Pf. Moore et al., CANINE CUTANEOUS EPITHELIOTROPIC LYMPHOMA (MYCOSIS-FUNGOIDES) IS A PROLIFERATIVE DISORDER OF CD8(-CELLS() T), The American journal of pathology, 144(2), 1994, pp. 421-429
Canine epitheliotropic lymphoma (mycosis fungoides [MF]) is a spontane
ous neoplasm of skin and mucous membranes that occurs in old dogs (mea
n age 11 years) and has no breed predilection. The lesions evolve from
a patch-plaque stage with prominent epitheliotropism into a tumor sta
ge in which distant metastasis is observed Unlike human MI;, epithelio
tropism of the lymphoid infiltrate is still prominent in tumor stage l
esions. Tropism of the lymphoid infiltrate for adnexal structures, esp
ecially hair follicles and aprocrine sweat glands, was marked in all c
linical stages of canine MF. Twenty-three cases of MF were subjected t
o extensive immunophenotypic analysis in which reagents specific for c
anine leukocyte antigens and fresh frozen tissue sections of the canin
e lesions were used. Canine MF proved to be a T cell lymphoma in which
the epitheliotropic lymphocytes consistently expressed CD3 (22 cases)
and CD8 (13 cases); CD3(+)CD4(-)CD8(-) lymphocytes predominated in th
e remaining 4 cases. In this regard, canine MF clearly differed from h
uman MF in which a CD4 immunophenotype predominates in the T cell infi
ltrate. Lack of expression of CD45RA by epitheliotropic T cells and in
tense expression of a beta 1 integrin (VLA-4-like) suggested that T ce
lls in canine MF belonged to the memory subpopulation, as has been sug
gested for T cells in human MF. Pan-T cell antigen loss or discordant
expression also proved useful as phenotypic indicators of neoplasia in
canine ML. Loss of CD5 was observed in epitheliotropic T cells in 63%
of cases. Discordance of neoplastic T cell Thy-1 expression was frequ
ently observed between epithelial and dermal or submucosal compartment
s. We conclude that canine MI;still represents a useful spontaneous an
imal disease model of human cutaneous T cell lymphoma, despite the imm
unophenotypic differences, which may reflect operational differences b
etween human and canine skin-associated lymphoid tissue.