A new marker for predicting sentinel lymph node metastasis in early (cT1-2N0) breast cancer: Tumor-infiltrating lymphocytes (TILs)

Abstract

Background: Tumor-infiltrating lymphocytes (TILs) are associated with lymph node metastasis and prognosis in breast cancer. Therefore, we explored the value of TILs in predicting sentinel lymph node metastasis (SLNM) in patients with early-stage (cT1-2N0) breast cancer and provided a new method for preoperative assessment of SLNM status.

Methods: This study included 337 patients with early-stage breast cancer who underwent surgery at our hospital from January 2022 to December 2023. The expression of estrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor receptor 2 (HER2), and Ki-67 in the patients was assessed using immunohistochemistry (IHC). TILs in the core needle biopsy samples were evaluated histopathologically, and patients were divided into high and low TILs groups based on the density of TILs. Statistical analysis was conducted, and a predictive model was established.

Results: The study found that patients with high TILs had a significantly lower rate of SLNM compared to those with low TILs (P < 0.001). The cT stage and the level of TILs were identified as independent predictive factors for SLNM. The ROC curve analysis indicated that the density of TILs has good predictive efficacy for SLNM. Based on the results of the multivariate regression analysis, a nomogram predictive model for SLNM was constructed.

Conclusions: Our study showed that the density of TILs and cT stage are independent predictive factors for SLNM in early-stage (cT1-2N0) breast cancer, and the predictive effect of TILs density on SLNM is significant in Luminal and triple-negative breast cancers.

Fig 1. Immunohistochemical test images (×200).

A: ER+; B:ER-; C: PR+; D:PR-; E: HER2+; F: HER2-; G: Ki67 high; H: Ki67 low; I: AR+; J: AR-; K: P53+; L: P53-.

Fig 2. Histopathologic analysis for tumor-infiltrating lymphocyte (TIL) density was performed on a single full-face hematoxylin and eosin-stained tumor section.

A: Tumor-infiltrating lymphocyte density is less than 10% (TILs < 10%) (Tumour percentage is 85%, and the TILs score is 1%) (H&E ×  200). B: Tumor-infiltrating lymphocyte density is greater than or equal to 10% and less than 50% (10% ≤ TILs < 50%) (Tumour percentage is 60%, and TILs score is 30%) (H&E ×  200). C: Tumor-infiltrating lymphocyte density is greater than or equal to 50% (TILs ≥ 50%) (Tumour percentage is 35%, and TILs score is 70%) (H&E ×  200).

Fig 3. Dividing the density of Tumor-Infiltrating Lymphocytes (TILs) at a threshold of 50% to categorize into lymphocyte predominant breast cancer (LPBC) and non-lymphocyte predominant breast cancer (nLPBC), patients with Triple-Negative Breast Cancer (TNBC) have a higher proportion of LPBC, followed by those with HER2-enriched breast cancer.

Patients with Luminal A (LA) and Luminal B (LB) breast cancer have the smallest proportion of LPBC (P <  0.001) (A). When dividing the density of Tumor-Infiltrating Lymphocytes (TILs) with a cutoff of 10% into high TILs and low TILs, there is no significant difference in TILs among Luminal A (LA), Luminal B (LB), HER2-overexpressing, and Triple-Negative (TN) breast cancer types (P = 0.902) (B).

Fig 4. Based on the multivariate analysis, a forest plot was constructed, showing that clinical T stage (P = 0.002, OR = 0.464) and the level of Tumor-Infiltrating Lymphocytes (TILs) (P < 0.001, OR = 4.549) are independent predictive factors for Sentinel Lymph Node Metastasis (SLNM).

Fig 5. Using t-tests and box plots to analyze the correlation between TILs (Tumor-Infiltrating Lymphocytes) density and SLNM (Sentinel Lymph Node Metastasis), it was found that in all patients, those with SLNM had a significantly lower TILs density than those without SLN metastasis (P = 0.001).

(A). When analyzing each subtype, only Luminal B (LB) breast cancer (P = 0.014) and Triple-Negative Breast Cancer (TNBC) (P < 0.001) showed statistical differences (C, E), while other subtypes did not exhibit significant differences (B, D).

Fig 6. Using the ROC (Receiver Operating Characteristic) curve to evaluate the value of TILs (Tumor-Infiltrating Lymphocytes) density in predicting Sentinel Lymph Node Metastasis (SLNM) in early-stage (cT1-2N0) breast cancer, the AUC (Area Under the Curve) was 0.624 (CI: 0.559-0.689), with a sensitivity of 0.440 and a specificity of 0.783.

The optimal cutoff value was 17.5%, indicating good predictive performance.

Fig 7. Based on the results of multivariate regression analysis, a nomogram model for Sentinel Lymph Node Metastasis (SLNM) was constructed using the R programming language.