The Role of Lymphovascular Invasion as a Prognostic Factor in Patients with Lymph Node-Positive Operable Invasive Breast Cancer

One of the most important issues for treating patients with early stage breast cancer is determining those who could benefit from adjuvant chemotherapy and which chemotherapy regimen would provide the best results. Prognostic factors can be used to predict the natural history of breast cancer, so they are included in the decision to apply adjuvant systemic chemotherapy for patients with breast cancer. Currently, the presence of axillary lymph node involvement, tumor size, nuclear grade, hormone receptor status, and patient age are well-known prognostic factors for patients with operable invasive breast cancer [1]. These prognostic factors are widely used to determine whether to apply adjuvant chemotherapy in patients with breast cancer. Among the prognostic factors, axillary lymph node involvement is the most significant [1, 2]. Because the presence of lymph node involvement shows that the cancer has already developed the ability for distant metastasis [2-4]. Thus, the presence of axillary lymph node involvement predicts the choice of adjuvant chemotherapy and radiation therapy after surgery for primary breast cancer [5].

Lymphovascular invasion (LVI) is defined as tumor emboli present within a definite endothelial-lined space in the breast surrounding invasive carcinoma [6, 7]. The existence of LVI may help identify who is at increased risk for axillary lymph node and distant metastasis [2, 3] Although LVI has been accepted as an important prognostic factor in patients with lymph node-negative invasive breast cancer [8], LVI is controversial regarding prognostic value in patients with lymph node-positive breast cancer [3, 9-11]. The aim of this study was to analyze the prognostic significance of LVI in a large cohort study of patients with lymph node-positive invasive breast cancer.

The mean age of the patients at the time of surgery was 48 years (range, 23-78 years), and the median follow-up period was 48 months (range, 12-78 months). The number of distant metastases with or without locoregional recurrence was 61 (17.5%). The number of locoregional recurrences or distant metastases was 72 (20.9%). Thirty-six of 343 (10.3%) patients died because of breast cancer at the time of the study. Five-year OS, MFS, and DFS were 93%, 89.1%, and 87%, respectively.

LVI was present in 192 of 349 (55%) patients and was significantly associated with patient age ≤40 years (p=0.009), high histological grade (grades II and III, p=0.007), tumor diameter >20 mm (p=0.001), number of involved lymph nodes ≥4 (p=0.001), and negative ER (p=0.001) tumors. But it was not related to tumor multiplicity, PR status, HER2 receptor status, or p53 status (Table 1). Patients with negative LVI tumors were more frequently treated with adjuvant endocrine therapy (52.5% vs. 47.5, p=0.001). Surgical treatment and adjuvant chemotherapy were not significantly associated with the presence of LVI (Table 2).

In a univariate survival analysis, both 5-year OS and 5 year DFS were significantly different in patients with or without LVI: 88.8% vs. 94.1% (p=0.007) for OS (Figure 1A) and 76.4% vs. 90.9% (p<0.001) for DFS (Figure 1B). Furthermore, 5-year MFS was shorter in patients with LVI: 80.1% vs. 91.5% than those without LVI (p<0.001) (Figure 2).

Figure 1
Overall survival and disease-free survival of whole patients. (A) Overall survival curves are shown according to presence or absence of lymphovascular invasion (LVI). (B) Disease-free survival curves according to presence or absence of LVI are shown.

• Click for larger image

Figure 2
Metastasis-free survival of whole patients. Metastasis-free survival curves are shown according to presence or absence of lymphovascular invasion (LVI).

• Click for larger image

In the univariate analysis, presence of LVI, tumor size (>20 mm), high histological grade (grades II and III), number of involved lymph nodes (≥4), hormone receptor status (negative ER and PR), and HER2 receptor status (positive) were associated with poor DFS. But age, tumor multiplicity, and p53 status did not influence DFS (Table 3).

Table 3
Univariate analysis of disease-free survival and clinicopathological data

• Click for larger image

LVI was an independent prognostic factor for DFS in all patients (relative risk [RR], 2.59; 95% confidence interval [CI], 1.35-5.53; p=0.005). Furthermore, histological grade II (RR, 3.20; 95% CI, 1.00-10.65; p=0.049), histological grade III (RR, 4.12; 95% CI, 1.25-13.57; p=0.02), and a high number of involved lymph nodes (≥4; RR, 2.73; 95% CI, 1.35-5.53; p=0.003) were independent predictors in all patients. However, age, tumor size, hormone receptor status, and HER2 status were not significant independent factors (Table 4).

Table 4
Disease-free survival: multivariate analysis in the whole group of patients

• Click for larger image

In the hormone receptor-positive group (n=218), positive LVI (p=0.021), number of involved lymph nodes (p=0.037) and histological grade III (p=0.023) were also independent factors for a poor prognosis (Table 5).

Table 5
Disease-free survival: multivariate analysis in the hormone receptor-positive group of patients

• Click for larger image

Previous well-designed studies have analyzed the prognostic factors in patients with invasive breast cancer. Axillary lymph node involvement, younger age, high histological grade, large tumor size, and negative hormone receptor status were significantly associated with a poor prognosis (DFS and OS) [1]. Previous studies have also shown that LVI is an independent poor prognostic factor in patients with invasive breast cancer [2, 3, 14]. Nevertheless, although LVI is an important prognostic factor in patients with lymph node-negative invasive breast cancer, its prognostic value in patients with lymph node-positive breast cancer is unclear and controversial [3, 9-11]. So, we report the analysis of the prognostic significance of LVI in a large cohort study of lymph node-positive patients with invasive breast cancer.

In our retrospective study, LVI was observed in 55% (n=192) of all patients (n=349), which was similar to other series of lymph node-positive cancer. LVI added prognostic information for patients with lymph node-positive breast cancer with a RR of 2.59 and a p-value of 0.003. These results contrast with those reported by Colleoni et al. [10] This discrepancy may be explained by the short-term follow-up period and analysis limitations in patients with 1 to 3 metastatic lymph nodes in their study.

In addition to LVI, our study demonstrated that high histological grade, number of lymph nodes involved, and both factors constituting the Nottingham Prognostic Index were other significant independent clinicopathological factors. Because LVI appears to be independent from these pathological factors, our results propose that LVI should be considered in the prognostic index for patients with lymph node-positive breast cancer as well as in the group of patients with lymph node-negative breast cancer [15]. However, tumor size and patient age were not significant independent prognostic factors in our study. This result may be explained by the shorter follow-up period compared to that in previous studies.

We also assessed whether LVI had an effect on the hormone receptor-positive subgroup of patients as an independent prognostic factor. This analysis was restricted to the large subgroup of 218 patients who were hormone receptor-positive. In this subgroup, with a RR of 2.48 and a p-value of 0.021, LVI was an significant prognostic factor in patients who were hormone receptor-positive. These results suggest that LVI and other pathological factors may be useful to determine the need for adjuvant treatment in some patients for whom adjuvant chemotherapy could be dangerous despite axillary lymph node involvement. Eventually this may be applicable to the elderly or fragile patients (patients with severe underlying disease) with other treatment options such as endocrine and trastuzumab therapy.

Our analysis had three limitations. First, our analysis had a short follow-up period, and it was retrospective in nature. Second, not all patients with HER2-positive breast cancer were treated with a trastuzumab-based regimen. Third, LVI was assessed by H&E staining, and the selective endothelial cell markers such as D2-40, CD34 were not used in the routine pathological evaluation, but use of such markers could potentially have improved the accuracy of detecting LVI [16].

In conclusion, this study emphasized the role of pathologic-almetastasis analysis in the prognostic evaluation of breast cancer, particularly those factors that cannot be subjected to molecular analyses, such as size, lymph node status, histological grade, and LVI. These pathological factors are useful to evaluate prognostic and predictive factors for more effective and convenient clinical decision-making tools during adjuvant treatment.

The presence of LVI was an independent significant prognostic factor in patients with lymph node-positive breast cancer as well as patients with lymph node-negative breast cancer. The existence of LVI alone cannot be used to decide to omit adjuvant chemotherapy in all patients with lymph-node positive breast cancer but, it may be considered in the adjuvant treatment decision in a specific subgroup of patients for whom chemotherapy is contraindicated. We suggest that patients with LVI-positive breast cancer require a shorter follow up and additional management.