Atypical Lymphoid Infiltrate with CD30+ Cells and Evolution to Granulomatous Dermatitis at the Site of Zoster Reactivation
University of Louisville School of Medicine
Click on each slide thumbnail image for an enlarged view
A 69-year-old male with a history of rheumatoid arthritis developed pruritis followed by a
self-described vesicular eruption of his right trunk two weeks after an injection of infliximab. He was
simultaneously receiving methotrexate. Six weeks later, he presented to a dermatologist complaining of
an asymptomatic plaque in a linear distribution in the same region that he had had "blisters" a few weeks
prior. A biopsy was performed.
Scanning magnification demonstrates a punch biopsy with a superficial and deep lymphoid infiltrate. Papillary dermal edema is present.
Low power demonstrates a somewhat nodular infiltrate.
Small aggregates of histiocytes form vague granulomas admixed with lymphocytes.
Periadnexal infiltration is prominent.
Large transformed lymphocytes and mitotic figures are present.
Large transformed lymphocytes in addition to a vessel with plump reactive endothelial cells. An eosinophil is present under the vessel.
Haloed lymphocytes line the dermal-epidermal interface and exocytosis is present. Some intraepidermal lymphocytes have convoluted nuclei.
Figure 8 - CD20
CD20 highlights a few admixed B-cells.
Figure 9 - CD4
The infiltrate is predominantly composed of CD4 + T-cells.
Figure 10 - CD8
Quite a few admixed cytotoxic T-cells are also present.
Figure 11 - CD30
CD30 highlights scattered cells throughout the infiltrate.
Figure 12 - CD30
A closer look demonstrates that CD30 is staining the large transformed lymphocytes.
Sections demonstrated a lichenoid tissue reaction with exocytosis of irregular lymphocytes, some with
halos, and a dense superficial and deep perivascular and interstitial lymphoid infiltrate. Papillary
dermal edema was present. The infiltrate is somewhat wedge-shaped with large, atypical lymphocytes
noted, as well as a focal aggregation of histiocytes reminiscent of granulomas. No viral cytopathic
effect was identified in the biopsy. No polarizable foreign bodies were identified.
Diagnostic Considerations and Ancillary Studies:
The histopathologic features and history of immunosuppression in an older patient bring to mind a
number of differential diagnostic considerations, both malignant and reactive. Because of the
wedge-shaped character of the infiltrate with atypical large cells, a diagnosis of lymphomatoid papulosis
was considered in addition to a lesion of mycosis fungoides with focal granulomatous features and an
EBV-associated lymphoproliferative disorder. Because of immunosuppression and patient described symptoms
consistent with zoster infection, a reactive etiology such as a post-zoster pseudolymphoma was in the
differential diagnosis. And finally, with the presence of somewhat granulomatous regions, fungal and
mycobacterial infectious etiologies had to be regarded. Special stains for these organisms were
Immunohistochemical evaluation further characterized the infiltrate as containing predominantly CD3
and CD4 immunopositive T-cells, and large transformed atypical lymphocytes were noted to be CD30
immunopositive. In-situ hybridization for EBER was negative. A diagnosis of an "atypical lymphoid
infiltrate with CD30+ cells" was made with a comment of favoring a pseudolymphoma but being "unable to
rule out a lesion of lymphomatoid papulosis". Tissue was sent for B and T cell gene rearrangements.
The patient was referred to our University Dermatology service and seen 4 weeks after the initial
biopsy was performed (12 weeks after the original skin eruption). He remained asymptomatic and physical
examination revealed an erythematous crusted plaque on the right flank with surrounding erythematous
papules in a dermatomal distribution. Gene cell rearrangement results had returned detecting no clonal
rearrangements. Additional biopsies were performed and demonstrated a granulomatous dermatitis featuring
necrobiotic palisaded granulomas, necrotizing granulomas, and sarcoidal-type granulomas. Multinucleated
giant cells and a minimal lymphoplasmacytic infiltrate were present. There was no evidence of viral
cytopathic effect. Special stains for fungal and mycobacterial elements were negative, and no
polarizable foreign bodies were identified. PCR testing for HSV1/2-DNA and VZV-DNA did not detect viral
DNA, and bacterial, fungal, and mycobacterial cultures were negative.
Based on the clinical and histopathologic information, the patient was diagnosed with a post-zoster
dermatitis initially presenting as a pseudolymphoma and eventuating into a granulomatous dermatitis. At
a 5-month follow-up visit, the lesion had completely resolved other than some post-inflammatory
This case is illustrative of three main points:
CD30 is a transmembrane protein that is part of the tumor necrosis factor/nerve growth factor receptor
superfamily. It is normally restricted to a small area in B-cell follicles of lymphoid tissue and to a
small number of stimulated memory T-cells. It used to be considered a marker for lymphoma, and while it
defines lymphomatoid papulosis and anaplastic large cell lymphoma (CD30+ lymphoproliferative disorders);
it is also seen in Hodgkin lymphoma, transformed mycosis fungoides, and other lymphoproliferative
disorders. However, it has been well characterized that CD30+ cells, often with atypical forms, are seen
in many inflammatory and reactive conditions as listed in Table 1. In general, one of the more common
diagnostic dilemmas faced is that of reactive lymphoid infiltrate vs. lymphomatoid papulosis (usually
Type A pattern). Table 2 lists some helpful features to aid in differentiating these two processes.
Herpesvirus infections not uncommonly contain large atypical lymphocytes; in fact, 2/3 of herpes
inflammatory infiltrates contain these cells. One-half of herpesvirus infiltrates will have a dense
- CD30+ atypical lymphocytes can be seen in herpesvirus infections in addition to other reactive dermatoses.
- The clinical and histopathologic features of herpesvirus infections are variable and characteristic viral cytopathic changes are not always identifiable in biopsy specimens.
- Several types of cutaneous lesions have been described to develop within resolved cutaneous herpes zoster lesions, and evolution to granulomatous dermatitis may occur.
Table 1. Inflammatory and Reactive Conditions That May Express CD30
|Herpes virus infections|
|Parasitic infections; especially scabies|
|Parapox virus infections|
|Pityriasis lichenoides et varioliformis acuta|
|Drug reactions with carbamazepine and cefuroxime|
|Arthropod bite reactions|
Table 2. Comparison between Lymphomatoid Papulosis and Reactive/Inflammatory Conditions with CD30
| ||Lymphomatoid Papulosis ||Reactive/Inflammatory|
|Clinical ||Waxing and waning grouped or disseminated papules at different stages ||Depends on clinical entity; PLEVA may be quite similar in clinical presentation|
|Atypical cells ||Not difficult to find; often clustered|
May be few to absent in Type B pattern
|Scattered singly throughout the infiltrate|
|CD30+ cells ||Should correlate with large atypical forms seen on H&E|
Should highlight clustering
|May include atypical cells and/or small to medium admixed lymphocytes|
Should highlight scattered single cells
|Clonal rearrangements ||Variable; approximately 50% of cases ||Very, very rare|
|Immunophenotype ||CD3, CD4, CD30, HLA-DR, CD25+|
Cytotoxic (TIA-1, granzyme) +
CD7 diminished or lost in most cases
Rare cases CD8 LyP
|CD3+ with mixture of both CD4 and CD8+ cells (note that PLEVA is a CD8 predominant process) |
CD7 should be retained
Herpesvirus infections can be clinically challenging in various stages mimicking allergic contact
dermatitis, folliculitis, and arthropod bite reactions to name a few. This is especially true in a
prevesicular or late postvesicular stage of disease. Furthermore, biopsy of lesions in either of these
stages may not show the characteristic diagnostic viral cytopathic features. This is especially true in
cases of zoster infection. Cases in which herpes is not clinically suspected and where typical changes
are not seen on biopsy have been referred to as herpes incognito. Step sections are very helpful in
locating a small focus of epithelium with characteristic viral changes in most cases. Especially
important is the evaluation of adnexal epithelium; in particular, follicular structures may reveal the
diagnostic changes. Even in the face of negative cytopathic changes, some feel that the diagnosis can
still be suggested if the following constellation of histopathologic features is present, especially if
the clinical suspicion for zoster infection is noted.
Additionally, PCR for viral DNA can be helpful detecting viral genomic material in some cases.
- Perivascular and interstitial infiltrate of lymphocytes with a patchy lichenoid component
- Deep inflammatory infiltrate
- Periadnexal infiltrate especially around follicles, sebaceous glands, eccrine glands, and nerves
- Exocytosis into the basal epithelium with mild spongiosis and variable basal vacuolar changes
- Necrotic keratinocytes in basal and spinous layers
- Edema of the papillary dermis
- Dilated blood vessels with extravasated erythrocytes
- Atypical lymphocytes
- Mixed cell inflammation with neutrophils and/or eosinophils
An interesting phenomenon of herpes zoster infections is that several reactive and neoplastic
conditions have been reported to occur in sites of resolved infections (Table 3). This phenomenon has
been described to take place concurrently, or from weeks to months, or even years following the initial
lesion. In fact, healed zosteriform sites are the most common precursor described in the isotopic
response: "the occurrence of a new skin disorder at the site of another, unrelated, and already healed
skin disease". An important caveat is that bonafide lymphomas and leukemias have been reported to occur
as post-zoster cutaneous reactions.
Also of interest in zosteriform eruptions is the histopathologic evolution to a granulomatous
dermatitis. Previous studies have reported that the granulomas may develop in response to lingering
viral protein particles that induce a delayed-type hypersensitivity reaction. This theory would fit with
the fact that post-zoster granulomatous cutaneous reactions tested by PCR for viral DNA are mostly
negative when the lesion is greater than four weeks old. Often, the granulomas are of differing types,
as seen in our case which consisted of palisaded necrobiotic, necrotizing, and sarcoidal-type granulomas.
Table 3. Cutaneous lesions reported to occur in resolved zoster sites
|Granuloma annulare ||Lichen sclerosus|
|Sarcoidal granulomas ||Morphea|
|Tuberculoid granulomas ||Lichenoid chronic graft-vs.-host disease|
|Granulomatous vasculitis ||Eosinophilic dermatitis|
|Granulomatous folliculitis ||Reactive perforating collagenosis|
|Granulomatous dermatitis, NOS ||Verruca plana|
|Xanthomatous changes ||Verruca vulgaris|
|Rosai-Dorfman disease ||Molluscum contagiosum|
|Acneiform eruptions ||Leukemia|
|Furunculosis ||Kaposi's sarcoma|
|Contact dermatitis ||Angiosarcoma|
|Nodular solar degeneration ||Basal cell carcinoma|
|Psoriasis ||Squamous cell carcinoma|
|Lichen planus ||Metastases from internal carcinoma|
- Kempf W. CD30+ lymphoproliferative disorders: histopathology differential diagnosis, new variants, and simulators. J Cutan Pathol. 2006;33(Suppl. 1): 58-70.
- Leinweber B, Kerl H, Cerroni L. Histopathologic features of cutaneous herpes virus infections (herpes simplex, herpes varicella/zoster). Am J Surg Pathol. 2006;30:50-58.
- Requena L, Kutzner H, Escalonilla P, Ortiz S, Schaller J, Rohwedder A. Cutaneous reactions at sites of herpes zoster scars: and expanded spectrum. Br J Dermatol. 1998;138:161-168.
- Cepeda LT, Pieretti M, Chapman SF, Horenstein MG. CD30-Positive atypical lymphoid cells in common non-neoplastic cutaneous infiltrates rich in neutrophils and eosinophils. Am J Surg Pathol. 2003;27:912-918.
- Resnik KS, Di Leonardo M. Herpes incognito. Am J Dermatopathol. 2000;22:144-150.
- Boer A, Herder N, Blodorn-Schlicht N, Falk T. Herpes incognito most commonly is herpes zoster and its histopathologic pattern is distinctive. Am J Dermatopathol. 2006;28:181-186.