Patients and treatments
In total, 84 early breast cancer patients who underwent complete resection between December 2015 and November 2016 were included. Patients with non-invasive breast cancer were excluded. A neoadjuvant chemotherapy (NAC) regimen comprised docetaxel (75 mg/m2, every 3 weeks) for four cycles, followed by FEC (5-fluorouracil, 500 mg/m2; epirubicin, 100 mg/m2; cyclophosphamide, 500 mg/m2, every 3 weeks) for four cycles. Patients with human epidermal growth factor receptor 2 (HER2)-positive breast cancer received trastuzumab (8 mg/m2 as the first dose and 6 mg/m2 thereafter) every 3 weeks together with docetaxel. Pathological complete response (pCR) was defined as the absence of invasive residual tumors in the primary lesion and axillary lymph nodes.30 This study was approved by the Ethics Committee of Hiroshima University and 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 invasive breast cancer tissue specimens were harvested via core needle biopsy, vacuum-assisted biopsy (Mammotome elite, Cincinnati, OH, USA), or surgery. Biopsy specimens of patients receiving NAC were harvested before treatment. Tumor specimens were harvested as follows: 3 to 5 samples with 16-gauge biopsy needles, >6 samples with 13-gauge Mammotome needles, or an area of at least 10 mm × 10 mm × 2 mm shaved with a razor during surgery. Fresh TILs were extracted using a previously described protocol.29 Fresh tissues were rapidly diced using tissue scissors and homogenized using a GentleMACS dissociator (Miltenyi Biotech, Bergisch Gladbach, Germany). TILs were harvested from the cell suspension.
Flow cytometry
Fresh TILs were washed with phosphate-buffered saline containing 2% fetal calf serum and APC-conjugated anti-CD4 mAb (BD Biosciences, Franklin Lakes, NJ, USA), V500-conjugated anti-CD8 mAb (BD Biosciences), FITC-conjugated anti-CD45RA mAb (BD Biosciences), and Fixable Viability Dye (eBioscience, San Diego, CA, USA). Intracellular FOXP3 was stained using an anti-FOXP3 mAb and FOXP3 Staining Buffer Set (eBioscience) in accordance with the manufacturer’s instructions. Cells were analyzed using LSRFortessa (BD Biosciences) and FlowJo software (Tree Star, Ashland, OR, USA).
Determination of the TIL component
TILs were categorized in accordance with previously described protocols.8,28 TILs were gated into CD4−CD8+ T cells and CD4+CD8− T cells, and the CD4+CD8− T cell fraction was further gated on the basis of FOXP3 and CD45RA expression as follows: naïve Tregs (FOXP3lowCD45RA+), eTregs (FOXP3highCD45RA−), and non-Tregs (FOXP3lowCD45RA−). The ratio of each TIL subpopulation to the total CD4+CD8− TIL subset was determined. Moreover, lymphocytes were harvested from PBMCs and lymphocytes from normal breast tissue (LNBT).
Treg suppression assay
Naïve Tregs, eTregs, and non-Tregs were isolated from TILs, LNBTs, and PBMCs, using a FACS Aria II system (BD Biosciences). Responder CD25−CD4+ T cells (control) were purified from PBMCs, labeled with carboxyfluorescein succinimidyl ester (CFSE), and mixed with purified individual components at a ratio of 1:3 (responder T cell:individual components). Treg Suppression Inspector reagent (Miltenyi Biotec) was added in accordance with the manufacturer’s instructions. Cells were cultured for 5 d, and the proliferation of CFSE-labeled cells was evaluated.
Pathological assessment and evaluation of stromal TILs
Histological characteristics including nuclear grade, estrogen receptor (ER) and HER2 status, Ki-67 labeling index, and stromal TILs, were assessed by two pathologists indipendently. ER and HER2 status was assessed in accordance with the American Society of Clinical Oncology/College of American Pathologists Guidelines33,33. Breast cancer subtypes were defined as ER(+), HER2(+), and triple-negative (TN; ER(-) and HER2(-)). The Ki-67 labeling index was scored as high (≥20%) and low (<20%). Stromal TILs were assessed via hematoxylin-eosin (HE)-stained slides of maximum tumor lesions, using the method of the International TILs Working Group 201434. Lymphocyte predominant breast cancer (LPBC) was defined as stromal TILs ≥60%.
Statistical analyses
Basic statistics for TIL subpopulations were expressed as the median and the interquartile range (IQR). The Wilcoxon rank sum test was performed for multiple pairwise comparisons. Kaplan-Meier curve analysis was performed for disease-free survival with the rog-rank test. Receiver operating characteristic curves were constructed to determine the cutoffs of parameters predicting pCR. Statistical significance was set at P < 0.05. All statistical analyses were performed using JMP Pro14 SAS software (SAS Institute Inc., Cary, NC, USA).