Correspondence

Case 34-1993: Kaposi's Sarcoma

N Engl J Med 1994; 330:146January 13, 1994

Article

To the Editor:

Kaposi's sarcoma was an enigma to Dr. Kaposi in 1872, and it is an enigma still. Case 34-1993 (Aug. 26 issue)1 concerning Kaposi's sarcoma ends with a pathological discussion. There is a long-standing scholarly debate about whether this sarcoma is a cancer or a reactive process that sometimes behaves aggressively. It is stated that in the patient described, Kaposi's sarcoma “metastasized” to hilar lymph nodes. The anatomical diagnosis is stated as “Kaposi's sarcoma of skin and oral mucosa, with extensive metastasis to lungs.” Is the word “metastasized” used here to mean the spread of a malignant cell to a noncontiguous site? The skin and oral mucosa cannot be considered to be the primary sites simply because they have little to hide. Are there data in the case inconsistent with the view that Kaposi's sarcoma arises multifocally from a nonclonal population of mesenchyma-derived cells, driven to proliferate by aberrant cytokine circuits possibly triggered by an infectious agent? A confirmed case of metastatic Kaposi's sarcoma would be of tremendous importance, because it would challenge current thinking about this biologically fascinating and clinically important disease.

Kathryn A. O'Connell, M.D., Ph.D.
Johns Hopkins University School of Medicine, Baltimore, MD 21205

1 References

1. 1

   Case Records of the Massachusetts General Hospital (Case 34-1993). N Engl J Med 1993;329:645-653
   Full Text | Web of Science | Medline

Author/Editor Response

Dr. Mark replies:

To the Editor: A recent review of premalignant nonepithelial lesions classifies Kaposi's sarcoma in patch stage as an early form of nonepithelial cancer1. The lymphangitic distribution of Kaposi's sarcoma in the lung in other cases is additional proof that the tumor spreads sequentially rather than arising randomly2. The mechanism of the malignant transformation is unproved.

One can speculate about nonclonal populations, cytokine circuits, and infectious agents as causes of sarcoma. The tumor could also have originated in many organs in the body not examined at autopsy. These are hypotheses. The statement in the Case Records that the disease was a sarcoma that had metastasized was based on the best available evidence rather than on a hypothesis. Two years previously, the patient had obvious Kaposi's sarcoma of the skin and mucosa. At that time, radiographic and pathological studies of the chest were performed that showed no evidence of Kaposi's sarcoma. After two years, the same tumor appeared in an organ where it had not been found earlier. This is consistent with the definition of metastasis as the transfer of disease from one part of the body to another with characteristic lesions at the new location, although the finding does not exclude multicentricity. The tumor in the lung at autopsy was malignant by virtue of its focal permeative invasion of arteries and cytologic characteristics.

I have always thought the ancient histologic description of Kaposi's sarcoma as malignant granulation tissue to be an intriguing concept, so perhaps Dr. O'Connell and I are not too far apart philosophically.

Eugene J. Mark, M.D.
Harvard Medical School, Boston, MA 02115

2 References

1. 1

   Seidman JD, Berman JJ. Premalignant nonepithelial lesions: a biological classification. Mod Pathol 1993;6:544-554
   Web of Science | Medline

2. 2

   Purdy LJ, Colby TV, Yousem SA, Battifora H. Pulmonary Kaposi's sarcoma: premortem histologic diagnosis. Am J Surg Pathol 1986;10:301-311
   CrossRef | Web of Science | Medline