KAI1/CD82 is a novel target of estrogen receptor-mediated gene repression and downregulated in primary human breast cancer (Int J Cancer 123: 2239-2246, 2008)

Abstract

With interest I read the article by Christgen et al. that was recently published under the Fast Track heading in the International Journal of Cancer.1 The major reported findings, nicely summarized in the article's title, are that (a) KAI1/CD82 is downregulated in 77% of human breast carcinomas and (b) KAI1 downregulation is strongly correlated with ER positivity. I do not question the accuracy of the reported data. However, I take exception to the authors' claim, repeated several times throughout their article, that their major findings are novel. The fact is that my colleagues and I published practically identical observations 3 years ago.2 Specifically, we reported downregulation of KAI1 expression in 74 of 98 (76%) archival breast carcinomas, and we mentioned that KAI1 downregulation was significantly more common in ER-positive (48 of 55, 87%) versus ER-negative (22 of 36, 61%) tumors (p = 0.0052). Although the focus of our manuscript was on new cell membrane biomarkers involved in breast oncogenesis, we very clearly detailed our observations on KAI1 expression in human breast cancer tissues and 9 cell lines including MCF-7, which Christgen used as a negative control. We published our data in Clinical Cancer Research, which can hardly be considered an obscure journal. It is hard to imagine that even a cursory review of the literature involving KAI1/CD82 and breast cancer would not have brought our article to Christgen et al.'s attention. A scientific finding does not become novel by ignoring its previous description by other investigators. It would have been appropriate for Christgen etal. to acknowledge our published data. Of course, in so doing they would have had to admit that most of their data are confirmatory rather than novel, notwithstanding the fact that they documented the inverse relationship between ER and KAI1 expression in greater detail. Christgen et al. also give the impression that their immunohistochemical staining method for KAI1 is a major technological advance. In fact, they used the same anti-KAI1 antibody and the same heat induced epitope retrieval method that we published in 2005.2 Already in 1999, we had demonstrated that KAI1 protein expression can reliably be evaluated in fixed tissues by immunohistochemistry using a different antibody.3 Christgen also confirms our original observation that KAI1 expression is not correlated with axillary lymph node involvement, tumor grade, or c-erbB-2 status.2 Christgen et al. describe several in vitro experiments as well. They show that the growth and migration of ER-positive, but not ER-negative, breast cancer cells is inhibited by fulvestrant, an antiestrogenic agent, and that this is correlated with up- or downregulation of 4 indicator genes. The authors acknowledge that those experiments confirm previously published observations by other research groups.4-7 It seems that Christgen et al.'s only original and novel contribution is the finding that KAI1 is upregulated by fulvestrant in 2 ER-positive breast cancer cell lines. The International Journal of Cancer may consider following the example of other scientific journals that discourage authors from inserting superlatives (“first”, “largest”, etc.) into their manuscripts. Some of those claims do not stand up to scrutiny. Evidently the reviewers assigned to Christgen's article were not sufficiently familiar with the literature to rebut the authors' exaggerated claims of novelty. In conclusion, I believe the data presented by Christgen et al. are interesting and a useful addition to the scientific literature. However, their major findings are confirmatory (rather than novel) in nature. It is inappropriate for the authors to suggest that their methodology and experimental results are groundbreaking, when they essentially replicate portions of an article that they unfortunately chose not to cite. Yours sincerely, Joseph Geradts MD