Correlations of neutrophil-to-lymphocyte, lymphocyte-to-monocyte and platelet-to-lymphocyte ratios with biomarkers of atherosclerosis risk and inflammatory response in patients with a history of breast cancer

Introduction

Predictive and prognostic biomarkers play an increasingly important role in the management of cancer patients [1]. The significance of inflammatory reaction that reflects the host response to neoplastic growth is being increasingly recognized [2]. A number of biomarkers of the inflammatory response have been introduced into the management of cancer patients over the last decades, including, for example, C-reactive protein (CRP) and neopterin. These biomarkers are increased in a wide range of disorders. For example, increased serum or urinary neopterin concentrations have been reported in disorders as diverse as cancer, viral infections, organ rejection or atherosclerosis and its complications [3–6].

In the last few years, considerable research has been focused on simple biomarkers obtained by calculating the ratios of lymphocytes to other cellular components of peripheral blood [7]. The neutrophil-to-lymphocyte ratio (NLR), lymphocyte-to-monocyte ratio (LMR) and platelet-to-lymphocyte ratio (PLR) have been demonstrated to represent independent prognostic biomarkers across a spectrum of solid tumors [8–12]. Because peripheral blood cell count (PBC) is determined repeatedly in virtually all cancer patients, a number of retrospective analyses on large cohorts were relatively easy to be performed using retrospective patient databases.

Similarly to some cancers of other primary location, breast cancer is currently characterized by a high cure rate. As a consequence of improved prognosis, competitive causes of mortality are of increasing importance. The complications of atherosclerosis currently represent the most important comorbidities and also the most common cause of non-cancer-related deaths. Both cancer and atherosclerosis are associated with inflammatory response. In fact, PBC-derived ratios have also been shown to be associated with biomarkers of atherosclerosis or its complications and to predict the risk of atherosclerosis-related events [13–19]. In contrast, more limited information is available on the correlation between PBC-derived ratios and other biomarkers of inflammatory response [20, 21].

In prior studies, we have investigated the association of intima-media thickness (IMT), a biomarker of atherosclerosis, with other biomarkers of atherosclerosis risk and inflammatory response in two independent cohorts of patients with a history of breast cancer [22, 23]. The aim of the present study was to retrospectively evaluate the correlations between PBC-derived ratios and these biomarkers.Although differential blood cell counts obtained by manual and automated methods could give different results [24], in the large cohorts of patients examined in the aforementioned studies, limited details regarding the method used were reported. Therefore, another aim of the present study was to compare PBC-derived ratios obtained by manual and automated counts.

Patients and methods

In the present study, NLR, LMR and PLR were calculated in cohorts of patients with a history of breast cancer (patients) and in control groups (controls). These cohorts of patients and controls in three studies as well as the methods used to determine laboratory biomarkers, body mass index (BMI) and IMT have been described in detail in prior reports [22, 23, 25, 26]. The investigations were performed at two centers: Palacký University Medical School and Teaching Hospital, Olomouc, Czech Republic (cohort A), and Charles University Medical School and Teaching Hospital, Hradec Králové, Czech Republic (cohort B). Because, with the exception of urinary neopterin, other biomarkers were assessed in different laboratories, the two cohorts were investigated separately. In total, PBC-derived ratios could be calculated in all 61 patients and 74 controls in cohort A and in 190 (99%) of 192 patients and in all 17 controls in cohort B. PBC was obtained by both manual and automated counts in 54 patients and 26 controls in cohort A, whereas only manual counts were available in cohort B. Survival information was updated in 71 patients with breast cancer treated at Charles University Medical School and Teaching Hospital who had complete data on automated peripheral blood cell count and urinary neopterin (cohort C) [24].

Differences between patients and control groups were analyzed by the Mann-Whitney U-test. Differences between paired measurements were evaluated using the Wilcoxon signed rank test. Correlations were analyzed using Spearman’s rank correlation coefficient (R_s). Survival was assessed using the Kaplan-Meier method. The cohort of patients was dichotomized according to the values of urinary neopterin, NLR, LMR and PLR, with cutoffs selected based on prior reports [8, 9, 25, 27]. The decision on statistical significance was based on a p-value of 0.05. The analyses were performed using the NCSS software (Number Cruncher Statistical Systems, Kaysville, UT, USA).

Results

Significant differences were observed between LMR obtained from automated and manual counts in patients, and between LMR and PLR obtained from automated and manual counts in controls (Table 1). In cohort A, LMR obtained from manual counts was significantly lower and PLR obtained from manual counts was significantly higher in patients than in controls, whereas when the values obtained from automated counts were evaluated NLR and PLR were significantly higher and LMR was significantly lower in patients than in controls. No significant difference was observed in NLR, LMR and PLR between the patients with a history of breast cancer and controls in cohort B (Table 2).

The correlations of PBC-derived ratios and biomarkers of atherosclerosis and/or inflammatory response in controls and patients with a history of breast cancer in cohorts A and B are shown in Tables 3–6. NLR, LMR and PLR obtained from manual and automated counts exhibited correlations in patients and controls in cohort A, with the exception of LMR in controls, but these correlations were not as strong as would be expected (Tables 3 and 4). NLR and PLR correlated positively with each other and negatively with LMR. In the control groups, occasional correlations, including the negative correlation of NLR with total cholesterol, lipoprotein(a) and IMT; the positive correlation of LMR with serum neopterin, urinary neopterin, triglycerides and IMT; the negative correlation of LMR with urinary albumin/creatinine ratio; and the negative correlation between PLR and fibrinogen, triglycerides and glucose, were noted in one of the cohorts (Tables 3 and 5). Consistently in both patient cohort A and patient cohort B, NLR exhibited a significant correlation with age, glucose and CRP concentrations, whereas LMR correlated negatively with CRP. Other correlations, including the positive correlation of NLR with BMI, uric acid, triglycerides, urinary albumin/creatinine ratio, fibrinogen, homocysteine and IMT; the negative correlation of NLR with total cholesterol, LDL cholesterol and albumin; the positive correlation of LMR with antithrombin and magnesium; the negative correlation of LMR with age, BMI and fibrinogen; the positive correlation of PLR with CRP and fibrinogen; and the negative correlation of PLR with hemoglobin and uric acid, were mostly weak and were present only in one of the patient cohorts (Tables 4 and 6).

In another cohort of 71 patients with breast cancer (cohort C), NLR and PLR correlated positively with each other and negatively with LMR, but no other correlations were observed (Table 7). Thirty patients in this cohort died and 41 were alive at the time of analysis, with the median follow-up of surviving patients in this cohort being 121 months. NLR≥3 was a significant predictor of poor survival, but neither urinary neopterin ≥205 μmol/mol creatinine, NLR≥150 nor LMR≥4.25 was significantly associated with survival (Table 8).

Discussion

Present data indicate that NLR and PLR could be increased and LMR decreased in patients with breast cancer. Moreover, increased NLR was associated with inferior survival. However, present data also indicate that the values of PBC-derived ratios may significantly differ based on the method used to obtained differential blood cell count.

Although numerous reports have been published on the use of PBC-derived ratios in patients with malignant tumors as well as with benign disorders, most of these studies have been retrospective and the details on the method used to obtain blood cell counts have been mostly scanty [8–19]. In prior studies comparing automated and manual peripheral blood cell count, a strong correlation was observed [24], but the correlation between PBC-derived ratios obtained by both methods in the present study was markedly weaker. These differences have obvious implications for the interpretation of the results. The method used to obtain PBC-derived ratios should always be reported, and the values obtained using one method may not be comparable to the results determined using a different methodology.

Urinary or serum neopterin concentrations and serum CRP are biomarkers associated with different primary tumors as well as with atherosclerosis. High neopterin concentrations have been reported in a wide range of disorders [4, 5], and increased urinary neopterin concentrations in cancer patient were associated with poor prognosis [6]. Alterations of PBC-derived ratios have been associated with a similar range of disorders [8–19]. A correlation between NLR and CRP has been reported in prior studies in patients with cancer [21] or complications of atherosclerosis [20].

We therefore expected that urinary and/or serum neopterin concentrations will exhibit a correlation with PBC-derived ratios. In agreement with earlier reports [20, 21], a correlation between NLR and CRP was observed in patients with a history of breast cancer, but with the exception of a correlation between LMR and urinary or serum neopterin concentrations in controls, no other correlation between neopterin concentrations and PBC-derived ratios was noted despite the fact that relatively large independent cohorts of patients and controls were studied. It can therefore be concluded that neopterin and PBC-derived ratios reflect different facets of the inflammatory response. This opens the way for combining neopterin and PBC-derived ratios in complex indices of inflammatory activity. In a subsequent retrospective analysis of the survival of a cohort of patients with breast cancer with a long follow-up, NLR, but not urinary neopterin or other PBC-derived ratios, was a significant predictor of survival.

A number of correlations were observed between PBC-derived ratios and biomarkers of atherosclerosis risk. These associations were more pronounced in patients with a history of breast cancer than in controls. In fact, some of these associations were in opposite direction in cancer patients and the control group. Improved prognosis of patients with breast cancer is the result not only of early diagnosis, but also of improved systemic therapy. It is, however, being increasingly recognized that systemic therapy may cause acceleration of atherosclerosis. Moreover, increased risk of atherosclerosis may also be linked to the systemic inflammatory response. Cardiovascular morbidity is an issue of great importance in breast cancer survivors. Present data support the hypothesis that inflammatory response in patients with breast cancer could be associated with the progression of atherosclerosis. Thus, PBC-derived ratios in patients with a history of breast cancer may be not only a prognostic biomarker, but also a biomarker of atherosclerosis risk.

Only NLR, but not urinary neopterin, LMR or PLR, was a prognostic biomarker in the present study. However, the number of patients and events was limited and these parameters should be investigated in a larger cohort of patients. The present study has several other limitations. The present analysis on rather heterogeneous cohorts of patients and controls was retrospective and should obviously be regarded as hypothesis generating. Similarly to CRP or neopterin, PBC-derived ratios reflect systemic immune and inflammatory phenomena. However, the outcome of the host-tumor interaction is dependent on the phenomena taking place in the microenvironment, which may be difficult to assess [28, 29]. The correlations observed could reflect the effect of other confounding factors. Because of the exploratory nature of the investigations, the Bonferroni correction for multiple comparisons was not performed as this would decrease the statistical power. As a consequence, some correlations observed could be spurious.

Prior studies have reported only moderately increased neopterin concentrations, with high neopterin levels present only in a minority of patients with breast cancer [30–33]. However, similarly to tumors of other primary locations, increased neopterin concentrations have been shown to be associated with poor prognosis [25, 31]. Increased urinary neopterin concentrations have been reported to correlate with higher grade or the presence of metastatic disease, and elevated neopterin level was a significant prognostic biomarker in both univariate and multivariate analyses [25, 31]. In other studies, significantly higher serum neopterin concentrations accompanied by increased concentrations of other biomarkers of inflammatory response have been reported in breast cancer survivors with chronic fatigue compared to patients without chronic fatigue [34–36], although these findings could not be reproduced in a subsequent study [37]. Future studies should investigate the prognostic significance of neopterin concentrations along with PBC-derived ratios in a larger cohort of patients.

In conclusion, NLR correlated with CRP and other biomarkers of atherosclerosis risk in patients with a history of breast cancer, but no consistent correlation was observed between urinary and serum neopterin concentrations and any of the PBC-derived ratios. In a cohort of breast cancer patients, a higher NLR predicted poor survival.