An update on reporting histopathologic prognostic factors in melanoma.
The understanding of prognostic factors and their clinical significance in cutaneous melanomas is continuously evolving. The American Joint Committee on Cancer (AJCC) adopted in 2001 a version of the melanoma staging system based on an analysis of 17600 patients in the AJCC Melanoma Staging Database. (2,3) In November 2009, a new version of the AJCC system for melanoma staging and classification was released, based on an expanded sample size of the melanoma staging database (more than 30 000 patients) and multivariate analysis of different independent prognostic factors. (4)
There is variability between different institutions and pathologists regarding the routine reports of histologic parameters in melanomas. Some of the reports contain only minimal information (such as tumor thickness and presence or absence of ulceration), while others are very comprehensive. In such detailed reports, some of the information provided might not be of immediate relevance to a given patient, but its importance might become apparent at a later date, especially when used in conjunction with information gathered from a large number of patients in prospective or retrospective studies. (5) As an example, mitotic count is now included as an essential histopathologic element in the newest version of AJCC's melanoma staging and classification system. (4)
In our practice, we are providing a template that includes all the histologic parameters that have been proved significant in determining the stage of a tumor and patient prognosis, as well as a few other details that may be helpful in further analysis of potential complete excisional biopsies or metastatic lesions (such as predominant type of tumor cells, presence or absence of desmoplastic component, or associated benign melanocytic lesions). The goal of this review article is to describe the histopathologic findings that we include in our reports, along with description of their potential practical significance or usefulness. As a note, we are also aware of the numerous drawbacks in reporting some of these histopathologic parameters. These include controversial or out-of-date criteria of diagnosis for some parameters, interobserver variability (which may be sometimes significant) in the assessment of some histologic details, and occasional lack of a perfectly reproducible method for quantifying these criteria.
[FIGURE 1 OMITTED]
HISTOLOGIC PARAMETERS REPORTED IN INVASIVE MELANOMA
Traditionally, the melanomas are classified into one of the several well-defined histogenetic subtypes: superficial spreading, lentigo maligna, acral lentiginous, and nodular (6,7) (Figure 1, A through C). However, justification for such a classification is controversial because of histologic overlap, because each of these types defines mainly a pattern of growth, and because of increasing recognition that classification of melanomas into histogenetic subtypes has no significant prognostic relevance. (8-10) In recent years, emerging genetic studies have suggested that there are distinctive patterns of chromosomal aberrations in melanomas arising on chronically sun-exposed skin versus melanomas arising in areas intermittently exposed to the sun or acral/mucosal melanomas. Understanding this molecular heterogeneity of melanoma is of great clinical importance because it is likely to result in separate targeted therapeutic approaches. (11) For example, acral lentiginous/mucosal melanoma can harbor KIT gene mutations and can potentially respond to targeted therapies with tyrosine kinase inhibitors.
It is also important to recognize that spindle and desmoplastic melanomas have higher propensity for neurotropism, thus resulting in high local recurrence. Moreover, in cases of pure desmoplastic melanoma (defined as having greater than 90% desmoplastic component), there is a lower incidence of lymph node metastasis and, in many practices, sentinel lymph node biopsy is not recommended. (12)
In 1970, Breslow (13) proposed that measuring the vertical thickness of melanoma is a good tool to predict the metastatic potential of a tumor. To date, this measurement is the single most important prognostic factor for melanoma clinical behavior. (4) Moreover, the evaluation of this parameter has the least interobserver variability. The Breslow thickness is measured from the top of the epidermal granular layer to the deepest melanocyte of the invasive component of melanoma. (14,15) It is important to mention that involvement of follicular or adnexal structures by melanoma in situ, even if they are situated deeper in dermis, should not be considered for the measurement of the Breslow thickness. On the other hand, the presence of deep perineural invasion by melanoma should be measured and reported. In our opinion, such areas should be included in the measurement of the Breslow thickness, but not all authors agree. When there is tumor-induced ulceration, measurement should be from the base of the ulceration to the deepest aspect of the invasive component of melanoma. Extensive involvement of follicular structures by melanoma is sometimes present and the invasive component is identified only in the vicinity of those structures. In these cases, the thickness of melanoma should be measured from the central portion of the follicular/adnexal structure to the furthest adjacent invasive melanoma cell. Some pathologists include in their report both Breslow thickness and the deeper invasion measured from the center of the hair follicle. For the measurement of thin or intermediate-thickness melanomas, we recommend using an ocular micrometer for a precise assessment. For thick lesions (beyond 2 mm) it may be acceptable to measure the tumor thickness by placing a ruler over the slide. If the invasive melanoma is present at the deep tissue edge of a biopsy specimen or at the deep margin of a resection specimen, the report should note this fact and indicate that the final Breslow thickness may be higher.
The anatomic levels of melanoma invasion were proposed by Clark as level I (melanoma in situ), level II (invasion into superficial papillary dermis), level III (invasive melanoma that fills and expands the papillary dermis), level IV (invasion into reticular dermis), and level V (infiltration into subcutaneous adipose tissue). (6) Assessment of Clark levels II, III, and IV has a very high interobserver variability. Moreover, in melanomas with polypoid growth pattern, the tumor can reach a thick Breslow thickness but still be a Clark level III tumor. In these cases, there is no significant prognostic value of the Clark level; ultimately, the thickness of the tumor dictates the clinical behavior. Historically, the level of invasion was correlated with survival but, on multivariate analyses, this prognostic factor loses its validity. However, in some institutions, detection of Clark level IV triggers examination of sentinel lymph nodes. (2,3) Currently, after 40 years as an integral component of melanoma staging, the Clark level is no longer recommended as a staging criterion, since it is not an independent prognostic factor when mitotic rate is included in the analysis. (4)
Radial (Nontumorigenic) and Vertical (Tumorigenic) Growth Phase
The definition of radial and vertical growth phases is based on the concept that tumor progression implicates different evolutionary steps that can be identified on histologic examination and predict the tumor metastatic potential. By definition, all in situ melanomas have only radial growth phase. When melanoma invades into dermis, it may still have only radial growth phase (single cells or small nests without mitotic figures). Vertical growth phase has been defined as "tumorigenic" growth when the dermal nests are larger than any nest in the junctional component of the lesion or when mitotic figures can be identified within the dermal melanocytes. An earlier report (16) has shown that the metastatic potential of a melanoma strongly correlates with the presence of vertical growth phase. Vertical growth may be related to tumor volume; a melanoma that has confluent nests of tumor cells in dermis will most likely have a greater metastatic potential than a tumor with rare single cells in dermis, even if they have identical Breslow thickness.
While there is no standard method for counting mitotic figures, this should be done only in the invasive dermal component of melanoma. Most authors recommend identifying "hot spots" that contain the highest number of mitoses and then counting adjacent fields. Until recently, the reported mitotic rate was expressed as the number of mitoses per 10 high-power fields, but since different microscopes have different field sizes, this method may be inaccurate. Currently, the preferred method is to report the mitotic count per square millimeter. Therefore, morphologists should measure the surface of a high-power field (x40) in their microscope and then determine how many fields are needed to include 1 square millimeter (usually about 4 1/2 high-power fields). Previous reports (17,18) have indicated that a high mitotic rate correlates with a poor survival rate. Most notably, in the latest version (2009) of the AJCC melanoma staging and classification, the mitotic rate has emerged as a powerful factor of survival. (4) The mitotic rate replaces the level of invasion (Clark) in defining T1 categories. The presence of 1 or more mitotic figures per square millimeter is now used as a primary criterion for defining T1b-stage melanoma. A possible drawback to the applicability of this criterion may be cases in which there is a very small dermal component and dermal mitoses are identified only after multiple serial sections.
The presence of ulceration is regarded as an independent prognostic factor for melanoma-associated survival. (19) Survival rates of patients with an ulcerated melanoma are lower than those of patients with a nonulcerated melanoma of similar thickness. (4) The presence of ulceration, and possibly its extent, should be always reported. It is also important to recognize the difference between tumor-related ulceration (due to "epidermal consumption and attenuation") and ulceration due to trauma (20) (Figure 2). Therefore, morphologists should be very cautious before reporting ulceration in reexcision specimens (in which the cause for ulceration may be surgical trauma). Recommendation for using ulceration status in defining TNM categories and stage grouping remains unchanged in the 2009 AJCC melanoma staging system.
Histologically, regression can be recognized by the presence of dermal fibrosis, vascular proliferation, inflammatory infiltrate, and presence of melanophages in association with complete or partial destruction of tumor cells. Overlying epidermal thinning with lost rete ridges may be also appreciated (Figure 3). The regression changes can range from focal to extensive. The correlation of regression with prognosis is controversial. Although most studies have not found a significant role for regression in determining survival, some studies showed that the metastatic rate is higher in thin melanomas with extensive regression. (21) In some institutions, sentinel lymph node biopsy is recommended if extensive regression is identified in melanomas with less than 1-mm thickness. At our institution we define "extensive" regression as those cases in which there is regression in more than 50% of the invasive component. Until more information is gathered regarding the possible prognostic utility of the thickness of the areas of regression, we do not advocate providing any estimate of such thickness in the histopathology report.
The presence of tumor cells in lymphovascular spaces is considered a major prerequisite for metastatic spread. Several authors (22-29) have reported that vascular invasion in melanoma may be associated with an increased risk of relapse, lymph node metastasis, and distant metastases and may affect overall and disease-free survival. However, other authors (30) have shown that this feature is not an independent factor in predicting prognosis in patients with melanoma. This apparent discrepancy may be due to the difficulty of identifying this feature on hematoxylineosin sections alone. Previous reports (31,32) on immunohistochemical detection of lymphovascular invasion and correlation with metastatic rate have yielded conflicting results. Particularly, Sahni et al (31) showed that although immunohistochemical staining for LYVE-1 (a lymphatic marker) can reliably demonstrate lymphatic vessel distribution and can detect melanoma cells within lymphatic vessels, it is unreliable in predicting melanoma metastasis, failing to detect metastatic spread in more than two-thirds of patients with regional node metastasis. More recently, we have shown that immunohistochemical detection of lymphovascular invasion with anti-D2-40, a monoclonal antibody against podoplanin, in melanomas thicker than 1 mm, correlates with sentinel lymph node status and survival. Therefore, we suggest that such analysis might be incorporated in the routine workup of primary cutaneous melanoma thicker than 1 mm. (29)
[FIGURE 2 OMITTED]
[FIGURE 3 OMITTED]
[FIGURE 4 OMITTED]
Infiltration of nerves by tumor cells should also be recorded (Figure 4). Some types of melanomas, such as desmoplastic or spindle cell melanomas, have a high propensity for perineural invasion. This feature is also relatively commonly seen in acral-lentiginous melanomas.
Microsatellites are defined as discrete tumor aggregates, arbitrarily designated by some authors as having a diameter larger or equal to 0.05 mm in largest dimension, that are separated from the main tumor by normal (not fibrosis or inflammation) dermal collagen or subcutaneous fat. (33) Only few studies have evaluated the role of microsatellites as prognostic factors in cutaneous melanomas. It is controversial whether their presence is an independent prognostic factor, but they do appear to correlate with a higher risk of local recurrence and with an increased frequency of regional lymph node metastasis (from 12% to 53%) in tumors thicker than 1.5 mm. (34)
The host immune response is recognized by the presence of a lymphocytic inflammatory infiltrate. It can be "brisk," "nonbrisk," and "minimal" ("absent") and is usually measured by assessing the extent of lymphocytic infiltrate surrounding the invasive dermal component of melanoma. The presence of a brisk inflammatory infiltrate has been reported that correlates with improved survival. However, the assessment of this feature is observer-dependent, mainly owing to the lack of a uniform definition of host response in terms of type and location of the infiltrate.
Associated Melanocytic Nevus
Recording this finding may be useful for epidemiologic studies or when a reexcision is performed and it is difficult to decide if the residual lesion is an associated benign melanocytic nevus or melanoma (especially in cases of melanomas arising in cellular blue nevi or deep-penetrating nevi).
In our practice, we report this feature mainly for its practical applicability, when the reexcision specimens or possible lymph nodes are evaluated. This is especially true for small cell or nevoid melanomas for which it may be difficult to assess small lymph node metastases. Moreover, if the primary tumor has spindle cell cytology, we may perform not only a pan-melanocytic cocktail (HMB45, anti-MART1, and anti-tyrosinase) but also an anti-S100 protein immunohistochemical study on the sentinel lymph nodes to rule out a metastasis, since it is recognized that spindle cell melanomas do not usually label for HMB 45 or MART1.
Margins of Resection
The presence or absence of invasive or in situ components at both peripheral and deep tissue edges must be reported. Furthermore, in our practice, we also note if the tumor-associated stroma is present at the tissue edges.
(1.) Ries LAG, Eisner MP, Kosary CL, et al, eds. SEER cancer statistics review, 1975-2002. National Cancer Institute Web site. Based on November 2004 SEER data submission. Posted 2005. http://seer.cancer.gov. Accessed March 2010.
(2.) Balch CM, Buzaid AC, Soong SJ, et al. Final version of the American Joint Committee on Cancer staging system for cutaneous melanoma. J Clin Oncol. 2001;19(16):3635-3648.
(3.) Balch CM, Soong SJ, Gershenwald JE, et al. Prognostic factors analysis of 17,600 melanoma patients: validation of the American Joint Committee on Cancer melanoma staging system. J Clin Oncol. 2001;19(16):3622-3634.
(4.) Balch CM, Gershenwald JE, Soong SJ, et al. Final version of 2009 AJCC melanoma staging and classification. J Clin Oncol. 2009;27(36):6199-206.
(5.) Calonje E. The histological reporting of melanoma. J Clin Pathol. 2000; 53(8):587-590.
(6.) Clark WH Jr, From L, Bernardino EA, Mihm MC. The histogenesis and biological behavior of primary human malignant melanoma of the skin. Cancer Res. 1969;29(3):705-727.
(7.) McGovern VJ, Mihm MC Jr, Bailly C, et al. The classification of malignant melanoma and its histologic reporting. Cancer. 1973;32(6):1446-1457.
(8.) Ackerman AB, David KM. A unifying concept of malignant melanoma: biologic aspects. Hum Pathol. 1986;17(5):438-440.
(9.) Flotte TJ, Mihm MC Jr. Melanoma: the art versus the science of dermatopathology. Hum Pathol. 1986;17(5):441-442.
(10.) Breuninger H, Kohler C, Drepper H, et al. Is acrolentiginous melanoma (ALM) more malignant than superficially spreading melanoma (SSM) at a high-risk site: a matched-pair comparison between 113 ALM and SSM within the scope of a multicenter study [in German]. Hautarzt. 1994;45(8):529-531.
(11.) Curtin JA, Fridlyand J, Kageshita T, et al. Distinctsets of genetic alterations in melanoma. N Engl J Med. 2005;353(20):2104-2107.
(12.) Pawlik TM, Ross MI, Prieto VG, et al. Assessment of the role of sentinel lymph node biopsy for primary cutaneous desmoplastic melanoma. Cancer. 2006;106(4):900-906.
(13.) Breslow A. Thickness, cross-sectional areas and depth of invasion in the prognosis of cutaneous melanoma. Ann Surg. 1970;172(5):902-908.
(14.) Balch CM, Murad TN, Soong SJ, Ingalls AL, Halpern NB, Maddox WA. A multifactorial analysis of melanoma: prognostic histopathologic features comparing Clark's and Breslow's staging methods. Ann Surg. 1978;188(6):732-742.
(15.) Keefe M, Mackie RM. The relationship between risk of death from clinical stage 1 cutaneous melanoma and thickness of primary tumour: no evidence for steps in risk. Scottish Melanoma Group. Br J Cancer. 1991;64(3):598-602.
(16.) Lefevre M, Vegier B, Balme B, et al. Relevance of vertical growth pattern in thin level II cutaneous superficial spreading melanomas. Am J Surg Pathol. 2003; 27(6):717-724.
(17.) Eldh J, Boeryd B, Peterson LE. Prognostic factors in cutaneous malignant melanoma in stage I: a clinical, morphological and multivariate analysis. Scand J Plast Reconstr Surg. 1978;12(3):243-255.
(18.) Schmoeckel C, Braun-Falco O. Prognostic index in malignant melanoma. Arch Dermatol. 1978;114(6):871-873.
(19.) Balch CM, Wilkerson JA, Murad TN, Soong SJ, Ingalls AL, Maddox WA. The prognostic significance of ulceration of cutaneous melanoma. Cancer. 1980; 45(12):3012-3017.
(20.) Scolyer RA, Shaw HM, Thompson JF, et al. Interobserver reproducibility of histopathologic prognostic variables in primary cutaneous melanomas. Am J Surg Pathol. 2003;27(12):1571-1576.
(21.) Ronan SG, Eng AM, Briele HA, et al. Thin malignant melanomas with regression and metastases. Arch Dermatol. 1987;123(10):243-255.
(22.) Dadras SS, Paul T, Bertoncini J, et al. Tumor lymphangiogenesis: a novel prognostic indicator for cutaneous melanoma metastasis and survival. Am J Pathol. 2003;162(6):1951-1960.
(23.) Kashani-Sabet M, Sagebiel RW, Ferreira CM, Nosrati M, Miller JR III. Vascular involvement in the prognosis of primary cutaneous melanoma. Arch Dermatol. 2001;137(9):1169-1173.
(24.) Massi D, Borgognoni L, Franchi A, Martini L, Reali UM, Santucci M. Thick cutaneous malignant melanoma: a reappraisal of prognostic factors. Melanoma Res. 2000;10(2):153-164.
(25.) Nagore E, Oliver V, Botella-Estrada R, Moreno-Picot S, Insa A, Fortea JM. Prognostic factors in localized invasive cutaneous melanoma: high value of mitotic rate, vascular invasion and microscopic satellitosis. Melanoma Res. 2005; 15(3):169-177.
(26.) Nguyen CL, McClay EF, Cole DJ, et al. Melanoma thickness and histology predict sentinel lymph node status. Am J Surg. 2001;181(1):8-11.
(27.) Niezabitowski A, Czajecki K, Rys J, et al. Prognostic evaluation of cutaneous malignant melanoma: a clinicopathologic and immunohistochemical study. J Surg Oncol. 1999;70(3):150-160.
(28.) Straume O, Akslen LA. Independent prognostic importance of vascular invasion in nodular melanomas. Cancer. 1996;78(6):1211-1219.
(29.) Petersson F, Diwan AH, Ivan D, et al. Immunohistochemical detection of lymphovascular invasion with D2-40 in melanoma correlates with sentinel lymph node status, metastasis and survival. J Cutan Pathol. 2009;36(11):1157-1163.
(30.) Clark WH Jr, Elder DE, Guerry D IV, et al. Model predicting survival in stage I melanoma based on tumor progression. J Natl Cancer Inst. 1989;81(24): 1893-1904.
(31.) Sahni D, Robson A, Orchard G, Szydlo R, Evans AV, Russell-Jones R. The use of LYVE-1 antibody for detecting lymphatic involvement in patients with malignant melanoma of known sentinel node status. J Clin Pathol. 2005;58(7): 715-721.
(32.) Shields JD, Borsetti M, Rigby H, et al. Lymphatic density and metastatic spread in human malignant melanoma. Br J Cancer. 2004;90(3):693-700.
(33.) Day CL Jr, Harrist TJ, Gorstein F, et al. Malignant melanoma: prognostic significance of "microscopic satellites" in the reticular dermis and subcutaneous fat. Ann Surg. 1981;194(1):108-112.
(34.) Harrist TJ, Rigel DS, Day CL Jr, et al. "Microscopic satellites" are more highly associated with regional lymph node metastases than is primary melanoma thickness. Cancer. 1984;53(10):2183-2187.
Doina Ivan, MD; Victor G. Prieto, MD, PhD
Accepted for publication June 23, 2010.
From the Departments of Pathology and Dermatology, University of Texas M. D. Anderson Cancer Center, Houston.
The authors have no relevant financial interest in the products or companies described in this article.
Reprints: Doina Ivan, MD, Departments of Pathology and Dermatology, Dermatopathology Section, University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Blvd, Unit 85, Houston, TX 77030-4009 (e-mail: DSIvan@mdanderson.org).
|Printer friendly Cite/link Email Feedback|
|Author:||Ivan, Doina; Prieto, Victor G.|
|Publication:||Archives of Pathology & Laboratory Medicine|
|Date:||Jul 1, 2011|
|Previous Article:||Melanocytic lesions: current state of knowledge--part III.|
|Next Article:||Update on fluorescence in situ hybridization in melanoma: state of the art.|