In Reply:—

I would like to make three points in respect to the comment made by Professor Kenneth E. Shepherd in his letter.

No doubt that there is more to postoperative metastasis risk than immunosuppression. Among other factors, the physical manipulation of the tumor may release tumor cells into the circulation, 1and the sudden drop in levels of tumor-derived angiostatic agents may promote the development of existing micrometastases. These additional risk factors may indeed exacerbate the consequences of the suppression of natural killer cells evident in our study, 2especially given the role of natural killer cells in controlling both the seeding of circulating tumor cells and the development of existing micrometastases.

Nevertheless, our study 2was concerned with the effects of hypothermia, rather than tumor removal, on natural killer activity and resistance to metastasis. Angiogenesis inhibitors such as angiostatin are not expected to play a role in these respects, and certainly could not be implicated for the enhancement of metastasis seen in our study, as no primary tumor was removed. The study of natural killer cell-mediated resistance to metastasis under this condition is advantageous in discerning their unique role.

In accordance with the suggestion to couple the impact of angiostatic agents and immunosuppression in studying the pathobiology of perioperative metastasis, we have now begun to use surgical removal of spontaneously metastasizing tumors to better simulate the clinical setting, and study the interaction of immunosuppression with other factors that promote metastasis.

Motomura K, Koyama H, Noguchi S, Inaji H, Kasugai T, Nagumo S: Malignant seeding of the lumpectomy cavity upon breast-conserving surgery. Oncology 1999; 57:121–6
Ben-Eliyahu S, Shakhar G, Rosenne E, Levinson Y, Beilin B: Hypothermia in barbiturate-anesthetized rats suppresses natural killer cell activity and compromises resistance to tumor metastasis: A role for adrenergic mechanisms. A nesthesiology 1999; 91:732–40