TOX Outperforms FOXP3, CD4 and GATA3 in Histopathological Diagnosis of Early Mycosis Fungoides
Mona Mostafa AHMED1 , Abdelmonem Awad HEGAZY2,3 , Ahmed EMBABY4 , Esraa Mohammad NAWWAR5 , Salwan Abdelmonem HEGAZY61 , Hanaa M. IBRAHIM1 , Mai Ahmed GOBRAN1
1Department of Pathology, Zagazig University, Faculty of Medicine, ZAGAZIG, EGYPT
2Department of Human Anatomy and Embryology, Zagazig University, Faculty of Medicine, ZAGAZIG, EGYPT
3Department of Medical Biotechnology, Misr University for Science and Technology (MUST), College of Biotechnology, SIX OF OCTOBER CITY, EGYPT
4Department of Internal Medicine, Faculty of Medicine, Zagazig University, ZAGAZIG, EGYPT
5Department of Dermatology, Al-Ahrar Teaching Hospital and Zagazig University Hospitals, ZAGAZIG, EGYPT
6Zagazig University, Faculty of Medicine, ZAGAZIG, EGYPT
Keywords: Mycosis fungoides, Benign inflammatory dermatoses, Differential histopathological diagnosis, Immunohistochemical markers
Mycosis fungoides (MF) is the most common type of cutaneous lymphoma. The early stage of MF is a difficult diagnostic case, as it
is often confused with many benign inflammatory dermatoses (BID). The study aimed to evaluate the diagnostic utility of TOX, FOXP3, CDD4
and GATA3 in differentiating early stages of MF from histologically overlapping BID lesions.
Material and Method: A retrospective cross-sectional study was performed, in which immunohistochemistry (IHC) was used to evaluate the
expression of TOX, FOXP3, CD4 and GATA3 in formalin-fixed paraffin-embedded (FFPE) sections of skin lesions from 30 cases with BID and
30 patients with early-stage MF.
Results: The association between TOX expression and early-stage MF was statistically significant (P <0.001). TOX had the highest sensitivity of
96.77% and accuracy of 85.71% in diagnosis of MF; followed by CD4 with sensitivity of 85.71% and accuracy of 78.95%; and then, GATA3 with
sensitivity of 76.7% and finally FOXP3 with sensitivity of 70.0%.
Conclusion: TOX is suggested to be of higher diagnostic value in the early stages of MF than the conventionally used CD4 and other markers
Mycosis fungoides (MF) is considered the most common
type of primary cutaneous T-cell lymphoma (CTCL) 1
is very difficult to diagnose, especially in the early stages. At
these stages, it may be indistinguishable from inflammatory
. Determining the challenging early stage of MF
is critical as it helps in managing cases without delay with
favorable subsequent prognosis 3
Clinically, the disease progresses slowly in three stages
including patches, plaques, and tumors 4. MF is histologically
characterized by a subcutaneous band-like infiltration
with epidermotropic CD4+ neoplastic lymphocytes
5. Therefore, histopathological and immunohistochemical
parameters have a pivotal role in its diagnosis. It has
been demonstrated that infiltrating epidermotropic neoplastic
lymphocytes in MF are predominantly CD4+, with
a minority of CD8+ T cells. However, CD4 is also expressed
by histiocytes and inflammatory T-lymphocytes, which can
be abundant in this context 2 and it must be kept in mind that CD4+ T lymphocytes also take place in inflammatory
skin diseases 6.
Transcription factor FOXP3 is used to identify normal
regulatory T cells (Tregs) that have been assigned a
variety of roles in immune regulation. An increase in
normal FOXP3+ Treg cells in dermal infiltrates of MF
has been linked to improved prognosis, similar to other
indolent lymphomas 7. These cells play a pivotal role
in maintaining immune homeostasis and preventing the
induction of potent antitumor immune responses 8.
FOXP3 Tregs are found in various lymphomas as well as
tumors 9. In a recent study, the majority of malignant T
cells in early MF expressed FOXP3 but the expression level
on average was much lower than that of non-malignant
Tregs 10. Moreover, FOXP3 is positive in the vast
majority of T cells of most dermatoses 11.
GATA-binding protein 3 (GATA3), a transcription factor
expressed in the epidermis, is essential for the regulation
of epidermal differentiation and development of lymphoid cells; GATA3 is required in T helper (Th) cell subset
differentiation into Th2 cells 12. GATA3 is expressed on
Th2 cells in dermatoses and constitutes the predominant
cell in late-stage MF, erythrodermic MF and Sézary
syndrome. Immunohistochemical stains for GATA3 is said
to be useful in discriminating MF from dermatoses 13,14.
Thymocyte selection-associated high mobility group box
factor protein (TOX), which belongs to DNA-binding
factors, is involved in the formation of T cells. TOX is
upregulated in the thymus during positive selection of
CD4+ CD8+ precursors to CD4+ T cells, but it shows a
decreased expression in mature CD4+ T cells after leaving
the thymus 15. The difficulty in the differential diagnosis
between early MF and BIDs may be partially caused by
the lack of tumor cell-specific markers. Early studies
have reported that TOX is a tumor cell-specific marker of
CTCLs including early MF based on immunohistochemical
findings, as it is expressed in tumor cells of cutaneous
T-cell lymphomas (CTCLs) but barely in inflammatory
infiltrate of BIDs 16. TOX is also expressed in infiltrating
lymphocytes in BIDs, although the frequency is not high
Therefore, in this work we aimed to evaluate the diagnostic
utility of common immunohistochemical markers,
including TOX, FOXP3, CD4 and GATA3, used in
diagnosing early-stage MF in order to avoid misdiagnosis
and thus ensure early proper management.
This retrospective cross-sectional study included 30 cases
of early MF and 30 cases of BID (15 cases of chronic
dermatitis, 10 cases of psoriasis and 5 cases of lichen
planus) collected from the archives of the Department of
Pathology, Zagazig Faculty of Medicine (2018-2021).
Patients’ consent was obtained; and the research was
reviewed and approved by the Institutional Research Board
(IRB), (ZU-IRB #6967/-2-6-2021). The research complied
with the Helsinki Declaration.
MF was diagnosed by looking at the histologic features
of sections stained with hematoxylin and eosin (H&E).
All of the included cases of MF had an active lesion in the
early stage; and BID cases were also with an active lesion.
Pregnant and lactating women and patients with the
tumor stage of MF or other types of cutaneous B or T-cell
lymphomas were excluded.
The diagnostic feature of MF was the predominance
of atypical lymphocytes whose nuclei are larger than
those of dermal lymphocytes. The atypical cells had an
irregular nuclear border surrounded by a clear halo and
were observed in the basal layer or in both stratum basalis
and stratum spinosum layers as single cells. Pautrier’s
microabscesses were also detected. Clinical findings
included the impressions of dermatologists, and the IHC
findings (pan T cell markers) were present in the reports.
Analysis for TOX, Fox and GATA3 was carried out on 5-μm
thick archived formalin-fixed paraffin-embedded (FFPE)
sections using the Leica Bond III autostainer. The primary
antibodies were applied to the slides. They included Rabbit
polyclonal anti-TOX antibody, as the optimal primary
antibody, at a dilution of 1:150 (Sigma-Aldrich, St. Louis,
MO, USA) and GATA3 mouse monoclonal antibody
(clone L50-823, 1:600, Cell Marque, Rocklin, California,
USA), CD4 (Dako; clone: 4B12, dilution 1:40), CD8 (Dako;
clone: C8/144B, dilution 1:50) and FOXP3 mAb (FOXP3
mAb, clone NB100–39002, dilution 1: 400; Novus, St
Charles, Missouri, USA). Appropriate negative and positive
controls were included. The slides were examined by three
The IHC evaluation was done by reporting the number of
cells with nuclear (FOX, TOX, GATA3) or cytoplasmicmembranous
(CD4) staining per 10 high power fields
(HPF) in both the epidermis and dermis; and the positive
cell percentage was evaluated as Grade 0 if no cells were
stained, Grade 1 if < 10%, Grade 2 if 11–50% and Grade 3 if
> 50% of cells stained 1,7,14.
Statistical analysis was assessed using SPSS software, version
11.5 (IBM SPSS Inc., Armonk, NY, USA). The Pearson χ2-
test and Fisher exact test were used for comparison of the
IHC results between the MF and nonmalignant groups.
The p value of ≤0.05 was accepted as statistically significant.
Sensitivity, specificity, false negative and positive rates (1−
sensitivity), and diagnostic accuracy of each marker were
Sixty cases (30 MF and 30 BID) were included in this
study. Evaluation revealed higher grades (2 and 3) of TOX
immunohistochemical marker expression in MF cases
(33.4% and 60% respectively) than in BID, in which 70% of
cases were negative for TOX expression. The results were
highly significant (p=0.001) (Table I; Figure 1).
Click Here to Zoom
|Figure 1: Photomicrographs
showing TOX immunoreactivity
in different lesions: A) A case
of chronic dermatitis showing
negative TOX immunoreactivity.
B) A case of chronic dermatitis
showing Grade 1 TOX
immunoreactivity. C) MF case
showing Grade 2 positivity
of TOX immunoreactivity.
D) MF case showing Grade
3 TOX immunoreactivity.
(Immunoperoxidase stain, x 400)
GATA3 expression assessment revealed that negative and
grade 3 GATA3 expression were higher in MF cases being
23.3% and 66.7%, respectively, compared with BID of
3.3% and 43.4%, respectively; and the results were highly
significant (p=0.002). Meanwhile, Grades 1 and 2 GATA3
immunohistochemical expressions were significantly higher
in BID (33.3% and 20% respectively) than found in MF
cases (6.7% and 3.3%), and the results were highly significant
(p=0.002) (Table I; Figure 2).
Click Here to Zoom
|Figure 2: Photomicrographs showing
GATA3 immunoreactivity in different
lesions. A) A case of MF showing negative
B) A case of chronic dermatitis showing
Grade 1 GATA3 immunoreactivity.
C) A case of MF showing Grade 2 GATA3
D) A case of MF showing Grade 3 GATA3
stain, x 400).
Regarding assessment of CD4 expression, we found that
40% of MF cases showed grade 3 expression while among
the BID cases only 6.67% showed it, and the results were
significant (p=0.0003). However, negative CD4 immunohistochemical
expression was significantly higher in BID
(63.3%) than in MF cases (16.7%) (Table I; Figure 3).
Click Here to Zoom
|Figure 3: Photomicrographs showing
CD4 immunoreactivity in different
lesions: A) A case of chronic dermatitis
with negative CD4 expression.
B) A case of chronic dermatitis with
Grade 1 CD4 expression, showing
infiltration of the dermis and the
epidermis by a small number of CD4
positive cells. C) A case of chronic
dermatitis with Grade 2 CD4 expression
showing infiltration of the dermis and
the epidermis by moderate numbers of
CD4 positive cells. D) A Case of MF
with Grade 3 CD4 expression showing
infiltration of the dermis and the basal
layer of the epidermis by large numbers of
CD4 positive cells. (Immunoperoxidase
stain, x 400).
Regarding FOXP3 expression, we found that grades 1 and
3 of FOXP3 expression were higher in MF cases (36.7%
and 50%, respectively) than in BID cases (23.3% and
20% respectively), and the results were highly significant
(p=0.002). However, negative and grade 2 FOXP3 expressions were significantly higher in BID (30.3% and 26.7%,
respectively) than in MF cases (13.3% and 0.0% respectively);
and the results were highly significant (p=0.002) (Table
I; Figure 4).
Click Here to Zoom
|Figure 4: FOX3
A) A case of chronic
B) A case of lichen
planus showing Grade 1
C) Chronic dermatitis
case showing Grade
2 positivity of FOX3
D) A case of MF
showing Grade 3 FOX3
stain, x 400)
The immunohistochemical assessment revealed higher
expression (grade 3) of TOX, GATA3, CD4 and FOXP3 in
MF than in BID. Negative expressions of TOX, CD4 and
FOXP3 were significantly higher in BID than MF cases,
but the reverse was true for GATA3 negative expression
as it was significantly higher in MF than BID. Moreover,
significant differences in other grades between MF and
BID were detected (p<0.05) (Table I; Figures 1-5).
Click Here to Zoom
|Figure 5: Scatter Chart showing that grade 3 of TOX, GATA3,
CD4 and FOXP3 expressions is significantly higher in MF than
BID. Grade 1 of GATA3 and TOX expressions is significantly
higher in BCID than MF.
Investigating the predictive value and sensitivity of TOX,
GATA3, CD4 and FOXP3 in MF revealed that TOX had
the highest sensitivity (96.77%) followed by CD4 (85.71%),
GATA3 (76.7%) and finally FOXP3 (70.0%). Regarding
the specificity, TOX had the highest specificity (76.92%)
followed by CD4 (73.17%), FOXP3 (13.3%) and finally
GATA3 (3.3 %). The marker with highest positive predictive
value (PPV) was TOX with 76.92% followed by CD4 (73.17%), FOXP3 (44.7%) and finally GATA3 (44.2%). The
marker with highest negative predictive value (NPV) was
TOX 96.77% followed by CD4 (85.71%), FOXP3 (30.8%)
and finally GATA3 (12.5%). Among the studied markers,
TOX had the highest diagnostic accuracy (85.71%) followed
by CD4 (78.95%), FOXP3 (41.7%) and GATA3 (40.0%). To
sum up, TOX had the highest sensitivity, specificity, PPV,
NPV and diagnostic accuracy followed by CD4. FOXP3
was following CD4 as regards specificity, PPV, NPV and
diagnostic accuracy but it was less sensitive than GATA3 in
distinguishing MF from BID. In other words, GATA3 had
the lowest specificity, PPV, NPV and diagnostic accuracy
but it was more sensitive than FOXP3 in MF diagnosis than
BID (Table II; Figure 6).
Click Here to Zoom
|Table II: Predictive value of TOX, GATA-3, CD4 and FOXP3 in MF detection
Click Here to Zoom
|Figure 6: Smooth line chart showing TOX with the highest
sensitivity, followed by CD4 with sensitivity and then GATA3 and
The early stage of MF is a challenging clinical condition, as
it is often confused with BID. The difficulty in differential
diagnosis between early MF and BID may be due in part to
a lack of tumor cell-specific markers 18
. Finding markers
that can distinguish MF from these lesions is a priority.
Moreover, the exact etiology of MF is still uncertain, which
is why analysis of various molecular markers that may be
related to the mechanism of its development may help to
determine the pathogenesis of MF. This work investigated
the diagnostic potency of TOX, FOXP3, CD4 and GATA3
immunohistochemical expressions in differentiation of MF
from the clinically and pathologically overlapping benign
TOX is one of the proteins that act as a gene expression
regulator by changing the structure of chromatin 14.
Regarding the development of T cells, TOX is upregulated
in the thymus during the positive selection of CD4+CD8+
precursors to CD4+ T cells, but mature CD4+ T cells
lose TOX expression and never re-express it again 19.
Consistent with previous results of Yu et al. 16 and
Huang et al. 20, our study showed that TOX is highly
expressed in MF compared to BID. This is because it was
expressed in most cases of MF while only observed in 30%
of BID. Therefore, TOX could be a potential tool for new
therapeutic lines for MF cases.
The FOXP3 gene encodes a transcriptional factor that
is vital for the function and development of regulatory
Tregs, which are differentiated from naive CD4+ T-helper
cells 21. Mutations in FOXP3 induce Treg abnormality
and a subsequent defect in immune function, either for
inflammatory diseases or tumors 9.
In line with previous reports, the current work revealed
that the difference between MF and BID regarding
FOXP3 expression was statistically significant 21. Wada
et al. 9, found that FOXP3 T cells are less regularly
detected in MF than in BID, and its dermal distribution is
probably associated with inflammatory components. On
the contrary, Fujimura et al. 21 reported that FOXP3
immunoreactivity was significantly weaker in eczematous
dermatitis than in MF.
The function of FOXP3 in MF development and progression
is quite puzzling. Manso et al. 22 reported that the Tregs
number is low at the tumor stage of MF compared to early
stages. At the same time, Perelman et al. 23 revealed that
FOXP3 expression in the dermis correlated with poor
response to treatment. Bhat et al. 24 reported that the
worse prognosis of cases of Sézary syndrome with high
FOXP3 expression is attributed to its possible role in
suppression of immune defenses.
Expression of GATA3 showed a significant difference
between MF and BID. This can be explained by the fact that
TH1 and not TH2 is the prominent intraepidermal cell in
early MF, and GATA3 stains the later one 25.
According to our results, TOX is recommended as a marker
for diagnosis of early-stage MF, helping to differentiate it
from the benign mimickers. This is because TOX had a
high sensitivity (96.77%) and accuracy (85.71%), followed
by CD4 for sensitivity (85.71%) and accuracy (78.95%). The
same results were obtained by Zhang et al. 15. Morimura
et al. (26) reported TOX as a tumor cell-specific marker of
CTCLs including early MF based on immunohistochemical
findings and TOX was expressed in tumor cells of CTCLs
but hardly in inflammatory infiltrates of BID.
The limitations of the study included the relatively
small numbers of cases seen at a single medical center.
Furthermore, the study did not include all types of BIDs.
Among the immunohistochemical markers examined,
TOX is suggested to be the most accurate in diagnosing
MF, which is difficult to diagnose especially in its early
stages when it is confused with BID. Further future studies
including larger numbers of patients are recommended to
confirm the diagnostic significance of TOX.
We would like to thank Dr Safaa A El-Naggar, Professor of Industrial
Medicine and Occupational Health, Faculty of Medicine, Zagazig
University for her assistance in the statistical analysis of the achieved
Conflict of Interest
The authors declare that there is no conflict of interest regarding the
publication of this paper.
No funding has been received.
Concept: MMA, MAG, HMA, Design: MMA, AAH, MAG,
Data collection or processing: MMA, HMI, MAG, Analysis or
Interpretation: MMA, HMI, MAG, AAH, Literature search: MMA,
AE, EMN, SAH, Writing: MMA, AAH, MAG, Approval: All
1) Mehdi SJ, Moerman-Herzog A, Wong HK. Normal and cancer
fibroblasts differentially regulate TWIST1, TOX and cytokine
gene expression in cutaneous T-cell lymphoma. BMC Cancer.
2) Ortonne N, Buyukbabani N, Delfau-Larue MH, Bagot M,
Wechsler J. Value of the CD8-CD3 ratio for the diagnosis of
mycosis fungoides. Mod Pathol. 2003;16:857-62.
3) Motamedi M, Xiao MZX, Iyer A, Gniadecki R. Patterns of gene
expression in cutaneous T-cell lymphoma: Systematic review of
transcriptomic studies in mycosis fungoides. Cells. 2021;10:1409.
4) Cerroni L. Mycosis fungoides-clinical and histopathologic
features, differential diagnosis, and treatment. Semin Cutan Med
5) Raghavan SS, Kim J. Histopathologic approach to epidermotropic
lymphocytic infiltrates. Semin Cutan Med Surg. 2018;37:56-60.
6) Miyagaki T. Diagnosis of early mycosis fungoides. Diagnostics.
7) Gjerdrum LM, Woetmann A, Odum N, Burton CM, Rossen K,
Skovgaard GL, Ryder LP, Ralfkiaer E. FOXP3+ regulatory T cells
in cutaneous T-cell lymphomas: Association with disease stage
and survival. Leukemia. 2007;21:2512-8.
8) Savage PA, Klawon DEJ, Miller CH. Regulatory T cell
development. Annu Rev Immunol. 2020;38:421-53.
9) Wada DA, Wilcox RA, Weenig RH, Gibson LE. Paucity of
intraepidermal FoxP3-positive T cells in cutaneous T-cell
lymphoma in contrast with spongiotic and lichenoid dermatitis.
J Cutan Pathol. 2010;37:535-41.
10) Clark RA, Shackelton JB, Watanabe R, Calarese A, Yamanaka K,
Campbell JJ, Teague JE, Kuo HP, Hijnen D, Kupper TS. Highscatter
T cells: A reliable biomarker for malignant T cells in
cutaneous T-cell lymphoma. Blood. 2011;117:1966-76.
11) Martig DS, Bridges AG, Macon WR. Assessing for disease:
Recurrent mycosis fungoides or cutaneous granulomatous
drug eruption after mogamulizumab therapy. J Hematopathol.
12) Hsi AC, Lee SJ, Rosman IS, Carson KR, Kelley A, Viele V, Pang
X, Musiek A, Schaffer A. Expression of helper T cell master
regulators in inflammatory dermatoses and primary cutaneous
T-cell lymphomas: Diagnostic implications. J Am Acad
13) Papadavid E, Economidou J, Psarra A, Kapsimali V, Mantzana
V, Antoniou C, Limas K, Stratigos A, Stavrianeas N, Avgerinou
G, Katsambas A. The relevance of peripheral blood T-helper 1
and 2 cytokine pattern in the evaluation of patients with mycosis
fungoides and Sézary syndrome. Br J Dermatol. 2003;148:709-18.
14) Collins K, Gu J, Aung PP, Nagarajan P, Curry JL, Huen A,
Ivan D, Prieto VG, Tetzlaff MT, Duvic M, Miranda RN, Vega
F, Torres-Cabala CA. Is immunohistochemical expression of
GATA3 helpful in the differential diagnosis of transformed
mycosis fungoides and primary cutaneous CD30-positive T cell
lymphoproliferative disorders? Virchows Arch. 2021;479:377-83.
15) Zhang Y, Wang Y, Yu R, Huang Y, Su M, Xiao C, Martinka M,
Dutz JP, Zhang X, Zheng Z, Zhou Y. Molecular markers of earlystage
mycosis fungoides. J Invest Dermatol. 2012;132:1698-706.
16) Yu X, Luo Y, Liu J, Liu Y, Sun Q. TOX acts an oncological role in
mycosis fungoides. PLoS One. 2015;10:e0117479.
17) McGirt LY, Degesys CA, Johnson VE, Zic JA, Zwerner JP,
Eischen CM. TOX expression and role in CTCL. J Eur Acad
Dermatol Venereol. 2016;30:1497-502.
18) Schrader AM, Jansen PM, Willemze R. TOX expression in
cutaneous T-cell lymphomas: An adjunctive diagnostic marker
that is not tumour specific and not restricted to the CD4(+)
CD8(-) phenotype. Br J Dermatol. 2016;175:382-6.
19) Xu J, Huang H, Wang S, Chen Y, Yin X, Zhang X, Zhang
Y. Molecular profiling of TOX-deficient neoplastic cells in
cutaneous T cell lymphoma. Arch Dermatol Res. 2020;312:513-
20) Huang Y, Su MW, Jiang X, Zhou Y. Evidence of an oncogenic
role of aberrant TOX activation in cutaneous T-cell lymphoma.
21) Fujimura T, Okuyama R, Ito Y, Aiba S. Profiles of Foxp3+
regulatory T cells in eczematous dermatitis, psoriasis vulgaris
and mycosis fungoides. Br J Dermatol. 2008;158:1256-63.
22) Manso R, Martínez-Magunacelaya N, Erańa-Tomás I, Monsalvez
V, Rodríguez-Peralto JL, Ortiz-Romero PL, Santonja C, Cristóbal
I, Piris MA, Rodríguez-Pinilla SM. Mycosis fungoides progression
could be regulated by microRNAs. PLoS One. 2018;13:e0198477.
23) Perelman M, Rozenblat M, Ellenbogen E, Geller S, Slutsky Bank E,
Eytan O, Gat A, Sprecher E, Goldberg I. FOXP3 predicts response
to treatment in mycosis fungoides. Acta Dermatovenerol Croat.
24) Bhat R, Khandpur B, Vonderheid EC, Hou JS. FoxP3-positive
T-regulatory cells in lymph nodes with mycosis fungoides and
Sézary syndrome. Lymphoma. 2014;2014:1-9.
25) Aslani FS, Safaei A, Ghezelbash H, Sepaskhah M. Value of TOX
immunoexpression in the diagnosis of early mycosis fungoides.
Middle East J Cancer. 2019;10:95-102.
26) Morimura S, Sugaya M, Suga H, Miyagaki T, Ohmatsu H, Fujita
H, Asano Y, Tada Y, Kadono T, Sato S. TOX expression in
different subtypes of cutaneous lymphoma. Arch Dermatol Res.