A prospective comparative clinical study of peripheral blood counts and indices in patients with primary brain tumors.

BACKGROUND: Elevation of the neutrophil to lymphocyte ratio (NLR) has been shown to be an indicator of poor prognosis in many malignancies including recurrent glioblastoma multiforme.
OBJECTIVES: This study was aimed at assessing if the NLR and other leukocyte counts and indices were deranged in treatment-naïve patients with primary brain tumors when compared with an age-matched healthy control group.
MATERIALS AND METHODS: This was a prospective comparative clinical observational study by design. A healthy control population was compared with treatment-naïve patients diagnosed with intra- and extraaxial brain tumors. Leukocyte counts (neutrophil, lymphocyte, monocyte, eosinophil, and basophil counts) as well as leukocyte ratios such as the NLR and the monocyte to lymphocyte ratio (MLR) were calculated. We also evaluated if the counts and indices were related to the tumor volume.
RESULTS: In all patients with tumors, the platelet and neutrophil counts were elevated when compared to the controls. In contrast, monocyte counts and the MLR were found to be decreased in patients with tumors when compared to the controls. The subset of patients with glioblastoma showed a significant increase in NLR when compared to the controls.
CONCLUSIONS: Significant changes in the neutrophil, monocyte, and platelet counts as well as NLR and MLR were observed. Prospective longitudinal studies are required to determine the prognostic and therapeutic implications of these findings.

Introduction

Brain tumors are classified based on their relation to the neuraxis — Those arising within the substance of the brain or spinal cord are termed intraaxial lesions, whereas those that arise outside are termed extraaxial. Several studies have established the occurrence of a systemic inflammatory response to brain tumors, similar to that seen with other malignancies.[12] The degree and type of inflammatory response would obviously depend on the location, antigenicity, and vascularity of the lesion — both in the brain and elsewhere in the body. Thus, it may reasonably be expected that the degree and direction of derangement of the systemic markers of inflammation vary between malignancies, based both on anatomic location and biologic aggressiveness of the neoplasm. The levels of several of these markers of inflammation (eg. C-reactive protein) appear to correlate with the prognosis.[3]

The neutrophil to lymphocyte ratio (NLR) is the ratio of the count (or percentage) of neutrophils to lymphocytes in peripheral blood and is a systemic marker of inflammation.[3] The NLR has been shown to have prognostic significance in pancreatic, liver, prostate, and bladder cancers, and more recently, in glioblastoma multiforme (GBM) as well.[345] An NLR>4 has been found to connote poor prognosis in patients with GBM.[5] However, it has not been clearly established whether the NLR and/or other leukocyte counts and indices are deranged in all patients with malignant intraaxial brain tumors and if so, the magnitude and direction of such derangement. Moreover, no data have been published documenting the derangement of peripheral counts and ratios in patients with low grade [World Health Organization (WHO) grade 1 and 2] intraaxial lesions and benign extraaxial lesions. In this study, peripheral blood counts and some leukocyte indices in a control population were compared with those of patients with low and high grade intraaxial lesions and benign extraaxial brain tumors.

Materials and Methods

This institutional review board (IRB)-approved study (JIP/IEC/SC/2014/10/706) was conducted prospectively over a period of 6 months (October 2014 to March 2015). The tumor group included treatment-naïve patients with primary brain tumors at the time of initial diagnosis. This group included patients with low grade and high grade intraaxial tumors and benign extraaxial tumors. Patients who were on steroids, those suffering from any endocrinopathy, and individuals with suspected or diagnosed infections or autoimmune disease were excluded. Control data were obtained from the hematology database. Age-matched controls were randomly selected from the hemogram data for those healthy individuals who had undergone tests for either pre-employment screening or as part of other routine check-up protocols. For all subjects who had undergone a hemogram study, 2 mL of blood was drawn in an ethylene diamine tetra-acetic acid tube and a complete blood count was obtained using the Sysmex 2000i hematology analyzer (Sysmex, Japan, Sysmex India Pvt Ltd. distributes them in India). The parameters measured included hemoglobin, total leukocyte count (TLC), differential count (including neutrophil, lymphocyte, basophil, eosinophil, and monocyte counts), and platelet count. Two ratios were calculated from the differential count — The NLR and the monocyte to lymphocyte ratio (MLR). In the tumor group, the histopathologic grade of the tumors was noted where possible. The linear dimensions of each tumor were measured in three orthogonal planes (axial, sagittal, and coronal) on Gadolinium contrast-enhanced T1 sequences of magnetic resonance (MR) images and the volume of the tumor was calculated using the formula v = (abc)/2. The data were analyzed using the Statistical Package for the Social Sciences (SPSS) (version 17, IBM, New York, USA) and graphs were plotted on MS Excel 2013.

Results

The control group had data pertaining to 216 healthy individuals, out of whom 37 were females. The mean age of individuals in this group was 25.95 (±7.5) years. The mean NLR for the control group was 1.97 ± 0.84 and the mean MLR was 0.18 ± 0.08. No mathematically definable relation was noted between age and the NLR or age and the MLR [Figure 1a and b]. An inverse linear relationship was noted between the neutrophil and lymphocyte counts and this was maintained across the tumor groups as well. No definite relationship was found between the monocyte and lymphocyte counts [Figures 1c and d] There was no significant difference in the mean NLR between men (2 ± 1.03) and women (1.97 ± 0.87, t = 0.084, P = 0.933) or in the mean MLR between men (0.17 ± 0.07) and women (0.18 ± 1.12) (t = -0.43, P = 0.665). A total of 115 patients were recruited to the tumor group — 70 with intraaxial tumors and 45 with extraaxial lesions.

Figure 1

The control group. (a) This graph plotted between age and the NLR shows an absence of any relationship between these two variables (b) This panel shows the plot of age versus MLR. There is no clear relation between the two (c) Plot of neutrophil count versus lymphocyte count. A definite negative linear relationship is noted (d) There was no definite relation between the lymphocyte and monocyte counts

Intraaxial tumors

All patients in the intraaxial lesion group had supratentorial gliomas — 36 of these patients were eventually found to have GBM (WHO grade 4) and 9 had WHO grade 3 lesions. Sixteen were low grade gliomas (WHO grades 1 and 2) and the tumor grade was not known for nine patients. The male to female distribution was 1.33: 1 and the mean age was 32.89(±16.3) years. The hematology values for this group are displayed in Table 1. There is a relative reduction in the hemoglobin level as well as a relative increase in platelet counts in the tumor group when compared to the control group. In the tumor group the neutrophil and eosinophil counts were elevated, whereas the monocyte count was reduced. There was no significant difference in the lymphocyte count between the tumor and control groups. The MLR was significantly reduced in the tumor group compared to the control group (P < 0.001). The NLR in the tumor group was elevated (2.44 ± 1.9) compared to the control group (1.97 ± 0.8); this elevation showed a trend toward statistical significance (t = -0.23, P = 0.051).

Table 1

Blood parameters — Intraaxials versus controls

Parameter	Group	N	Mean	Standard deviation	P value on t-test
Hb	Controls	216	14.53	1.59	0.005
	Intraaxial tumors	68	13.681	2.2649
RBC	Controls	124	5.48	4.85	0.296
	Intraaxial tumors	60	4.82	0.59
TLC	Controls	216	7794.63	1774.60	0.185
	Intraaxial tumors	70	8315.71	3099.53
N	Controls	216	57.60	9.05	0.021
	Intraaxial tumors	70	61.31	12.25
L	Controls	216	32.46	8.21	0.595
	Intraaxial tumors	70	31.83	9.99
B	Controls	216	0.32	0.35	0.742
	Intraaxial tumors	70	0.39	1.61
E	Controls	214	4.25	3.41	0.029
	Intraaxial tumors	70	5.74	5.30
M	Controls	216	5.39	2.10	<0.001
	Intraaxial tumors	68	0.79	1.18
Platelets	Controls	215	223595.35	75330.480	<0.001
	Intraaxial tumors	70	276028.57	67531.63
NLR	Controls	216	1.97	0.84	0.051
	Intraaxial tumors	70	2.44	1.92
MLR	Controls	216	0.18	0.08	<0.001
	Intraaxial tumors	70	0.03	0.06

Hb = Hemoglobin; RBC = Red blood corpuscles; TLC = Total leukocyte count; N = Neutrophils; L = Lymphocytes; M = Monocytes; E = Eosinophils; B = Basophils; NLR = Neutrophil to lymphocyte ratio; MLR = Monocyte to lymphocyte ratio

We then analyzed the data for the 36 patients with GBM separately. These data are presented in Table 2. The same overall trends were seen as with the entire intraaxial group, with one notable difference. The NLR in the GBM group (2.67 ± 2.2) was significantly elevated when compared to the control group (1.97 ± 0.84, t = -0.97, P = 0.028).

Table 2

Blood parameters — GBM versus controls

Parameter	Group	N	Mean	Standard deviation	P value on t-test
Hb	Controls	216	14.53	1.59	0.002
	GBM	36	13.68	1.73
RBC	Controls	124	5.48	4.85	0.05
	GBM	36	4.87	0.41
TLC	Controls	216	7794.63	1774.597	0.15
	GBM	36	8623.61	3337.77
N	Controls	216	57.60	9.04842	0.01
	GBM	36	63.36	12.34
L	Controls	216	32.46	8.21	0.41
	GBM	36	30.94	10.53
B	Controls	216	0.32	0.35	0.74
	GBM	36	0.42	1.78
E	Controls	214	4.25	3.41	0.32
	GBM	36	5.03	4.49
M	Controls	216	5.39	2.11	<0.001
	GBM	36	0.54	1.01
Platelets	Controls	215	223595.35	75330.48	<0.001
	GBM	36	274916.67	60654.23
NLR	Controls	216	1.97	0.84	0.028
	GBM	36	2.67	2.20
MLR	Controls	216	0.18	0.08	<0.001
	GBM	36	0.02	0.04

Hb = Hemoglobin; RBC = Red blood corpuscles; TLC = Total leukocyte count; N = Neutrophils; L = Lymphocytes; M = Monocytes; E = Eosinophils; B = Basophils; NLR = Neutrophil to lymphocyte ratio; MLR = Monocyte to lymphocyte ratio; GBM = Glioblastoma multiforme

Extraaxial tumors

In the extraaxial group, 27 patients had vestibular schwannomas and the remaining 18 had supratentorial meningiomas. All lesions in this group were WHO Grade 1. The male to female distribution was 1: 2.29; the mean age was 46.18 (±9.65) years. The hematology values for this group are displayed in Table 3. In the extraaxial group as well, the platelet and neutrophil counts were significantly elevated when compared to the controls. In contrast, the monocyte count and the MLR were found to be significantly decreased (P < 0.01). There was no significant difference in the NLR between the controls and the extraaxial tumor group. Furthermore, the proportion of basophils was found to be significantly decreased in the extraaxial tumor group when compared to the control group (P < 0.001). No significant differences in the values of these counts/indices were noted between schwannomas and meningiomas.

Table 3

Blood parameters — Extraaxials versus controls

Parameter	Group	N	Mean	Standard deviation	P value on t-test
Hb	Controls	216	14.53	1.59	<0.001
	Extraaxial tumors	42	12.53	1.83
RBC	Controls	124	5.48	4.85	0.267
	Extraaxial tumors	40	4.62	0.51
TLC	Controls	216	7794.63	1774.60	0.284
	Extraaxial tumors	42	8136.19	2394.76
N	Controls	216	57.60	9.05	0.003
	Extraaxial tumors	41	62.17	8.76
L	Controls	216	32.46	8.21	0.714
	Extraaxial tumors	41	32.98	8.24
B	Controls	216	0.32	0.35	<0.001
	Extraaxial tumors	41	0	0
E	Controls	214	4.25	3.41	0.694
	Extraaxial tumors	41	4.02	2.92
M	Controls	216	5.39	2.11	<0.001
	Extraaxial tumors	41	0.71	1.33
Platelets	Controls	215	223595.35	75330.48	0.014
	Extraaxial tumors	40	256925.00	91773.24
NLR	Controls	216	1.97	0.84	0.34
	Extraaxial tumors	41	2.11	0.97
MLR	Controls	216	0.18	0.08	<0.001
	Extraaxial tumors	41	0.03	0.05

IA = Intraaxial; Hb = Hemoglobin; TLC = Total leukocyte count; N = Neutrophils; L = Lymphocytes; M = Monocytes; E = Eosinophils; B = Basophils; NLR = Neutrophil to lymphocyte ratio; MLR = Monocyte to lymphocyte ratio; G = Granulocytes; AGr = Agranulocytes

The mean tumor volume in the intraaxial group was 46.14 ± 36.68 cm3 (median 42.07 cm3) and the mean volume of the extraaxial lesions was 26.226 ± 26.34 cm3 (median 16.21 cm3). No significant correlation was found between the tumor volume and any of the leukocyte counts/indices in either the intra- or extraaxial group.

Discussion

This study generated several interesting findings. Certain alterations in the peripheral counts and indices were seen to occur in all patients with brain tumors (low grade and high grade intraaxial tumors as well as benign extraaxial lesions). These include a reduction in the hemoglobin (Hb), relative increase in platelet count, a relative increase in neutrophil counts, and a reduction in the monocyte counts and the MLR. We included patients with benign extraaxial lesions in this study since there was no published data regarding the derangements in hematologic indices associated with these lesions. This study documents the occurrence of hematologic alterations in patients with low grade gliomas as well as benign extraaxial tumors. This finding implies that brain tumors, irrespective of grade and location, produce some alterations in the hematopoietic system and immune status.

The occurrence of tumor-related thrombocytosis has been documented to occur and to indicate poor prognosis in several solid organ tumors.[6] The putative mechanism for paraneoplastic thrombocytosis could be aberrant paracrine signaling by tumor cells that produce interleukin-6 (IL-6), which in turn stimulates the liver to produce thrombopoietin, leading to an increase in megakaryocyte production in the bone marrow.[7] In the current study, we observed relative thrombocytosis in patients with benign (intra- and extraaxial tumors) and high grade intraaxial lesions. It is possible, therefore, that either the mechanism underlying thrombocytosis in patients with brain lesions is different from the IL-6 pathway or that the same mechanism is operational in all brain tumors, intra- and extraaxial.

The other notable finding emerging from this study was the significant reduction in the monocyte count and the MLR in all patients with brain tumors. Recent studies on tumor immunology have shown that glioblastomas are heavily infiltrated with monocytes. These tumor monocytes can account for up to 10-30% of the viable cell mass within a tumor.[8910] It is thought that these tumor monocytes are derived from systemic monocytes and not locally transformed microglia. These tumor monocytes are thought to be responsible for suppressing the immune response to the tumor.[11] The decreased MLR was due to a reduction in monocyte counts, indicating a suppression of cell-mediated immunity in patients with brain tumors. This finding is in keeping with the current concepts of monocyte-mediated immune suppression in GBM. However, surprisingly, we found similar changes in patients with low grade gliomas and benign extraaxial lesions. The implication of a reduced monocyte count in low grade gliomas and extraaxial tumors is still unclear.

Bambury et al. in their study found that an NLR>4 was an independent prognostic factor in patients with glioblastoma.[5] In this study, we found that the NLR in patients with intraaxial tumors was elevated compared to the control group (trend toward significance, P = 0.05). When patients with GBM were analyzed separately, it was found that the elevation in NLR when compared to the controls was statistically significant (P = 0.02). The elevation in NLR was not significant in the extraaxial tumor group. Furthermore, we found that this increase in the NLR in the intraaxial tumor group was due to an increase in the neutrophil count rather than a decrease in the lymphocyte count. It is thus evident that there is an increase in the phagocytic cells of inflammation in patients with intraaxial tumors. Thus, it appears that in all patients with GBM, there is an elevation in the NLR. The subset of patients who have an NLR>4 may have aggressive forms of the disease and may be prone to earlier recurrence/progression. Larger longitudinal studies will be required to establish the significance of this fact.

The occurrence of alterations of the peripheral blood counts and indices in patients with benign extraaxial tumors and low grade gliomas point to the existence of an immune response to these tumors as well. The major limitation of this study is the lack of follow-up and outcome data. A correlation between the MLR/NLR and disease-free survival would make any analysis of significance more robust and credible.

Conclusion

This study provides some baseline data about the hematologic derangements that occur in patients with both extraaxial and intraaxial tumors. The derangement of these counts and indices in patients with extraaxial tumors as well as in patients with low grade intraaxial brain tumors is a curious finding.

Financial support and sponsorship

This project was partly funded by the intramural Golden Jubilee Short-Term Research Award for Undergraduate Students (GJ-STRAUS) scheme awarded by JIPMER in the year 2014-2015.

Conflicts of interest

There are no conflicts of interest.