Two Articles by Clark

 
Also in 1995, two articles relevant to matters germane to this discussion, one by Clark alone and the other by Clark, Tucker, and Goldstein, appeared in two separate issues of Acta Oncologica. To analyze those publications thoroughly would require more space than is allowable and, for that reason, we have elected to confine our comments to the titles and abstracts of them (Figures 10 and 11).
 

"The Nature of Cancer: Morphogenesis and Progressive [Self]-disorganization in Neoplastic Development and Progression."

 
Clark WH Jr. Acta Oncologica 34(1):5–21, 1995.
 
"The aberrant forms of life, neoplasia and cancer, are discussed under the events at the beginning of neoplasia and under five classes of neoplastic lesions. The lesional classes are: 1) The precursor state; 2) Intermediate lesions; 3) Primary cancer; 4) Metastasis; and 5) Metastasis from metastasis. The events at the beginning are a diverse group of agents and mechanisms that induce the lesions of the precursor state, not cancer. The lesions and events produced by induction are similar regardless of the agent. Thus, there must be similar biological principles and mechanisms operative in different neoplastic systems. The classes of neoplastic lesions and cancer are described and a theory derived therefrom. The theory is: Any perturbation that alters a cell or a group of cells and their stroma so that they no longer respond appropriately to the forces of tissue, organ, and organismal maintenance, may induce a neoplastic system. The sequential progression of lesions of the induced neoplastic system is the result of a successive series of flaws in the continuum of reciprocal interactions between a group of cells and their stroma. The flaws, appearing seriatim, produce progressive (self)-disorganization of the lesions and progressive loss of response to the forces of tissue and organ maintenance."
 
Let readers of this abstract decide for themselves whether the content of it illuminates their concept of the nature of cancer.
 
Clark: "The aberrant forms of life, neoplasia and cancer,. . ."
 
Critique: Cancer is the malignant epithelial expression of neoplasia.
 
Clark: "The lesional classes are: 1) The precursor state; 2) Intermediate lesions; 3) Primary cancer; 4) Metastasis; and 5) Metastasis from metastasis."
 
Critique: The five classes of neoplastic lesions set forth by Clark in 1995 represent his most recent view of the "steps of tumor progression", a scheme first proposed by him in 1966 and then again in 1984 (see page 49 of this essay). The entire construct of five "lesional classes" of "neoplasia and cancer" is flawed conceptually and is entirely speculative, devoid of any scientific validity, and, furthermore, never the same in successive articles of Clark himself. For one example, in this same article from Acta Oncologica, No. 1, 1995, Clark included under "Precursor state" (another speculation defined obliquely by him as involving "an entire organ" and consisting of "all proliferative lesions and atypical cells of a neoplastic system that are confined to a single tissue compartment and only grow for a limited time..."), "dysplasia" which he then proceeded to define differently, yet again, as follows: "Dysplasia is a focal proliferation of cells showing abnormal tissue organization and scattered cells with cytological atypia similar to some cells seen in a primary cancer. Dysplastic lesions grow for a while and then growth ceases or becomes quite indolent: growth is temporally restricted. With rare exceptions the lesions are confined to a single tissue compartment." In a second example, Clark claimed that "The prototype of intermediate lesions is carcinoma in situ", but, in 1990, in Human Pathology, he had written that: "The diagnosis of carcinoma in situ (melanoma in situ, malignancy in situ) is a contradiction in terms, the prototype of an oxymoron"14.
 
Clark: "The events at the beginning are a diverse group of agents and mechanisms that induce the lesions of the precursor state, not cancer."
 
Critique: A "diverse group of agents and mechanisms" do not qualify as "events".
 
Clark: "The lesions and events produced by induction are similar regardless of the agent. Thus, there must be similar biological principles and mechanisms operative in different neoplastic systems."
 
Critique: This is not only wholly speculative, but a non-sequitur.
 
Clark: "The classes of neoplastic lesions and cancer are described and a theory derived therefrom."
 
Critique: "Primary cancer" is one of the "five classes of neoplastic lesions" as advocated by Clark and is not separate from them.
 
Clark: "The theory is: Any perturbation that alters a cell or a group of cells and their stroma so that they no longer respond appropriately to the forces of tissue, organ, and organismal maintenance, may induce a neoplastic system."
 
Critique: A "perturbation" has capability for inducing a neoplasm, but it cannot induce a "neoplastic system". In addition, the theory put forth by Clark turns on what "respond appropriately" means, and that is never made clear by him.
 
Clark: "The sequential progression of lesions of the induced neoplastic system is the result of a successive series of flaws in the continuum of reciprocal interactions between a group of cells and their stroma."
 
Critique: The idea of flaws in the relationship between a group of cells and its stroma as being responsible for the development of cancer was suggested as long ago as 1961 by Van Scott and Reinertson22.
 
Clark: "The flaws, appearing seriatim, produce progressive (self)-disorganization of the lesions and progressive loss of response to the forces of tissue and organ maintenance."
 
Critique: The meaning of this sentence alludes us.
 

"Parenchymal-stromal Interactions in Neoplasia. Theoretical Considerations and Observations in Melanocytic Neoplasia."

 
Clark WH Jr, Tucker MA, Goldstein AM. Acta Oncologica 34(6):749–57, 1995.
 
"The paper briefly reviews the reciprocal and continuous reciprocal interactions between epithelia, mesenchyme, and extracellular matrix in the development and maintenance of organismal form in multicellular organisms in the animal kingdom and describes the progressive changes in the parenchymal-stromal interactions in melanocytic neoplastic development and progression. In addition to the parenchymal stromal form in non-lesional skin seven different and unique stromal patterns are described. These have been termed: 1) The stroma (diffregress) of programmed differentiation leading to lesional regression characteristic of common nevi; 2) Concentric eosinophilic fibroplasia (cef), the hallmark of precursor nevi (dysplastic nevi) with and without melanocytic nuclear atypia; 3) Fibroplasia with angiogenesis (fa) commonly seen in superficial spreading melanoma without metastatic competence (SSM); 4) Lamellar fibroplasia (lf) seen in precursor nevi and melanomas with and without metastatic competence; 5) Diffuse fibroplasia with angiogenesis (dfa), 6) Narrow, uniform concentric eosinophilic fibroplasia (nucef), 7) No parenchymal-stromal interaction (nopsi); the last three being seen in the heterogeneous stroma of melanomas of the superficial spreading type with metastatic competence. The changes in neoplastic stroma proceed in concert with the changes in the parenchyma characteristic of melanocytic tumor progression."
 
The language employed in this abstract of one paragraph, like so much of Clark's writing about the subject of "tumor progression" for more than a decade and about "melanocytic dysplasia" and "melanocytic neoplasia" for nearly two decades, contributes to making the message impenetrable. In this paragraph, for example, the phrase "reciprocal and continuous reciprocal interactions" is unfathomable; and the terms "parenchymal stroma", "metastatic competence", and "neoplastic stroma" are not readily comprehensible. Last, the abbreviations and acronyms "diffregress", "cef", "fa", "lf", "dfa", "nucef", and "nopsi" are distracting at best and opaque at worst. Parenthetically, not all of the so-called patterns of fibroplasia are really distinguishable from one another, e.g., "concentric eosinophilic fibroplasia" from "narrow, uniform, concentric eosinophilic fibroplasia" or "fibroplasia with angiogenesis" from "diffuse fibroplasia with angiogenesis". Our position that "concentric eosinophilic fibroplasia" and "lamellar fibroplasia" are a continuum has been stated by us previously23.

View Figure  		Fig. 10  Acta Oncologica, No. 1, 1995.

View Figure  		Fig. 11  Acta Oncologica, No. 6, 1995.
 
Clark et al. attempted in this article to give legitimacy to the concept of "tumor progression in the melanocytic system", i.e., from common nevus to metastatic melanoma, by demonstrating what they contend to be repeatable, distinctive changes in the stroma at every stage in that chronological sequence. That the attempt was doomed to fail can be judged from their definitions of two words integral to their thesis, to wit, "parenchyma" and "stroma". This is how Clark and coworkers defined those pivotal terms in the Glossary at the end of the article:
 
Parenchyma: "The parenchyma of a neoplasm consists of the distinguishing cells of the tumor; the phenotypic cells of the tumor, from which it derives its name."
 
Stroma: "Neoplastic stroma consists of the collagens, the elastic fibers, the glycosoaminoglycans, the proteoglycans, the non-collagenous adhesive proteins such as fibronectin, fibrocytes, other cellular elements such as masts cells, vascular structures, and nerves. As a rule the organization and resultant form of the neoplastic stroma is different from the parent stroma of the tumor site and is tumor-specific, at times uniquely so."
 
The definitions just recorded of Clark are at variance with those of classic pathology as expressed in Dorland's Medical Dictionary, with which we agree, in these words:
 
Parenchyma: "The essential elements of an organ; used in anatomical nomenclature as a general term to designate the functional elements of an organ as distinguished from its framework, or stroma."
 
Stroma: "The supporting tissue or matrix of an organ, as distinguished from its functional element, or parenchyma."
 
To illustrate "parenchymal-stromal interactions in tissue maintenance", Clark proposed the "cutaneous scar" as his model and, amazingly, wrote about the "phenomenon" thus: "The epidermis (the parenchyma of a scar) and dermis (the stroma of a scar) have a different and unique appearance when compared with the form of skin where there is no scar, an appearance that is irrevocable, lasting as long as the organism." Epidermis of a scar does not qualify as parenchyma and dermis of a scar is not stroma.
 
In sum, Clark's premise, [i.e. "Progression from precursor lesions to melanoma (cancer) is accompanied pari passu by significant changes in tumor parenchyma and tumor stroma"] is faulty, his definitions (e.g., parenchyma, stroma) are wrong, his observations (seven unique stromal patterns are encountered in the course of progression from common nevus to metastatic melanoma) are inaccurate, and his conclusions ("The common nevi are dominated by the stroma of differentiation with regression; the precursor nevi by concentric eosinophilic fibroplasia; melanoma without competence for metastasis (radial growth phase superficial spreading melanoma) by fibroplasia with angiogenesis; and melanoma with competence for metastasis (vertical growth phase superficial spreading melanoma) by stromal heterogeneity"), as is expected in the context of a so seriously flawed premise, definitions and observations, are equally wrong. Last, they are at variance with Clark's previous statements about the same subjects. For example, in 1984, lamellar fibroplasia was said by Clark4 to be one of the "five distinctive histologic features" of "melanocytic nevi with dysplasia", but, in 1995, Clark concluded that only "concentric eosinophilic fibroplasia", and not "lamellar fibroplasia", was present in precursor lesions.