ARTICLE
Lichenoid
tissue reactions have been reported in a variety
of dermatoses [1, 2]. Graft-versus-host disease-like eruptions may occasionally
occur in the absence of any graft or blood products. This has been reported
in association with certain drugs [3], disseminated carcinoma [4] and also
in HIV-1 infections [5].
More recently the same type of reaction was observed in a patient with
a malignant B-cell lymphoma [6] and also in systemic T-cell lymphoma [7].
In the present study we report on a peculiar ulcerative and erosive
lichenoid dermatosis in a male patient who developed a nodal CD30+
lymphoma. This unusual type of skin reaction has only been reported once
before to our knowledge [8].
Its warrants a thorough search for an underlying malignancy and close
follow-up when the dermatologist is confronted with this type of lesion.
Case report
A 48-year old Caucasian man was admitted for evaluation of an extensive
erosive dermatosis of 2 years duration, which was refractory to topical
and systemic treatment.
Crude coal tar, topical steroids, acitretin, PUVA therapy and azathioprine
all yielded poor results. Only systemic prednisone (50 mg daily) resulted
in a substantial decrease of the skin lesions. However, after discontinuation
of prednisone treatment the skin lesions were promptly exacerbated. The
patient experienced itching, particularly during the night. No drugs were
used by the patient which could be responsible for the eruption.
On admission the skin revealed generalised erythema with a livedo reticularis
pattern on the legs. The skin of the trunk and arms was indurate with
superficial erosions and hemorrhagic crusts. Poikiloderma with atrophy,
telangiectasia, hypo- and hyperpigmentation was present (Fig.
1).
The lesions did not resemble lichen planus, no Wickham's striae were
seen. The scalp showed patchy alopecia. On the palms and soles moderate
hyperkeratosis was present. The oral mucosa revealed no abnormalities.
Multiple punch biopsies of involved but non-eroded skin showed acanthosis
with compact hyperkeratosis and focal parakeratosis. Solitary clusters
of dead keratinocytes, several colloid bodies and lymphocytes were seen
(Fig. 2).
Biopsies from superficial erosions showed histiocytes and neutrophils.
The dermal-epidermal transition zone was obscured by a dense infiltrate
of normal looking lymphocytes. In the dermis lymphocytic infiltrates with
many eosinophils were observed perivascularly with scattered plasma cells.
Immunophenotyping revealed reduction of Langerhans cells (CD1) in the
epidermis. The vast majority of mononuclear cells in the lichenoid infiltrate
showed T-cell markers (CD2, 3 and 5). T lymphocytes with a helper cell
phenotype (CD4) just outnumbered lymphocytes with a suppressor/cytotoxic
cell phenotype (CD8). A few lymphocytes expressed a B-cell marker (CD20).
No CD30+ cells were seen.
Immunofluorescence of lesional skin showed no immunopathology. Colloid
bodies were easily detected and stained non-specifically. Routine blood
examinations were unremarkable. Serologic tests for HIV-1, EBV and hepatitis
C were negative.
During admission the skin lesions exacerbated following discontinuation
of prednisone therapy (Fig.
3). The patient developed fever up to 40° C and weight loss.
An infectious focus could not be detected. The fever persisted despite
treatment with several antibiotics and antimycotics. Anaemia and thrombopenia
developed. Bone marrow aspiration showed normal erythropoiesis and thrombopoiesis.
Mild hematuria developed. No abnormalities on ultrasound and computer
tomography of the abdominal organs were seen. Subsequently cervical lymphadenopathy
developed. Histologic examination of a jugular lymph node revealed a disturbed
architecture with diffuse infiltration by small- and large lymphoid cells.
The large cells had highly lobulated hyperchromatic nuclei and large
amounts of eosinophilic cytoplasm (Fig.
4). They were CD4+ and often CD30+. Both cell
types showed mitoses.
A diagnosis of nodal pleomorphic CD30+ malignant lymphoma,
stage I according to the Ann Arbor staging system, was made.
Therapy consisted of prednisone and intravenous etoposide 200 mg, adriamycin
80 mg and cyclophosphamide 800 mg. Hereafter the condition of the patient
deteriorated. Thrombocytopenia increased and anaemia worsened despite
multiple blood transfusions. Fever declined following administration of
the cytostatic drugs. This supported the assumption that the fever was
caused by the non-Hodgkin lymphoma. However, shortly thereafter fever
recurred. An infectious focus could not be found. Exhaustion led to food
refusal, weight loss and emaciation. The patient became somnolent and
died.
At autopsy large lymphocytes with highly lobulated nuclei were present
in the retroperitoneal lymph nodes. Moreover in the gastrointestinal tract,
kidney and myocard septic foci with Candida were found. The skin showed
the same erosive and ulcerative lichenoid picture as was observed in the
biopsies described above. No atypical lymphocytes were found in the skin.
Clonality assays on skin biopsies and lymphnode material were not performed.
We concluded the cause of death was heart failure due to cachexia, anaemia,
Candidasepsis, prolonged prednisone use and non-Hodgkin lymphoma.
Discussion
Lichenoid and erosive skin lesions which can be regarded as a paraneoplastic
phenomenon have rarely been reported in malignant lymphoma.
Parkes et al. described a patient who developed a GVHD-like eruption
with superficial skin necrosis 2 years after the diagnosis of a high-grade
large B-cell lymphoma [6]. This patient also had oral, genital and conjunctival
involvement. Scarisbrick et al. reported on a patient with a systemic
T-cell lymphoma who simultaneously had erosive skin lesions with histologic
features of GVHD [7]. No involvement of mucous membranes was seen.
An extensive ulcerative and erosive lichenoid dermatosis in a patient
who simultaneously presented with a centroblastic-centrocytic B-cell lymphoma
was reported by Lee et al. [8]. Clinically and histologically the
lesions had the features of lichen planus, including Wickham's striae.
The patient reported here also had extensive ulcerative erosive and
lichenoid changes without mucosal involvement. His skin lesions did not
resemble lichen planus. No cause could be detected. Two years later he
developed a nodal CD30+ lymphoma.
In these patients the temporal relationship between the presentation
of erosive lichenoid eruptions and the occurrence of a non-Hodgkin lymphoma
was suggestive of an association between the two. Oral and cutaneous lesions
markedly improved after chemotherapy in the patient of Lee and Parkes.
The patient described by Scarisbrick et al. improved well but unfortunately
died of septicaemia.
Because the patient reported here also died
shortly after starting chemotherapy, the association could not be confirmed
with certainty. However, the skin had shown remarkable improvement during
prednisone therapy. The relationship of the skin lesions to the lymphoma
is therefore only speculative. More observations are needed to confim
lichenoid dermatosis to be a paraneoplastic phenomenon.
Differential diagnosis of GVHD-like eruptions include drug reactions
[3], preceding viral infections [5], mycosis fungoides (MF) [9] and paraneoplastic
pemphigus [10]. The patient described here did not use any drug prior
to the outbreak of his skin lesions. No virus infection could be detected.
The skin lesions did not resemble MF clinically nor histologically based
on multiple skin biopsies. The patient did not fulfil the diagnostic criteria
for paraneoplastic pemphigus as proposed by Anhalt [10].
The pathogenesis of GVHD-like eruptions in lymphoma or other malignancies
is not clear. More observations are needed to confirm lichenoid dermatosis
to be a paraneoplastic phenomenon.
The dermatologist should be aware and search for internal malignancies
when confronted with an erosive and ulcerative lichenoid dermatosis.
Article accepted on 25/3/02
CONCLUSION
REFERENCES
1. Patterson JW. The spectrum of lichenoid dermatitis. J Cutan
Pathol 1991; 18: 67-74.
2. Oliver GF, Winkelmann RK, Muller SA. Lichenoid dermatitis:
a clinicopathologic and immunopathologic review of sixty-two cases. J
Am Acad Dermatol 1989; 21: 285-92.
3. Osawa J, Kitamura K, Saito S. Immunohistochemical study of
graft-versus-host reaction type drug eruptions. J Dermatol 1994;
21: 25-30.
4. Graham-Brown RAC, Jones JAG, Shaw PV. A graft-versus-host
disease-like syndrome with carcinomatosis. Br J Dermatol 1987;
116: 249-52.
5. Smith KJ, Skelton HG, Yeager J, Baxter D, Angritt P, Johnson
S, Oster CN, Wagner KF. Clinical features of inflammatory dermatosis in
human immunodefriciency virus type I disease and their correlation with
Walter Reed stage. J Am Acad Dermatol 1993; 28: 167-73.
6. Parkes IR, Zaki I, Stevens A, Davies JM, Allen BR. Graft-versus-host
disease-like eruption in a patient with non-Hodgkin's lymphoma. Br
J Dermatol 1997; 137: 137-9.
7. Scarisbrick JJ, Wakelin SH, Russell-Jones R. Cutaneous graft-versus-host-like
reaction in systemic T-cell lymphoma. Clin Exp Dermatol 1999; 24:
382-4.
8. Lee YS, Fong PH. Extensive ulcerative and erosive lichenoid
dermatosis in a patient with malignant lymphoma. Am J Dermatopathol
1993; 15: 576-80.
9. Guitart J, Peduto M, Caro WA, Roenigk HH. Lichenoid changes
in mycosis fungoides. J Am Acad Dermatol 1997; 36: 417-22.
10. Anhalt GJ. Paraneoplastic pemphigus. Advances in Dermatology
1997; 12: 77-97.
|
Printable version
