Dermatology: Practical and Conceptual


Original Article | Dermatol Pract Concept. 2023;13(4):e2023284 1

Dermoscopy in the Diagnosis of Mycosis Fungoides: 
Can it Help?

Sarah Hamdy Soliman1, Wafaa Mohammad Ramadan1, Amal Ahmed Elashmawy1, 
Sameh Sarsik1, Aimilios Lallas2

1 Dermatology Department, Faculty of Medicine, Tanta University, Tanta, Egypt

2 First Department of Dermatology, Aristotle University, Thessaloniki, Greece

Key words: mycosis fungoides, diagnosis, dermoscopy, histopathology, immunohistochemistry

Citation: Soliman SH, Ramadan WM, Elashmawy AA, Sarsik S, Lallas A. Dermoscopy in the Diagnosis of Mycosis Fungoides: Can it Help? 

Dermatol Pract Concept. 2023;13(4):e2023284. DOI: https://doi.org/10.5826/dpc.1304a284

Accepted: June 10, 2023; Published: October 2023

Copyright: ©2023 Soliman et al. This is an open-access article distributed under the terms of the Creative Commons Attribution-
NonCommercial License (BY-NC-4.0), https://creativecommons.org/licenses/by-nc/4.0/, which permits unrestricted noncommercial use, 
distribution, and reproduction in any medium, provided the original authors and source are credited.

Funding: None.

Competing Interests: None.

Authorship: All authors have contributed significantly to this publication.

Corresponding Author: Sarah Hamdy Soliman, MSc, Dermatology Department, Faculty of Medicine, Tanta University, Tanta, 31527, 
Egypt. E-mail: sara.hamdy@med.tanta.edu.eg

Introduction: The diagnosis of mycosis fungoides (MF) is challenging since it can mimic a variety of 
benign skin conditions. Multiple biopsies for histopathologic and immunohistochemical examination 
are required to diagnose MF. Dermoscopy is an affordable, non-invasive device with expanding indi-
cations in dermatology,

Objectives: To investigate the dermoscopic morphology of MF variants and assess the correlation 
between dermoscopic criteria, histopathologic, and immunohistochemical findings,

Methods: We included 88 patients with several MF variants (classic, hypopigmented, hyperpigment-
ed, poikilodermatous, erythrodermic, and folliculotropic). The diagnosis was histopathologically and 
immunohistochemically confirmed. Dermoscopic findings were collected, statistically analyzed, and 
correlated with the results of histopathology and immunohistochemistry,

Results: All patients had MF diagnosis in H&E-stained sections. The majority revealed positive stain-
ing with CD3, 4, 8 and negative CD7. Orange-red areas of discoloration, short linear, and spermato-
zoa like blood vessels are the most frequent dermoscopic findings, while an analysis per MF variant 
was also performed. The frequently observed dermoscopic structures in classic MF were patchy whit-
ish scales, dotted, short linear vessels, and spermatozoa-like vessels,

Conclusions: Dermoscopy reveals a repetitive dermoscopic pattern in MF (non-homogenous pink to 
erythematous background, patchy areas of orange discoloration, patchy whitish scales, dotted and 
short linear blood vessels with some variations according to the clinical variant.

ABSTRACT



2 Original Article | Dermatol Pract Concept. 2023;13(4):e2023284

Introduction

Cutaneous T-cell lymphoma is the most frequent extra-nodal 

non-Hodgkin lymphoma. It comprises a wide variety of 

types, the most frequent being mycosis fungoides (MF) and 

Sezary syndrome (SS) [1]. MF is characterized by a significant 

variability of clinical manifestations. Classically, it evolves 

in 4 sequential stages (patch, plaque, tumor, and erythro-

dermic), but several less frequent clinical variants do exist 

(eg hypopigmented, poikilodermatous) [2]. The  diagnosis of 

MF is very often challenging and necessitates the integration 

of information acquired by clinical histopathologic and im-

munohistochemical assessment [1].

In addition to its usefulness in the assessment of pig-

mented cutaneous lesions [3], dermoscopy gains a role in the 

diagnosis of non-pigmented skin diseases as well, including 

inflammatory and infectious diseases [4–6]. Allowing the vi-

sualization of vascular structures, scales, color variegations, 

follicular abnormalities, and other structures, dermoscopy 

provides a link between clinical evaluation and dermatopa-

thology and is considered an essential device for dermatol-

ogists in their daily practice [5]. MF has been reported to 

display a different dermoscopic pattern compared to chronic 

eczema, which represents the main differential diagnosis and 

dermoscopy was suggested to facilitate MF diagnosis by 

guiding to a more accurate biopsy for histopathologic exam-

ination and immunohistochemistry [6].

Objectives

This study aimed to investigate the dermoscopic features of 

MF, to assess the distribution of these features per clinical 

variant and to and to correlate these features with histo-

pathologic and immunohistochemical alterations.

Methods

This cross-sectional study included a total of 88 patients 

with several clinical variants of MF whose diagnosis was 

based on the clinical presentation of skin lesions and con-

firmed by histopathological and immunohistochemical ex-

aminations. After obtaining the research ethics committee 

approval from our faculty of medicine, we included all pa-

tients diagnosed with MF at the outpatient clinic, cancer in-

stitute, and private clinics, in the period between November 

2013 and May 2016.

Inclusion criteria were the following:

1. A definitive diagnosis of MF established histopatholog-

ically and immunohistochemically. A combination of at 

least 3 of the following histopathological criteria is the 

basis for diagnosis of the studied cases: atypical lympho-

cytes, epidermotropism with Pautrier microabscesses, 

some atypical lymphocytes exhibited arrangement along 

the epidermal side of the dermo-epidermal junction 

and atypical dermal infiltrate with a variable density as 

proposed by Smoller et al [7].  Immunohistochemically 

stained tissues using CD3,4,8, and 7 were performed 

and assessed according to the percentage of stained 

cells considering the positive stain if the percentage ≥ 

30% whereas percent ≤ 30% is considered negative [8]. 

CD3  positive staining with positive CD4 or CD8, and 

negative staining for CD7 were the immunohistochemi-

cal features of most of the included cases.

2. The presence of clinical manifestations (even mild) at 

the time of inclusion. Persistent erythematous patches, 

plaques, and or nodules with necrosis and ulceration were 

the criteria of classic MF. While multiple hypopigmented 

well- defined macules on the trunk with prevalence on 

the lower back were the characteristic for hypopigmented 

variant, hyperpigmented variant represented black col-

ored hyperpigmented patches without scaling nor atrophy 

or telangiectasias. Localized or diffuse patches comprised 

of atrophy, telangiectasia, hypo or hyperpigmentation 

were the clinical criteria in the poikilodermatous variant. 

Regarding erythrodermic variant, there was generalized 

erythroderma while acneiform lesions, cystic comedones, 

alopecia, and follicular papules were the most frequent 

presentations of folliculotropic variant.

3. No prior treatment or no treatment within the last 

6 months. All patients who accepted to participate in the 

study signed written consent. Exclusion criteria were the 

application of any topical treatment, phototherapy, or 

systemic treatment within the last 6 months, pregnancy 

or lactation, and the co-existence of other dermatological 

diseases.

The patients have been classified into the following 

 clinical variants; classic patch stage MF; the commonest vari-

ant with gradual progression from patches to more infiltrated 

plaques and eventually tumors, hypopigmented MF which is 

rare and characterized by solely hypopigmented patches or 

in combination with erythematous patches or plaques, poi-

kilodermatous MF identified by localized or diffuse large 

macules and patches of mottled hypopigmentation and hy-

perpigmentation with atrophy and prominent telangiectasia, 

folliculotropic MF; a rare variant frequently presented by 

folliculotropic infiltrates usually involving the head and neck 

area, erythrodermic MF which is infrequent with intractable 

progressive erythroderma, and hyperpigmented MF; an ex-

tremely rare variant with hyperpigmented plaques without 

poikiloderma.



Original Article | Dermatol Pract Concept. 2023;13(4):e2023284 3

All patients were subjected to a thorough dermatological 

examination to evaluate the morphology and distribution of 

skin lesions and to exclude any co-existent dermatological 

disease. The dermoscopic examination was performed using 

Dermlite II PRO HR (3Gen) using polarized light, magnifi-

cation x10. Digital images of both clinical and dermoscopic 

features of the same MF biopsied lesions were taken by dig-

ital camera Sony: Cyber-Shoot DSC-W690. The selection 

of the dermoscopic variables to be included in the analy-

sis was based on pre-existing literature on dermoscopy of 

MF,  adjusted to a consensus paper on the terminology of 

 dermoscopy in general dermatology [5].

Skin biopsies were taken from all patients; formalin-fixed 

routinely processed paraffin-embedded tissue sections (3-5 µ) 

and prepared on charged glass slides for confirmation of the 

diagnosis of MF using light microscopy (Leica DM 500). Im-

munohistochemical staining of tissue sections was conducted 

using antibodies against CD3, CD4, CD8, and CD7 and then 

repeated on a second freshly obtained biopsies after recruit-

ing study participants. All biopsies were obtained from the 

same sites of dermoscopic examination.  Ethical Committee 

in our institution approved the study.

Statistical Analysis

We calculated the sample size according to Raosoft and all 

statistical calculations were done using SPSS (statistical pack-

age for the social science version 26.00) statistical program. 

at 0.05, 0.01, and 0.001 level of probability. Qualitative (cat-

egorical) data were presented by frequency and percentage 

was done using chi-square, Pearson linear correlation coef-

ficient (r) was estimated to show the relationship between 

histopathological features (atypical lymphocytes or epider-

motropism with Pautrier micro-abscesses or alignment of 

atypical lymphocytes or atypical dermal infiltrate) and der-

moscopic features [9].

Results

Clinical Results

Out of 88 patients, 48 patients (54.6%) had classic MF, among 

them 32 (36.4%) presented with the patch stage, and 16 

(18.2%) presented with both patch the tumor stages. Twenty 

patients (22.8%) had hypopigmented MF, 8 patients (9.1%) 

had poikilodermatous MF, and the study also included 12 pa-

tients; 4 patients (4.5%) from each of the following: follicu-

lotropic, erythrodermic, and hyperpigmented MF (Table 1).

The patients were 44 males (50%) and 44 females 

(50%). Their ages varied from 10-65 years. The duration 

of MF lesions ranged from 3 months-20 years. The num-

ber of patients who had previous treatment was 32 (36.4%), 

those patients were either recurrences or who stopped their 

treatment due to side effects but still have lesions, while the 

other 56 (63.6%) received no treatment (Table 2).

Histopathological Examination

Histopathological examination of tissue sections stained 

with H&E from MF lesions revealed; all patients (100%) 

had atypical lymphocytes, 84 patients (95%) had epidermot-

ropism escorted with pautrier micro-abscesses, some atyp-

ical lymphocytes showed arrangement along the epidermal 

side of the dermo-epidermal junction in 48 patients (54.5%) 

and all patients (100%) had atypical dermal infiltrate with 

variable density (Figure 1). Overall, all patients presented 

with the typical morphological histopathological criteria 

recommended by Smoller et al [7].

Immunohistochemical Examination

To begin with CD3, all the patients (100%) revealed positive 

staining. About CD4, 76 (86.4%) patients showed positivity 

Table 1. Classification of the patients according 
to mycosis fungoides clinical variant.

Mycosis fungoides variant No. %

Classic mycosis fungoides 
(patch stage)

32 36.4

Classic mycosis fungoides  
(patch/ tumor stages)

16 18.2

Hypopigmented mycosis fungoides 20 22.8

Poikelodermatic mycosis fungoides 8 9.1

Follicultropic mycosis fungoides 4 4.5

Erythrodermic mycosis fungoides 4 4.5

Hyperpigmented mycosis 
fungoides

4 4.5

Table 2. Distribution of the patients according 
to demographic data.

No. %

Sex

  Male 44 50.0

  Female 44 50.0

Age/ year

  ≤30 32 36.4

  31 – 40 28 31.8

  >40 28 31.8

Duration/ year

  ≤ 1 yrs 32 36.4

  > 1 yrs – 6 yrs 48 54.5

  > 6 yrs 8 9.1

Previous treatment

  No 56 63.6%

  Yes 32 36.4%



4 Original Article | Dermatol Pract Concept. 2023;13(4):e2023284

a. A pink to the erythematous background was found in 

72 patients (81.8%), a non-homogenous background in 

68 patients (77.3%), and patchy areas of orange discol-

oration in 68 patients (77.3%).

Vascular Pattern:

b. In 64 patients (72.7%) dermoscopy revealed dotted and 

short linear vessels, while spermatozoa-like vessels were 

detected in 44 patients (50%) and arborizing vessels in 

16 patients (18.2%).

Scales:

c. Fifty-six patients (63.6%) had patchy whitish scales, 

while perifollicular and interfollicular scaling was found 

in 4 patients (4.5%).

Other Structures:

d. Blue-gray and dark globules were seen in 32 patients 

(36.4%), 20 patients (22.7%) showed focal white struc-

tureless areas in 20 patients (22.7%), while white glob-

ules were found in 16 patients (18.2%), and red globules 

in 12 patients (13.6%).

The analytic results of the dermoscopic evaluation according 

to the MF variant are shown in (Table 4). In all of the 48 patients 

with classic MF, dermoscopy revealed a pink to erythematous 

background with non-homogeneous distribution. The most fre-

quently observed dermoscopic structures were patchy whitish 

scales (44/48, 91.7%), dotted and short linear vessels (44/48, 

91.7%), spermatozoa-like vessels (32/48, 66.7%), and patchy 

areas of orange discoloration, (24/48, 50%) (Figure2).

Concerning the most frequent findings in other 

rare MF variants, the erythrodermic variant exhibited 

while 12 (13.6%) were with negative staining. 65 (73.9%) 

of our patients had positive CD8 stains and 23 (26.1%) were 

negative. 84 (95%) patients had a loss of CD7 expression. 

Finally, the CD4/CD8 ratio was more than 4 in 72 (81.9%) 

of the studied patients.

Dermoscopic Examination

The analytic results of the dermoscopic evaluation are shown 

in Table 3. The most frequent dermoscopic features in the to-

tal study population were (by category):

Background Changes:

Figure 1. H&E of skin biopsy from classic mycosis fungoides (x10); 

dermal infiltrate with atypical lymphocytes with alignment at dermo 

epidermal junction, epidermotropism with Pautrier microabscess.

Table 3. Dermoscopic findings of the studied mycosis fungoides patients.

No. %

Background changes Non-homogenous background 68 77.3

Pink to erythematous background 72 81.8

Patchy areas of orange discoloration 68 77.3

Vascular pattern Dotted blood vessels 64 72.7

Short linear blood vessels 64 72.7

Spermatozoa like blood vessels 44 50.0

Arborizing blood vessels 16 18.2

Scales Patchy whitish scales 56 63.6

Perifollicular and interfollicular scaling 4 4.5

Globules Blue-gray globules 32 36.4

Dark globules 32 36.4

Patchy red globules 12 13.6

Pigmentation Multifocal pigmentation 4 4.5

Foggy whitish areas 20 22.7

Hypopigmented areas 20 22.7

Pearly white globules 16 18.2

Structures Comedo like openings 4 4.5



Original Article | Dermatol Pract Concept. 2023;13(4):e2023284 5

Ta
b

le
 4

. R
el

at
io

n 
be

tw
ee

n 
cl

in
ic

al
 v

ar
ia

nt
s 

of
 m

yc
os

is
 f

un
go

id
es

 a
nd

 d
er

m
os

co
pi

c 
fe

at
ur

es
.

D
er

m
o

sc
o

p
ic

 
fe

at
u

re
s

C
la

ss
ic

 M
F 

(p
at

ch
 s

ta
g

e)
 (

n
=

48
)

O
th

er
 M

F 
va

ri
an

ts
 (

n
=

40
)

C
la

ss
ic

- 
p

at
ch

 s
ta

g
e 

(n
=

32
)

C
la

ss
ic

- 
p

at
ch

 
st

ag
e 

&
  

tu
m

o
r 

(n
=

16
)

To
ta

l
%

Er
yt

h
ro

d
er

m
ic

(n
=

4)

H
yp

o
p

ig
m

en
te

d

(n
=

20
)

Po
ik

el
o

d
er

m
at

ic

(n
=

8)
Fo

lli
cu

lo
tr

-o
p

ic
 

(n
=

4)

H
yp

er
p

ig
-

m
en

te
d

 
(n

=
4)

To
ta

l
%

P

N
on

-h
om

og
en

ou
s 

B
ac

kg
ro

un
d

10
0%

10
0%

48
10

0.
0

10
0%

0.
00

%
10

0%
10

0%
10

0%
20

50
.0

<0
.0

01

Pa
tc

hy
 w

hi
ti

sh
 s

ca
le

s
87

.5
0%

10
0%

44
91

.7
10

0%
0.

00
%

0.
00

%
10

0%
10

0%
12

30
.0

<0
.0

01

Pe
ri

fo
lli

cu
la

r 
an

d 
in

te
rf

ol
lic

ul
ar

 s
ca

lin
g

0.
00

%
0.

00
%

0
0.

0
0.

00
%

0.
00

%
0.

00
%

10
0%

0.
00

%
4

10
.0

0.
02

5

Pi
nk

 t
o 

er
yt

he
m

at
ou

s 
ba

ck
gr

ou
nd

10
0%

10
0%

48
10

0.
0

10
0%

20
.0

0%
10

0%
10

0%
10

0%
24

60
.0

<0
.0

01

Pa
tc

hy
 a

re
as

 o
f 

or
an

ge
 d

is
co

lo
ra

ti
on

87
.5

0%
10

0%
44

91
.7

10
0%

20
.0

0%
10

0%
10

0%
10

0%
24

60
.0

<0
.0

01

B
lu

e 
gr

ay
 g

lo
bu

le
s

37
.5

0%
75

.0
0%

24
50

.0
0.

00
%

0.
00

%
0.

00
%

10
0%

10
0%

8
20

.0
0.

00
4

D
ar

k 
gl

ob
ul

es
37

.5
0%

75
.0

0%
24

50
.0

0.
00

%
0.

00
%

0.
00

%
10

0%
10

0%
8

20
.0

0.
00

4

Pa
tc

hy
 r

ed
 g

lo
bu

le
s

12
.5

0%
25

.0
0%

8
16

.7
10

0%
0.

00
%

0.
00

%
0.

00
%

0.
00

%
4

10
.0

0.
36

4

M
ul

ti
fo

ca
l 

pi
gm

en
ta

ti
on

0.
00

%
0.

00
%

0
0.

0
0.

00
%

0.
00

%
0.

00
%

0.
00

%
10

0%
4

10
.0

0.
02

5

Fo
gg

y 
w

hi
ti

sh
 a

re
as

0.
00

%
0.

00
%

0
0.

0
0.

00
%

10
0%

0.
00

%
0.

00
%

0%
20

50
.0

<0
.0

01

C
om

ed
o 

lik
e 

op
en

in
gs

0.
00

%
0.

00
%

0
0.

0
0.

00
%

0.
00

%
0.

00
%

10
0%

0.
00

%
4

10
.0

0.
02

5

H
yp

op
ig

m
en

te
d 

ar
ea

s
0.

00
%

0.
00

%
0

0.
0

0.
00

%
10

0%
0.

00
%

0.
00

%
0.

00
%

20
50

.0
<0

.0
01

Pe
ar

ly
 w

hi
te

 g
lo

bu
le

s
12

.5
0%

0.
00

%
4

8.
3

0.
00

%
60

.0
0%

0.
00

%
0.

00
%

0.
00

%
12

30
.0

0.
00

9

D
ot

te
d 

bl
oo

d 
ve

ss
el

s
87

.5
0%

10
0%

44
91

.7
10

0%
0.

00
%

10
0%

10
0%

10
0%

20
50

.0
<0

.0
01

Sh
or

t 
lin

ea
r 

bl
oo

d 
ve

ss
el

s
87

.5
0%

10
0%

44
91

.7
10

0%
0.

00
%

50
.0

0%
10

0%
10

0%
20

50
.0

<0
.0

01

Sp
er

m
at

oz
oa

 li
ke

 
bl

oo
d 

ve
ss

el
s

50
.0

0%
10

0%
32

66
.7

10
0%

0.
00

%
50

.0
0%

10
0%

0.
00

%
12

30
.0

0.
00

1

A
rb

or
iz

in
g 

bl
oo

d 
ve

ss
el

s
0.

00
%

0.
00

%
0

0.
0

10
0%

0.
00

%
10

0%
10

0%
0.

00
%

16
40

.0
<0

.0
01

a P 
p-

va
lu

e 
fo

r 
C

hi
-s

qu
ar

e 
te

st
 f

or
 c

om
pa

ri
ng

 b
et

w
ee

n 
th

e 
di

ff
er

en
t 

ca
te

go
ri

es
, s

ig
ni

fi
ca

nt
 a

t 
p-

va
lu

e 
< 

0.
05

, h
ig

hl
y 

si
gn

if
ic

an
t 

at
 p

-v
al

ue
 <

 0
.0

01
.



6 Original Article | Dermatol Pract Concept. 2023;13(4):e2023284

Conclusions

Our study provides novel insights into the dermoscopic mor-

phology of MF according to the clinical variant. Overall, our 

results are consistent with pre-existing evidence [10].

Classic patch stage MF was dermoscopically typified by a 

pink to erythematous background, patchy orange-yellowish 

color, patchy whitish scales, dotted and fine short linear ves-

sels. These results are in perfect agreement with the previous 

study by Lallas et al [10]. However, a pink to erythematous 

background, which results from dilatation of dermal blood 

vessels, was found to be not specific for MF, since it is also 

frequent in psoriasis, chronic dermatitis, and lichen pla-

nus [11].

Patchy areas of orange discoloration which represent se-

rum oozing from erosions [12], were observed in the patch 

stage of classic MF. This feature was previously suggested as 

a significant criterion for the differential diagnosis between 

MF and chronic dermatitis [10]. In our study, we did observe 

this feature in all clinical variants of MF without significant 

differences among them.

Dotted and short linear blood vessels, that correlate with 

vertical vessels [12] and tumor angiogenesis [11] respectively, 

were observed mostly in patients with patch stage of classic 

MF. Also, the peculiar vascular structure consisting of a dot-

ted and a linear vessel (spermatozoa like blood vessels) were 

observed in 66.7% of classic MF patients, and 30% of other 

rare MF variants including erythrodermic, folliculotropic, 

and poikilodermatous variants. This feature was previously 

described as a dermoscopic characteristic of classic MF, but 

it has not been reported in other MF variants.

We detected patchy whitish scales, that correspond to 

areas of parakeratosis [11], in classic, erythrodermic, fol-

liculotropic, and hyperpigmented variants of MF. This is in 

line with the study by Saleh MA and Abdel Halim DM who 

non-homogenous pink to erythematous background (4/4, 

100%), dotted, short linear, spermatozoa like and arboriz-

ing blood vessels (4/4, 100%). (Figure 3) The hypopigmented 

variant was characterized by foggy whitish areas (20/20, 

100%), hypopigmented areas (20/20, 100%), and pearly 

white globules (12/20, 60%). The poikilodermatous variant 

revealed patchy areas of orange discoloration (8/8, 100%), 

dotted (8/8, 100%), and arborizing blood vessels (8/8, 

100%), folliculotropic variant (Figure 4) presented perifol-

licular and interfollicular scaling (4/4, 100%), comedo-like 

openings (4/4, 100%), dotted, short linear, spermatozoa like 

(Figure 5) and arborizing blood vessels (4/4, 100%), hyper-

pigmented variant showed blue-gray globules (4/4, 100%), 

dark globules (4/4, 100%), dotted and short linear blood 

vessels (4/4, 100%). Regarding the correlation between the 

frequently encountered dermoscopic features of MF with his-

topathological findings (Table 5); non-homogeneous back-

ground, showed a statistically significant moderately positive 

correlation with the presence of atypical dermal infiltrates 

(ADI) (P < 0.001), while showed a weak positive correla-

tion with atypical lymphocytes (AL), epidermotropism and 

Pautrier microabscesses (E and PMA) and alignment of atyp-

ical lymphocytes (AAL) with high statistical significance ( P 

0.003, 0.33 and 0.006, respectively). Patchy whitish scales 

showed a statistically significant strong positive correlation 

with AL, E, PMA, AAL, and ADI (P < 0.001). Pink to er-

ythematous background showed a strong positive correla-

tion with AL and E, PMA (P < 0.001), on the other hand, 

showed a weak positive correlation to AAL and ADI (P = 

0.03). Patchy areas of orange discoloration and short linear 

blood vessels showed a strong positive correlation with all 

findings which was highly statistically significant (P < 0.001). 

Spermatozoa-like blood vessels showed a strong positive cor-

relation with AL, E, and PMA (P < 0.001) and a moderately 

strong positive correlation to both AAL and ADI (P < 0.001).

Figure 2. (A) Male patient presented with classic mycosis fungoides, showing multiple erythematous scaly plaques on trunk. 

(B) Dermoscopic features of lesions on the trunk; patchy whitish scales (black arrows), spermatozoa like blood vessels 

( yellow arrow), and short linear blood vessels (blue arrow) on pink to erythematous background, polarized light x10.



Original Article | Dermatol Pract Concept. 2023;13(4):e2023284 7

Blue-gray globules were also detected in folliculo-

tropic variant, which is a novel finding in the literature. 

In  addition, the folliculotropic variant was characterized 

by perifollicular, interfollicular scaling, and comedo-like 

openings that represent keratin filled invagination of the 

epidermis [13] and could be found in many scalp and hair 

disorders [14]. Still, they are not specific for MF diagnosis, 

they may differentiate folliculotropic variant from other 

variants.

The hypopigmented variant of MF showed whitish areas 

that represent areas of hypomelanosis [12], and pearly white 

observed these scales in classic and hyperpigmented variants. 

As scales were observed in MF as well as some inflamma-

tory skin diseases, they were not specific for MF diagnosis 

[10,11].

Dermoscopic examination of hyperpigmented MF re-

vealed multifocal pigmentation and dark and blue-gray 

globules that represent epidermal or horny layer melanin 

or blood and melanophages in the papillary dermis respec-

tively [11]. This was observed previously by Saleh and Abdel 

Halim. Thus, it can be concluded that these globules are of 

value in the diagnosis of the hyperpigmented variant.

Figure 3. (A) Female patient presented with erythrodermic mycosis fungoides, with erythema and scales on the face. (B) Dermoscopic 

features of lesions of the face; dotted blood vessels (red arrows), patchy whitish scales (yellow arrow), and spermatozoa like blood vessels 

(blue arrows), all on non-homogenous pink background, polarized light x10. (C) The same patient with erythema on the back of the legs. 

(D) Dermoscopic features of lesions of the legs; dotted blood vessels (yellow arrow), and spermatozoa like blood vessels (white arrow) on 

non-homogenous pink background, polarized light x10.



8 Original Article | Dermatol Pract Concept. 2023;13(4):e2023284

Figure 4. (A) Male patient presented with folliculotropic mycosis fungoides, showing erythematous scaly plaques behind the left ear. 

(B)  Dermoscopic features of the postauricular area; pink to erythematous background with areas of orange discoloration (black ar-

rows), patchy white scales (yellow arrows), arborizing blood vessels (white arrows), spermatozoa like blood vessels (green arrow) and 

short linear blood vessels (blue arrow), polarized light x10. (C) The same patient with hairless plaque on the scalp. (D) Dermoscopic 

features of the scalp lesions; inter and perifollicular scaling (red arrows) and large blue-gray globules (yellow arrow). Polarized light x10.

Figure 5. Dermoscopic features of poikilodermatous mycosis fungoides showing spermatozoa like blood vessels 

(black arrows).



Original Article | Dermatol Pract Concept. 2023;13(4):e2023284 9

Table 5. Correlation between dermoscopic and histopathological features (H&E stain).

Dermoscopic Features Histopathological Features

Atypical 
Lymphocytes

Epidermotropism 
With Pautrier 

Microabscesses

Alignment 
of Atypical 

Lymphocytes
Atypical Dermal 

Infiltrate

R P r P r P r P

Non-homogenous Background 0.312 0.003 0.227 0.033 .289 0.006 0.627 0.000

Patchy whitish scales .889 0.000 .883 0.000 .930 0.000 .889 0.000

Perifollicular and interfollicular 
scaling

.219 0.040 .227 0.033 .308 0.003 .965 0.000

Pink to erythematous background .965 0.000 .963 0.000 .308 0.003 .308 0.003

Patchy areas of orange 
discoloration

1.000 0.000 1.000 0.000 .937 0.000 .937 0.000

Blue-gray globules .646 0.000 .671 0.000 .937 0.000 .937 0.000

Dark globules .646 0.000 .671 0.000 .690 0.000 .690 0.000

Patchy red globules .384 0.000 .399 0.000 .594 0.000 .646 0.000

Multifocal pigmentation .219 0.040 .227 0.033 .482 0.000 .482 0.000

Foggy whitish areas .502 0.000 .521 0.000 .339 0.001 .339 0.001

Hypopigmented areas .219 0.040 .227 0.033 .690 0.000 .690 0.000

Pearly white globules .502 0.000 .521 0.000 .690 0.000 .963 0.000

Dotted blood vessels .446 0.000 .463 0.000 .594 0.000 .646 0.000

Short linear blood vessels .963 0.000 .962 0.000 .855 0.000 .772 0.000

Spermatozoa like blood vessels .963 0.000 .962 0.000 .482 0.000 .446 0.000

Arborizing blood vessels .772 0.000 .802 0.000 .289 0.006 0.627 0.000

dermoscopic images offer a horizontal assessment. Therefore, it 

is always hard to correlate both tools, however, the horizontal 

perception which is added by dermoscopy is complementary to 

histopathology [11]. Dermoscopy is similarly valuable in barely 

visible vascular components and color deviations so it may as-

sociate clinical examination and dermatopathology [17].

Our study has several limitations. First, we included 

only MF cases and, therefore, our results do not provide 

information on the usefulness of described criteria for the 

differential diagnosis between MF and clinical simulators. 

Besides, some rare clinical variants included in the study 

were represented by a small sample, which did not allow a 

more robust statistical analysis to assess differences among 

clinical variants.

In conclusion, our study confirms previous evidence on 

dermoscopy of MF and provides novel information on the 

distribution of the dermoscopic criteria according to the clin-

ical variant, which might be clinically relevant.

References

1. Huang Y, Litvinov IV, Wang Y, Set al. Thymocyte selection 

-associated high mobility group box gene (TOX) is aberrantly 

over-expressed in mycosis fungoides and correlates with poor 

prognosis. Oncotarget. 2014;5(12):4418-4425. DOI 10.18632 

/oncotarget.2031. PMID: 24947046. PMCID: PMC4147334.

globules. These features, that were not reported before in 

any previous study, can differentiate hypopigmented variant 

from other variants of MF.

Regarding the pearly white globules, an interesting ob-

servation in the present study in some patients who previ-

ously received phototherapy; was observed before in the 

active vitiligo lesions [15,16] and it was described by Jha 

et al as “tapioca sago” appearance [15]. We suggest that it 

may be a sign of photodamaging. This relates to Friedland R 

et al who stated that MF patients who were treated with 

phototherapy may acquire an eruption resembling idiopathic 

guttate hypomelanosis as a side effect [16].

Another noteworthy finding of our analysis was that all 

MF variants except hypopigmented MF were characterized 

by a non-homogenous distribution of structures and colors. 

Malvehy et al stated that the non-homogenous distribution 

of structures and colors was not observed previously in the 

dermoscopic examination of benign skin disorders [12]. 

Thus, it could be added as a dermoscopic feature indicative 

of malignancy.

Our analysis revealed a very good correlation between 

histopathologic alterations and dermoscopic structures. The 

combination of dermoscopy with histopathology results in a 

better tumor examination with two distinct but balancing tools. 

While histopathology reflects the vertical sight of the lesions, 



10 Original Article | Dermatol Pract Concept. 2023;13(4):e2023284

10. Lallas A, Apalla Z, Lefaki I, et al. Dermoscopy of early stage myco-

sis fungoides. J Eur Acad Dermatol Venereol. 2013;27(5):617-621. 

DOI: 10.1111/j.1468-3083.2012.04499.x. PMID: 22404051.

11. Saleh MA, Abdel Halim DM. Dermoscopy: an easy, nonin-

vasive tool for distinguishing mycosis fungoides from other 

inflammatory mimics. J Egypt Women’s Dermatologic Soc. 

2014;11(3):215–219.

12. Malvehy J, Puig S, Braun RP, Marghoob AA, Kopf AW. Hand 

book of Dermoscopy. Taylor & Francis, 2006:1–6.

13. Marghoob AA, Malvehy J, Braun RP. Atlas of Dermoscopy. 
 Second. Informa Healthcare, 2012:12–150.

14. Rudnicka L, Olszewska M, Rakowska A. Atlas of trichoscopy: 

Dermoscopy in hair and scalp disease. Springer London Heidel-

berg New York Dordrecht. 2012:3–11.

15. Kumar Jha A, Sonthalia S, Lallas A, Chaudhary RKP. Der-

moscopy in vitiligo: diagnosis and beyond. Int J Dermatol. 

2018;57(1):50-54. DOI: 10.1111/ijd.13795. PMID: 29076154.

16. FFriedland R, David M, Feinmesser M, Fenig-Nakar S, Hodak 

E. Idiopathic guttate hypomelanosis-like lesions in patients 

with mycosis fungoides: a new adverse effect of phototherapy. 

J Eur Acad Dermatol Venereol. 2010;24(9):1026-1030. DOI: 

10.1111/j.1468-3083.2010.03571.x. PMID: 20180893.

17. Bosseila M, Sayed Sayed K, El-Din Sayed SS, Abd El Monaem 

NA. Evaluation of Angiogenesis in Early Mycosis Fungoides 

Patients: Dermoscopic and Immunohistochemical Study. Derma-

tology. 2015;231(1):82-86. DOI: 10.1159/000382124. PMID: 

26044542.

2. Al Hothali GI. Review of the treatment of mycosis fungoides 

and Sézary syndrome: A stage-based approach. Int J Health 

Sci (Qassim). 2013;7(2):220-239. PMID: 24421750. PMCID: 

PMC3883611.

3. AArgenziano G, Soyer HP, Chimenti S, et al. Dermoscopy of pig-

mented skin lesions: results of a consensus meeting via the Inter-

net. J Am Acad Dermatol. 2003;48(5):679-693. DOI: 10.1067 

/mjd.2003.281. PMID: 12734496.

4. Lin MJ, Pan Y, Jalilian C, Kelly JW. Dermoscopic character-

istics of nodular squamous cell carcinoma and keratoacan-

thoma. Dermatol Pract Concept. 2014;4(2):9-15. DOI: 10.5826 

/dpc.0402a02. PMID: 24855567. PMCID: PMC4029247.

5. Lallas A, Giacomel J, Argenziano G, et al. Dermoscopy in gen-

eral dermatology: practical tips for the clinician. Br J Dermatol. 

2014;170(3):514-526. DOI: 10.1111/bjd.12685. PMID: 24266695.

6. Zalaudek I, Argenziano G. Dermoscopy subpatterns of inflam-

matory skin disorders. Arch Dermatol. 2006;142(6):808. DOI: 

10.1001/archderm.142.6.808. PMID: 16785400.

7. Smoller BR, Bishop K, Glusac E, Kim YH, Hendrickson M. Re-

assessment of histologic parameters in the diagnosis of myco-

sis fungoides. Am J Surg Pathol. 1995;19(12):1423-1430. DOI: 

10.1097/00000478-199512000-00009. PMID: 7503364.

8. Florell SR, Cessna M, Lundell RB, et al. Usefulness (or lack 

thereof) of immunophenotyping in atypical cutaneous T-cell 

infiltrates. Am J Clin Pathol. 2006;125(5):727-736. DOI: 

10.1309/3JK2-H6Y9-88NU-AY37. PMID: 16707374.

9. Hardle W, Simar L. Applied Multivariate Statistical Analysis. 

Springer; 2007:430.