The ratio of CD8+ lymphocytes to tumor-in ltrating suppressive FOXP3+ effector regulatory T cells is associated with treatment response in invasive breast cancer

Takayuki Kadoya (  takayukikadoya@gmail.com ) Hiroshima University Noriko Goda Hiroshima University Shinsuke Sasada Hiroshima University Hideo Shigematsu National Hospital Organization Kure Medical Center and Chugoku Cancer Center Norio Masumoto Hiroshima University Koji Arihiro Hiroshima University Hiroyoshi Nishikawa National Cancer Center Shimon Sakaguchi Osaka University Morihito Okada Hiroshima University


Introduction
Breast cancer is a common malignancy that leads to morbidity and mortality in women worldwide [1][2][3]. According to the Global Cancer Statistics report in 2021, approximately 2.2 million women were diagnosed with breast cancer with nearly 700,000 deaths [2].
FOXP3, detected by immunohistochemical staining, is a general Treg marker. However, FOXP3+ cells are functionally heterogeneous and can be classi ed into three fractions using ow cytometry based on the expression levels of FOXP3 and naïve T cell marker CD45RA: naïve Treg, effector Treg (eTreg), and non-Treg. Only eTreg cells have a suppressive function, and the other fractions are nonsuppressive and secrete in ammatory cytokines [10,30]. In colorectal cancer, it has been reported that the variation of tumor-in ltrating Treg component is caused by immunologically relevant genes and that these variations affect disease prognosis [31]. However, in ltration of only eTregs from total FOXP3+ cells cannot be detected using FOXP3 immunohistochemical staining. Moreover, the variation in eTreg in ltration has not been investigated in breast cancer, and the role of eTregs in the disease remains unclear. Herein, we examined the association between tumor-in ltrating eTregs and CD8+TILs and the clinical outcomes of patients with invasive breast cancer.

Patients and treatments
In total, 84 patients with early breast cancer who underwent complete resection between December 2015 and November 2016 were included. Patients with noninvasive diseases were excluded from the study. The N-acetylcysteine (NAC) regimen consisted of four cycles of docetaxel (75 mg/m2, every three weeks), followed by four cycles of FEC (500 mg/m2 5uorouracil, 100 mg/m2 epirubicin, and 500 mg/m2 cyclophosphamide; every three weeks). Patients with human epidermal growth factor receptor 2 (HER2)-positive breast cancer received trastuzumab (8 mg/m2 for the rst dose and 6 mg/m2 thereafter) every three weeks together with docetaxel. Pathological compete response (pCR) was de ned as the absence of invasive residuals in the primary lesion and axillary lymph nodes [32]. The protocol of this study was approved by the Ethics Committee of Hiroshima University and was conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all patients.

Breast cancer tissue collection and extraction of TILs
Fresh samples of invasive breast cancer tissues were collected via core needle biopsy, vacuum-assisted biopsy (Mammotome Elite; Mammotome, Cincinnati, OH, USA), or surgery. Biopsy specimens from patients who had received NAC were collected before treatment. Tumor samples were collected as follows: three to ve samples using 16-gauge biopsy needles, >6 samples using 13-gauge Mammotome needles, or an area of at least 10 mm × 10 mm × 2 mm shaved using a razor during surgery. Fresh TILs were extracted as described previously [31]. Fresh tissues were rapidly cut into small pieces using tissue scissors and homogenized using a GentleMACS dissociator (Miltenyi Biotech, Bergisch Gladbach, Germany), and the TILs were collected from the cell suspensions.

Statistical analyses
Basic statistics for the TIL subpopulations are expressed as the median and interquartile range (IQR). The Wilcoxon ranksum test was performed for multiple pairwise comparisons. Kaplan-Meier curve analysis was performed for assessment of disease-free survival with the log-rank test. Receiver operating characteristic curves were constructed to determine the cutoff values of parameters predicting pCR. Statistical signi cance was set at P < 0.05. All statistical analyses were performed using JMP Pro14 SAS software (SAS Institute Inc., Cary, NC, USA).   (Figure 1c). Among the CD4 + TILs, PBMCs, and LNBTs, the median percentage of total FOXP3 + cells was the highest, followed by that of eTregs, non-Tregs, and a low proportion of naïve Tregs; however, the proportion of eTreg cells was higher in the TILs compared with the PBMCs and LNBTs (Figure 1d).

Results
The median percentage of CD8 + TILs to CD4 + TILs was 124% (IQR, 87.5-140).  subtypes achieved higher pCR rates than those with the luminal subtype (P < 0.05). There was no signi cant difference in pCR rates between patients with high and low eTregs. Patients displaying a high CD8 + /eTreg ratio achieved signi cantly higher pCR than those displaying a low CD8 + /eTreg ratio (P = 0.001), whereas pCR rates did not differ signi cantly according to CD8 + /FOXP3 + ratio ( Figure 2). The factors predicting pCR were assessed based on some cut-offs (Table 3). LPBC and high-CD8 tumor were signi cant predictors for pCR (P = 0.006 and P = 0.020, respectively). After investigating the balance of CD8+ CTLs and Treg subpopulations, CD8/eTreg ratio was the most promising parameter for predicting pCR and the positive predictive value was 91.7%. In the multivariate analysis, both LPBC and high-CD8/eTreg ratio were independent predictors for pCR (odds ratio [OR] 10.1, P = 0.043 and OR 18.7, P = 0.034, respectively) ( Table 4).

Discussion
In this study, we focused on the immunological functional heterogeneity of TILs and evaluated the balance of immunoprogressive and immunosuppressive power of TILs in patients with breast cancer. To the best of our knowledge, this study is the rst to assess eTregs in breast cancer TILs and verify the association between the functional balance of TILs and clinical outcome of breast cancer. We demonstrated the functional heterogeneity of FOXP3+ Tregs using fresh TIL samples. Tregs are components of CD4+ T cells and essential effector cells that maintain immune homeostasis [9,10,37].
Although Tregs are considered to be homogeneously immunosuppressive TILs in breast cancer, we demonstrated that only eTregs had an immunosuppressive effect. We found that eTreg cells were more abundant in breast cancer tissues than in PBMCs and LNBTs, which is consistent with previous reports on other cancer types, such as colorectal [31]and gastric cancer [38]. The difference in the in ltration of Treg subpopulations indicates that T cells are activated and acquire functions in the tumor microenvironment. It is worth noting that patients with low levels of eTreg to total FOXP3+TILs can overestimate their immunosuppressive function by the FOXP3 immunohistochemical staining method. Thus, these patients may lead to con icting results in some studies that are based on immunohistochemical evaluation of FOXP3 for breast cancer prognosis. LPBC is considered as a strong biomarker for pathological response to NAC regardless of molecular subtypes [4,6] and survival in triple-negative and HER2+ subtype [5]. Our ndings indicate that the CD8+/eTreg ratio is a more promising predictor of pCR and progression-free survival compared to the CD8+/FOXP3+ ratio. We only showed the likelihood of better prognosis of the CD8+/FOXP3+ ratio, and the low numbers of patients may be attributable to a lack of power to show signi cant difference. Moreover, approximately 90% of patients were in the early stage. It has been reported that CD15s antibodies can speci cally identify eTreg cells [40]. The dual immunohistochemical staining of FOXP3 and CD15s may be a promising method for detecting eTreg in ltration with a large sample size.
Despite the ndings of this study, it has several limitations. First, the number of cases was small, and the follow-up period was short. Our ndings cannot elucidate the mechanism underlying increased eTregs in the tumor microenvironment.
Moreover, we did not evaluate the other TIL components. Recently, an exhaustive and promising evaluation of breast cancer TILs by single-cell RNA sequencing has been reported [41]. The genomic background or mechanism that causes these variations in TIL components will be elucidated in the near future. In addition, eTregs express immune checkpoint molecules, such as cytotoxic T-lymphocyte-associated protein 4 (CTLA-4) and programmed cell death-1 (PD-1), suggesting that controlling tumor-in ltrating Tregs may be a potential target for cancer immunotherapy [42][43][44][45]. Our ndings may be linked to other studies evaluating the response to immune checkpoint inhibitors in patients with breast cancer to build on their signi cance.
In summary, our study indicates that the functional heterogeneity of FOXP3+ TILs represents speci c variations in the immunological balance in invasive breast cancer. A high CD8/eTreg ratio enhances treatment response in invasive patients with breast cancer. Further studies are warranted to validate these ndings.

Conclusion
The CD8+/eTreg ratio is a simple and optimal marker for cancer immunity and an increase in this ratio suggests an enhanced prognosis and treatment response in patients with invasive breast cancer; however, further large-scale studies are warranted to validate the present ndings.