The modifier "atypical" first was applied to Spitz's nevus by Huarte in 1969 in an article devoted to "Nevus atípico de Spitz (melanoma juvenil). Estudio clínico-patológico de nueve casos [Atypical Spitz's nevus (juvenile melanoma). Clinico-pathologic study of new cases]"¹² (Figs. 10, 11) in which he justified the choice of that designation as follows:

Paradoxically, what Helwig called atypical fibroxanthoma and for 25 years insisted was benign turned out to be malignant fibrous histocytoma and that Spitz asserted was malignant melanoma turned out to be a nevus.

In 1975, Reed, Ichinose, Clark, and Mihm, in an article given to "Common and Uncommon Melanocytic Nevi and Borderline Melanomas," spawned the dizzying notion of a "malignant variant" of Spitz tumor and advocated the equally untenable concept of "benign metastases" (the latter flawed concept having been introduced in 1939 by Steiner in regard to fibroleiomyoma). These lines come directly from the publication of Reed and his coworkers:

The example of "benign metastasis" provided by Reed et al. (their Figs. 27 and 28) was purported by them to be a "cellular blue nevus" that gave rise to a large nodule in the parenchyma of a lymph node, which they characterized morphologically in the legend to their Fig. 28 thus: "[it is] cytologically benign and has a pattern identical to that seen in the primary." The coworkers concluded, therefore, that the nodule in the node represented a "benign metastasis" from the "cellular blue nevus," rather than a metastasis to a lymph node from a melanoma primary in the skin, which, in reality, was the case, as is patent in the photomicrographs presented. Reed, Ichinose, Clark and Mihm used "Spitz tumor" interchangeably with "spindle and epithelioid cell nevus," the justification for that synonymy being offered by them in these words:

Today the term "Spitz tumor" is substituted for "Spitz nevus" only by those of the Clark school of which Reed, and Mihm are exemplars. Curiously, Reed, one of the most vocal of that school, in an editorial in 1999 in Human Pathology (vida infra) called into question, properly, the applicability of his own term "tumor" as a replacement for nevus in what nearly everyone else refers to as "Spitz's nevus," that distinctive proliferation qualifying as a melanocytic nevus because it is a benign neoplasm constituted of nests and/or fascicles of abnormal melanocytes.

In 1976, Helwig, then chief of the Skin Branch of the Armed Forces Institute of Pathology (AFIP), contributed substantially to bewilderment about the behavior biologically of what he called "spindle and epithelioid cell nevus," which in the 1970s was the most popular term for what today is known nearly universally as Spitz's nevus. In a pivotal article concerning melanomas in children¹⁴ (Figs. 12, 13), Helwig reported on 23 young persons, all of whom had an unquestionable melanoma as evidenced by metastases from it. In no instance was a diagnosis of melanoma suspected clinically in the 23 patients studied by Helwig. This is what he found:

He opined that some of those melanomas resembled "spindle and epithelioid nevus" histopathologically, and he maintained further that that particular kind of melanoma in young persons had a more favorable prognosis than other melanomas for the reason solely that when viewed by conventional microscopy it simulated so closely an authentic spindle and epithelioid nevus. The data on which Helwig predicated hopeful outlook for such melanomas were as follows:

Spitz had made a similar observation nearly 30 years before, both she and Helwig seeking to communicate that living for decades with metastases somehow implies that the primary neoplasm was less malignant than usual. Nothing could be further from the truth.

All 23 patients developed metastasis, but some melanomas that seemed to involve regional nodes only were difficult to identify as the malignant melanomas they truly were on the basis of findings histopathologically in the neoplasm primary in the skin. According to Helwig, the primary neoplasms had an architectural pattern and cytopathological attributes very much like those of a spindle and epithelioid cell nevus. These observations led Helwig to state that "On the basis of the behavior of the lesions that show spindle and epithelioid cell characteristics, it seems probable that this group has a more favorable prognosis."

One such primary melanoma that mimicked a Spitz's nevus histopathologically developed on the helix of the right ear of a 2-year-old girl and is depicted in our Fig. 14 (Helwig's Fig. 5); a metastasis from it to a lymph node is shown in Fig. 15 (Helwig's Fig. 6). That this child happened to have only regional lymph node metastases and was alive for 3 years and that 6 other children with metastases seemingly to regional lymph nodes only were discovered to be alive after 18, 9, 9, 9, 9, 7, and 2 years (at the time of the publication of the article from the AFIP), prompted Helwig to these conclusions:

A melanocytic neoplasm that metastasizes is a malignant melanoma and it matters not a whit whether or not it resembles histopathologically a "spindle and epitheioid nevus." Nor does the diagnosis change an iota if the patient survives for scores of years with metastasis from it or whether metastases are found only in "regional lymph nodes" but not at "distant sites." Incontrovertible is the fact that a primary cutaneous neoplasm of melanocytes that metastasizes is a melanoma. Parenthetically, Helwig extended his concept of favorable prognosis for a patient with metastasis limited to regional lymph nodes to certain cellular blue nevi that he believed had capability for metastasis and then exercised it. In the same article, "Malignant melanoma in children," Helwig had this to say in 1976 about metastasizing blue nevi:

One year later, in 1977, Weedon and Little gave credence to the belief that "atypical Spitz's nevus" was a bona fide diagnosis by setting forth their criteria for it histopathologically and for differentiation of it from melanoma.¹⁵ They did that in these lines:

By 1989, the concept of "atypical Spitz's nevus" had become entrenched firmly in the literature of dermatology, general pathology, and dermatopathology, a phenomenon illustrated well by Palazzo and Durey in their Table 2 (our Fig. 16) in which a diagnosis of "atypical Spitz's nevus" was made in Cases 1 and 3 and in their discussion of criteria they suggested be utilized for making the diagnosis of it histopathologically.¹⁶ This is what those associates advised:

That same year, 1989, KJ Smith and her colleagues in the Skin Branch of the AFIP expanded on Helwig's concept of a more favorable prognosis for persons, usually very young ones, whose melanoma bore striking similarities histopathologically to Spitz's nevus.¹⁷ They attempted to justify both the idea and the designation of "malignant Spitz's nevus," a daunting task because the phrase "malignant Spitz's nevus" is unsettingly incongruous; no nevus is malignant and, therefore, no nevus ever metastasizes. A melanocytic nevus may be either (1) a hamartoma made up of abnormal melanocytes and often of other abnormal constituents at a particular anatomic site, like terminal follicles where normally reside only vellus follicles, such is the case for a congenital giant hairy nevus, or (2) a benign neoplasm of abnormal melanocytes, as is the situation for a Spitz's nevus acquired after birth (rarely Spitz's nevus may be congenital, presenting itself then usually as agminated or systematized). The work of Smith et al., which derived in large part from the speculations of Helwig in 1976 (Helwig in 1976 and Smith et al. in 1989 having been stationed at the same Branch of the same Institute where Arthur Allen had been a staff pathologist prior to his coming to Memorial Hospital in New York City, namely, the Skin Branch of the AFIP) was published in the American Journal of Surgical Pathology under the heading, "Spindle cell and epithelioid cell nevi with atypia and metastases (malignant Spitz's nevus)" (Figs. 17, 18). Of 32 patients with what the coworkers claimed was a Spitz's nevus, 6 underwent lymph node dissection and in all 6 was found evidence of metastasis of melanoma. Photomicrographs of 3 of the 32 lesions were published and even a glance at them at scanning magnification enables a diagnosis of melanoma to be rendered reflexively and with confidence (Figs. 19, 20); a sheet of neoplastic melanocytes extends throughout the dermis and far into the subcutaneous fat, a finding indicative of melanoma and not of a nevus of any kind. Despite those changes, dramatic as they are, Smith and her colleagues came to these positions:

In brief, the spirit of Helwig (Allen by then was long forgotten) was very much alive at the AFIP 13 years after publication of his supposition to the effect that melanomas [in children] have a better prognosis because they resemble so closely spindle and epithelioid cell nevus histopathologically. The notions of "malignant Spitz's nevus" and "metastasizing Spitz's nevus," however, are an affront to basic principles of classic pathology. A Spitz's nevus is a nevus and, like melanocytic nevi of all kinds, is incapable of metastasis. Moreover, the idea that "these lesions [of malignant Spitz's nevus] have the ability to metastasize to local lymph nodes but are not capable of widespread metastases" also is contrary to fundamental precepts of pathology; metastasis, by definition, is widespread, no matter whether it is said to be a "satellite," "in transit," or "regional." Conceptually, a "satellite" metastasis is a "distant" metastasis.

Just two years later, in 1991, Barnhill and his cohorts added a new dimension to the ever-more murky issue that attracts our attention here, writing as they did in Human Pathology about what they called "atypical pigmented spindle-cell nevus (pigmented spindle-cell nevus with architectural and/or cytologic atypia)."¹⁸ They titled their article, "The histologic spectrum of pigmented spindle cell nevus: A review of 120 cases with emphasis on atypical variants." In it, they fired a fusillade of mixed messages, among those being that pigmented spindle cell nevus was benign (in fact, they designated it a nevus in every one of scores of references made to it), but they implied that it might presage development of melanoma because of "atypical" features inherent in it (those purported atypical attributes being "abnormal architectural findings;" nuclear atypia was only "slight to moderate" in the vast majority of their 95 "cases" (Figs. 21, 22). But no measuring stick for "abnormal architectural findings" was provided by them, to say nothing of criteria for "normal architectural findings;" "normal" architectural pattern or "architectural order" is a contradiction in terms for any a process deemed to be pathological. "Architectural disorder" has never been defined because it is fictitious. The authors went on to state that pigmented spindle cell nevus is different histopathologically from Spitz's nevus, yet they conceded it has features in common with Spitz's nevus. They advised that pigmented spindle cell nevus may be "typical" or "atypical," and averred that the nevus is confused often with melanoma histopathologically, even though it does not behave like melanoma. And as if to underscore the sinister aspect of "atypical pigmented spindle-cell nevus," they reminded readers that, in 1975, when Richard J. Reed first described "pigmented spindle-cell tumor," he believed it to be a type of "borderline" melanoma. The muddle created by Barnhill et al.^{18,21,23,24,27,29,31,33,34} in regard to "atypical variants" of pigmented spindle cell nevus was not of their own invention alone; their forebears, from Darier and Civatte¹ through Packand Anglem,⁴ Spitz,^2,3,8 Allen,^7,8 Huarte,¹² Reed et al.,¹³ Weedon and Little,¹⁵ Palazzo and Durey,¹⁶ Smith et al.,¹⁷ to Reed alone,³² paved the way for the untenable concept of "atypical Spitz's nevus" as it was advanced trenchantly by Barnhill himself and by Barnhill in company with associates during the last decade of the 20th Century. Just how opaque was the idea of "atypical spindle-cell nevus" can be appreciated by taking note of this quotation, directly, from Barnhill et al.¹⁸ in their article in 1991 in Human Pathology:

In a memorial to his late wife, Sophie Spitz, Allen, in 1991, in an article titled, "Classics in oncology: Introduction to melanomas of childhood by Spitz,"¹⁹ recalled her/their work on melanomas in children. This is some of what Allen wrote then:

Unfortunately, Spitz,³ Allen,^7,19 and Allen and Spitz⁸ never came even close to admitting the indubitable error in misinterpretation of Spitz in her original article. As he had done, time and again, for more than 50 years, Allen continued to take unbridled liberties with facts, in this case not only with what Pack had written and affirmed, but what Spitz and he, together, had written and affirmed. For one example, Spitz's article on "melanomas of childhood" was not "the culmination of a series of articles" by them; it was her first article about the subject.² For another example, it was Pack who, earlier than Spitz (and surely before Allen), had realized that "pre-pubertal melanoma of the skin" was benign biologically, but Allen never mentioned that incontrovertible fact in his paean to Spitz (and to himself). For a third example, Allen gave no hint that in 1948, Spitz asserted, without equivocation, that juvenile melanoma was a malignant melanoma. On the contrary, he declared wrongly that Spitz and he understood from the very outset that juvenile melanoma was of no more or less concern than any other nevus. Would that this sorry performance of revision of history by Allen could be attributed to dotage, but it was characteristic of him throughout his entire professional life.

In 1992, Casso and others wrote, in an article about "Spitz's nevi,"²⁰ of their frustration in endeavoring to comprehend the meaning of "atypical Spitz's nevus." That sense of discouragement is apparent in these words:

In 1993, Barnhill returned again to what he termed the "atypical variant of Spitz's nevus," which, by then, he regarded as "premalignant" because of its tendency to "progress to melanoma."²¹ This is what he said:

The impenetrable language that typifies phrases like "intermediate range of (premalignant) architectural and cytologic abnormalities" that "may be evidence of an atypical process with some potential for progression to melanoma" makes the effort by Barnhill, alone, in 1993²¹ no less futile than that of Barnhill et al.¹⁸ in 1991 at elucidating what was meant by "atypical variants" of pigmented spindle cell nevus, and for the same reason: incomprehensibility. No matter how careful heed is paid to the language employed by Barnhill, it simply is impossible to fathom the "concept" he attempted to set forth; what, exactly, is meant by "intermediate range," "premalignant," "architectural abnormalities," "atypical process," and "some potential for progression to melanoma?" Barnhill did not define a single one of those terms and phrases, which, in the final analysis, constitute mere verbiage devoid of meaning entirely. What seems to have meaning in this context is that a nevus is benign, a melanoma is malignant, and approximately 5–10% of all melanomas in man (when all races are taken into account) begin in association with a pre-existing nevus. In the vast majority of instances of that particular affiliation, there is no evidence that abnormal melanocytes of the nevus (the vast majority of which are Clark's type) housed in the epidermis or situated at the dermo-epidermal junction actually turn into abnormal melanocytes of melanoma. In the case of Clark's nevi, a melanoma develops in contiguity/continuity with it, the source of the abnormal melanocytes of melanoma seeming to be melanocytes that are cytologically "normal" and disposed as solitary units at the dermo-epidermal junction; in this sense, the melanoma that develops next to the nevus actually begins de novo.

Perkocha, in 1994, in the course of a communication dedicated to classification of melanoma in children and in adults, gave credence to a notion like that expressed by others before him and one that strains credulity seriously, to wit, a Spitz's nevus can metastasize, such a Spitz's nevus being qualified by him as "atypical."²² He even ventured a term for a neoplasm with such propensity, namely, "atypical dermal melanocytic lesion with features of Spitz nevus." This is how Perkocha couched his thesis:

Of course, Perkocha's "atypical dermal melanocytic lesion with features of Spitz's nevus" with "the ability to metastasize" was, and is, a melanoma.

Also in 1995, Barnhill et al. returned once more to the subjects of "atypical Spitz's tumors" and "metastasizing Spitz's tumors," this time by way of reporting on the results of a study of 23 children 15 years of age or younger who were referred to Children's Hospital in Boston with a diagnosis of melanoma.²³ In an article captioned, "Cutaneous melanoma and atypical Spitz tumors in childhood," the authors disclosed how they had separated the neoplasms in the 23 children into four groups as follows: "small cell melanoma," "adult-like melanoma," "Spitz-like melanoma," and "atypical Spitz tumors,"/"metastasizing Spitz tumors." This is the way they conveyed their thinking about what characterized "atypical Spitz-like tumors" and "metastasizing Spitz tumors":

Among the children in their series were two whose melanocytic neoplasm metastasized to regional lymph nodes. Barnhill et al. considered those neoplasms to be "metastasizing Spitz tumors." The coworkers described and illustrated findings histopathologically in the primary neoplasm they claimed to be a "metastasizing Spitz tumor"(Fig. 23; their Fig. 7). The changes noted by conventional microscopy were described by them thus:

"Two patients developed lymph node metastases but were alive and well as of this writing on limited follow-up of 3 and 9 years, respectively, after excision of the primary tumor and lymphadenectomy. Histologically, these tumors exhibited features of Spitz nevus but also asymmetry, deep extension, confluent nodules, prominent cellularity, diminished maturation, and some degree of cytologic atypia (Fig. 7). Cytologically, the cells in both tumors contained abundant eosinophilic cytoplasm with a ground glass appearance as noted in typical Spitz nevi (Fig.7b). The cells also had polyangular contours and exhibited nuclei with enlarged and generally vesicular nuclei. The tumor developing on the arm of the 2-year-old female showed discohesive sheets of multinucleated giant cells, many of which were frankly bizarre in appearance (Fig. 7b). Neither lesion could be distinguished from the group of nine atypical Spitz-like tumors."

The legend to Fig. 7 of Barnhill et al. (our Fig. 23) tells of "Metastasizing Spitz tumor diagnosed as malignant melanoma," a sentence that can only boggle; a metastasizing Spitz tumor, by definition, is a melanoma, and, therefore, the diagnosis, rendered initially, of a benign neoplasm was dead wrong. Despite acknowledgment by Barnhill et al. of findings histopathologically of melanoma in the neoplasms primary in the skin and despite metastases and death from one of them, they insisted, nevertheless, that the diagnosis, in fact, was "metastasizing Spitz tumor." For reasons inexplicable, Barnhill et al. failed, yet again, to recognize those neoplasms that metastasized to be the melanomas they unquestionably were, one of them causing death. The statement that these patients "were alive and well as of this writing on limited follow-up of 3 and 9 years, respectively, after excision of the primary tumor and lymphadenectomy," is meant to imply that prognosis for a "metastasizing Spitz tumor" is better than that for a metastasis of an "ordinary" melanoma, a throwback to the surmise of Helwig in 1976 and Lupton as recently as 2002 (infra vide). The time of survival while bearing metastases is not a criterion for determining "degree" of malignancy of a neoplasm; survival for long with metastases more likely has to do more with factors pertinent to the status of the person who harbors them, rather than to the character of the neoplastic cells themselves. The fact of metastasis verifies malignancy and no histopathologist, no matter how adroit at employing sophistic arguments, including a host of "risk factors," can predict with any accuracy how long a patient with metastases of melanoma will survive; all assessments of prognosis are pure guess. That being so, those speculations have no place in the mind-set of a pathologist, the task of whom is diagnosis, not prognosis. A diagnosis of "benign" or of "malignant" issued by a pathologist communicates to the physician managing a patient that which is of importance, namely, no capability for metastases of the former and potential for metastases of the latter.

The nine "atypical Spitz tumors" reported on by Barnhill et al. were diagnosed first as "melanomas," but were reclassified later as "atypical Spitz tumors" because "they were judged to have features insufficient for unequivocal melanoma and were considered atypical tumors with characteristics of Spitz's nevus." (Fig. 24) On the basis of the findings histopathologically in those "atypical Spitz tumors," as they were described and illustrated by Barnhill et al., it becomes apparent that the diagnosis of melanoma issued initially was correct. Barnhill et al. said this about the "tumors" that they considered to be different from "ordinary Spitz's nevi":

In an endeavor to give added validity to their thesis, Barnhill et al. advised that the two Spitz tumors that metastasized and the 9 "atypical Spitz tumors" resembled histopathologically the "malignant Spitz nevi" reported on 6 years previously by Smith et al. at the AFIP. Curiously, Barnhill and compatriots continued to maintain the stance that the neoplasms under discussion by them in some way were related to Spitz's nevus, even though they acknowledged that some of the changes in them, as assessed by conventional microscopy, were those of melanoma. This is what they wrote in that regard:

"Examination of our two metastasizing Spitz tumors and nine atypical Spitz lesions revealed morphological features similar to those reported by Smith et al. In general, these tumors had features of Spitz nevus but also additional atypical features including large size, significant depth, cellularity, cellular atypia, dermal, and deep and, occasionally atypical mitoses. Many of the latter features have specifically been cited as indicative of melanoma."

Just as the neoplasms reported on by Smith et al.¹⁷ were melanomas, not Spitz's nevi, all of them having metastasized, so, too, it was for two of the neoplasms in children reported on by Barnhill et al. The conclusions of Barnhill et al. that "atypical Spitz tumors [that] metastasize to regional lymph nodes (without further progression)," are "simulates of melanomas," and "the ultimate gold standard [for diagnosing melanoma in childhood] is distant metastases and/or death of the patient" is plain wrong; all of them were melanomas. Any melanocytic neoplasm that metastasizes is a melanoma, not a simulator of melanoma, and, moreover, a metastasis of melanoma anywhere, i.e., to a node (or even to skin nearby), is just as compelling evidence of malignancy as is metastasis widely. Although Barnhill et al. noted peculiarities in what they called "atypical Spitz tumors" and were alert to the potential of those "tumors" for metastasis, they persisted in maintaining that the neoplasms were not melanomas. This is the argument as they expressed it:

When the findings histopathologically recorded by Barnhill et al. are summed, they compute to melanoma, not to Spitz's nevus, including the type of that nevus known curiously as pigmented spindle cell tumor. Furthermore, characterization of certain attributes displayed by the neoplasms under consideration by them, for example, "multifocal or plexiform finger-like infiltration" is befuddling, as is the statement that that particular pattern morphologically "suggested some degree of order or growth control rather than loss of growth control in melanoma." Gobbledygook. In reality, no judgment of worth about "order" or "growth control" (or lack of it) can be reached by virtue of observations made through the lens of a conventional microscope. Even metastasis did not deter Barnhill, and those who joined him in a slew of publications, from trumpeting, again and again, the idea, illogical consummately, that the "atypical Spitz tumors" studied by them were authentic Spitz's nevi and not melanomas.

In 1995, Busam and Barnhill made recommendations concerning terminology of proliferations of abnormal melanocytes in which melanoma was a serious consideration histopathologically, but in which diagnosis of that malignant neoplasm could not be made by them with utter surety.²⁴ This is the solution the coworkers proposed:

The suggestion of Busam and Barnhill not only is impractical because the language they recommend be employed is unintelligible, but it evades the single issue of import, namely, the essential nature of the neoplasm that they term "atypical Spitz tumor." In every instance, that "tumor" must be either a Spitz's nevus, which is benign, or a melanoma, which is malignant. The profusion of words that typify all of the publications of Barnhill seems designed to obscure and evade, thereby skirting the one matter of consequence to a patient and to a physician managing that patient: Is an "atypical Spitz tumor" a nevus or a melanoma?

In an article devoted to "Malignant Spitz's nevus in a 2-year-old Japanese child,"²⁵ Nitta and associates, in 1995, told of a large nodule above the Achilles tendon that they diagnosed by conventional microscopy as being a "malignant Spitz's nevus." This is how they described the findings in that neoplasm:

The conclusion of Nitta et al. is illogical; "prominent lymphatic vessel invasion by melanocytes," for practical purposes, is virtually synonymous with metastasis of a malignant neoplasm and, in fact, the child reported on by the Japanese coworkers died in 2001 of metastases of melanoma (personal communication, Kazuo Hara, M.D.). But by their own admission, Nitta et al. had reason to doubt the legitimacy of their own thesis that the neoplasm under consideration by them truly was a malignant Spitz's nevus. They expressed those reservations in these lines:

"Our case showed many features of spindle cell and epithelioid cell nevus (Spitz's nevus), but several findings were out of the ordinary: the large size (27 x 17 mm), invasion of the subcutis, mitotic figures deep in the tumor, no maturation of melanocytes, and prominent lymphatic vessel invasion by melanocytes. These findings, obviously beyond the spectrum of an ordinary Spitz's nevus, could lead one to consider the case under study as malignant melanoma . . . Since our case presented features quite similar to those described by Smith, et al., we considered the child to be another typical case of MSN [malignant Spitz's nevus]."

Instead of concluding that Smith et al.¹⁷ at the AFIP had opted wrongly in 1989 when they set forth the notion of a Spitz' nevus that metastasized ("malignant/metastasizing Spitz's nevus"), Nitta et al.²⁴ followed them docilely down the same cul de sac and with the same unfortunate results. The contagion generated by the AFIP, beginning with Helwig¹⁴ and followed by Smith et al.¹⁷, spread as far and wide as Boston (Barnhill et al.^18,21,23) and Nagoya (Nitta et al.²⁵).

In 1997, Crotty brought a refreshing point of view to the subject of "metastasizing Spitz's nevi"²⁶ when she wrote as follows:

In short, Crotty understood, rightly, that if any neoplasm made up of melanocytes metastasizes, it is a melanoma. The last sentence of hers just quoted, however, would have been more accurate had she deleted the phrase, "especially beyond local lymph nodes," the reason being that a metastasis is indicative of dissemination of neoplastic cells widely, and that being the case, arbitrary schemes for classifying metastases, e.g., "satellite," "in-transit," "regional," and "distant," with the aim of grading severity of prognosis according to those artificial boundaries, are without merit. A satellite metastasis indicates distant metastasis. There is no such thing as "local metastasis;" all metastasis is widespread.

In 1998, Barnhill, alone, wrote still another time, in an article titled "Childhood melanoma,"²⁷about the same two "metastasizing Spitz tumors" and nine neoplasms in children that he referred to as "atypical Spitz's lesions," and also as "atypical Spitz tumors" as he had done in his article titled, "Cutaneous melanoma and atypical Spitz tumors in childhood," published in 1995.²³ Just as he had before, Barnhill insisted that "metastasizing Spitz tumors" and "atypical Spitz lesions" were akin to the "malignant Spitz's nevus" reported on by Smith and coworkers in 1989. This is what he said:

"Examination of the two metastasizing Spitz tumors and nine atypical Spitz lesions in this series revealed morphological features similar to those reported by Smith et al. In general, these tumors had features of Spitz nevus in addition to atypical features, including large size; significant depth; cellularity; cellular atypia; and dermal, deep, and occasionally atypical mitoses. Many of the last mentioned features have specifically cited as indicative of melanoma."

The "atypical features" cited by Barnhill are those of melanoma, and the criticisms of the concepts of "atypical Spitz tumor," "atypical Spitz's nevus," "atypical Spitz lesion," "malignant (metastasizing) Spitz tumor" and "malignant (metastasizing) Spitz's nevus" enunciated by us earlier in this work apply equally to the ideas restated by Barnhill in 1998.

Also in 1998, N. M. Smith et al. told of cytogenetic studies performed on what they diagnosed as an "atypical Spitz's nevus" that had metastasized to a lymph node.²⁸ On the basis of the outcome of those studies, they were forced to the following statement and conclusion:

It is difficult to fathom why the natural history of melanocytic neoplasms that metastasize is uncertain; the only uncertainty is the length of time that will transpire, if the patient does not die of something else, until death results from the effects of metastases. It is difficult equally to comprehend why a melanoma that has metastasized would be referred to vapidly as a "group of melanocytic lesions."

In 1999, Spatz et al. proposed a system for grading proliferations of abnormal melanocytes that they, like Barnhill et al.,^18,21,23,27 thought shared "some but not all of the classic features of Spitz's nevi or display[ed] unusual features," a condition they also chose to designate "atypical Spitz tumors." In an article titled, "Spitz tumors in childhood: A grading system for risk stratification,"²⁹ they recorded the results of a study of 30 patients, each of whom had an "atypical Spitz tumor" which was graded by them as being of low, intermediate, or high risk. They selected 5 parameters for the purpose of grading systematically, those being, "age," "diameter in millimeters," "fat involvement," "ulceration," and "mitotic activity per square millimeter." The parameters and the justification for them were established by the authors in these words:

"In order to provide reproducible criteria to categorize the lesions, only 5 quantitative histological or well-defined parameters were evaluated. The lesion diameter (< le >10.0 vs. > 10.0 mm) was measured on the slide by micrometer. The tumor thickness (< le > 3.00 vs. > 3.00 mm) was measured by micrometer according to the method of Breslow. Ulceration was defined as an interruption of the whole epidermis, including the stratum corneum. The mitotic rate in the dermal component (0–5, 6–8, or > 8 mm²) was recorded after the lesion had been scanned for hot spots. The anatomical level (I-IV vs. V) was recorded according to the method of Clark, et al."

From this system of Spatz et al. for grading Spitz's "tumors" in children came a method for scoring the probability with which an "atypical Spitz tumor" would metastasize, but the very notion of a Spitz tumor/nevus, whether modified by the word "atypical" or not, that metastasizes does violence to rudimentary principles on which pathology as a discipline is founded. It is dizzying that as recently as the 1990s, article after article lent support to a concept which at best is untenable and at worst preposterous. But the charade went on interrupted; no one exclaimed, demurred, or even muttered, "The emperor has no clothes!"

According to Spatz et al., a score of 0–2 indicated "low-risk" for metastasis, 3–4 "intermediate risk," and 5–11 "high risk." Where, precisely, the 30 "tumors" actually fit in regard to the three risk categories was shown in their Table 4 (our Fig. 25); ¹¹ "atypical Spitz tumors" metastasized. One of those neoplasms (pictured in our Fig. 26, their Fig. 1) had been placed by Spatz et al. in the "low risk" category, yet the patient, a 15-year-old boy, died of widely disseminated melanoma seven months after the diagnosis of "atypical Spitz tumor" had been made. The legend to that figure identified the neoplasm as an "atypical Spitz tumor with metastasis and death." It would have been far more accurate to have termed the neoplasm melanoma, directly and unequivocally, rather then resorting to a maneuver that failed abjectly as a cover for woeful inconsistency. The same can be said for Fig. 27 (their Fig. 2) which depicts, on the leg of a 14-year-old boy, a lesion said to be an "atypical Spitz's tumor;" it, too, is a melanoma.

Based on their observations of the 30 melanocytic neoplasms, Spatz et al. came to these conclusions:

In fact, Spatz et al. provided no useful information of any kind, only a gemisch that could not possibly redound to the benefit of patients.

In 1999, Shimek and Golitz, in a celebration in the Archives of Dermatology of the "golden anniversary of the Spitz's nevus,"³⁰ offered their own criteria for identifying neoplasms that they referred to as "high-risk atypical Spitz's nevi." These were they:

The changes referred to by Shimek and Golitz are those of melanoma, not of Spitz's nevus. For example, Spitz's nevus, like all benign neoplasms in the skin, never ulcerates, (unless it has been traumatized); tends to be less than 1.0 cm in greatest diameter; rarely extends deep, such as far into the subcutaneous fat; shows signs of maturation, both in terms of nests of melanocytes and of nuclei of individual melanocytes becoming smaller with progressive descent, and, as a rule, is not associated with many mitotic figures, mitoses at the very base of the neoplasm, or abnormal mitotic figures. In short, Shimek and Golitz leapt aboard the bandwagon of a concept devoid of authenticity, namely, "high risk atypical Spitz nevi."

In 1999, Barnhill et al. returned yet another time to the subject of how difficult is distinction histopathologically of Spitz's nevus from melanoma. They did that in an article in Human Pathology titled, "Atypical Spitz nevi/tumors: Lack of consensus for diagnosis, discrimination from melanoma, and prediction of outcome."³¹ In that exercise, the authors set out to gauge how accurately a panel of 10 dermatopathologists, judged by them to be expert (Argenyi, From, Glass, Maize, Mihm Jr., Rabkin, Ronan, White, and Piepkorn), was in diagnosing 30 vexing lesions in which the choices for diagnosis were one of the following:

Barnhill et al. provided the following guidelines ahead of time in an effort to assist the participating experts in coming to an accurate diagnosis:

Inherent in the difficulty of coming to one of the 5 possible diagnoses was the way the choices were phrased, as well as the manner in which the guidelines were to be followed. Indeed, an "expert" might well be compelled to query, "What, exactly, is the difference between a Spitz nevus and a Spitz tumor?" and "What are the findings morphologically that qualifies a Spitz nevus/tumor as being "stereotypical?"; "What, precisely, are the criteria for characterizing a Spitz nevus/tumor as 'atypical,' particularly in the context of the mentality of Sophie Spitz, herself, who considered the lesion to be so atypical that for her, originally, it was a true 'malignant melanoma?'"; "What, strictly speaking, is a spitzoid melanoma, morphologically and biologically?"; "What, in particular, is a tumor of unknown biologic potential?" and "Is it not true that no pathologist is able to predict the behavior accurately of any malignant neoplasm?"; and "What, really, are the 'unclassifiable melanocytic lesions'?" As is apparent even at cursory first reading, the "only histopathological guidelines discussed [by the experts] before the review [of the 30 cases]" were woefully inadequate for the task they were about to undertake, thereby rendering the exercise futile before it began. As but a single example of a flaw that doomed irrevocably the endeavor of Barnhill and all the others was serious misapprehension concerning what they conceived to be "conventional" criteria (as they are encountered in the literature) for diagnosis histopathologically of Spitz's nevi, atypical Spitz's nevi, and melanoma. In actuality, there are no agreed on criteria for any of those (and especially for "atypical Spitz's nevus" which, until this day, has never been shown to exist for real). And it is this utter lack of unanimity about criteria for diagnosis and differential diagnosis of basic melanocytic neoplasms that contributes to the fact, established beyond doubt, that panels of "expert" dermatopathologists never agree about diagnosis, with specificity, of melanocytic neoplasms that deviate a scintilla from consummate conventionality. After review of the 30 cases by the appointed experts, Barnhill et al. came to these conclusions about what they referred to as "Spitz nevi/atypical Spitz tumors":

As had been true ever since the idea of "atypical" Spitz's nevus and "atypical variants of Spitz nevus," to say nothing of "malignant" Spitz's nevus and "metastasizing" Spitz's nevus, had been given impetus in the 1980s, the irrationality that animated these concepts ran more and more amok. No fiction, no matter how startling, concocted by the proponents of those concepts met with any opposition at all, thereby encouraging the confabulating advocates to ever greater flights of fancy. The unreality of all this seemed to reach a zenith in the statements of Barnhill et al. just quoted. That there was "no consensus" among the experts should not have come as any surprise to those who truly are expert. An expert (and even a rank amateur) knows that the 19 neoplasms that metastasized were melanomas, not nevi of any kind. And that is what Barnhill et al. should have diagnosed each of them to be, straightforwardly and unambiguously. And that is what those who reviewed the manuscript submitted for publication should have demanded be the diagnosis. And that is what the editor-in-chief should have insisted on, irrespective of the opinion of the reviewers. Where was rigorous, incisive, critical thought on the part of any of the reviewers who recommended that particular article, and many forbearers of it, for publication, and of the editor-in-chief who acquiesced to those recommendations? That turn of mind, or lack of it, sad to say, is missing entirely, as a glance at the photomicrographs and the snippet of history provided convey tellingly. For example, Fig. 28 (their Fig. 2) shows a melanocytic neoplasm that developed on the ear of a 7-year-old boy and metastasized to cervical lymph nodes. That neoplasm was not an "atypical Spitz nevus/tumor" as alleged, but a melanoma. Fig. 29 (their Fig. 3) pictures a melanocytic neoplasm that developed on the thigh of a girl, then age 12, from which metastases became apparent 14 years later. That neoplasm, too, was not an "atypical Spitz nevus," as claimed, but a melanoma; the patient died of disseminated metastases. Amazingly, none of the expert dermatopathologists picked by Barnhill et al. considered the neoplasm in that woman, on histopathological grounds alone, to be a melanoma; the majority of the experts, eight of them, favored a diagnosis of "Spitz nevus/tumor" and "atypical Spitz nevus/tumor." In Fig. 30 (their Fig. 4), Barnhill et al. depicted a melanocytic neoplasm on the arm of a 2-year-old girl and, despite the fact that it metastasized to "a regional lymph node," establishing it as a melanoma undeniably, characterized it in the legend to the figure as an "atypical Spitz tumor." That Barnhill et al. not only were confused greatly by all this, but were ignorant of principles fundamental to pathology, to wit, those which qualify a neoplasm as being malignant, as opposed to being benign, may be inferred from a question they raised and from the somewhat enigmatic answer they, themselves, provided:

All of this is surreal to us. From our vantage, the answer to the question posed by Barnhill et al. is that a "Spitz-like lesion" that metastasizes is a melanoma, pure and simple, and the reason is that in classic pathology a malignant neoplasm is defined as one that has capability to kill by destruction locally or metastasis widely. Using that definition as arbiter, the so-called atypical Spitz's (nevus) tumor is a melanoma; it has capability for metastasis, it does metastasize, and it kills. Res ipsa loquiter!

In a letter to the editor of Human Pathology in 1999³² concerning the article by Barnhill et al. that appeared in that journal earlier that year, Richard J. Reed asked, "What is the meaning of the designation 'atypical Spitz nevus/tumor'?" and, having posed the question, then proceeded to answer it himself thus:

"What is the meaning of the designation 'atypical Spitz nevus/tumor'? Of all the words that might elicit the virtual images of a lesion with bulk in 3 dimensions, 'tumor' is the least specific; it is even less specific than 'nevus'. In their [Barnhill and coworkers' article in 1999 in Human Pathology; volume 30, pages 513–20], usage of 'tumor,' is it the authors' intention to assign special attributes (intentions) that are not embodied in definitions in common textbooks on general pathology? What attributes of the 2 words, nevus and tumor, alone or in combination, convey the threat of regional or distant metastases?"

Barnhill and Piepkorn³³ replied to the queries of Reed in the form of justifications thus:

The long established name for a neoplasm of melanocytes that metastasizes is "melanoma"! That term is reasonable, and ample data of substance, referred to colloquially as "hard," gives legitimacy to melanoma as a thoroughly "suitable" name, available for more than a century, for a malignant neoplasm of melanocytes.

Rapini had a go at this same subject in 1999 in an article that led off with a question: "Spitz nevus or melanoma?"³⁴ Some of his thoughts about the matter of "malignant Spitz's nevi" were expressed by him in these lines:

In the sentences preceding, Rapini gave approval tacitly to the idea of "malignant Spitz's nevus," a notion that deserves no approval because it is an oxymoron; there is as much chance of there being a "malignant Spitz's nevus" as there is of a generous miser. The translation of "malignant Spitz's nevus" is melanoma.

Rapini explained why it is that he, himself, employs the term "atypical Spitz nevus." This is how he did that:

In one sense, Rapini, who is to be applauded for his candor, is right; the terms "atypical Spitz nevus" and "atypical melanocytic proliferation" are "hedges," that is, evasions from a specific diagnosis of "nevus" or "melanoma." Rather than hedge, however, we believe that it is preferable to state, simply and unashamedly, "I don't know," and to write a note that conveys to a clinician the reasons for the uncertainty, the diagnosis deemed to be most consonant with the findings present, and the stipulation that such a neoplasm must be excised in toto with a narrow margin. Consultation with a respected colleague is appropriate.

Spatz and Barnhill took on the matter of criteria for distinguishing histopathologically between Spitz's nevus and melanoma in an article in 1999 that they captioned, "The Spitz tumor 50 years later: Revisiting a landmark contribution and unresolved controversy."³⁵ This is some of what the authors said then:

The statement that "The typical Spitz tumor almost never metastasizes" is arresting; an authentic Spitz's nevus never metastasizes because it is a nevus and a nevus, being benign, never, ever metastasizes. Every so-called metastasizing Spitz's nevus is a melanoma that originally was misdiagnosed histopathologically. The preceding quotation from Spatz and Barnhill consists of half-baked, ill-conceived notions that impede any effort at coming to a diagnosis with specificity of Spitz's nevus or of melanoma. Statements said by them to be "reasonable" are unreasonable because the premises on which they are based are wholly without foundation.

In 1999, Toussaint and Kamino, in an article titled promisingly, "Dysplastic changes in different types of melanocytic nevi: A unifying concept,"³⁶ (Fig. 31) ended up recommending that the term "dysplastic" be affixed to the already established designations for various types of melanocytic nevi that showed features histopathologically considered by them to fulfill criteria published previously for "dysplastic nevi of the dysplastic nevus syndrome [that] can be seen either in association with a dermal component characteristic of different types of melanocytic nevi or overlapping with features of other variants of nevi." Theirs, they posited in the subtitle of their article, was an attempt to provide a "unifying concept" of nevi of different kinds which they regarded to be "dysplastic." To that end, they examined 2,164 examples of compound melanocytic nevi and identified six types that they thought qualified as "dysplastic": "1) dysplastic nevus (original); 2) dysplastic nevus with a congenital pattern; 3) dysplastic Spitz's nevus; 4) dysplastic combined blue nevus; 5) dysplastic halo nevus; and 6) dysplastic neuronevus" (Fig. 32). Of the more than 2000, they considered 67 to be "dysplastic Spitz's nevi" and for these reasons:

The premise on which the "unifying concept" of Toussaint and Kamino was constructed proved to be completely wrong: there is no "regular" Spitz's nevus (any more than there is a "normal" mole or a "metastasizing nevus") and the "changes" said to be "typical of dysplastic nevi" may be encountered in nevi other than Clark's (so-called dysplastic) type and, episodically, in melanomas. In short, Toussaint and Kamino simply used the word "dysplastic" as a modifying prefix to a variety of different kinds of benign proliferations of melanocytes, one of which is Spitz's nevus. Not only did the effort at forging a "unifying concept" of melanocytic nevi of different kinds characterized by dysplasia fail because of the two false premises already mentioned, but because the word "dysplasia" has yet to be defined in a cogent, lucid, repeatable way (and which Toussaint and Kamino never attempted to define) and what is called "Spitz's nevus" likely is more than a single pathologic process, probably being several of them, just as is the case for those nevi currently named "blue" and "dysplastic (Clark's)." In sum, the notion of "dysplastic Spitz's nevus" is as unhelpful as that of "atypical Spitz's nevus" (Fig. 33); the words "dysplastic" and "atypical" not only fail to enlighten, they make a morass even more murky.

In 2000, Edwards and Blessing, in a review of "Problematic pigmented lesions: Approach to diagnosis,"³⁷ engaged, in detail, the issue of "atypical Spitz lesions" and recommended that they be called "atypical spitzoid lesions of uncertain malignant potential." This is what they said:

Edwards and Blessing, like the advocates of "atypical Spitz lesions" before them, got it all wrong. They argued that the "problem of the atypical Spitz" has yet to be "resolved," but that simply is not true; the problem of the "atypical Spitz nevus" has been solved—there is no problem because there is no "atypical Spitz (tumor, nevus, lesion)." What has been termed "atypical Spitz nevus" either is a melanoma that was misdiagnosed as a Spitz's nevus or a Spitz's nevus. Those are the possibilities; there are no others.

In an article about "Melanoma in children"³⁸ that appeared in 2000, Patterson et al. were deficient in failing to present any findings by conventional microscopy of what they called "atypical Spitz's nevus" and Spitz's nevus with "malignant" histopathologic features. This is part of what they wrote:

Patterson et al. seemed to accept wholesale the assertions of Barnhill et al.^{20,22,26,28,30,31,33,35} about the matter of atypical Spitz's nevus and the inclination of it to metastasize, that, despite the fact so much had been written during the previous five decades about the disparity at times between striking atypia cytopathologically and benign behavior biologically, the best representative of that phenomenon among neoplasms being some examples of Spitz's nevus and among inflammatory diseases being dermatofibroma with "monster cells." For nearly half a century, the seeming disconnect between cytopathologic attributes and biologic behavior was emphasized repeatedly as a serious peril to diagnosis accurately. The phenomenon itself even was given a name, to wit, "pseudomalignancy," the cliché for which was "histologically malignant, biologically benign." In reality, the idea of a condition being malignant histopathologically by virtue of striking nuclear atypia alone is as flawed as is the obverse, namely, a condition being benign because of the absence of nuclear atypia, the illogic of the latter precept being "pseudobenignancies," such as illustrated by lymphocytes in the patch stage of mycosis fungoides (which, as a rule, are small and monomorphous) and the mesenchymal cells of dermatofibrosarcoma protuberans at any stage of the process (which are thin, wavy, and monomorphous). In short, for 5 decades, reflective pathologists have understood, full well, that there was a vast difference between diagnosis of neoplasms histopathologically and diagnosis of them cytopathologically, the former being predicated not simply on nuclear characteristics but on architectural pattern (silhouette). The principle just articulated is disregarded dismissively by statements such as "Spitz's nevi exhibit histological features that range from completely benign to malignant" and that "patient age, diameter [of the neoplasm], fat involvement, ulceration, and mitotic activity" are "parameters" in a "grading system" designed "to assist in the clarification of malignant potential." In actuality, none of the five "parameters," or all of them together, enables differentiation to be accomplished histopathologically of a benign from a malignant neoplasm, the reason being that that distinction turns on a combination of findings, namely, silhouette and nuclear attributes.

Parenthetically, the notion of "histologically malignant, biologically benign" is given the lie by what transpired in the last quarter of the 20th Century to what were considered to be 4 stereotypical examples of pseudomalignancy, namely, keratoacanthoma, atypical fibroxanthoma, proliferating tricholemmal cyst, and lymphomatoid papulosis. That tetrad was claimed, universally, to be leading examples of pseudomalignancy. Today, it is accepted, equally universally, that keratoacanthoma is one type of squamous-cell carcinoma, atypical fibroxanthoma is malignant fibrous histiocytoma, proliferating tricholemmal cyst is proliferating tricholemmal cystic carcinoma, and lymphomatoid papulosis is one expression of CD 30 T-cell lymphoma. So much for pseudomalignancy—and for "atypical Spitz's nevus" and "dysplastic Spitz's nevus" (to say nothing of "malignant Spitz's nevus" and "metastasizing Spitz's nevus"), all of which are doomed to the same unhappy end as "pseudomalignancy" and "active junctional nevus."

In sum and in short, no direct relationship exists between conspicuous nuclear atypia on one hand and malignancy on the other, just as no direct relationship exists between lack of nuclear atypia and benignancy. Recognition of that reality and appreciation of the implication of it for diagnosis histopathologically eliminates from consideration seriously notions like "atypical Spitz's nevus" and "Spitz's nevi with malignant features."

Groben et al.³⁹, in 2000, reported on their studies of 33 patients who had blue nevi of "epithelioid type," but with no evidence of signs of "Carney complex." In the section of their article given to "Discussion," the authors made the following enigmatic comments with respect to "metastasizing Spitz's nevus" and "malignant Spitz's nevus":

Just as there can be no "malignant" or "metastasizing" Spitz's nevus, there can be no such thing as "limited metastases;" it, too, is an oxymoron. Once neoplastic cells enter vascular channels, either those of lymph or blood, they are carried away, destined to be disseminated.

In 2001, in a text titled, "The Melanocytic Proliferations: A Comprehensive Textbook of Pigmented Lesions,"⁴⁰ Crowson, Magro, and Mihm addressed the issues of "atypical Spitz tumor," "severely atypical superficial compound Spitz tumor," and a neoplasm to which they gave the intriguing designation, "severely atypical Spitzian proliferation." The coauthors asked the question, "Is it atypical Spitz tumor or Spitzian melanoma?", and this was the answer they gave:

Statements such as those by general pathologists in the 21st Century are nothing less than astounding.

Of "severely atypical superficial compound Spitz tumor," Crowson, Magro, and Mihm had this to say:

If kind, the preceding lines may be considered garble that defy attempts, no matter how determined, to comprehend them; if plain direct, they are tommyrot.

Of "severely atypical Spitzian proliferation" Crowson, Magro, and Mihm expressed these thoughts:

In our view, the statements by Crowson, Magro, and Mihm represent "pathobabel," a Towel of Babel in pathology, that, by virtue of language which is utterly unintelligible and totally incomprehensible, makes communication between physicians impossible, a situation that redounds badly to patients. But, as evidenced by what has emanated during the past several decades from a single school of dermatopathology, peopled mostly by general pathologists, about matters as disparate as "melanocytic dysplasia" and "melanocytic neoplasm of uncertain biologic potential," it is apparent that acolytes of that school, and they are many, are not bothered by it; they relish it. That being the case, the future of pathology is certain: bleak.

In 2001, in a chapter devoted to "Atypical Spitz nevus?" in a volume titled Resolving Quandaries in Dermatology, Pathology and Dermatopathology, second volume,⁴¹ Ackerman and Mones made their position about the matter under discussion here crystal clear in these words:

In fact, Ackerman had begun to decry the idea of "malignant" and "metastasizing" Spitz's nevus soon after the first volley of those concepts was launched, and he continues to combat those fictions in print to the present day.^42–44 That his criticism, sometimes trenchant and even harsh, has had some little salutary effect may be inferred from the title of a presentation given by Lupton, current Head of the Dermatopathology Branch of the AFIP, at a course, chaired by Ackerman concerning Controversies in Dermatopathology at the International Congress of Dermatology in Paris in 2002, that title being, "Spitz's nevus cannot metastasize." In his lecture, Lupton told of follow-up of 4 of the 6 patients reported on in 1989 by K.J. Smith et al. from the AFIP, those being what was then a 7-year-old who is alive with metastases 15 years later, a then 18-year-old who is alive with metastases 21 years later, a then 16-year-old with metastasis who is alive 15 years later, and a then 2-year-old who is alive with metastasis 18 years later. In the judgment of Lupton, neoplasms diagnosed as "malignant Spitz's nevus" at the AFIP and written about in 1989, were, in actuality, "large atypical Spitzoid tumors" if they had not yet metastasized and were something other than conventional melanomas if they had metastasized, what he called "malignant melanomas of the special Spitzoid type." But the arguments of Lupton are a rear guard action; the neoplasms of melanocytes metastasized and, perforce, are melanomas. No amount of sophistry can change that reality. It matters not a jot how long a patient with metastases lives in harmony with them; they still are metastases of melanoma irrespective of the longevity of the person who harbors them, and they presage a grim outcome; just when the demise will occur no mortal can say.

In 2001, Zuckerman et al.⁴⁵ reflected on the ramifications of 2 melanocytic neoplasms that had been diagnosed originally by "expert consultants" as "atypical Spitz nevus" and "cellular neurothekeoma with focal mucinous alteration and focal marked cytologic atypia," both of those neoplasms having metastasized to a sentinel lymph node. The patient said to have an "atypical Spitz nevus" was an 8-year-old girl whose pigmented lesion, positioned above the left elbow, is pictured in Fig. 34 (their Fig. 1); the patient diagnosed initially as having a "cellular neurothekeoma" was a 14-year-old boy who had a "blue cutaneous lesion" on the left ear. Based on the finding of metastases in these two patients, Zuckerman et al. concluded, correctly, that, in fact, the neoplasms were melanomas, which they acknowledged forthrightly, without a trace of hedging or evading in this simple sentence: "The positive [sentinel lymph node] biopsies then established the diagnosis of melanoma." Logic and rationality triumphed!

Also, in 2001, Fabrizi and Massi⁴⁶ wrote about what they called melanomas of the "spitzoid" type in teenagers. A melanoma that appeared on the leg of a 15-year-old boy was diagnosed histopathologically as "Spitz nevus." Five years later, the patient was found to have metastases of melanoma to regional lymph nodes, lungs, and brain, and he died 8 years after the erroneous diagnosis of Spitz's nevus had been made. The other patient was a 19-year-old man whose lesion on the back was diagnosed by conventional microscopy as a "spindle cell-type Spitz naevus." Fifteen years later, he developed signs of metastases of melanoma in the skin of the back, the retroperitoneum, and the bones of the pelvis. Based on their experience with these two patients, Fabrizi and Massi offered these observations:

Both teenage patients of Fabrizi and Massi had a melanoma that initially was misdiagnosed histopathologically as Spitz's nevus or an equivalent of it. Both lived with their metastases for years, one for 8 years and the other for at least 15 years. In contrast to the conclusion of Helwig at the AFIP 25 years previously, and of Lupton at the AFIP now, to the effect that melanomas in children that resemble Spitz's nevus histopathologically have a better prognosis than those that by microscopy bear no resemblance to Spitz's nevus, it can be stated authoritatively that a melanoma in a young person, irrespective of how closely the primary melanoma resembles a Spitz's nevus histopathologically, has the same potential for metastasis as does a melanoma in an adult, and that those metastases result eventually in death. Why a large percentage of young persons with primary cutaneous melanoma seem to live for many years in equilibrium with metastases from it is not known. It is highly likely, however, that that phenomenon is independent of any resemblance histopathologically of the primary melanoma to Spitz's nevus.

In 2002, Lohmann et al.⁴⁷ assessed the utility of sentinel lymph node biopsy for a group of lesions they referred to as "diagnostically controversial Spitzoid melanocytic tumors" which they believed "represent[ed] a morphologically heterogeneous group of melanocytic tumors but have in common the fact that expert pathologists reviewing their histology discuss the distinction between Spitz nevus and melanoma and disagree on the final diagnosis." There were 10 such lesions in their series; 5 of them diagnosed previously as "atypical Spitz tumors" were found to have metastasized to a sentinel lymph node (Figs. 35 –38). Lohmann et al. concluded, rightly, that the lesions were not "atypical nevi," but were melanomas. This is how they put it:

Also in 2002, Mones and Ackerman, in a treatise titled, "Melanomas in prepubescent children: Review comprehensively, critique historically, criteria diagnostically, and course biologically,"⁴⁸ told of their findings in 11 children 10 years of age or younger whose melanoma (not a single one of which was diagnosed correctly clinically) had been shown to have metastasized to regional lymph nodes (Fig. 39). One child died of metastatic melanoma. In over 40% of the 11 neoplasms, diagnosis of the original histopathologist was "atypical Spitz's nevus," an error that resulted in delay in diagnosis in one patient for about 10 months, at which time nodal metastases were detected, a happening which prompted reversal of the initial diagnosis to melanoma. This is what Mones and Ackerman declared with respect to the diagnosis of "atypical Spitz nevi":

In that publication⁴⁸, Mones and Ackerman also set forth criteria in their Tables 1 for the differentiation histopathologically of Spitz's nevus from melanoma in general and in their Table 6 criteria for differentiation of melanoma in pre-pubescents from Spitz's nevi. Criteria essential for differentiation histopathologically of Spitz's nevus from melanoma, irrespective of the age of the patient are contrasted in Table 1.

In 2002, van Dijk et al.⁴⁹ employed Allelic Imbalance (AI) Analysis, a method that utilizes principles of molecular biology, in an attempt to test its worth in diagnosing 55 melanocytic lesions, those being said to be 12 Spitz nevi, 9 atypical Spitz tumors, 17 "suspect spitzoid melanomas," and 17 primary spitzoid melanomas. Unfortunately, van Dijk et al. did not provide any information about the criteria they employed for diagnosis of Spitz nevi, "suspect spitzoid melanomas" and "spitzoid melanomas," the justification given by those coworkers being that "all lesions had been previously classified by the Dutch melanoma panel, which includes two of the authors (DJR, WJM), using established criteria." It was only for "atypical Spitz tumor" that van Dijk et al. provided measuring sticks whereby they came to that diagnosis histopathologically. These were the criteria as stated by them:

The criteria of van Dijk et al. are not applicable to a nevus of any kind, but rather to melanoma: asymmetry and uneven spacing of nests of melanocytes at the dermo-epidermal junction, confluence of nests, obliteration of adnexal structures, "compact growth pattern of the dermal component," severe atypia cytopathologically, uneven distribution of melanin, and an increased number of mitoses. That being so, the neoplasms purported by van Dijk et al. to be "atypical Spitz tumors" really were melanomas, which surely do occur in very young children, even those as young as one year of age. That also being so, melanoma cannot be excluded as a possibility diagnostically simply because a patient is younger than 15 years of age.

Two neoplasms diagnosed at first by van Dijk et al. as" typical Spitz naevi" demonstrated allelic imbalance. One of them, which was removed from the forearm of a 10-year-old boy, had such an imbalance on chromosome 6 and at a region of chromosome 9p. The other, located on the knee of a 9-year-old boy and thought initially by van Dijk et al. to be a Spitz's nevus, had an allelic imbalance on chromosome 8. Review of the two neoplasms histopathologically, in conjunction with the results of analysis of allelic imbalance, motivated van Dijk et al. to reclassify both of them as "atypical Spitz tumors." Another lesion, situated on the foot of a 63-year-old man (their case 23), was interpreted originally to be an "atypical Spitz tumor suspicious for melanoma," but it metastasized. Nowhere in their article, however, do van Dijk et al. state that they reclassified the neoplasm that metastasized as the melanoma it was. Because molecular biologic techniques employed by van Dijk et al. did not permit a diagnosis of Spitz's nevus or of melanoma to be made with certainty, the coworkers were forced to the following admission:

van Dijk et al. did seek to provide an explanation for the diversity of their results and in that effort they invoked the theory of tumor progression thus:

Gurbuz et al.,⁵⁰ in 2002, reported on their experience with a neoplasm that developed in the region of the left subscapula of a 4-year-old girl and that they diagnosed as an "atypical Spitz tumor." Based on that diagnosis, the child was subjected to a "wide peripheral excision and sentinel lymph node dissection;" there was no evidence of metastatic disease in the lymph node and the child was said to be free of metastases after a follow-up of 22 months. Gurbuz et al. endeavored to define what they meant by "atypical Spitz tumor" and "malignant Spitz nevus," and they did that in these lines:

Despite the try at definition, it can only be concluded that Gurbuz et al. failed miserably for reasons that have been stated by us repeatedly throughout this work, reasons that have been elementary to pathologists for more than 150 years, to wit, a neoplasm that metastasizes, by definition, is malignant, in this instance melanoma, and not a nevus, Spitz type or otherwise. Also dismaying is the attempt by Gurbuz et al. to make a distinction between "atypical Spitz tumor" and "malignant Spitz nevus," both of those terms reflecting ideas that are bankrupt. Also disheartening are the efforts by the collaborators at distinguishing between "the ability to metastasize to regional lymph nodes but not further" and "widespread metastases." As has been stated by us ad nauseum, there is no such thing, either in concept or in practice, as a "local" metastasis; metastasis implies dissemination. As startling as are the comments by Gurbuz et al. just quoted verbatim, they are not hyposensitizing to the reaction engendered by this statement by them in regard to melanomas in children:

By now it hardly needs repeating that "atypical tumors," with or without spitzoid features, that metastasize are melanomas which at first were misdiagnosed.

In a "Commentary" to the article by Gurbuz et al. titled, "'Safe Spitz' and its alternatives," published in the journal, Pediatric Dermatology, LeBoit⁵¹ put forth the idea of "safe Spitz" versus "unsafe Spitz." The use of the name "Spitz" implies that LeBoit was referring to Spitz's nevus. The following lines provide insight into LeBoit's thinking about the paradigm of "safe" and "unsafe" Spitz's nevus:

"The prevailing attitude among dermatologists and pathologists is that melanomas are malignant, Spitz nevi are benign, and a lesion interpreted as a Spitz nevus that resulted in metastasis must be a melanoma, and reflect a misdiagnosis. That belief system may need to be revised if enough evidence accumulates that there are melanocytic proliferations that have features intermediate between those of Spitz nevus and melanoma."

LeBoit seems to be advocating the notion of a "borderline" neoplasm, one that is somewhere between benign and malignant. But as will become obvious in the quotations from LeBoit that follow, "safe Spitz" for him is a Spitz's nevus that he has no difficulty diagnosing histopathologically, whereas, "unsafe Spitz" for him is a melanocytic neoplasm that because it has attributes histopathologically in common with Spitz's nevus, poses problems in diagnosis. In reality, there is nothing "borderline" in regard to the notion, phrased enigmatically, of "safe" and "unsafe" Spitz; the former is a Spitz's nevus and the latter either a Spitz's nevus or a melanoma. If one could ask an "unsafe Spitz" to disclose its true nature, it would not respond "borderline!" LeBoit defined "safe Spitz" thus:

And this is what LeBoit had to say about "unsafe Spitz":

We have endeavored to convince readers of this monograph that the only possibilities for what are called "atypical Spitz nevus" and "unsafe Spitz" are Spitz's nevus and melanoma. Criteria for diagnosis histopathologically of those two neoplasms, so dissimilar clinically, histopathologically, and biologically, are available and, in the vast majority of instances they work. In some instances, mistakes in diagnosis are made, not because the criteria fail but because a human brain fails to apply them properly. Whenever a histopathologist is unable, for whatever reason, to resolve the conundrum of "Spitz's nevus" or "melanoma," it is mandatory that that uncertainty be conveyed in writing in the most unequivocal way. LeBoit seems to agree, as evidenced by these lines with which he concluded his Commentary:

"Unsafe Spitz"? Unsafe for whom? The patient, the dermatologist, the dermatopathologist? It is highly unlikely that a patient or a managing physician will be cheered by a diagnosis of "unsafe Spitz" and it is equally unlikely that a jury will find an explanation for it by a dermatopathologist, no matter how articulate, compelling. We encourage colleagues to eschew the terms "atypical Spitz" and "unsafe Spitz" and to issue diagnoses of "Spitz's nevus" and "melanoma," or to acknowledge uncertainty directly as "I don't know." Of course, any neoplasm that could be melanoma on the basis of characteristics of it histopathologically must be removed completely with a narrow margin.

In a study of 22 "persistent (recurrent) nevi" published in 2002, Harvell et al.⁵² identified a single lesion that was diagnosed originally as an "atypical melanocytic proliferation with some features of a Spitz nevus," but that metastasized to regional lymph nodes. In straightforward fashion, the coworkers reclassified the neoplasm as a melanoma and pictured it as their Figs. 6 A-C.

In 2002, Murphy et al.⁵³ assessed treatment surgically by dermatologists, practicing privately, of 61 patients who had been given a diagnosis histopathologically of "Spitz's nevus" and 28 whose diagnosis was "atypical Spitz's nevus". The collaborators did not publish any photomicrographs. They characterized "atypical Spitz nevus" histopathologically as follows:

The language employed by Murphy et al. is reminiscent of that pertaining to "minimal deviation" and "borderline," both of those terms being attempts at escape from a diagnosis of "nevus" or "melanoma." Murphy et al. seem to acknowledge their evasion in the form of "atypical Spitz lesion" in these concluding remarks:

"Complete removal by initial biopsy is ideal because it prevents recurrent Spitz nevi that have different histologic characteristics from Spitz nevi and can be confused with malignant melanoma . . . Spitz nevi and atypical Spitz lesions can pose diagnostic challenges to the dermatopathologist. In some cases of atypical Spitz nevi, the biologic behavior is uncertain and ranges from completely benign to the rare malignant melanoma."

In 2002, Zaenglein et al.⁵⁴ told of a 4-month-old boy who had two pigmented lesions on the right side of the parietal region of the scalp and said to have been present since birth (Fig. 40). The larger lesion had darkened, enlarged, and developed an elevation within it. Because of those changes, melanoma was suspected clinically. Examination of sections of tissue from the biopsy specimen taken from the larger lesion showed it to be a "compound congenital Spitz nevus." The smaller lesion was diagnosed histopathologically as a "compound melanocytic nevus with some features of a Spitz nevus." The experience with that larger lesion induced Zaenglein et al. to comment as follows in regard to Spitz's nevus and melanoma:

The terms "atypical variants of Spitz nevus" and "atypical pigmented lesions" take a histopathologist nowhere because they lack content utterly; "atypical" in this context is uninformative in the extreme. Not only is "atypical" in this context meaningless; never has it been defined. Putting the word "atypical" or "dysplastic" in front of Spitz's nevus, so-called blue nevus, and other kinds of nevi is posturing transparently. To those who are discerning, that ploy is designed as a smokescreen, a maneuver, an evasion from committing to a diagnosis of "benign" or "malignant" straightforwardly. It, being weasel-like, should be condemned—and we, on behalf of patients and managing physicians, condemn it damningly.

LeBoit⁵⁵, in his role as editor of the American Journal of Dermatopathology, wrote a commentary on a photomicrograph "shot" at scanning magnification and printed in color on the cover of the February issue for 2003. He titled the piece, "Pictures of a unicorn?" In it, he elaborated on the concept of "malignant Spitz nevus" and offered a hypothesis alternative to that of a Spitz's nevus actually being malignant. This is how LeBoit phrased his idea:

Humbly offered herein are three cases that I believe could be examples of melanoma arising within a preexistent Spitz's nevus."

In our judgment, cases 1 and 2 of LeBoit shown in our Figs. 41 and 42 (his Fig. 1 and 2), are melanomas and his case 3, our Fig. 43 (his Fig. 3), is a Spitz's nevus. Moreover, the concept of melanoma arising in a Spitz's nevus reminds of the stance adopted by Allen and Spitz after they had first declared that "juvenile melanoma," in actuality, was an authentic malignant melanoma that behaved in benign fashion, asserting then that although "juvenile melanoma" looked exactly like melanoma histopathologically, it did not behave like melanoma biologically, and, later, that "juvenile melanoma" was not melanoma at all, but that it could convert to melanoma. The "conversion/transformation" posture seems to us to be a defensive measure employed in an effort, futilely, to synthesize that which is not integratable, that worthy aim only being achievable once assaults on basic principles of pathology cease. There can be no true "malignant Spitz's nevus," yet some neoplasms diagnosed by microscopy as Spitz's nevus metastasize. That is the maelstrom for LeBoit and he attempted extrication from it in this manner:

We resolve the quandary that confronted LeBoit in the guise of a unicorn by, at the same time, dispatching mythology and employing principles fundamental to pathology, namely, diagnosis predicated on reliable and repeatable criteria histopathologically and in synchrony, over time, with behavior biologically. In our estimation, given the limitations imposed by assessing photomicrographs printed on paper, the three lesions pictured in photomicrographs by LeBoit are melanoma in two instances and Spitz's nevus in one; none of them to our eye are melanoma in conjunction with Spitz's nevus (although that combination does occur exceedingly rarely) and surely none are "malignant Spitz's nevus."

Su et al.,⁵⁶ in 2003, published the results of their study of 18 patients who underwent sentinel node biopsy for "problematic" and "diagnostically challenging" melanocytic lesions to which they gave 4 designations as follows: "atypical Spitz tumor, markedly atypical Spitz nevus, atypical epithelioid melanocytic proliferation of uncertain biologic potential, or borderline epithelioid melanocytic neoplasm." Metastases were identified in sentinel nodes in 8 of the 18 patients, 44% of them (Figs. 44 –49). Based on these findings, Su et al. presented their rationale for undertaking sentinel lymph biopsy in these sentences:

We agree with Su et al. that a melanocytic neoplasm that metastasizes is a melanoma, but disagree with the concepts advanced by them of "indeterminate malignant potential," "borderline . . . proliferation," "atypical . . . proliferation of unknown biologic potential," and "atypical Spitz tumor;" all of them which we regard as spurious and, therefore, cannot possibly serve either physicians or patients well. The uncertainty is only in the mind of the histopathologist and that then becomes important to the patient for reasons that are obvious. If a histopathologist is able to make a diagnosis accurately of Spitz's nevus or melanoma and then issues a report to that effect, his/her task has been performed admirably. If the diagnosis is Spitz's nevus, the behavior of it can be predicted with confidence: benign; if the diagnosis is melanoma no longer in situ or no longer exceedingly "thin," then behavior is mere guess and pathologists should avoid scrupulously engaging in a game of supposition. Parenthetically, sentinel node biopsy has no benefit whatsoever for patients with primary cutaneous melanoma or with melanoma that has arrived in a node; it is useless exercise and, that being so, should be abandoned forthwith.⁵⁷ In the study of Su et al., almost half of the neoplasms, by dint of their behavior, to wit, metastasis, made it known beyond doubt that they were not "indeterminate," "borderline," or of "uncertain biologic potential;" they were melanomas! Yet, for reasons enigmatic to us, histopathologists, even in 2003, continue to use terms that fly headlong in the face of principles fundamental to pathology, to say nothing of plain logic and common sense, chief among those terms being "malignant Spitz nevus" and "metastasizing Spitz nevus," rather than "melanoma" as they actually are.

On the cover of the American Journal of Dermatopathology for August 2003 and in an essay titled, "What do these cells prove" and directed at what was pictured on the cover, LeBoit gave assent to the concept that abnormal melanocytes of Spitz's nevus could be found in lymph nodes (Fig. 50). ⁵⁸ This is what he wrote: