doi: 10.1111/1346-8138.

13875 Journal of Dermatology 2017; : 1–10

ORIGINAL ARTICLE
Neutrophil–lymphocyte ratio, platelet–lymphocyte ratio and
mean platelet volume in Japanese patients with psoriasis and
psoriatic arthritis: Response to therapy with biologics
Akihiko ASAHINA, Naoko KUBO, Yoshinori UMEZAWA, Hiromi HONDA,
Koichi YANABA, Hidemi NAKAGAWA
Department of Dermatology, The Jikei University School of Medicine, Tokyo, Japan

ABSTRACT
Recent studies indicate the presence of systemic inflammation in psoriatic patients, and this inflammatory status
is significantly associated with a range of comorbidities. The aim of this study was to evaluate the clinical signifi-
cance of novel inflammatory biomarkers, neutrophil–lymphocyte ratio (NLR), platelet–lymphocyte ratio (PLR) and
mean platelet volume (MPV) in Japanese patients with plaque-type psoriasis (PsV) and psoriatic arthritis (PsA).
One hundred and eighty-six patients with PsV and 50 patients with PsA treated with biologics, including inflix-
imab, adalimumab and ustekinumab, were retrospectively analyzed before and after treatment. At baseline, NLR
and PLR, as well as C-reactive protein (CRP), were significantly higher in PsA patients than those in PsV patients,
and a significant correlation was found between NLR and PLR. In PsV patients, the NLR-high and PLR-high sub-
groups exhibited significantly higher Psoriasis Area and Severity Index scores compared with the NLR-low and
PLR-low subgroups, respectively, and the NLR-high subgroup also showed higher CRP levels. MPV value was
negatively associated with the presence of arthritis, but its association with inflammation was less clear than that
of NLR or PLR. After treatment of the patients with biologics for up to 12 months, NLR and PLR decreased
promptly in parallel with a decrease of CRP, irrespective of the type of biologics used. Altogether, these results
indicate that both NLR and PLR may be useful markers to evaluate systemic inflammation in psoriatic patients.
They may serve as simple, convenient and cost-effective biomarkers to monitor the disease course after systemic
therapy.

Key words: biologics, mean platelet volume, neutrophil–lymphocyte ratio, platelet–lymphocyte ratio,
psoriasis.

INTRODUCTION may be evaluated based on the results of simple blood cell

Psoriasis is a chronic immune-mediated skin disorder.1,2
Recent studies indicate the presence of subclinical systemic
inflammation in psoriatic patients, which is associated with a
range of comorbidities, including metabolic diseases, such as
diabetes and cardiovascular diseases.2 Arthritis is observed in
as many as 7–26% of psoriatic patients,3 and may also reflect
systemic inflammation. It is desirable to have a simple means
to objectively measure the extent of inflammation in psoriasis,
not only to estimate the severity of disease, but also to
evaluate the benefit of systemic treatment.
In our previous study,4 we found a significant elevation of
serum C-reactive protein (CRP) levels in Japanese psoriatic
patients, particularly in those with severe cutaneous involve-
ment and/or active arthritis. CRP is one of the most sensitive
biomarkers of systemic inflammation, and it can be measured
easily and routinely. On the other hand, systemic inflammation
analysis alone. The neutrophil–lymphocyte ratio (NLR) and
platelet–lymphocyte ratio (PLR) have both emerged as good
indicators of subclinical systemic inflammation and poor prog-
nosis in a variety of diseases such as cancer,5 cardiovascular
diseases,6 metabolic syndrome7,8 and autoimmune inflamma-
tory diseases.9,10 NLR and PLR are relatively stable, as com-
pared with individual blood cell parameters, which are easily
fluctuated by dehydration/overhydration, diluted blood speci-
mens and blood specimen handling.9 Furthermore, in routine
blood cell analysis, mean platelet volume (MPV) has also been
demonstrated to be associated with several inflammatory
diseases.11 These novel biomarkers, however, have not been
well investigated in psoriasis, and their time-course changes
after treatment with biologics have not previously been
addressed. The present study was undertaken to assess the
association of NLR, PLR and MPV with the extent of systemic
inflammation in Japanese psoriatic patients with and without

Correspondence: Akihiko Asahina, M.D., Department of Dermatology, The Jikei University School of Medicine, 3-25-8 Nishi-Shimbashi,
Minato-ku, Tokyo 105-8461, Japan. Email: asahina-tky@umin.ac.jp
Received 9 September 2016; accepted 14 March 2017.

© 2017 Japanese Dermatological Association 1

A. Asahina et al.
arthritis. We also analyzed time-course changes of these novel
biomarkers in comparison with that of serum CRP levels after
treatment with different biologics.
METHODS
Patients and study design
Male and female patients aged 20 years or older with the diag-
nosis of either plaque-type psoriasis (psoriasis vulgaris: PsV) or
psoriatic arthritis (PsA), diagnosed according to the Classifica-
tion Criteria for Psoriatic Arthritis criteria, were included in this
cross-sectional retrospective study. Patients were enrolled
when they received treatment with one of the following three
biologics available in Japan: infliximab, adalimumab or ustek-
inumab, for over 3 months, at the psoriasis special clinic of
The Jikei University School of Medicine between 2011 and
2015. Available data of blood cell counts, MPV and serum
CRP were collected by routine blood tests before the treatment
(–1 to 0 months) and at any of the following time points up to
12 months: at 3 months (2–4 months), 6 months (5–7 months)
and 12 months (11–13 months) of treatment. We utilized a con-
ventional method for CRP quantification, and CRP values of
“less than 0.04 mg/dL” were handled as “0.04 mg/dL” in this
analysis for descriptive purposes. In addition, serum levels of
lactate dehydrogenase (LDH), uric acid, total cholesterol, low-
density lipoprotein cholesterol (LDL-C), high-density lipoprotein
cholesterol (HDL-C) and hemoglobin A1c (HbA1c) were
recorded wherever available. The severity of cutaneous lesions
was evaluated by Psoriasis Area and Severity Index (PASI)
scores at baseline and also at the initial evaluation time point
of 3–4 months (10–14 weeks for infliximab and at 14–18 weeks
for adalimumab and ustekinumab); patients with missing PASI
scores at baseline were excluded from this study. Body mass
index (BMI) was recorded at baseline.
Treatments
Infliximab was administrated i.v. (5 mg/kg) at weeks 0, 2, 6
and thereafter every 8 weeks. Adalimumab was administrated
s.c. (80 mg) at week 0 and thereafter at 40 mg every 2 weeks.
Adalimumab dose was allowed to increase up to 80 mg for
patients showing an insufficient response. In some patients
with a good response, the administration interval was later
changed from every 2 weeks to every 3 weeks at the discre-
tion of the prescriber. Ustekinumab was administrated s.c.
(45 mg) at weeks 0, 4 and thereafter every 12 weeks. Ustek-
inumab dose was allowed to increase up to 90 mg for patients
showing insufficient response. Patients were excluded from
this study if any deviation from the above-mentioned protocol
had been noted during the treatment, or if the treatment had
been unexpectedly discontinued within the observation period.
In addition, the patients were excluded if they showed any
symptoms of infection at the time of blood examination includ-
ing cellulitis or pneumonia.
Statistical analysis
Data are expressed as mean  standard deviation for continu-
ous variables, with a median value where indicated. To
2 © 2017 Japanese Dermatological Association
compare parameters between two groups, an unpaired t-test
or Mann–Whitney U-test was used for those with or without
normal distribution, respectively. The normality of the distribu-
tion was analyzed using the Kolmogorov–Smirnov test. The v2-
test was used for categorical variables. To evaluate changes of
variables between the baseline and the designated time point,
a two-tailed Wilcoxon rank sum test was used. Correlation
analysis was performed by calculating the Spearman or Pear-
son correlation coefficient. Statistical analysis was performed
using StatMate IV (ATMS, Tokyo, Japan) software. P < 0.05
was considered significant.
RESULTS
Comparison between PsV and PsA
A total of 236 patients (186 with PsV and 50 with PsA) were
included in this study. This population largely overlaps with that
in our previous study.4 As shown in Table 1, the male : female
ratio was 140:46 for PsV and 38:12 for PsA, reflecting the male
predominance of the Japanese psoriatic population, and PsV
patients were older on average than PsA patients (54.6  15.0
vs 48.8  13.1 years, P < 0.05). PASI scores were statistically
higher in PsV patients compared with those in PsA patients
(14.5  9.3 vs 10.6  9.4, P < 0.005), while CRP levels were
statistically lower in PsV patients compared with those in PsA
patients (0.22  0.35 vs 1.22  1.81 mg/dL, P < 0.001), indi-
cating that PsA patients have an inflammatory burden aside
from skin eruptions. Total white blood cell counts, neutrophil
counts and platelet counts were significantly different between
PsV and PsA patients. Both NLR and PLR were significantly
lower in PsV patients than those in PsA patients (NLR,
2.71  1.66 vs 3.53  1.84, P < 0.001; PLR, 153.5  67.7 vs
178.7  89.4, P < 0.05). Interestingly, MPV was significantly
higher in PsV patients than in PsA patients in this study
(10.09  0.75 vs 9.79  0.83, P < 0.05). Other parameters,
including lymphocyte counts, BMI, LDH, uric acid, total choles-
terol, HDL-C, LDL-C and HbA1c were not significantly different
between the two groups.
Correlation among the parameters
As shown in Figure 1 and as calculated by the correlation
coefficient, there was a significant linear relationship between
NLR and PLR at baseline (Pearson correlation: PsV, r = 0.744,
P < 0.001; PsA, r = 0554, P < 0.001; PsV + PsA, r = 0.696,
P < 0.001). We then analyzed the Spearman correlation of CRP
with other parameters (Table 2). For both PsV and PsA
patients, NLR and PLR each showed a significant, weak to
moderate positive correlation with CRP. However, this weak
correlation may not be unexpected, given that we did not uti-
lize highly sensitive CRP measurement and that a majority of
patients had low CRP values in this study. Furthermore, a
moderate negative correlation was observed between CRP
and MPV in PsA patients.
Our next approach was to subdivide the patients according
to the parameters of interest. There is no validated consensus
on the normal values for NLR and PLR, and no patients devi-
ated from the normal range of MPV (8.4–12.8 fl) determined
 NLR, PLR and MPV in psoriatic patients

Table 1. Comparison between PsV and PsA patients 700

PsV (n = 186) PsA (n = 50) PsV
Mean  SD Mean  SD 600
Median Median P PsA

Sex (M/F) 140/46 38/12 ns 500 r = 0.744

Age, years 54.6  15.0 48.8  13.1 <0.05*
53.5 47.0
400
WBC (9103/mL) 6.57  1.91 7.87  1.97 <0.001*

PLR
6.20 7.55 r = 0.554
Neutrophils 4.25  1.64 5.48  1.87 <0.001* 300
(9103/mL) 3.95 5.25
Lymphocytes 1.76  0.63 1.76  0.61 ns
200
(9103/mL) 1.70 1.60
Platelets 242.0  60.1 280.2  74.0 <0.001*
(9103/mL) 236.5 267.5 100
CRP (mg/dL) 0.22  0.35 1.22  1.81 <0.001*
0.08 0.31
0
NLR 2.71  1.66 3.53  1.84 <0.001*
0 5 10 15 20
2.34 3.10
PLR 153.5  67.7 178.7  89.4 <0.05* NLR
137.9 167.4
Figure 1. Scatter-dot graphics showing the correlation
MPV (fl) 10.1  0.75 9.79  0.83 <0.05*
between neutrophil–lymphocyte ratio (NLR) and platelet–lym-
10.00 9.75
phocyte ratio (PLR). Closed circles (●) and open triangles (M)
BMI (kg/m2) 24.5  4.3 24.7  4.1 ns
indicate patients with plaque-type psoriasis (PsV) and psoriatic
23.8 24.0
arthritis (PsA), respectively. Linear regression lines are also
PASI 14.5  9.3 10.6  9.4 <0.005*
shown with the values for the Pearson correlation coefficient
12.6 10.5
(solid line for PsV and dotted line for PsA).
LDH (IU/L) 189.2  34.1 179.2  34.4 ns
184.0 175.5
Uric acid (mg/dL) 6.35  1.51 6.55  1.64 ns
Table 2. Spearman correlation of serum CRP levels with other
6.50 6.40
parameters
Total cholesterol 207.0  43.3 198.2  55.2 ns
(mg/dL) 207.5 192.0 PsV PsA
HDL-C (mg/dL) 58.2  14.3 55.6  12.3 ns
57.0 55.0 rs P rs P
LDL-C (mg/dL) 123.1  36.2 116.0  50.8 ns WBC (9103/mL) 0.467 <0.001* 0.426 <0.01*
120.0 116.0 Neutrophils (9103/mL) 0.493 <0.001* 0.445 <0.01*
HbA1c (%) 5.88  0.98 5.76  0.46 ns Lymphocytes (9103/mL) 0.002 ns 0.039 ns
5.60 5.70 Platelets (9103/mL) 0.229 <0.005* 0.438 <0.01*
MPV (fl) 0.011 ns 0.333 <0.05*
*P < 0.05 was considered statistically significant. Data represent NLR 0.343 <0.001* 0.363 <0.05*
mean  standard deviation (SD). The median value is also shown below. PLR 0.162 <0.05* 0.294 <0.05*
BMI, body mass index; CRP, C-reactive protein; HbA1c, hemoglobin BMI (kg/m2) 0.144 ns 0.016 ns
A1c; HDL-C, high-density lipoprotein cholesterol; LDH, lactate dehydro-
genase; LDL-C, low-density lipoprotein cholesterol; MPV, mean platelet
PASI 0.359 <0.001* 0.175 ns
volume; NLR, neutrophil–lymphocyte ratio; ns, not significant; PASI,
Psoriasis Area and Severity Index; PLR, platelet-lymphocyte ratio; PsA, *P < 0.05 was considered statistically significant. rs indicates Spearman
psoriatic arthritis; PsV, plaque-type psoriasis (psoriasis vulgaris); WBC, correlation coefficient. BMI, body mass index; CRP, C-reactive protein;
white blood cell. MPV, mean platelet volume; NLR, neutrophil–lymphocyte ratio; ns, not
significant; PASI, Psoriasis Area and Severity Index; PLR, platelet–lym-
phocyte ratio; PsA, psoriatic arthritis; PsV, plaque-type psoriasis (psori-
in our facility. Therefore, we used the median value among all asis vulgaris); WBC, white blood cell.
psoriatic patients as the cut-off value for NLR, PLR and MPV,
respectively. As shown in Table 3, for PsV patients, NLR-high differences in both CRP and PASI scores, as well as NLR
and PLR-high subgroups exhibited significantly higher PASI and PLR. We also subdivided the patients according to the
scores compared with NLR-low and PLR-low subgroups, PASI scores (for PsV, n = 63 [PASI < 10], n = 89
respectively (NLR-high vs -low, 16.5  11.3 vs 12.9  7.0, [10 ≤ PASI ≤ 20] and n = 34 [PASI > 20]; for PsA, n = 23
P < 0.05; PLR-high vs -low, 16.5  9.6 vs 12.8  8.7, [PASI < 10] and n = 27 [10 ≤ PASI]) (Fig. 2). For PsV patients,
P < 0.005). In addition, the NLR-high subgroup showed a CRP, NLR and PLR increased with increasing PASI scores,
higher CRP, and the PLR-high subgroup showed a lower and significant differences were found as indicated. Again, no
MPV, both of which were significant. Subdivision of the differences in MPV values were found irrespective of the PASI
patients according to the MPV values did not show significant scores.

© 2017 Japanese Dermatological Association 3

A. Asahina et al.

Table 3. Subdivision of patients according to NLR, PLR and MPV values

(a) NLR
PASI CRP (mg/dL) PLR MPV (fl)
Mean  SD Mean  SD Mean  SD Mean  SD
n Median P Median P Median P Median P
PsV ≤2.53 103 12.9  7.0 i 0.16  0.25 i 121.8  39.2 i 10.05  0.77 i
11.9 <0.05* 0.05 <0.001* 117.3 <0.001* 9.90 ns
>2.53 83 16.5  11.3 0.29  0.43 192.8  74.9 10.14  0.72
14.5 0.12 182.0 10.10
PsA ≤2.53 15 7.2  5.3 i 0.67  1.25 i 110.8  21.6 i 9.89  0.60 i
4.4 ns 0.11 ns 107.6 <0.001* 10.00 ns
>2.53 35 12.1  10.4 1.46  1.97 207.8  91.8 9.75  0.91
11.1 0.53 200.0 9.70

(b) PLR
PASI CRP (mg/dL) NLR MPV (fl)
Mean  SD Mean  SD Mean  SD Mean  SD
n Median P Median P Median P Median P
PsV ≤142 99 12.8  8.7 i 0.17  0.21 i 2.02  0.79 i 10.20  0.82 i
11.6 <0.005* 0.07 ns 1.83 <0.001* 10.20 <0.05*
>142 87 16.5  9.6 0.28  0.45 3.50  2.01 9.96  0.64
14.7 0.10 2.94 9.90
PsA ≤142 19 7.4  5.4
i 0.76  1.19
i 2.46  1.87
i 10.06  0.94
i
4.4 ns 0.21 ns 1.91 <0.001* 10.00 ns
>142 31 12.6  10.7 1.50  2.07 4.18  1.51 9.63  0.72
12.0 0.53 4.25 9.70

(c) MPV (fl)

PASI CRP (mg/dL) NLR PLR
Mean  SD Mean  SD Mean  SD Mean  SD
n Median P Median P Median P Median P
PsV ≤10 94 15.5  10.2 i 0.24  0.41 i 2.66  1.78 i 154.7  65.2 i
14.2 ns 0.08 ns 2.21 ns 144.1 ns
>10 92 13.6  8.3 0.19  0.26 2.77  1.53 152.2  70.6
11.7 0.08 2.50 135.4
PsA ≤10 34 11.0  10.1 i 1.63  2.05 i 3.46  1.72 i 191.3  99.2 i
12.1 ns 0.72 ns 3.04 ns 187.7 ns
>10 16 9.7  7.7 0.35  0.44 3.68  2.13 151.9  57.9
8.6 0.23 3.37 142.6

*P < 0.05 was considered statistically significant. Data represent mean  standard deviation (SD). The median value is also shown below. n indicates
the number of patients. CRP, serum C-reactive protein; MPV, mean platelet volume; NLR, neutrophil–lymphocyte ratio; ns, not significant; PASI, Psori-
asis Area and Severity Index; PLR, platelet–lymphocyte ratio; PsA, psoriatic arthritis; PsV, plaque-type psoriasis (psoriasis vulgaris).

Time-course change of parameters after treatment
with biologics
Table 4 shows the change of parameters following treatment
of the patients with each of the three different biologics. Com-
pared with the baseline values, not only CRP, but also NLR
and PLR, decreased after treatment with infliximab and
adalimumab. In PsV patients, decrease of CRP was small and
not significant at 3 and 12 months after treatment with ustek-
inumab. In contrast, NLR and PLR both decreased as early as
3 months after treatment with ustekinumab, and remained at
low levels for up to 12 months, with a significant decrease at
every time point. Regarding MPV, its changes overall were

Figure 2. Serum C-reactive protein (CRP) levels, neutrophil–lymphocyte ratio (NLR), platelet–lymphocyte ratio (PLR) and mean pla-
telet volume (MPV) according to the Psoriasis Area and Severity Index scores in patients (a) with plaque-type psoriasis (PsV) and (b)
psoriatic arthritis (PsA), respectively. Graph bars represent mean  standard deviation (SD). P-values were calculated using the
Mann–Whitney U-test (CRP, NLR, PLR) or unpaired t-test (MPV). *P < 0.05 was considered significant.

4 © 2017 Japanese Dermatological Association

 NLR, PLR and MPV in psoriatic patients

(a)
CRP NLR
1.4 8
P<0.001 P<0.01
1.2 7
P<0.05 P<0.001 ns P<0.01
1 6
0.8
5
0.6
4
0.4
3
0.2
0 2
1
0
<10 10-20 20< <10 10-20 20<
PLR MPV
400 12
P<0.01 ns
350 ns P<0.05 11.5 ns ns
300 11
250 10.5
200 10
150 9.5
100 9
50 8.5
0 8
<10 10-20 20< <10 10-20 20<

(b)
CRP NLR
5
8
ns
4 7 ns
6
3
5
2 4
1 3
2
0
1
0
<10 10 = < <10 10 = <
PLR MPV
400 12
350 P<0.05 11.5 ns
300 11
250 10.5
200 10
150 9.5
100 9
50 8.5
0 8
<10 10 = < <10 10 = <

© 2017 Japanese Dermatological Association 5

 6
Table 4. Time-course changes of inflammatory parameters after treatment with biologics

Baseline 3M 6M 12M
Mean ± SD n Mean ± SD Mean ± SD Mean ± SD
Parameter Category Medication Median Median n P Median n P Median n P
A. Asahina et al.

CRP All (PsV + PsA) IFX 0.67 ± 1.38 (74) 0.10 ± 0.12 (55) <0.001* 0.17 ± 0.28 (56) <0.001* 0.10 ± 0.12 (51) <0.001*
(mg/dL) 0.15 0.04 0.06 0.04
ADA 0.49 ± 0.97 (68) 0.30 ± 0.62 (46) <0.001* 0.21 ± 0.56 (64) <0.001* 0.15 ± 0.27 (60) <0.001*
0.14 0.06 0.04 0.04
UST 0.20 ± 0.32 (94) 0.18 ± 0.32 (61) ns 0.15 ± 0.25 (81) <0.05* 0.17 ± 0.25 (85) ns
0.07 0.07 0.05 0.06
PsV IFX 0.31 ± 0.50 (50) 0.09 ± 0.11 (39) <0.001* 0.18 ± 0.31 (36) <0.05* 0.11 ± 0.12 (30) <0.005*
0.10 0.04 0.06 0.04
ADA 0.19 ± 0.20 (49) 0.16 ± 0.34 (30) <0.05* 0.11 ± 0.13 (45) <0.005* 0.11 ± 0.13 (42) <0.005*
0.12 0.07 0.04 0.04
UST 0.18 ± 0.30 (87) 0.18 ± 0.33 (56) ns 0.15 ± 0.26 (74) <0.05* 0.17 ± 0.26 (79) ns
0.06 0.07 0.05 0.05
PsA IFX 1.43 ± 2.16 (24) 0.12 ± 0.15 (16) <0.01* 0.15 ± 0.24 (20) <0.01* 0.10 ± 0.11 (21) <0.01*
0.53 0.05 0.04 0.04
ADA 1.28 ± 1.57 (19) 0.55 ± 0.91 (16) <0.01* 0.44 ± 0.99 (19) <0.01* 0.25 ± 0.44 (18) <0.01*
0.31 0.04 0.08 0.04
UST 0.35 ± 0.52 (7) 0.19 ± 0.22 (5) ns 0.14 ± 0.07 (7) ns 0.23 ± 0.13 (6) ns
0.11 0.12 0.12 0.26
NLR All (PsV + PsA) IFX 2.97 ± 1.46 (74) 1.73 ± 0.75 (55) <0.001* 1.88 ± 1.11 (57) <0.001* 1.89 ± 0.96 (54) <0.001*
2.74 1.57 1.72 1.69
ADA 3.14 ± 2.16 (68) 2.30 ± 1.48 (47) <0.001* 2.17 ± 1.15 (64) <0.001* 2.07 ± 1.23 (61) <0.001*
2.69 1.90 1.82 1.61
UST 2.64 ± 1.55 (94) 2.10 ± 0.94 (65) <0.001* 2.14 ± 1.27 (83) <0.001* 2.06 ± 1.02 (86) <0.001*
2.14 1.89 1.80 1.98
PsV IFX 2.84 ± 1.47 (50) 1.69 ± 0.70 (39) <0.001* 1.94 ± 1.21 (36) <0.001* 1.91 ± 0.97 (33) <0.001*
2.61 1.57 1.75 1.69
ADA 2.90 ± 2.11 (49) 2.46 ± 1.65 (30) <0.01* 2.15 ± 1.19 (45) <0.001* 2.05 ± 1.30 (43) <0.001*
2.50 1.89 1.75 1.65
UST 2.53 ± 1.47 (87) 2.02 ± 0.87 (60) <0.001* 2.09 ± 1.24 (76) <0.001* 2.01 ± 0.93 (80) <0.001*
2.11 1.87 1.78 1.98
PsA IFX 3.24 ± 1.42 (24) 1.82 ± 0.90 (16) <0.01* 1.78 ± 0.93 (21) <0.001* 1.85 ± 0.98 (21) <0.005*
2.83 1.51 1.65 1.67
ADA 3.74 ± 2.25 (19) 2.03 ± 1.09 (17) <0.01* 2.22 ± 1.09 (19) <0.01* 2.12 ± 1.10 (18) <0.01*
3.24 2.00 1.94 1.57
UST 3.96 ± 2.03 (7) 3.01 ± 1.38 (5) ns 2.70 ± 1.60 (7) <0.05* 2.65 ± 1.86 (6) <0.05*
4.29 3.00 2.56 2.17
PLR All (PsV + PsA) IFX 169.4 ± 83.5 (74) 118.9 ± 41.1 (55) <0.001* 126.4 ± 58.1 (57) <0.001* 124.1 ± 46.8 (54) <0.001*
161.5 116.8 112.0 126.6
ADA 155.5 ± 68.4 (68) 131.1 ± 62.2 (47) <0.001* 122.9 ± 46.6 (64) <0.001* 122.3 ± 50.3 (61) <0.001*
146.9 107.2 115.3 113.8
UST 152.9 ± 68.0 (94) 120.8 ± 40.8 (65) <0.001* 128.5 ± 53.4 (83) <0.001* 120.0 ± 42.1 (86) <0.001*

© 2017 Japanese Dermatological Association

 Table 4. (continued)

Baseline 3M 6M 12M
Mean ± SD n Mean ± SD Mean ± SD Mean ± SD
Parameter Category Medication Median Median n P Median n P Median n P
134.2 116.1 119.1 115.5
PsV IFX 162.5 ± 64.6 (50) 119.1 ± 38.2 (39) <0.001* 129.8 ± 62.7 (36) <0.001* 124.5 ± 47.3 (33) <0.001*
165.4 116.8 116.0 126.7
ADA 151.3 ± 72.2 (49) 138.7 ± 71.8 (30) <0.005* 122.0 ± 48.4 (45) <0.001* 124.4 ± 54.2 (43) <0.001*
137.8 110.6 110.0 115.6
UST 149.5 ± 67.2 (87) 118.2 ± 38.5 (60) <0.005* 127.6 ± 54.9 (76) <0.001* 118.4 ± 40.6 (80) <0.001*
131.2 114.9 117.4 115.5
PsA IFX 183.8 ± 113.7 (24) 118.3 ± 48.8 (16) <0.01* 120.5 ± 50.1 (21) <0.001* 123.4 ± 47.2 (21) <0.005*
152.6 106.2 111.9 109.4
ADA 166.3 ± 58.0 (19) 117.6 ± 38.6 (17) <0.01* 125.0 ± 43.3 (19) <0.01* 117.2 ± 40.6 (18) <0.01*
175.6 104.6 118.6 110.9
UST 194.5 ± 68.9 (7) 152.3 ± 58.4 (5) ns 138.1 ± 33.9 (7) <0.05* 140.8 ± 59.3 (6) ns

© 2017 Japanese Dermatological Association

221.7 136.7 135.0 128.4
MPV (fl) All (PsV + PsA) IFX 9.81 ± 0.65 (74) 9.88 ± 0.58 (55) ns 10.01 ± 0.66 (57) <0.005* 9.93 ± 0.64 (54) <0.05*
9.80 9.80 9.80 9.95
ADA 10.13 ± 0.89 (68) 10.17 ± 0.95 (47) ns 10.21 ± 0.89 (64) <0.05* 10.17 ± 0.87 (61) <0.05*
10.10 10.20 10.00 9.90
UST 10.12 ± 0.74 (94) 10.16 ± 0.74 (65) ns 10.14 ± 0.75 (83) ns 10.15 ± 0.72 (86) ns
10.10 10.10 10.10 10.10
PsV IFX 9.82 ± 0.60 (50) 9.90 ± 0.55 (39) ns 9.98 ± 0.68 (36) <0.05* 9.84 ± 0.59 (33) ns
9.80 9.90 9.80 9.80
ADA 10.21 ± 0.83 (49) 10.12 ± 0.88 (30) ns 10.27 ± 0.88 (45) ns 10.21 ± 0.84 (43) ns
10.20 10.20 10.10 10.20
UST 10.17 ± 0.74 (87) 10.19 ± 0.76 (60) ns 10.17 ± 0.77 (76) ns 10.19 ± 0.73 (80) ns
10.20 10.10 10.20 10.10
PsA IFX 9.78 ± 0.75 (24) 9.83 ± 0.65 (16) <0.05* 10.07 ± 0.65 (21) ns 10.08 ± 0.70 (21) <0.05*
9.85 9.75 10.10 10.10
ADA 9.91 ± 1.02 (19) 10.24 ± 1.09 (17) <0.05* 10.07 ± 0.91 (19) ns 10.09 ± 0.97 (18) <0.05*
9.70 10.10 9.90 9.80
UST 9.53 ± 0.44 (7) 9.72 ± 0.39 (5) ns 9.77 ± 0.32 (7) ns 9.67 ± 0.45 (6) ns
9.50 9.90 9.80 9.60

*P < 0.05 was considered statistically significant.

The parameters were measured before treatment (baseline) and at 3 months (3M), 6 months (6M) and 12 months (12M) of treatment with the biologics, infliximab (IFX), adalimumab (ADA) or usteki-
numab (UST). Data represent mean ± standard deviation (SD), and the median value is also shown below. n indicates the number of patients used for calculation. P-values were calculated versus
baseline according to the Wilcoxon signed rank test. CRP, C-reactive protein; MPV, mean platelet volume; NLR, neutrophil–lymphocyte ratio; ns, not significant; PLR, platelet–lymphocyte ratio;
PsA, psoriatic arthritis; PsV, plaque-type psoriasis (psoriasis vulgaris).

7
NLR, PLR and MPV in psoriatic patients
A. Asahina et al.

Table 5. Average reduction rate of NLR and PLR at 3 months Psoriasis is accompanied by systemic inflammation, espe-
of treatment according to the response of treatment cially when patients have arthritis and/or extensive cutaneous
lesions. Our previous study4 suggested that serum CRP level
PASI score improvement†
serves as a reliable systemic inflammatory biomarker in Japa-
<75% (low- ≥75% (high- nese psoriatic patients. Besides CRP, there are other biomark-
Category Medication responder) responder)‡ ers in psoriatic patients, such as serum levels of inflammatory
PsV IFX (n) (7) (32) cytokines,12 but they are not routinely measured and are time-
NLR 29.0% 32.4% and cost-consuming. NLR has been proposed as a novel indi-
PLR 8.1% 20.9% cator of systemic inflammatory status in several comorbidities
ADA (n) (9) (20) known to be associated with severe psoriasis.2,13 Neutrophils
NLR 4.8% 27.1% are increased in the blood14,15 as well as in active skin lesions
PLR 11.2% 19.2%
of psoriatic patients, and they are involved in the pathogenesis
UST (n) (13) (47)
of psoriasis16 by producing many inflammatory cytokines, free
NLR 8.3% 8.8%
PLR 12.0% 9.9% oxygen radicals and lytic enzymes.10,15 PLR has also been pro-
PsA IFX (n) (3) (13) posed as another indicator of systemic inflammation, although
NLR 45.0% 41.1% less frequently investigated than NLR in the published work.
PLR 32.3% 35.1% Platelets are a rich source of inflammatory cytokines, and play
ADA (n) (4) (13) an active role in inflammation while having regulatory effects
NLR 35.6% 44.8% on immune cells.10
PLR 17.1% 30.6% The present study demonstrated that NLR and PLR showed a
UST (n) (1) (4) close mutual correlation, both in PsV and PsA patients, and that
NLR 0.7% 22.3%
these biomarkers were positively associated, at least partly, with
PLR 4.2% 22.9%
systemic inflammation represented by serum CRP levels. Other
n indicates the number of patients. One patient in the PsV-ADA group
reports have also revealed that NLR is significantly increased in
had no record of PASI score after treatment and was excluded from the psoriatic patients in accordance with an increase of PASI
analysis. For both neutrophil–lymphocyte ratio (NLR) and platelet–lym- scores.13,14,17 Although one report failed to find an association
phocyte ratio (PLR), the reduction rate was calculated in each patient
between NLR and PASI scores,18 our study suggests that this dis-
as the percentage of reduction at 3 months of treatment in comparison
with those at baseline. †PASI score improvement was assessed at 3– crepancy may be due to the inclusion criteria of PsA patients with
4 months of treatment with infliximab (IFX), adalimumab (ADA) or ustek- low PASI scores in their study. With regard to PLR, two studies
inumab (UST). ‡According to the Wilcoxon signed rank test, the differ- failed to find its increase in psoriatic patients,14,17 but a positive
ence was not statistically significant between the subgroups in
each row. PsA, psoriatic arthritis; PsV, plaque-type psoriasis (psoriasis correlation was observed between PLR and PASI scores,14 similar
vulgaris). to our study. According to Yurtdas et al.,17 PLR, as well as NLR
and CRP, was correlated with predictors of subclinical atheroscle-
small, but a significant increase was observed at some time rosis, aortic velocity propagation and carotid intima-media thick-
points after treatment with infliximab or adalimumab, particu- ness in psoriatic patients. This is interesting in light of the evidence
larly in PsA patients. of platelet activation in psoriatic patients,19 and platelet activation
We then divided the patients according to the improvement may be linked to cardiovascular comorbidities associated with
of PASI score at the initial evaluation point of 3–4 months. The psoriasis.20
high-responder subgroup that achieved over 75% reduction Another important finding is the time-course decrease of NLR
from baseline in PASI score (PASI-75) basically showed a and PLR after treatment with biologics. To the best of our knowl-
numerically better reduction rate of both NLR and PLR at edge, this is the first study showing the effect of biologics on
16 weeks compared with the low-responder subgroup that these novel biomarkers in psoriatic patients, and this decrease is
failed to achieve PASI-75, especially following treatment with clearly linked to inhibition of systemic inflammation represented
infliximab and adalimumab, although the differences were not by CRP. Because there is no validated consensus regarding the
statistically significant possibly because of the small number of normal range or cut-off values for NLR and PLR, these biomark-
patients included (Table 5). ers may be better used to monitor the effect of treatment or to
evaluate subclinical inflammation after treatment in the same
patients. Indeed, treatment responders in this study showed a
DISCUSSION
trend for a better reduction of NLR and PLR, especially by inflix-
The most characteristic aspect of our study is the retrospective imab and adalimumab. Recently, Erek Toprak et al.21 reported
enrollment of Japanese psoriatic patients receiving biologics. that phototherapy for 3 months did not affect NLR in psoriatic
This excludes a majority of patients who can be controlled by patients. This is probably due to the presence of residual inflam-
topical therapies alone.1 Furthermore, we examined PsV and mation, as serum CRP levels still remained elevated after pho-
PsA patients separately, with a total number of patients suffi- totherapy.15,22 Of note, tumor necrosis factor (TNF)-a
ciently large enough for analysis. NLR, PLR and MPV were antagonists, infliximab and adalimumab showed greater CRP
simultaneously compared with CRP for the first time, and were inhibition than the interleukin (IL)-12/23p40 antagonist ustek-
also followed time-dependently after treatment with biologics. inumab.4 In contrast, both NLR and PLR decreased promptly

8 © 2017 Japanese Dermatological Association


and significantly after treatment with any of these three biolog-
ics, although the average reduction rate was numerically smaller
for ustekinumab than those for infliximab and adalimumab. NLR
and PLR may therefore respond more sharply than CRP to ame-
lioration of systemic inflammation. Alternatively, inhibition of
CRP by TNF-a antagonists may be primarily dependent on their
direct effect on TNF-a, because TNF-a is required for the synthe-
sis of CRP. In one study,23 the activity of neutrophils decreased
after treatment of psoriatic patients with biologics, and no differ-
ence was found between infliximab and ustekinumab.
This study also evaluated the implication of MPV as another
biomarker related to platelet activation, and indicated for the
first time that MPV was significantly lower in PsA patients than
in PsV patients. Furthermore, a moderate negative correlation
was observed between CRP levels and MPV values in PsA
patients, and MPV values increased, if at all, after treatment
with infliximab or adalimumab. These observations support the
hypothesis that pro-inflammatory cytokines and acute-phase
reactants may interfere with megakaryopoiesis and, alterna-
tively, large platelets may be intensively consumed at sites of
inflammation.24 Conversely, however, some studies failed to
find any significant benefits of MPV measurement in psoriatic
patients.21,25 More surprisingly, conflicting data have been
reported arguing that MPV was higher in psoriatic patients than
in healthy controls,11,20,26,27 and was positively correlated with
PASI scores.11,20,27 MPV was even higher in PsA patients com-
pared with that in PsV patients.11 MPV may be elevated in an
inflammatory situation by the release of immature platelets
from bone marrow as a result of coagulopathy or increased
platelet turnover,20,26 but the exact reason for the discrepancy
among clinical studies is not known. Interestingly, studies on
the measurement of MPV in patients with rheumatoid arthritis
also show conflicting results.28–31 Overall, unlike PLR, MPV
may be an inappropriate indicator of inflammation32 or platelet
activation25 by itself. MPV also requires standardization of its
measurement system and protocols to overcome possible
technical issues.29,32 Further sophisticated evaluation is neces-
sary to characterize this biomarker.
A limitation of this study is that comorbidities of the patients
and medications for these comorbidities, which may influence
inflammatory biomarkers, were not considered. Serum CRP
was measured by a conventional method. Additionally, this
study was performed in a retrospective fashion in a single facil-
ity, and prospective evaluation with a larger number of patients
is warranted.
In conclusion, we demonstrated that both NLR and PLR
have a positive association with PASI scores and/or arthritis in
patients with psoriasis, and that they reflect systemic inflam-
mation, at least partly. MPV appears to be negatively associ-
ated with the presence of arthritis, but its correlation with
inflammation is less clear. NLR and PLR can also be utilized to
monitor the disease course, particularly to evaluate inhibition of
systemic inflammation after treatment. These biomarkers are
simple, easily calculated in routine practise and cost-effective.
More studies are necessary to integrate these novel biomark-
ers into other conventional biomarkers for proper management
of psoriatic patients.
© 2017 Japanese Dermatological Association
NLR, PLR and MPV in psoriatic patients
CONFLICT OF INTEREST: A. A., Y. U. and H. N. have
received consultant and speaker fees from AbbVie Japan, Mitsubishi
Tanabe Pharma, Eisai and Janssen Pharmaceuticals.
REFERENCES
1 Boehncke WH, Scho € n MP. Psoriasis. Lancet 2015; 386: 983–994.
2 Reich K. The concept of psoriasis as a systemic inflammation:
implications for disease management. J Eur Acad Dermatol Vener-
eol 2012; 26(Suppl 2): 3–11.
3 Prey S, Paul C, Bronsard V et al. Assessment of risk of psoriatic
arthritis in patients with plaque psoriasis: a systematic review of the
literature. J Eur Acad Dermatol Venereol 2010; 24(Suppl 2): 31–35.
4 Asahina A, Umezawa Y, Yanaba K, Nakagawa H. Serum C-reactive
protein levels in Japanese patients with psoriasis and psoriatic
arthritis: long-term differential effects of biologics. J Dermatol 2016;
43: 779–784.
5 Proctor MJ, Morrison DS, Talwar D et al. A comparison of
inflammation-based prognostic scores in patients with cancer. A
Glasgow Inflammation Outcome Study. Eur J Cancer 2011; 47:
2633–2641.
6 C ß icß ek G, Acß ıkgoz SK, Bozbay M et al. Neutrophil-lymphocyte ratio
and platelet-lymphocyte ratio combination can predict prognosis in
patients with ST-segment elevation myocardial infarction undergo-
ing primary percutaneous coronary intervention. Angiology 2015;
66: 441–447.
7 Buyukkaya E, Karakas MF, Karakas E et al. Correlation of neutrophil
to lymphocyte ratio with the presence and severity of metabolic
syndrome. Clin Appl Thromb Hemost 2014; 20: 159–163.
8 Akboga MK, Canpolat U, Yuksel M et al. Platelet to lymphocyte
ratio as a novel indicator of inflammation is correlated with the
severity of metabolic syndrome: a single center large-scale study.
Platelets 2016; 27: 178–183.
9 Qin B, Ma N, Tang Q et al. Neutrophil to lymphocyte ratio (NLR)
and platelet to lymphocyte ratio (PLR) were useful markers in
assessment of inflammatory response and disease activity in SLE
patients. Mod Rheumatol 2016; 26: 372–376.
10 Uslu AU, Ku €cß u
€k A, S
ß ahin A et al. Two new inflammatory markers
associated with disease activity score-28 in patients with rheuma-
toid arthritis: neutrophil-lymphocyte ratio and platelet-lymphocyte
ratio. Int J Rheum Dis 2015; 18: 731–735.
11 Canpolat F, Akpinar H, Eskiog  lu F. Mean platelet volume in psoriasis
and psoriatic arthritis. Clin Rheumatol 2010; 29: 325–328.
12 Dowlatshahi EA, van der Voort EA, Arends LR, Nijsten T. Markers of
systemic inflammation in psoriasis: a systematic review and meta-
analysis. Br J Dermatol 2013; 169: 266–282.
13 Sen BB, Rifaioglu EN, Ekiz O, Inan MU, Sen T, Sen N. Neutrophil to
lymphocyte ratio as a measure of systemic inflammation in psoria-
sis. Cutan Ocul Toxicol 2014; 33: 223–227.
14 Kim DS, Shin D, Lee MS et al. Assessments of neutrophil to lym-
phocyte ratio and platelet to lymphocyte ratio in Korean patients
with psoriasis vulgaris and psoriatic arthritis. J Dermatol 2016; 43:
305–310.
15 Coimbra S, Oliveira H, Reis F et al. C-reactive protein and leucocyte
activation in psoriasis vulgaris according to severity and therapy. J
Eur Acad Dermatol Venereol 2010; 24: 789–796.
16 Lynde CW, Poulin Y, Vender R, Bourcier M, Khalil S. Interleukin
17A: toward a new understanding of psoriasis pathogenesis. J Am
Acad Dermatol 2014; 71: 141–150.
17 Yurtdasß M, Yaylali YT, Kaya Y, Ozdemir M, Ozkan I, Aladag  N. Neu-
trophil-to-lymphocyte ratio may predict subclinical atherosclerosis in
patients with psoriasis. Echocardiography 2014; 31: 1095–1104.
18 Ataseven A, Bilgin AU, Kurtipek GS. The importance of neutrophil
lymphocyte ratio in patients with psoriasis. Clin Med Res 2014; 3:
40–43.
19 Tamagawa-Mineoka R, Katoh N, Kishimoto S. Platelet activation in
patients with psoriasis: increased plasma levels of platelet-derived
9
A. Asahina et al.
microparticles and soluble P-selectin. J Am Acad Dermatol 2010;
62: 621–626.
20 Chandrashekar L, Rajappa M, Revathy G et al. Is enhanced platelet
activation the missing link leading to increased cardiovascular risk
in psoriasis? Clin Chim Acta 2015; 446: 181–185.
21 Erek Toprak A, Ozlu E, Uzuncakmak TK, Yalcınkaya E, Sogut S,
Karadag AS. Neutrophil/lymphocyte ratio, serum endocan, and nes-
fatin-1 levels in patients with psoriasis vulgaris undergoing pho-
totherapy treatment. Med Sci Monit 2016; 22: 1232–1237.
22 Strober BE, Poulin Y, Teller C, Wang Y, Williams DA, Goldblum OM.
Changes in C-reactive protein in patients with moderate-to-severe
psoriasis switched to adalimumab therapy after suboptimal
response to etanercept, methotrexate or phototherapy. J Eur Acad
Dermatol Venereol 2014; 28: 1701–1706.
23 Yamanaka K, Umezawa Y, Yamagiwa A et al. Biologic therapy
improves psoriasis by decreasing the activity of monocytes and
neutrophils. J Dermatol 2014; 41: 679–685.
24 Gasparyan AY, Sandoo A, Stavropoulos-Kalinoglou A, Kitas GD.
Mean platelet volume in patients with rheumatoid arthritis: the effect
of anti-TNF-a therapy. Rheumatol Int 2010; 30: 1125–1129.
25 Saleh HM, Attia EA, Onsy AM, Saad AA, Abd Ellah MM. Platelet
activation: a link between psoriasis per se and subclinical
10
atherosclerosis–a case-control study. Br J Dermatol 2013; 169: 68–
75.
26 Karabudak O, Ulusoy RE, Erikci AA, Solmazgul E, Dogan B, Har-
manyeri Y. Inflammation and hypercoagulable state in adult psoriatic
men. Acta Derm Venereol 2008; 88: 337–340.
27 Kim DS, Lee J, Kim SH, Kim SM, Lee MG. Mean platelet volume is
elevated in patients with psoriasis vulgaris. Yonsei Med J 2015; 56:
712–718.
28 Yazici S, Yazici M, Erer B et al. The platelet indices in patients with
rheumatoid arthritis: mean platelet volume reflects disease activity.
Platelets 2010; 21: 122–125.
29 Kim DA, Kim TY. Controversies over the interpretation of changes of
mean platelet volume in rheumatoid arthritis. Platelets 2011; 22: 79–
80.
30 Kisacik B, Tufan A, Kalyoncu U et al. Mean platelet volume (MPV)
as an inflammatory marker in ankylosing spondylitis and rheumatoid
arthritis. Joint Bone Spine 2008; 75: 291–294.
31 Tekeog  lu _I, Gu
€rol G, Harman H, Karakecß e E, C ß i _IH. Overlooked
ß iftc
hematological markers of disease activity in rheumatoid arthritis. Int
J Rheum Dis 2016; 19: 1078–1082.
32 Leader A, Pereg D, Lishner M. Are platelet volume indices of clinical
use? A multidisciplinary review Ann Med 2012; 44: 805–816.
© 2017 Japanese Dermatological Association