Search:

Return to Teaching Cases menu

TEACHING CASE 6

Relevant clinical history:
A 32 year-old, HIV-positive man with a history of cutaneous Kaposi sarcoma, presents to the ER with symptoms of small bowel obstruction. Physical exam and CT scan revealed an abdominal mass. At laparotomy, a 12 cm proximal jejunal tumor was excised, along with attached segment of bowel.

Gross images:

Gross examination shows a 13-cm, transmural mass (left) that occupies the entire wall of the intestine. The cut surface is nodular and partially necrotic (right).

H & E images:


Low power

Medium power

High power
Hematoxylin and eosin stained sections of the tumor show a malignant tumor eroding through the mucosal surface (left), composed of dishesive sheets of anaplastic cells with clear nuclei and large nucleoli. Numerous mitotic figures, apoptotic bodies and binucleated tumor cells are noted. The cytoplasm is markedly eosinophilic (center, right).

The morphologic findings of this malignant tumor are not characteristic of a specific cell type, but rather are those of undifferentiated malignant neoplasm. The known patient's HIV status strongly suggests a hematolymphoid neoplasm, possibly anaplastic large cell lymphoma, but other malignancies should also be considered. Accordingly, the 'first pass' immunohistochemical panel was designed to address this very question, with the following results:

Cytokeratin (antibody OSCAR): Negative
The tumor cells are negative.
CD45: Negative
CD30: Negative
CD43: Negative
CD20: Negative
CD3: Negative

Additional studies showed the following results:

CD79a: Negative
CD15: Negative
TdT: Negative
EBV (EBER): Variably positive
The tumor cells show the characteristic, variegated nuclear signal, detecting EBV genomic RNA.
Myeloperoxidase: Negative
CD34: Negative
WT-1: Negative
Melanoma antibody cocktail (gp 100, tyrosinase, MART-1): Negative
Broad Spectrum Cytokeratin (OSCAR antibody clone): Negative
Herpes Simplex Virus 8: Uniformly positive
Classic, 'speckled' nuclear HHV-8 signal is noted on the vast majority of tumor cells.

At this point, by these results, some of the strongly considered diagnostic entities are argued against, including anaplastic large cell lymphoma (negative CD30 and CD43), syncytial variant of NS Hodgkin's lymphoma (negative CD15, CD30), diffuse large B-cell lymphoma (negative CD20 and CD79a), granulocytic sarcoma (negative CD34, myeloperoxidase and WT-1), poorly differentiated carcinoma and melanoma. Even with the negative lymphoid markers, co-expression of EBV and HHV8 still points in that direction.

Additional studies showed the following results:

CD138: Variably positive
Diffuse cytoplasmic expression of CD138 is noted on the majority of tumor cells
Kappa light chain: Variably positive
The majority of tumor cells also show diffuse cytoplasmic expression of kappa light chain.
Lambda light chain: Negative
All tumor cells in this illustration show lack of lambda light chain expression.

FINAL DIAGNOSIS: HHV8-associated anaplastic lymphoma with plasmacytoid features.

DISCUSSION:
Lymphomas that are almost exclusive to HIV-positive patients are primary effusion lymphoma and oral cavity plasmablastic lymphoma. The tumor at hand could arguably be classified as immunoblastic lymphoma with plasmacytoid features; however, large nucleoli are not prominent features. Also, lack of expression of CD79a (and CD30) argues against the latter. Another HIV-related entity that shows some overlapping features is Burkitt's lymphoma with plasmacytoid differentiation. Again, lack of expression of CD20, CD79a (and CD10) does not provide support for the latter. It is reasonable to suggest, that because the only positive finding is CD138 expression and kappa light chain restriction, this tumor should be classified as anaplastic plasmacytoma. While the association with EBV should not be surprising, the detection of HHV8 genome on the tumor cells is uncharacteristic (1). In fact, a group of investigators reported three cases of HHV8-associated lymphomas with anaplastic large cell morphology (2). From a morphologic and immunophenotypic standpoint, this case appears to be part of that spectrum, even though it lacks CD30 expression. It is possible that this patient may already have bone marrow involvement with serum monoclonal gammopathy (3). At the time of uploading this case, work-up was underway to determine the extent of systemic involvement.

Extramedullary plasmacytoma in HIV-positive patients is rare, and probably under-reported. Recognition of this entity in the young HIV positive patient population, which is now only limited to sporadic case reports, is thus important (3-5). This tumor shows the unique association with HHV8 and EB viruses.

References:

  1. Delecluse HJ, et al. Plasmablastic lymphomas of the oral cavity: a new entity associated with the human immunodeficiency virus infection. Blood 1989;1413-1420.
  2. Katano H, et al. HHV8-associated solid lymphomas that occur in AIDS patients take anaplastic large cell morphology.
  3. Miranda EG, et al. Breast plasmacytoma in a patient with human immunodeficiency virus. J Clin Oncol 2001 Jul 1;19(13):3290-3291.
  4. Lallemand F, et al. Multiple myeloma in an HIV-positive man presenting with primary cutaneous plasmacytomas and spinal cord compression. J Am Acad Dermatol 1998;39:506-508.
  5. Nosari AM, et al. Multiple myeloma associated to HIV infection: Report of two patients. Eur J Haematol 1996;56:98-99.

    Back to top


Last Updated February 1, 2005