. 2017 Dec;86(6):436-443.
Pernot S, Terme M, Radosevic-Robin N, Castan F, Badoual C, Marcheteau E, et al. Infiltrating and peripheral immune cell analysis in advanced gastric cancer according to the Lauren classification and its prognostic significance. Gastric Cancer. 2020 Jan;23(1):73-81.
Wang X, Zhou L, Li B, Hu X, Dong R, Zhang X. Expression of CD39 in head and neck squamous cell carcinoma tissues and its prognostic value. Chin J Cancer Biother, 2020. 27(4): p. 396-402.
Zhou L, Wang X, Yu W, Yang L, Wei F, Sun Q, et al. Expression level of PD-L1 is involved in ALDH1A1-mediated poor prognosis in patients with head and neck squamous cell carcinoma. Pathol Res Pract. 2020 Sep;216(9):153093.
Forastiere AA, Zhang Q, Weber RS, Maor MH, Goepfert H, Pajak TF, et al. Long-term results of RTOG 91-11: a comparison of three nonsurgical treatment strategies to preserve the larynx in patients with locally advanced larynx cancer. J Clin Oncol. 2013 Mar 1;31(7):845-852.
Hitt R, López-Pousa A, Martínez-Trufero J, Escrig V, Carles J, Rizo A, et al. Phase III study comparing cisplatin plus fluorouracil to paclitaxel, cisplatin, and fluorouracil induction chemotherapy followed by chemoradiotherapy in locally advanced head and neck cancer. J Clin Oncol. 2005 Dec 1;23(34):8636-8645.
Posner MR, Hershock DM, Blajman CR, Mickiewicz E, Winquist E, Gorbounova V, et al. Cisplatin and fluorouracil alone or with docetaxel in head and neck cancer. N Engl J Med. 2007 Oct 25;357(17):1705-1715.
Lorch JH, Goloubeva O, Haddad RI, Cullen K, Sarlis N, Tishler R, et al. Induction chemotherapy with cisplatin and fluorouracil alone or in combination with docetaxel in locally advanced squamous-cell cancer of the head and neck: long-term results of the TAX 324 randomised phase 3 trial. Lancet Oncol. 2011 Feb;12(2):153-159.
Vermorken JB, Remenar E, van Herpen C, Gorlia T, Mesia R, Degardin M, et al. Cisplatin, fluorouracil, and docetaxel in unresectable head and neck cancer. N Engl J Med. 2007 Oct 25;357(17):1695-1704.
Janoray G, Pointreau Y, Garaud P, Chapet S, Alfonsi M, Sire C, et al. Long-term Results of a Multicenter Randomized Phase III Trial of Induction Chemotherapy With Cisplatin, 5-fluorouracil, ± Docetaxel for Larynx Preservation. J Natl Cancer Inst. 2015 Dec 16;108(4):djv368.
Blanchard P, Bourhis J, Lacas B, Posner MR, Vermorken JB, Cruz Hernandez JJ, et al. Taxane-cisplatin-fluorouracil as induction chemotherapy in locally advanced head and neck cancers: an individual patient data meta-analysis of the meta-analysis of chemotherapy in head and neck cancer group. J Clin Oncol. 2013 Aug 10;31(23):2854-2860.
Roselli M, Cereda V, di Bari MG, Formica V, Spila A, Jochems C, et al. Effects of conventional therapeutic interventions on the number and function of regulatory T cells. Oncoimmunology. 2013 Oct 1;2(10):e27025.
Garnett CT, Schlom J, Hodge JW. Combination of docetaxel and recombinant vaccine enhances T-cell responses and antitumor activity: effects of docetaxel on immune enhancement. Clin Cancer Res. 2008 Jun 1;14(11):3536-3544.
Kodumudi KN, Weber A, Sarnaik AA, Pilon-Thomas S. Blockade of myeloid-derived suppressor cells after induction of lymphopenia improves adoptive T cell therapy in a murine model of melanoma. J Immunol. 2012 Dec 1;189(11):5147-5154.
Podrazil M, Horvath R, Becht E, Rozkova D, Bilkova P, Sochorova K, et al. Phase I/II clinical trial of dendritic-cell based immunotherapy (DCVAC/PCa) combined with chemotherapy in patients with metastatic, castration-resistant prostate cancer. Oncotarget. 2015 Jul 20;6(20):18192-18205.
Ishii H, Chikamatsu K, Igarashi S, Takahashi H, Sakamoto K, Higuchi H, et al. Establishment of Synergistic Chemoimmunotherapy for Head and Neck Cancer Using Peritumoral Immature Dendritic Cell Injections and Low-Dose Chemotherapies. Transl Oncol. 2018 Feb;11(1):132-139.
Maugeri G, D'Amico AG, Rasà DM, La Cognata V, Saccone S, Federico C, et al. Nicotine promotes blood retinal barrier damage in a model of human diabetic macular edema. Toxicol In Vitro. 2017 Oct;44:182-189.
Clambey ET, McNamee EN, Westrich JA, Glover LE, Campbell EL, Jedlicka P, et al. Hypoxia-inducible factor-1 alpha-dependent induction of FoxP3 drives regulatory T-cell abundance and function during inflammatory hypoxia of the mucosa. Proc Natl Acad Sci U S A. 2012 Oct 9;109(41):e2784-e2793.
Savas P, Virassamy B, Ye C, Salim A, Mintoff CP, Caramia F, et al. Single-cell profiling of breast cancer T cells reveals a tissue-resident memory subset associated with improved prognosis. Nat Med. 2018 Jul;24(7):986-993.
Mann JE, Smith JD, Birkeland AC, Bellile E, Swiecicki P, Mierzwa M, et al. Analysis of tumor-infiltrating CD103 resident memory T-cell content in recurrent laryngeal squamous cell carcinoma. Cancer Immunol Immunother. 2019 Feb;68(2):213-220.

Table 1 Patient characteristics and association between the abundance of Foxp3⁺TILs and clinicopathological factors

Characteristic	n	Foxp3⁺TILs		P
Characteristic	n	low	high	P
Gender				0.017*
Male	67	30	37
Female	18	14	4
Age (years)				0.478
< 60	51	28	23
≥ 60	34	16	18
Sites 1				0.082*
Non-laryngopharynx	76	42	34
Laryngopharynx	9	2	7
Sites 2				0.001
Non-oral cavity	47	17	30
Oral cavity	38	27	11
Sites 3				0.013
Non-larynx	49	31	18
Larynx	36	13	23
T stage				0.350
T1	35	16	19
T2–T4	50	28	22
N stage				0.103
N0	59	34	25
N1–N3	26	10	16
TNM stage				0.169
Ⅰ	31	13	18
Ⅱ–Ⅳ	54	31	23
Poor differentiation
Absent	61	35	26	0.057
Mixed	21	7	14
Smoking index				0.006
< 1000	65	39	26
≥ 1000	20	5	15
Drinking				0.299
Non and mild	49	21	15
Constant	36	23	26
Preoperative chemotherapy				0.169
Absent	31	13	18
Received	54	31	23
CD3⁺TILs				0.165
Low	39	17	22
High	46	27	19
CD8⁺TILs				0.247
Low	26	11	15
High	59	33	26
CD103⁺cells				0.430
Low	39	22	17
High	46	22	24

* Fisher’s exact test

Table 2 Multivariate logistic regression analysis of the association between the abundance of Foxp3⁺TILs and clinicopathological factors

Factors	β	Wald χ²	Exp (B) (95% CI)	P
Gender (male vs. female)	-0.557	0.674	0.573 (0.151–2.166)	0.412
Sites 1 (Non-laryngopharynx vs. laryngopharynx)	1.064	1.304	2.897 (0.467–17.992)	0.254
Sites 2 (Non-oral cavity vs. oral cavity)	0.872	1.135	2.391 (0.481–11.879)	0.287
Sites 3 (Non-larynx vs. larynx)	0.761	1.990	2.140 (0.744–6.158)	0.158
N stage (N0 vs. N1–N3)	1.438	6.288	4.212 (1.369–12.958)	0.012
TNM stage (I vs. Ⅱ–Ⅳ)	-0.726	0.636	0.484 (0.139–1.684)	0.254
Poor differentiation (absent vs. mixed)	0.553	3.896	1.739 (0.431–7.010)	0.437
Smoking index (< 1000 vs. ≥ 1000)	2.444	11.946	11.515 (2.880–46.033)	0.001
Preoperative chemotherapy (absent vs. received)	-1.099	5.640	0.333 (0.135–0.825)	0.018
CD3⁺TILs (low vs. high)	-0.712	2.801	0.491 (0.213–1.130)	0.094

Table 3 Univariate and multivariate Cox regression analyses of clinicopathological factors related to recurrence-free survival

Factors	Univariate regression analysis		Multivariate regression analysis
Factors	HR (95% CI)	P	HR (95% CI)	P
Gender
(male vs. female)	0.500 (0.194–1.287)	0.151	0.437 (0.161–1.185)	0.104
Age (years)
< 60 vs. ≥ 60	1.086 (0.560–2.107)	0.807	-	-
Sites 1
Non-laryngopharynx vs. laryngopharynx	3.359 (1.529–7.539)	0.003	1.032 (0.369–2.885)	0.952
Sites 2
Non-oral cavity vs. oral cavity	0.642 (0.325–1.268)	0.202	-
Sites 3
Non-larynx vs. larynx	0.824 (0.421–1.610)	0.570	-
T stage
T1 vs. T2–T4	3.438 (1.564–7.588)	0.002	2.841 (1.109–7.281)	0.030
N stage
N0 vs. N1–N3	4.021 (2.077–7.784)	0.000	2.508 (1.013–6.207)	0.047
TNM stage
Ⅰ vs. Ⅱ–Ⅳ	2.682 (1.221–5.891)	0.014	1.413 (0.545–3.663)	0.477
Poor differentiation
Absent vs. mixed	2.088 (1.036–4.207)	0.039	1.184 (0.533–2.630)	0.679
Smoking index
< 1000 vs. ≥ 1000	1.393 (0.672–2.890)	0.373	-	-
Drinking
No and mild vs. constant	1.517 (0.789–2.916)	0.211	-	-
Preoperative chemotherapy
Absent vs. received	1.382 (0.691–2.765)	0.360	-	-
CD3⁺TILs
Low vs. high	0.549 (0.283–1.066)	0.077	0.870 (0.414–1.831)	0.715
CD8⁺TILs
Low vs. high	0.311 (0.161–0.601)	0.001	1.251 (0.480–3.258)	0.647
CD103⁺Cells
Low vs. high	0.362 (0.183–0.717)	0.004	0.300 (0.145–0.620)	0.001
Foxp3⁺TILs
Low vs. high	3.098 (1.518–6.319)	0.002	3.014 (1.236–7.348)	0.015

Table 4 Univariate and multivariate Cox regression analyses of clinicopathological factors related to overall survival

Factors	Univariate regression analysis		Multivariate regression analysis
Factors	HR (95% CI)	P	HR (95% CI)	P
Gender
(male vs. female)	0.562 (0.217–1.451)	0.234	-	-
Age (years)
< 60 vs. ≥ 60	1.081 (0.546–2.141)	0.822	-	-
Sites
Non-laryngopharynx vs. laryngopharynx	4.361 (1.942–9.794)	0.000	1.737 (0.690–4.371)	0.241
Sites 2
Non-oral cavity vs. oral cavity	0.716 (0.358–1.430)	0.343	-
Sites 3
Non-larynx vs. larynx	0.685 (0.339–1.385)	0.292	-
T stage
T1 vs. T2–T4	3.636 (1.581–8.363)	0.002	2.239 (0.865–5.796)	0.097
N stage
N0 vs. N1~N3	4.062 (2.061–8.007)	0.000	2.960 (1.250–7.005)	0.014
TNM stage
Ⅰ vs. Ⅱ–Ⅳ	2.844 (1.237–6.537)	0.014	1.532 (0.571–4.115)	0.397
Poor differentiation
Absent vs. mixed	2.054 (0.995–4.240)	0.052	0.942 (0.382–2.323)	0.896
Smoking index
< 1000 vs. ≥ 1000	1.226 (0.572–2.627)	0.601	-	-
Drinking
No and mild vs. constant	1.242 (0.633–2.436)	0.528	-	-
Preoperative chemotherapy
Absent vs. received	1.179 (0.584–2.383)	0.646	-	-
CD3⁺TILs
Low vs. high	0.602 (0.306–1.186)	0.142	0.907 (0.434–1.894)	0.794
CD8⁺TILs
Low vs. high	0.336 (0.170–0.661)	0.002	1.160 (0.446–3.019)	0.761
CD103⁺Cells
Low vs. high	0.313 (0.152–0.645)	0.002	0.253 (0.117–0.546)	0.000
Foxp3⁺TILs
Low vs. high	2.732 (1.330–5.610)	0.006	2.802 (1.226–6.399)	0.015

04SupplementaryTILinHNSCC.doc

Download PDF

Version 1

posted

You are reading this latest preprint version

Influencing Factors and Prognostic Significance of TIL Subset Abundance in Head and Neck Squamous Cell Carcinoma

Status:

Version 1

Abstract

Figures

1 Introduction

2 Patients And Methods

2.1 Patients

2.2 IHC

2.3 Statistical analysis

3 Results

3.1 Patients’ clinical and demographic characteristics

3.2 Staining and abundance of immunocyte subsets in HNSCC

3.3 Correlation between immunocyte subset abundance and clinicopathological features

3.4 Associations of immunocytes subset abundance with prognoses in HNSCC patients

4 Discussion

5. Conclusions

Abbreviations

Declarations

References

Tables

Supplementary Files

Status:

Version 1