American Journal of Dermatopathology:
Letter to the Editor
Department of Pathology, Long Island Jewish Medical Center New Hyde Park, NY.
To the Editor:
There has been long-standing controversy and heated debate concerning the validity of the term dysplastic nevus. In examining the pros and cons of the argument, one will find that the debate is not really about what one observes with his or her eyes both clinically and microscopically, but what one believes. Those who believe in the sequential multistep tumorigenesis theory will defend the term and concept of dysplastic nevus with might and passion. By contrast, those who do not believe in the multistep tumorigenesis theory will dismiss dysplastic nevus as a bogus term and concept.
The multistep tumorigenesis theory assumes that tumor develops via a sequential multistep process, such as from normal melanocytes to hyperplasia to nevus to dysplastic nevus and eventually to melanoma. Many believe this is true and hold the view that the multistep tumorigenesis theory has been proven scientifically. It is therefore appropriate to examine the literature critically in this regard.
When the term dysplastic nevus was first introduced by Greene et al in 1980, they believed that dysplastic nevi were precursors of cutaneous melanoma and stated that they “fit nicely into the schema of progression from hyperplasia to dysplasia to neoplasia that is accepted in many epithelial tumor systems, both experimental and human.”1 Here “the schema of progression” refer to the sequential multistep tumorigenesis theory often attributed to Leslie Foulds. However, acceptance does not equate with scientific proof. The reference Greene et al cited in support of their above statement is a review article dedicated to Leslie Foulds by Farber and Cameron published in 1980.2 If one reads this article critically, one will find that there was really no solid data in support of the multistep tumorigenesis theory. It was more speculative rather than evidence based.
The frequently cited contemporary article in support of the multistep tumorigenesis theory is a paper published by Fearon and Vogelstein in 1990.3 Vogelstein and colleagues have devoted decades in an attempt to validate the multistep tumorigenesis theory at the molecular level using colorectal neoplasia as a model system. They proposed a genetic model for colorectal tumorigenesis with a diagram depicting progression from normal epithelium to hyperproliferative epithelium to early adenoma to intermediate adenoma to late adenoma to carcinoma and then to metastasis with corresponding genetic alternations (Fig. 3 in the paper). This diagram has been reproduced and cited as evidence in support of the multistep tumorigenesis theory in popular textbooks, such as Rosai and Ackerman's Surgical Pathology.4 However, if one carefully reads the original article,3 one will find this article is really miscited, the data did not validate the multistep tumorigenesis theory at all. The authors stated in that article that their proposal of the genetic model for colorectal tumorigenesis is “rudimentary” and actually indirectly refuted the multistep tumorigenesis theory by declaring that “accumulation, rather than order, is most important.”
I was a firm believer of the multistep tumorigenesis theory. However, after critical review of literature, I came to a conclusion that there are no solid data supporting this theory. Most, if not all, tumors develop via a nonsequential stochastic process. At the molecular level, neoplasms are basically genetic diseases, involving alteration in DNA, such as deletion, substitution, insertion, or translocation. These genetic alterations are nonsequential and stochastic. There are no intermediate lesions between nevus and melanoma, only lesions that we are unable to classify into either of these categories based on histopathologic criteria. Dr. Bernard Ackerman summed it up well in these words: “In the realm of melanocytic neoplasms, there are only four possible answers: nevus, melanoma, melanoma in association with a nevus, and ‘I don't know.’”5
Sheng Chen, MD, PhD
Department of Pathology, Long Island Jewish Medical Center New Hyde Park, NY
1. Greene MH, Clark WH Jr, Tucker MA, et al. Precursor naevi in cutaneous malignant melanoma: a proposed nomenclature. Lancet. 1980;2:1024.
2. Farber E, Cameron R. The sequential analysis of cancer development. Adv Cancer Res. 1980;31:125-226.
3. Feason ER, Vogelstein B. A genetic model for colorectal tumorigenesis. Cell. 1990;61:759-767.
4. Rosai J. Rosai and Ackerman's Surgical Pathology. 9th ed. Edinburgh, UK: Mosby; 2004;808.
5. Ackerman AB, Cavegn BM, Robinson MJ, et al. Ackerman's Resolving Quandaries in Dermatology, Pathology, and Dermatopathology. Philadelphia, PA: Williams & Wilkins; 1995;48.
© 2010 Lippincott Williams & Wilkins, Inc.