Anaplastic large cell lymphoma: a potential pitfall in the differential diagnosis of melanoma.
We present a case in which a remote history of metastatic melanoma complicated the diagnosis of a second primary neoplasm, ultimately correctly diagnosed as a primary cutaneous CD30-positive anaplastic large cell lymphoma (cALCL). We illustrate the clinical, histologic, and immunohistochemical findings that contributed to our error of perception, and discuss strategies to avoid such pitfalls in routine pathology practice.
REPORT OF A CASE
A 69-year-old white man with a remote history (20 years prior) of malignant melanoma of the foot, metastatic to inguinal lymph nodes (clinical stage IIIA), presented with a new lesion on the ipsilateral thigh at a routine follow-up visit. Clinical examination was remarkable for a 1-cm, firm, mobile, subcutaneous nodule in the anterior aspect of the right thigh, favored to be a benign cyst. Because of the patient's history, there was some concern for metastatic/recurrent melanoma, and fine-needle aspiration was performed. Cytologic evaluation of aspirated material identified malignant cells, but was not conclusive. S100, HMB45, and A103 immunohistochemical stains were noncontributory; therefore, an excisional biopsy was performed to better define the tumor.
Histologic examination of the tissue sections identified a broad and deep, nodular infiltrate of large, pleomorphic cells arranged as cohesive sheets in the dermis and subcutis (Figure 1). The large, atypical cells were characterized by abundant cytoplasm, bizarre nuclei with intranuclear cytoplasmic protrusions, large eosinophilic nucleoli, and numerous mitotic figures (Figure 2). Tumor emboli were noted within adjacent vascular structures (Figure 1, inset). Given the history of metastatic melanoma to ipsilateral regional nodes 20 years previously, and because neither slides nor microscopic description of the previous tumor were available for review, immunohistochemical stains for melanocyte differentiation markers were performed, including S100, Melan-A, HMB-45, PNL-2, and microphthalmia-associated transcription factor. CD117 staining was also performed, and CD117 was the only marker significantly positive within an initial panel (Figure 3, A). Microphthalmia-associated transcription factor demonstrated a cytoplasmic blush in tumor cells. CD68, leukocyte common antigen, and pancytokeratin were negative (Figure 3, B). A second panel of immunostains showed negativity for CD34, terminal deoxynucleotidyl transferase, myeloperoxidase, muramidase (lysozyme), CD1a, and desmin, prompting consideration of a microphthalmia-associated transcription factor-positive, CD117-positive metastatic melanoma that had otherwise lost antigen expression. However, given unusual cytologic features of the tumor, further immunohistochemical stains were performed, which showed the tumor to be strongly positive for CD30 (Figure 3, C), CD4 (Figure 3, D), and CD43, and negative for anaplastic lymphoma kinase, most suggestive of cALCL. Because CD3 was found to be negative in the neoplastic cells, T-cell gene rearrangement studies were performed, which identified clonal beta and gamma T-cell receptor gene rearrangements, confirming the diagnosis.
The patient was subsequently evaluated by positron emission tomography/computed tomography scan, which did not demonstrate any further cutaneous or extracutaneous disease. Because the presentation was limited to a solitary nodule, fitting the diagnosis of primary cALCL, no further staging was performed, as per current recommendations.1 The patient is currently alive and well with no evidence of lymphoma, 3 years after presentation.
Although most melanoma recurrences from primary cutaneous melanoma occur within 10 years of initial diagnosis (and the majority within 3 years), (2,3) "late" metastases, those that occur after 10 years of quiescence, have a reported incidence ranging from 0.93% to 6.7%. (3-6) Late metastases may even affect as many as 25% of patients who survive beyond 10 years, supporting the recommendation for yearly follow-up even after 10 years. "Ultralate" metastases, those that occur after 15 years of quiescence, have been estimated to occur in at least 2% of melanoma survivors (3); most of these occur at a disease-free interval of 18 to 20 years. It is because of these findings that the American Joint Committee on Cancer recommendation has been for yearly follow-up after 6 years for stage III melanoma patients. (2) The fallout from this long-term follow-up, for the pathologist, is found in a "guilty until proven innocent" approach to new tumors in such patients. Even after long disease-free intervals, the pathologist must maintain a threshold of suspicion for metastasis in a poorly differentiated tumor, and sometimes within tumors that do not exhibit straightforward diagnostic morphologic and immunohistochemical profiles. In our case, suspicion was further heightened by the submitting clinician's clinical diagnosis reported as "nodule, history of melanoma," which guided the initial pathologic approach to the tumor.
Histologic examination of our patient's tumor at low power revealed a dermal nodular aggregate of large cells with involvement of adjacent vascular structures (Figure 1). Metastatic melanoma, like other metastatic tumors, most often presents as a nodular aggregate of cells. It is not uncommon to find intravascular tumor emboli near aggregates of melanoma; therefore, the low-power features were in keeping with the clinical diagnosis. The same low-power features are typical of many cALCLs, in which lymphocytes may aggregate around and within small vessels and endothelial lined spaces. (7)
Higher-power findings, for example recognition of the horseshoe- or kidney bean-shaped "hallmark cells" of anaplastic large cell lymphoma (ALCL), that were present in our case (Figure 2), in retrospect, could have added more insight into the tumor type prior to immunohistochemical workup. Many of the cytologic features of ALCL, however, such as rounded or wreathlike cells, angulated basophilic nucleoli, abundant amphophilic cytoplasm, and numerous mitoses, (7,8) are also seen in melanoma. Melanoma is, in fact, a well described consideration in the cytomorphologic differential diagnosis of ALCL. (7) It seems clear that, even with the observation of hallmark cells, in this case, concern for melanoma would persist, and immunohistochemistry would have been requested.
Metastatic melanoma, like primary melanoma, is known to express certain associated cytoplasmic and membranous antigens that can be detected by immunohistochemical methods. All melanomas, however, do not equivalently label with such typical markers, even given equal laboratory conditions. (2,9,10) One explanation for this phenomenon is the "genetic dysregulation of melanoma cells." Another commonly recognized confounding feature of some melanoma metastases is the loss of classic melanocyte differentiation markers such as S-100, HMB-45, Melan-A, and tyrosinase, sometimes described as an "antigenic shift" in tumor progression. (11) Occasionally all such markers are lost. This was the possibility we worried about in our diagnostic process.
CD117 positivity was also a confounding finding in our patient's tumor. The expression of CD117/c-kit protein, a transmembrane receptor tyrosine kinase involved in melanocyte development, has been well described in both benign melanocytes and nevi, (12-14) and has been noted to be increased in primary and metastatic melanoma. (14) There is controversy about whether or not c-kit expression is preserved in cutaneous metastases. (12,13,15) Although CD117 appears to be a fairly sensitive marker for melanoma, (16) the specificity is not high. Other CD117 positive tumors include clear cell sarcoma, gastrointestinal stromal tumors, (16) mastocytoma, and myeloid leukemia. The literature on CD117 expression in cALCL is not straightforward. Approximately half of all cases in 2 small case series of cALCL were found to express CD117, (17,18) but a recent study reports a complete absence of c-kit expression (19) in cALCL. In our case, terminal deoxynucleotidyl transferase, CD34, myeloperoxidase, CD68, CD163, muramidase, and CD45 were negative within the large cell population, assisting in ruling out a CD4+ leukemia cutis. The finding of clonal T-cell receptor gene rearrangements further mitigated against the diagnosis of an unusual myeloid, myelomonocytic, or related precursor neoplasm. Cutaneous gastrointestinal stromal tumor was not considered a high clinical probability for this patient. Ultimately, the use of CD117 was not helpful in our diagnosis, and probably incorrectly encouraged our suspicion of a melanocytic neoplasm.
Regarding microphthalmia-associated transcription factor, the expected finding in melanoma is a nuclear labeling pattern. Cytoplasmic staining by microphthalmia-associated transcription factor in our case was regarded as nonspecific, and could not be used to determine melanocyte differentiation.
An important feature of cALCL is its not-uncommon lack of expression of the leukocyte marker LCA and T-cell marker CD3, (7) as in our case. cALCL may be null for CD4 or CD8 (8); therefore, a high index of suspicion may be needed to pursue this diagnosis. CD45RO can be helpful, as it appears to be more sensitive than CD3 in cALCL. (7) Also useful is the cutaneous lymphoid antigen HECA-452, which is positive in 50% of cALCLs. Epithelial membrane antigen is usually negative in cALCL. (8) Melanoma can be epithelial membrane antigen positive (7); this is notable, as CD30 has been reported in as many as 4% of melanomas. (8) Fortunately, melanoma is typically CD45 and CD3 negative. cALCL almost never harbors the anaplastic lymphoma kinase translocations that are common to nodal ALCL; accordingly, we did not test for those molecular alterations. Neither did we test for molecular mutations associated with melanoma. Ultimately the histologic, immunohistochemical, and molecular findings identified in our case supported a diagnosis of cALCL rather than melanoma.
This case illustrates a few potential pitfalls in the evaluation of a tumor in a patient with known metastatic melanoma and emphasizes the need for a carefully selected, broad-spectrum immunohistochemical panel in those cases that fail to demonstrate adequate evidence of melanocytic differentiation, even in the context of prior metastasis. The diagnosis of ALCL should be considered in the differential for some such tumors, and the possibility of an aberrant immunophenotype should guide the selection of a complete panel of ancillary studies. Antigen loss or aberrant expression is important to recognize, but should not be so concerning as to override judicious use of immunohistochemical stains. Specifically, we caution against the use of CD117 in the diagnosis of melanoma, and emphasize the possible need for multiple lymphoid markers in the identification of an ALCL. Perhaps most importantly, morphologic findings really must be most considered at the earliest stages of this diagnostic process, and clinical information should be received with a healthy grain of salt.
(1.) Kempf W, Pfaltz K, Vermeer MH, et al. EORTC, ISCL, and USCLC consensus recommendations for the treatment of primary cutaneous CD30-positive lymphoproliferative disorders: lymphomatoid papulosis and primary cutaneous anaplastic large-cell lymphoma. Blood. 2011;118(15):4024-4035.
(2.) Poo-Hwu WJ, Ariyan S, Lamb L, et al. Follow-up recommendations for patients with American Joint Committee on Cancer stages I-III malignant melanoma. Cancer. 1999;86(11):2252-2258.
(3.) Tsao H, Cosimi AB, Sober AJ. Ultra-late recurrence (15 years or longer) of cutaneous melanoma. Cancer. 1997;79(12):2361-2370.
(4.) Callaway MP, Briggs JC. The incidence of late recurrence (greater than 10 years): an analysis of 536 consecutive cases of cutaneous melanoma. Br J Plast Surg. 1989;42(1):46-49.
(5.) Bouffard D, BarnhillRL, Mihm MC, Sober AJ. Verylatemetastasis (27years) of cutaneous malignant melanoma arising in a halo giant congenital nevus. Dermatology. 1994;189(2):162-166.
(6.) Crowley NJ, Seigler HF. Late recurrence of malignant melanoma: analysis of 168 patients. Ann Surg. 1990;212(2):173-177.
(7.) Ng WK, Ip P, Choy C, Collins RJ. Cytologic and immunocytochemical findings of anaplastic large cell lymphoma. Cancer. 2003;99(1):33-43.
(8.) Stein H, Foss HD, Durkop H et al. CD30+ anaplastic large cell lymphoma: a review of its histopathologic, genetic, and clinical features. Blood. 2000;96(12): 3681-3695.
(9.) Xu X, Chu AY, Pasha TL, Elder DE, Zhang PJ. Immunoprofile of MITF, tyrosinase, melan-A and MAGE-1 in HMB-45 negative melanomas. Am J Surg Pathol. 2002;26(1):82-87.
(10.) Miettinen M, Franssila K. Immunohistochemical spectrum of malignant melanoma: the common presence of keratins. Lab Invest. 1989;61(6):623-628.
(11.) Jager E, Ringhoffer M, Altmannsberger M, et al. Immunoselection in vivo: independent loss of MHC class I and melanocyte differentiation antigen expression in metastatic melanoma. Int J Cancer. 1997;71(2):142-147.
(12.) Guerriere-Kovach P, Hunt EL, Patterson JW, Glembocki DJ, English JC, Wick MR. Primary melanoma of the skin and cutaneous melanomatous metastases: comparative histologic features and immunophenotypes. Am J Clin Pathol. 2004;122(1):70-77.
(13.) Janku F, Novotny J, Julis I, et al. KIT receptor is expressed in more than 50% of early-stage malignant melanoma: a retrospective study of 261 patients. Melanoma Res. 2005;15(4):251-256.
(14.) Isabel Zhu Y, Fitzpatric JE. Expression of c-kit(CD-117)in Spitznevus and malignant melanoma. J Cutan Pathol. 2006;33(1):33-37.
(15.) Montone KT, van Belle P, Elenitsas R, Elder DE. Proto-oncogene c-kit expression in malignant melanoma: protein loss with tumor progression. Mod Pathol. 1997;10(9):939-944.
(16.) Miettinen M, Sobin LH, Sarlomo-Rikala M. Immunohistochemical spectrum of GISTs at different sites and their differential diagnosis with a reference to CD117 (KIT). ModPathol. 2000;13(10):1134-1142.
(17.) Pinto A, Gloghini A, Gattei V, Aldinucci D, Zagonel V, Carbone A. Expression of the c-kit receptor in human lymphomas is restricted to Hodgkin's disease and CD30+ anaplastic large cell lymphomas. Blood. 1994;83(3):785-792.
(18.) Brauns TC, Schultewolter T, Dissemond J, Maschke J, Goos M. C-Kit expression in primary cutaneous T-cell lymphomas. J Cutan Pathol. 2004;31(9): 577-582.
(19.) Rassidakis GZ, Georgakis GV, Oyarzo M, Younes A, Medeiros LJ. Lack of c-kit (CD117) expression in CD30+ lymphomas and lymphomatoid papulosis. Mod Pathol. 2004;17(8):946-953.
Melissa Pulitzer, MD; Mary Sue Brady, MD; Elen Blochin, MD; Bijal Amin, MD; Julie Teruya-Feldstein, MD
Accepted for publication February 23, 2012.
From the Departments of Pathology (Drs Pulitzer, Blochin, and Teruya-Feldstein) and Surgery (Dr Brady), Memorial Sloan-Kettering Cancer Center, New York, New York;and the Department of Pathology (Dr Amin), Montefiore Medical Center, New York, New York.
The authors have no relevant financial interest in the products or companies described in this article.
Reprints: Melissa Pulitzer, MD, Department of Pathology, Memorial Sloan Kettering Cancer Center, 1275 York Ave, New York, NY 10065 (e-mail: firstname.lastname@example.org).
Please note: Illustration(s) are not available due to copyright restrictions.
|Printer friendly Cite/link Email Feedback|
|Author:||Pulitzer, Melissa; Brady, Mary Sue; Blochin, Elen; Amin, Bijal; Teruya-Feldstein, Julie|
|Publication:||Archives of Pathology & Laboratory Medicine|
|Article Type:||Case study|
|Date:||Feb 1, 2013|
|Previous Article:||Microcystic/reticular schwannoma of the proximal sigmoid colon: case report with review of literature.|
|Next Article:||Diagnostic complexities of eosinophilia.|