Clin Transl Oncol (2012) 14:864–869
DOI 10.1007/s12094-012-0872-5
RESEARCH ARTICLE
Neutrophil to lymphocyte ratio (NLR) as an indicator of poor
prognosis in stage IV non-small cell lung cancer
S. Cedrés • D. Torrejon • A. Martı́nez •
P. Martinez • A. Navarro • E. Zamora •
N. Mulet-Margalef • E. Felip
Received: 28 November 2011 / Accepted: 24 January 2012 / Published online: 19 July 2012
Ó Federación de Sociedades Españolas de Oncologı́a (FESEO) 2012
Abstract
Background Neutrophil to lymphocyte ratio (NLR), an
index of systemic inflammation, has been associated with
worse survival for many types of cancer. The aim of this
study is to investigate the clinical significance of the blood
NLR as a prognostic factor in non-small cell lung cancer
(NSCLC) patients.
Methods and patients Stage IV NSCLC patients diag-
nosed in our institution between April 2004 and March
2009 were retrospectively reviewed. Potential prognostic
factors such as histology, gender, performance status,
response to chemotherapy and NLR were analyzed. NLR
was assessed baseline and during chemotherapy treatment.
Overall survival (OS) and progression free survival (PFS)
were calculated by the Kaplan–Meier method.
Results A total of 171 patients were included in the study
and 60 patients (35.1 %) presented a NLR C5. Median
survival for the entire cohort was 9.3 months. We found
that patients with undifferentiated carcinoma and patients
with NLR C5 had a worse survival. Median PFS of patients
with NLR \5 was 5.62 months and in patients with NLR
C5 was 3.25 months (p = 0.098), and OS was 11.1 versus
5.6 months for patients with NLR\5 and NLR C5,
respectively (p = 0.017). During the chemotherapy treat-
ment, patients who normalized NLR after one cycle pre-
sented better outcomes (OS 8.7 vs. 4.3 months, p = 0.001,
for patients who normalized NLR and for patients who
S. Cedrés (&)
D. Torrejon
A. Martı́nez
P. Martinez

A. Navarro
E. Zamora
N. Mulet-Margalef
E. Felip
Medical Oncology Service, Vall d’Hebron University Hospital,
Barcelona, Spain
e-mail: susanacedres@yahoo.es
S. Cedrés
A. Martı́nez
P. Martinez
Universidad Autonoma de Barcelona, Barcelona, Spain
123
remained persistently elevated). After multivariate analy-
sis, histology and NLR remained independent predictors of
survival (p \ 0.05).
Conclusion In our analysis, elevated NLR is a predictor
of shorter survival in patients with advanced NSCLC and
the variation of NLR during the first cycle of treatment
predicts survival. NLR is an easily measured, reproducible
test that could be considered to be incorporated in the
routine practice in NSCLC patients.
Keywords Non-small cell lung cancer (NSCLC)

Neutrophil to lymphocyte ratio
Prognostic factor

Inflammation
Introduction
Non-small cell lung cancer (NSCLC) represents 85 % of
patients diagnosed of lung cancer. Despite continues efforts
and progress in diagnoses and treatment, the overall sur-
vival (OS) is still poor [1]. Several prognostic factors to
better prediction of survival had been reported and tumor
stage and performance status (PS) are the most important
factors [2, 3].
Inflammation seems to play a critical role in the devel-
opment and progression of numerous cancers by promoting
cancer cell proliferation and survival, angiogenesis, tumor
metastasis and impacting tumor response to systemic
therapies [4]. It has been postulated that inflammatory cells
in the tumor microenvironment have significant effects on
tumor development, and markers of systemic inflammation
may provided useful information for prognostication.
Absolute white blood counts (WBC) as well as the neu-
trophil to lymphocyte ratio (NLR) have been investigated
for their prognostic role in some cancer populations [5–10].
Clin Transl Oncol (2012) 14:864–869
Walsh et al. [7] have shown a NLR C5 to be a marker of
survival in colorectal cancer patients. Halazun [8] found
that NLR C5 was an independent predictor of recurrence
and poor OS in patients with colorectal liver metastases.
Gomez [9] demonstrated the efficacy of the NLR in pre-
dicting outcomes in patients undergoing liver resection for
hepatocelular carcinoma. Kao [10] demonstrated NLR C5
is a predictor of shorter survival in patients with unresec-
table malignant pleural mesothelioma with systemic che-
motherapy, and patients whose NLR normalized during
chemotherapy were found to have significantly longer
survival than those who remained abnormal.
However, the prognostic role of inflammatory markers
in NSCLC has not been well defined [11–14]. Two studies
including surgically treated patients demonstrated opposite
results with regard to the prognostic role of NLR in lung
cancer [12, 13]. In a Japanese study, with previously not
treated stage IIIB–IV NSCLC patients, NLR was a weak
predictor of survival [11].
In the present study, we assessed the prognostic value of
NLR in predicting outcomes in patients with stage IV
NSCLC and the impact on variation of NLR following
systemic treatment on survival.
Materials and methods
Patients
One hundred and seventy-one consecutive NSCLC patients
diagnosed at Vall d’Hebron University Hospital between
April 2004 and March 2009 were retrospectively reviewed.
All patients presented clinical stage IV and histologi-
cally proven diagnostic of squamous cell carcinoma, ade-
nocarcinoma, bronchiololoalveolar carcinoma, large cell
carcinoma or undifferentiated carcinoma.
The baseline staging work-up included a complete his-
tory and physical examination, complete bloods counts,
serum biochemistry tests, computed tomography of thorax
and upper abdomen, brain imaging study (TC or MRI) and
pulmonary functional tests. The tumor stage was defined
according to the sixth edition of TNM classification [15].
All patients were surveillance at discretion of the treating
oncologist, before each cycle of chemotherapy and with CT
scan every 2–3 months until progression disease. Gluco-
corticoids were not chronically used for these patients, but
some chemotherapy regimens included short course of
corticoids as part of the necessary premedication. More-
over, no patients received G-CSF as part of the chemo-
therapy treatment.
Potential prognostic factors that were analyzed included
gender (male vs. female), histology (adenocarcinoma/
bronchioloalveolar, squamous carcinoma, large cell
865
carcinoma and undifferentiated carcinoma), ECOG per-
formance status (0–1/2–3) and NLR.
All patients were evaluated for chemotherapy treatment
and cases with solitary metastases (suprarenal and brain
localization) were considered for surgery treatment.
Methods
Peripheral blood samples were recorded at baseline, previ-
ous to initiate any treatment, and after one and two cycles of
systemic therapy. The first blood analysis was performed
between 14 and 1 day before first cycle of chemotherapy; the
second and third samples were obtained on the day before or
the same day of cycles two and three of chemotherapy.
Baseline WBC and lymphocytes were recorded for the
analysis. NLR was defined as the absolute neutrophil count
divided by the absolute lymphocyte count. An NLR of 5 or
greater was considered elevated in accordance with earlier
report [7, 10, 16]. During chemotherapy treatment, a nor-
malization of the NLR was defined as patients with NLR C5
at baseline who changed to NLR\5 during chemotherapy.
Statistical analyses
Categorical variables were analyzed using Fisher exact
tests and Chi-square tests, as appropriate. Continues vari-
ables were analyzed using 2-sides tests.
The duration of overall survival (OS) was calculated
from date of pathologic diagnosis and the death due to any
cause or until the date of the last follow-up visit for patients
still alive. Patients were censored at last follow-up if still
alive or lost to follow-up. Progression free survival (PFS)
was defined as the time of the beginning of the first treat-
ment until first event (progression or death due to any
cause). The OS, PFS and 95 % confidence intervals (CI)
were calculated by the Kaplan–Meier method. Curves were
compared using the log-rank test. Multivariable survival
analysis was performed using Cox regression performed to
assess for patient characteristics associated with high NLR.
Factors with a prognostic association in the univariate
analysis were entered into a multivariate Cox regression
model. Results of the Cox regression modeling are pre-
sented as hazard ratios (HR) and associated 95 % confi-
dence intervals (CI). Two-sided p values of \0.05 were
considered significant. All analyses were performed with
PASW statistics 17.0.2.
Results
A total of one hundred and seventy-one stage IV patients
were included in the study and all of them available for
123
866 Clin Transl Oncol (2012) 14:864–869

analysis of survival. The baseline characteristics of the (40.5 %). Thoracic surgery was performed in 13 patients
patient population included in this study are outlined in (7.6 %) that presented synchronic brain metastases treated
Table 1. The median age was 63 years (30–81 years), the with radiotherapy. During the follow-up 31 patients
majority of patients were male (83.6 %), current or former (11.2 %) developed brain metastases. All patients were
smokers (90 %), ECOG performance status 1 and 2 (42 and initially considered for systemic treatment, but 22 patients
38.6 %, respectively) and histology of adenocarcinoma did not received chemotherapy due to different causes
(clinical deteriorate, patient’s decision, serious co-mor-
Table 1 Baseline patients characteristics
bidities). The chemotherapy regimens were platinum dou-
Characteristics N % bles with taxanes in 63.2 %, with gemcitabine in 21.8 %,
pemetrexed in 10.5 % and other combinations in 0.6 %. At
Number of patients 171 100
the moment of the blood analysis no patients showed
Median age (range) 63 (30–81)
clinical signs of sepsis or other inflammatory illnesses.
Gender
Full blood counts were unobtainable for six patients.
Male 143 83
The median WBC for the study population was 70.7 9 109
Female 28 17
(range 16–201 9 109), median lymphocytes 17.6 9 109
Performance status
(range 4–90 9 109) and the NLR 4.8 (range 0.4–16.3).
0 0 0
Sixty patients (35.1 %) had a NLR C5. The association
1 72 42.0
between patients and tumor characteristics and NLR is
2 66 38.6 shown in Table 2. In univariate analysis, only tumor size
Histologic features and lymph nodes metastases were associated with baseline
Adenocarcinoma 69 40.5 NLR. The mean NLR for T1 was 4.3, for T2: 4.4, for T3:
Squamous cell carcinoma 31 18.3 5.4 and for T4: 5.4. The NLR was C5 in 31.6 % of T1
Large cell 36 21.2 patients and 51.4 % of T4 and NLR \5 in 68.4 % of T1
Undifferentiated 6 3.5 patients and 48.6 % of T4 (p = 0.022). Similarly, NLR
Other 28 16.4 was associated with lymph nodes metastases, with NLR C5
Smoking history
Former/current smoker 153 90 Table 2 Baseline characteristics based on NLR groups
Never smokers 18 10
Baseline characteristic NLR \5n % NLR C5n % P
T-factor
T1 19 11.1 Median age 62 63
T2 71 41.5 Gender
T3 31 18.1 Male 87 91 52 86.4 0.51
T4 38 22.2 Female 18 9 8 13.6
N factor Performance status
N0 28 16.4 1 50 59.5 21 39.6 0.08
N1 26 15.2 2 31 36.9 28 52.8
N2 75 43.9 3 3 3.3 4 7.5
N3 36 21.1 Histology
Metastatic site Adenocarcinoma 44 45.3 24 51 0.18
Lung 49 28 Squamous 15 16.4 13 27.7
Bone 22 13 Large cell 28 30.7 8 17
Nervous system 19 11 Undifferentiated 4 4.4 2 4.3
Liver 13 8 Tumor size
Other 23 13 T1 13 13.1 6 10.7 0.022
Multiple localization 45 26 T2 50 50.5 18 32.1
NLR median (range) 4.8 (0.4–16.3) T3 18 18.1 13 23.2
NLR C5 60 35.1 T4 18 18.1 19 33.9
NLR \5 105 61.4 Nodal metastases
Response to chemotherapy N0 19 18.8 7 11.9 0.03
Partial response 62 36.8 N1 20 19.8 5 8.5
Stable disease 33 19.3 N2 43 42.6 30 50.8
Progression disease 53 31 N3 19 18.8 17 28.8

123
Clin Transl Oncol (2012) 14:864–869
in 26.9 % of N0 patients and 47.2 % of N3 versus NLR\5
in 73.1 % of N0 and 52.8 % in N3 (p = 0.030). No asso-
ciation was found between the NLR with the number of
metastatic sites, performance status, type of chemotherapy
either use or not of glucocorticoids during chemotherapy.
Median follow-up for all patients was 9.1 months (range
1–70.37 months). At the moment of the analysis 164 patients
progressed and 159 patients had died. The median PFS for all
the population was 5.29 months (4.58–5.99 months) and the
median OS was 9.3 months (6.8–10.1 months). In univariate
analysis the variables associated with better survival were
histology (adenocarcinoma 17.3 vs. 3.3 months for undiffer-
entiated, HR 4.2, 95 % CI 1.8–9.8; p = 0.001) and NLR C5
(HR 1.4, 95 % CI 1.1–2.1; p = 0.017). We did not found
association of survival with gender either response, or type of
chemotherapy used. In our series PS was not statistically
associated with survival. Median survival for patients PS1 was
13 months, for PS2: 8.4 months and for PS3: 3.9 months
(p [ 0.05). However, when we compared PS1 versus PS2 and
3, the difference was statistically significant (P = 0.01). We
found that patients with elevated NLR had significantly worse
survival compared with those with NLR\5. The median PFS
in patients with low/high NLR was 5.62 versus 3.25 months
(p = 0.098) and the median OS in patients with low/high
NLR was 11.1 versus 5.6 months (p = 0.017) (Fig. 1).
NLR was evaluated during chemotherapy treatment.
After one cycle of chemotherapy NLR was normalized in
50 % of patients with NLR C5 at baseline, and after two
cycles in an additional 10 % of patients. In total, 60.3 % of
patients normalized NLR after 1 or 2 cycles of chemother-
apy. We found that patients who normalized NLR after one
cycle of chemotherapy had better PFS (4.86 vs. 2.3 months,
p = 0.106) and OS (8.77 vs. 4.3 months, p = 0.001).
Patients who normalized NLR after two cycles of chemo-
therapy presented an improve in PFS and OS (4.0 versus 2.6
months; p = 0.429 and 9.5 versus 3.9 months; p = 0.91
respectively). Patients who normalized NLR at any moment
of the evaluation (after one or two cycles of chemotherapy)
had a PFS of 4.9 versus 2.0 months (p = 0.083) and OS of
9.1 versus 3.9 moths (p = 0.67). Figure 2 shows that a
normalization of NLR was associated with significantly
longer survival than whose NLR remained elevated with a
survival difference of around 4 months.
Multivariate analysis revealed that histology (HR 4.1,
95 % CI 1.8–9.6, p = 0.001) and NLR less than 5 (HR 1.5,
95 % CI 1.1–0.2.1, p = 0.015) remained as independent
predictors of length of survival.
Discussion
The results of this study confirm our hypothesis and sug-
gest that an elevated NLR is an independent predictor for
867
poor survival in patients with stage IV NSCLC. These
results are consistent with previous observations on the
association between NLR and variety of cancers such as
colorectal cancer, resected intrahepatic cholangiocarci-
noma, renal cancer, ovarian cancer, pancreatic cancer,
gastric cancer and pleural mesothelioma [7–10, 17–20].
In lung cancer three previous reports had evaluated the
association between NLR and outcome. Sarraf et al. in a
retrospective review of 178 patients undergoing complete
resection for NSCLC, found that increasing NLR on pre-
operative blood tests was associated with higher stage and,
on multivariate analysis, increasing NLR, expressed in
tertiles, remained as an independent prognostic factor.
Teramukai, in a Japanese study examined the impact of
pretreatment NLR count (expressed in quartiles) on sur-
vival in patients with advanced NSCLC. A total of 338
chemonaive patients with stage IIIB–IV were evaluated.
They found that the pretreatment neutrophil count was
statistically significant associated with short overall sur-
vival and PFS. In multivariate analysis, they found linear
association between pretreatment elevated neutrophil count
and shorter OS and PFS, but the relationship between NLR
and OS was some degree weak and non-linear. Recently, a
negative study was published in a retrospective review of
23 patients who underwent a complete resection for pul-
monary adenocarcinoma and they did not found differences
in the risk of recurrences according to the NLR. These
studies suggest that further evaluation should be considered
for NLR in NSCLC patients.
In our study, we detected increasing T and N staging
was associated with increasing NLR. In our series,
patients with T4 and N3 presented significantly higher
NLR than patients with T1 and N0 (p \ 0.05). Thus, the
NLR may be higher in patients with more advanced dis-
ease. This is consistent with previous studies on colon
cancer and NSCLC where increasing stage was associated
with increasing NLR [7, 12]. However, when we analyzed
the association between NLR and tumor burden, assessed
by number of metastatic sites, no association was
detected.
We detected NLR was associated with poor prognosis in
stage IV NSCLC patients. After adjusting for known
prognostic factors such as histological subtype and gender,
patients with NLR C5 presented a decrease in overall
survival of around 4 months respects patients with a NLR
\5 at baseline. In addition, we found that patients whose
NLR decrease during chemotherapy treatment presented
better survival. These results agree with a previous study in
mesothelioma patients in which patients whose NLR nor-
malized after one cycle of systemic therapy were found to
have significantly longer survival than those whose NLR
remained abnormal. However, in another study of meta-
static renal cell carcinoma patients treated with sunitinib,
123
868 Clin Transl Oncol (2012) 14:864–869

a b
NLR < 5: 5.6 months
NLR> 5: 3.25 months NLR < 5: 11.1 months
P=0.09 NLR> 5: 5.6months
P=0.01

Progression Free Survival (months) Overall Survival (months)

Fig. 1 Kaplan–Meier progression free survival (a) and overall survival (b) according to NLR in all population

a b

NLR < 5: 4.9 months

NLR < 5: 8.8 months
NLR> 5: 2.3 months
NLR> 5: 4.3 months
P=0.11
P=0.01

Progression Free Survival (months) Overall Survival (months)

Fig. 2 Kaplan–Meier progression free survival (a) and overall survival (b) according to variation NLR after one cycle of chemotherapy

the NLR after first cycle of treatment was not found
associated with response rate and survival [21]. At our
knowledge, previous studies analyzing the NLR as a
potential predictor of chemotherapy response in NSCLC
treated with chemotherapy had not been reported.
This study has its limitations. This is a retrospective
single centre study with 171 patients without validation in
an independent series of patients. We chose dichotomize
NLR C/\5 to be consistent with prior studies that have
used the same cutoff value. However, it is unclear whether
a different cutoff value would have changed our results.
Our series was treated heterogeneously with different
chemotherapy combinations and the majority of patients
presented multiple metastatic sites. We observed that PS
was a prognostic factor when compared PS1 versus PS2
and PS3, but not in multivariate analysis.

123
Clin Transl Oncol (2012) 14:864–869
The NLR is an inexpensive, reproducible, and widely
available blood test. Several reports have found NLR to be
an important indicator of adverse prognostic in colorectal,
gastric, pancreatic, ovarian cancer and malignant pleural
mesothelioma [5–10]. Recently, an increasing neutrophil
count has also been identified as an independent predictor
of death in patients with surgically treated NSCLC and
more controversial in advanced NSCLC [11–13].
Our clinical observation that the pretreatment NLR may
be associated with survival in patients with metastatic
NSCLC may contribute to treatment decisions. We con-
cluded that patients with stage IV NSCLC with NLR C5
have worse prognosis than patients with a NLR normal,
and this prognosis is better if the ratio is normalized during
chemotherapy treatment. The NLR is an easily measured
and reproducible test that could be considered to be
incorporated into the routine clinical practice to contribute
to treatment decisions in NSCLC patients.
Acknowledgments We would like to thank to Eva López Guerrero
for statistical support, Marta Lara for collecting data and Francisco J.
Sánchez Vélez for graphic design.
Conflict of interest None.
References
1. Jemal A, Siegel R, Ward E et al (2009) Cancer Statistics, 2009. CA Cancer J
Clin 59:225–249
2. Groome PA, Bolejack V, Crowley JJ et al (2007) The IASLC Lung Cancer
Staging Project: validation of the proposal for revision of the T, N and M
descriptors and consequent stage groupings in the forthcoming (seventh) edition
of the TNM classification of malignant tumors. J Thorac Oncol 2(8):694–705
3. Hoang T, Xu R, Schiller JH et al (2005) Clinical model to predict survival in
chemo-naive patients with advanced non-small cell lung cancer treated with
third generation chemotherapy regimen based on eastern cooperative oncology
group data. J Clin Oncol 23(1):175–183
4. Mantovani A, Allavena P, Sica A, Ballkwill F (2008) Cancer-related inflam-
mation. Nature 454:436–444
869
5. Leitch E, Chakrabarti M, Crozier J, McKee RF, Anderson JH, Horgan PG et al
(2007) Comparison of the prognostic value of selected markers of the systemic
inflammatory response in patients with colorectal cancer. Br J Cancer
97:1266–1270
6. McMillan D (2009) Systemic inflammation, nutritional status and survival in
patients with cancer. Curr Opin Clin Nutr Metab Care 12:223–226
7. Walsh SR, Cook EJ, Goulder F, Justin TA, Keeling NJ (2005) Neutrophil
lymphocyte ratio as a prognostic factor in colorectal cancer. J Surg Oncol
91:181–184
8. Halazun KJ, Aldoori A, Malik HZ, Al-Mukhtar A, Prasad KR, Toogood GJ et al
(2008) Elevated preoperative neutrophil to lymphocyte ratio predicts survival
following hepatic resection for colorectal liver metastases. Eur J Surg Oncol
34:55–60
9. Gomez D, Morris-Stiff G, Toogood G, Lodge JP, Prasad KR (2008) Impact of
systemic inflammation on outcome following resection for intrahepatic chol-
angiocarcinoma. J Surg Oncol 97:513–518
10. Kao S, Pavlakis N, Harvie R, Vardy J, Boyer M et al (2010) High blood
neutrophil-to-lymphocyte ratio is an indicator of poor prognosis in malignant
mesothelioma patients undergoing systemic therapy. Clin Cancer Res
16(23):5805–5813
11. Teramukai S, Kitano T, Kishida Y et al (2009) Pretreatment neutrophil counts as an
independent prognostic factor in advanced non-small cell lung cancer—an analysis
of Japan multinational trial organization LC00-03. Eur J Cancer 45:1950–1958
12. Sarraf KM, Belcher E, Raevsky E et al (2009) Neutrophil/lymphocyte ratio and
its association with survival after complete resection in non-small cell lung
cancer. J Thorac Cardiovasc Surg 137:425–428
13. Sakai T, Tsushima T, Kimura D, Hatanaka R, Yamada Y et al (2011) A clinical
study of the prognostic factors for postoperative early recurrence in patients who
underwent complete resection for pulmonary adenocarcinoma. Ann Thorac
Cardiovasc Surg 17(6):539–543
14. Pacsmans M, Sculier JP, Libert P et al (1995) Prognostic factors for survival in
advanced non-small-cell lung cancer: univariate and multivariate analyses
including recursive partitioning and amalgamation algorithms in 1052 patients.
J Clin Oncol 13:1221–1230
15. Greene FL (2002) American Joint Committee on Cancer, American Cancer
Society, AJCC Cancer Staging Manual, 6th edn. Springer, New York
16. Kishi Y, Kopetz S, Chun YS, Palavecino M, Abdall EK, Vauthey JN (2009)
Blood neutrophil-to-lymphocyte ratio predicts survival in patients with colo-
rectal liver metastases treated with systemic chemotherapy. Ann Surg Oncol
16:614–622
17. Ohno Y, Nakashima J, Ohori M, Hatano T, Tachibana M (2010) Pretreatment
neutrophil-to-lymphocyte ratio as an independent predictor of recurrence in
patients with non-metastatic renal cell carcinoma. J Urol 184:873–878
18. An X, Ding PR, Li YH et al (2010) Elevated neutrophil-to lymphocyte ratio
predicts survival in advanced pancreatic cancer. Biomarkers 15:516–522
19. Cho H, Hur HW, Kim SW et al (2009) Pre-treatment neutrophil to lymphocyte
ratio is elevated in epithelial ovarian cancer and predicts survival after treat-
ment. Cancer Immunol Immunother 58:15–23
20. Yamanaka T, Matsumoto S, Teramukai S et al (2007) The baseline ratio of
neutrophil to lymphocyte is associated with patients prognosis in advanced
cancer gastric. Oncology 73:215–220
21. Keizman D, Ish-Shalom M, Huang P, Eisenberger M, Pili R et al (2011) The
association of pre-treatment neutrophil to lymphocyte ratio with response rate,
progression free survival, and overall survival of patients treated with sunitnib
for metastatic renal cell carcinoma. Eur J Cancer 48(2):202–208
123