Usefulness of neutrophil-lymphocyte ratio and platelet-lymphocyte ratio as a predictor of disease-free survival in breast cancer: A cross-sectional study

Introduction

Worldwide, breast cancer is the most diagnosed cancer type among women¹. Animal models suggest “immune-editing” in which activation of immune mechanisms control the tumor, but over time lead to the selection of tumor cells that escape the immune pressure and grow progressively². Most tumor antigens identify as non-mutated self-antigens. Tumors are heterogeneous, and the antigens on cells of one tumor are variable, even within the same patient, so the down-regulation of major histocompatibility complex molecules and other components of the antigen-presentation process can occur¹. Tumors also do not express the ligands recognized by innate immune cells that microbes express or the co-stimulatory ligands necessary to stimulate adaptive T cells². The expression of Fas ligand by some tumor cells help to maintain a state of immune privilege that induce apoptosis. Tumor cells lead to the release of many cytokines and soluble factors, such as prostaglandin E2, that are not conducive to antitumor immunity. Cancer-associated factors have been shown to inhibit the production and stimulatory capacity of tumor cells³. T-helper cell responses skewed toward a Th2 phenotype, lead to inhibition of the Th1 response and cellular immunity that mediates tumor rejection⁴.

The evidence of the relationship between inflammation and cancers prognosis has increased in past years, especially gastrointestinal tumors and non-small-cell lung cancers, and even breast cancers. The systemic inflammatory responses may mimic biochemical or hematological markers, such as raised C-reactive protein and the elevation of white blood cells, neutrophils, and platelets. Elevation of neutrophil-lymphocyte ratio (NLR) and platelet-lymphocyte ratio (PLR) in breast cancer patients is not well studied in Iraq. This relationship is a complex and multifactorial process that is still poorly understood, but a high NLR may reflect systemic inflammation in enhancing angiogenesis, tumor growth and development of metastasis. Therefore, this study aimed to estimate the relationship between NLR, PLR and disease-free survival (DFS) in breast cancer patients in Iraq.

Methods

Results

Out of a total of 1167 case files only 102 patients fit the eligibility criteria and completed the study, with a mean age of 50.4 ± 11.7 years (range, 23–75 years, the rest of the disease characteristic illustrated in (Table 1). Local recurrence had the lowest rate, while combined local and distant recurrence had the highest rate (Table 1). The median DFS was 62 months with 5 years DFS in 52.5% patients, patients with stage N0 had a significantly higher DFS compared to stage N1, and those patients with a positive hormonal status have a significantly better DFS compared to those with a negative hormonal status (Table 2; Figures 1A–D). NLR had the highest discrimination ability to predict recurrence (since AUC between 0.7 – 0.79), while the rest of the variables show poor discrimination ability, as illustrated in Table 3 and Table 4 and Figures 1E and F. NLR fair specificity (76.9%) with lower sensitivity (62.2%), with optimal cut point of >2.194 to predict all recurrence. NLR also showed similar predictability for distant metastasis, while for local recurrence NLR had poor ability to predict local recurrence (Table 5 and Table 6; Figure 1G and Figure 2). Table 7 shows uni- and multivariate analysis of predictors of DFS with a significant p- value (p=0.007) for NLR, which means that it is an independent risk factor (Figure 3).

Figure 1. Kaplan Meier curves for disease free survival (DFS).

A: DFS for all patients, B: median DFS (MDFS) by T staging, C: MDFS by N staging, D: MDFS by hormonal status, E: NLR MDFS by months, F: PLR MDFS by months, G: Kaplan–Meier estimator of DFS using NLR cut point (2.194).

Figure 2. ROC curve of neutrophil-lymphocyte ratio (NLR) to predict recurrence.

Figure 3. Disease-free survival according to neutrophil-lymphocyte ratio (NLR; cut off value: 2.194) after performing multivariate analysis.

Discussion

This is the first study to show an association between high NLR and poor prognosis in Iraqi breast cancer patients. A higher NLR independently reflected a higher risk of local recurrence and distant metastasis in women with early breast cancer with a cutoff point of 2.194.

We found significant differences in DFS (16 months) according to our NLR cut off point with significant p-value (p=0.004). The role of lymphocytes in cancer is exemplified by the strong association between high densities of tumor-infiltrating lymphocytes and better responses to both cytotoxic treatments and outcome in patients^5,6. Two meta-analysis studies have confirmed the association between elevated NLR and poor prognosis for breast cancer^7,8; however, studies about these values are rare and not done in Iraq. In this study, we validated the usefulness of high NLR in early stage breast cancer up to stage IIA and to estimate DFS for at least 36 months follow-up.

In our univariate and multivariate analysis, we found hazard ratio (HR) for NLR 2.5 with significant p-value (p=0.007), while in the univariate analysis the nodal status and hormonal status were significant as dependent prognostic factors. Comparing our result with Ethier et al., a meta-analysis which comprised patients with reported HRs for DFS, and included only non-metastatic cases⁸, our result has the same significance with good sample size and follow-up period; however, we couldn't calculate overall survival (OS). Although another study shown that PLR was not related with DFS or OS in women⁹, in our study PLR was neither sensitive nor specific with non-significant p-value, so it will not considered as a prognostic index.

Conclusion

The role of NLR is a prognostic marker and elevated NLR is correlated with poor DFS in early breast cancer patients.

Data availability