Improved Diagnostic and Prognostic Strategies For Desmoplastic Melanoma

Natalia Jaimes, MD
Fundación Cáncer de Piel Colombia
Dermatology Service
Aurora Skin Cancer Center
Medellín, Colombia

Ashfaq A. Marghoob, MD
Clinical Associate Professor
Dermatology Service
Memorial Sloan-Kettering Cancer Center
New York, NY

Diagnosis of desmoplastic melanoma (DM) has long been a challenge for clinicians and pathologists alike. However, new clues and strategies are emerging to aid in early detection and diagnosis and to better determine prognosis. DM is an uncommon variant of spindle cell melanoma, believed by some to represent a sarcomatoid variant of melanoma. It accounts for less than 4 percent of primary cutaneous melanomas, and the overall incidence rate has been reported to be 2.0 per million US persons, with an annual 4.6 percent increase.1,2 The definition and classification of DM has evolved through the years. It was first used in 1971 by Conley, et al3 to refer to the association of invasive tumor cells with abundant stromal collagen. Eight years later, Reed and Leonard4 introduced the term neurotropic melanoma to describe a variant of DM with prominent neural involvement (neurotropism). Subsequently, in 2004, Busam, et al5 further classified DMs into pure DM (pDM) and mixed DM (mDM), based on the degree of desmoplasia (growth of dense connective tissue or stroma) present in the tumor; pDMs have more and mDMs less than 90 percent desmoplasia. The distinction between pDM and mDM appears to have clinical, therapeutic, and prognostic significance; pDM is associated with better overall survival (OS) and less frequent metastasis to regional lymph nodes than mDM. In contrast, mDM appears to display similar biologic behavior to other, non-desmoplastic melanoma subtypes, such as superficial spreading melanoma.

Risk Factors

DM has been associated with risk factors such as male gender, older age, chronic ultraviolet radiation exposure, and sun-damaged skin2 (Table 1). In general, men have about twice the risk of developing DM as women, and this disparity grows even higher in the older population.


Diagnosis of DM can be difficult, not only clinically but histologically. Clinically, early diagnosis of DM is challenging, since it often presents with a non-specific or non-descript morphology. While it may manifest a morphology commonly associated with malignant tumors (e.g., basal cell carcinoma, squamous cell carcinoma, and amelanotic melanoma), it is often mistaken for a benign lesion (e.g., scar, dermatofibroma, neurofibroma, cyst, sclerosing melanocytic nevus).6 In a recent study at eight dermatology clinics for highrisk patients,7 at least 27 percent of DM cases were initially diagnosed as benign lesions and treated with cryotherapy, lasers, or intralesional steroids before the definitive diagnosis of DM was confirmed by biopsy.7

Efforts to enhance clinical recognition of suspected DM are ongoing. Improved knowledge of its risk factors as well as its clinical and dermoscopic characteristics may provide clues, leading to earlier biopsy.

1. Clinical and Dermoscopic Characteristics

DM often manifests features commonly associated with banal lesions, and frequently presents as amelanotic, firm, palpable, or indurated lesions with ill-defined borders. Since it most often appears on sun-damaged skin, especially on the head and neck,2,7 it is not uncommon for patients to present with a history of a non-specific or scarlike lesion located on chronically sundamaged skin that developed without any antecedent trauma.

The primary clinical morphology of DM alone provides insufficient criteria for recognition. However, two case series have shown that dermoscopy may provide additional information indiscernible to the unaided eye; it thus can prompt clinical concern and further testing.8,7 Dermoscopically, DMs may appear as featureless lesions or may reveal one or more lanoma-specific structures, in particular atypical vascular structures, granularity, blue-white veils, atypical globules, and crystalline structures or atypical networks. In addition, dermoscopic features of lentigo maligna melanoma (LMM), such as annular granular pattern and polygonal lines, can be seen in up to one third of DM cases7 (Table 2).

Jaimes, Chen, Dusza, et al7 recently described the clinical and dermoscopic features of DM as a function of the histopathological subtypes. Their research demonstrated that pDM and mDM cannot be clinically distinguished from one another; however, with dermoscopy some differences were noted: pDMs more often reveal a monomorphous vessel pattern, with dotted vessels being the most common. In contrast, mDMs tend to present with a polymorphous vascular pattern and a vascular blush.

Since DMs have a prominent dermal stromal component, palpation remains an important part of the clinical examination. Accordingly, DM should be considered in the differential diagnosis of any firm lesion encountered on chronically sun-damaged skin, even if under dermoscopy it appears to be featureless. In addition, since DMs can be associated with LMM, all clinically suspect LMMs should also be palpated to ensure there are no subcutaneous firm nodules within. For any such firm areas discovered, if a benign diagnosis cannot be assured, a biopsy should be performed. Furthermore, if a firm lesion also reveals any of the dermoscopic structures described in Table 2, a biopsy should be strongly contemplated.

2. Histopathology of DM

On routine hematoxylin and eosinstained sections, DMs are characterized by the presence of an invasive melanoma with abundant collagenous matrix. Usually, pDMs are pauci-cellular with a prominent component of desmoplasia throughout the majority of the tumor (>90 percent). In contrast, mDMs have a higher cell density, which may be observed throughout the tumor, or may manifest as nodules of solid spindle and/or epithelioid melanocytes against a background of classic pauci-cellular DM.5

In approximately one third of DMs, there is no identifiable in situ elanoma, and histopathologic recognition of the malignant dermal spindle cells may be challenging. DMs associated with an in situ melanoma in the epidermis and/or follicular epithelium are easier to recognize histopathologically. LM is the most common type of melanoma associated with DM, followed by superficialspreading melanoma (SSM).5,7,9 It has been suggested that mDMs may be easier to recognize clinically, based on the presence of a superficial component consisting of either an LM or an SSM. In contrast, early pDMs may be relatively inconspicuous, since these tumors usually appear as dermal nodules or plaques and generally lack epidermal clues such as pigmentation.

Immunohistochemical studies are often needed for optimal assessment of tumor thickness and to distinguish DMs from non-melanocytic mimickers. S-100 is one of the most valuable diagnostic markers for the disease. However, caution is warranted when evaluating scars for possible residual DM. Scars can contain S-100-positive cells, which may be confused with DM; however, these S-100 positive cells tend to be scattered and isolated in scars, whereas they are strongly positive in DMs.6 Tumors of Schwann cells, such as Schwannoma, neurofibroma, or malignant peripheral nerve sheath tumors, cannot be distinguished from DM by S-100. Other immunohistochemical studies that can be useful in select cases include antibodies, nerve growth factor receptor, and Sox 10.10,11 Other melanocyte differentiation antigens, such as gp100, Melan-A/ Mart-1, tyrosinase, and microphthalmia transcription factor are usually negative in DM.12

DMs, especially deeply infiltrating tumors located in the head and neck region,2,13,14 can infiltrate and extend along nerves (neurotropism). Neurotropic DMs (NDM) have been associated with deeper tumors, greater mitotic activity, local infiltration, multiple recurrences, and more metastases than DMs without neurotropism.1,4 When NDM involves the head and neck region, it may give rise to trigeminal and facial nerve neuropathies, and may result in higher rates of CNS metastases due to direct tumor extension.15,16 As a result, some have advocated adjuvant radiation therapy to improve local control.17

3. Molecular findings

There is a paucity of information regarding the molecular profile of DM. Some studies have demonstrated a decrease in the expression of genes involved in melanin synthesis and increased expression of the glycoprotein clusterin, which is involved in physiological processes such as cell adhesion and tissue and fibrous remodeling.18 Unfortunately, fluorescence in situ hybridization has had limited diagnostic value in differentiating DM from benign melanocytic lesions.19 However, arraycomparative genomic hybridization may prove helpful in select cases – for example, in distinguishing DM from sclerosing Spitz nevus through detection of a copy number gain of chromosome 11p in the absence of any other chromosomal changes. Such a finding would support the diagnosis of a benign nevus and, for all practical purposes, exclude the diagnosis of DM.


Five-year and 10-year DM-specific survival have been reported to be 84.8 and 79.2 percent respectively, with five-year OS between 67 percent and 89 percent.2,20 Factors associated with an increased risk for DM-specific death include male gender, advancing age, and location on the head or neck.2

Whether the prognosis of DM is more or less favorable than for non-DMs remains unclear. Some studies have reported no significant difference in patient survival rates.1 Other studies have shown a poorer prognosis for DM,28,29 while still others have demonstrated a more favorable prognosis for DM compared to non-DMs of similar thickness.15 Some studies have even suggested that desmoplasia may confer better survival for patients with tumors greater than 4mm.21,22

The classification of DMs into the two histologic subtypes (pDM and mDM) has demonstrated prognostic differences in disease-free survival.5 Patients with mDM have shown a 3.5-fold greater risk for metastasis or death, and a shorter time to recurrence,14 while pDMs have a lower frequency of regional lymph in NDM.1,22 node involvement (1 percent) as compared with mDM (10 percent) or other melanoma subtypes (6 percent; P<0.05, pDM vs. other types).23

The impact of neurotropism onsurvival continues to be studied. Someresearchers have shown that NDM is associated with a 30 percent decrease in 8-year survival,15 while others have reported no association of neurotropism with a worse OS, despite the observation of a higher local recurrence rate in patients with NDM.1,22

Sentinel Lymph Node Biopsy (SLNB)

The impact of regional lymph node status on the survival of patients with DM remains uncertain and somewhat controversial. Regional lymph node involvement is reported in 0 to 18.8 percent of cases, which is less frequent than observed for other cutaneous melanomas.15,31-33 Unlike non-DMs, the nodal status of patients with DMs does not appear to predict prognosis, with the 5-year OS being virtually the same for node-positive and node-negative patients (65 percent for node-negative, 64 percent for node-positive, p=0.86).20,24 Nevertheless, some studies show that the risk of death is higher for patients with DM-positive lymph nodes (95 percent CI: 1.94-4.65).2 These inconsistencies may be due to differences in definitions used for selecting DM cases for inclusion in studies, or to the use of data acquired prior to our understanding of the differences between pDM and mDM.

Overall, the consensus of experts in the field is that the risk of developing nodal involvement in DM is in fact lower than for patients with non-DM melanoma. Given this lower risk and the questionable prognostic value of knowing the patient’s SLNB status, the utility of SLNB in management of DM patients has been brought into question, with some authors now recommending against routine SLNB.20,25,26 This is particularly true for patients with pDMs, since significant differences in lymph node positivity have been observed between patients with pDM and those with mDM (1.4 percent in pDM patients staged by SLNB compared to 18.5 percent in mDM patients.)26 In fact, Mohebati, et al27 recently reported that staging pDMs of the head and neck by SLNB may not be necessary, given the preponderance of negative LNBs and the low incidence of both lymphovascular invasion and recurrence.

Weighing all the facts, many have suggested that rather than eliminating SLNB altogether, selective SLNB should still be considered and encouraged in patients with mDM, as well as in pDM patients with additional high-risk factors,such as younger age, presence of neurotropism, high mitotic rate, and tumor ulceration.22

Distant Metastases

Systemic metastasis occurs in 7-44 percent of DM cases, with lung, liver,and bone the most commonly involved areas.13 Risk for distant metastasis can vary depending on histopathological type, the risk higher for patients with mDM than for those with pDM.28


Wide Local Excision (WLE) remains the main line of treatment for DM. The extent of surgical resection, independent of tumor thickness, has been shown to be a predictor for survival.20 Patients undergoing WLE with margins greater than 1cm have a better OS than those undergoing excisions of less than 1cm (67 percent vs. 60 percent; p=0.029).20 In addition, wide margins can reduce the rate of local recurrence resulting from persistent disease present focally near the excision margin and/or along nerves (neurotropism). The recurrence rate is as high as 20 percent in neurotropic DM cases vs. only 6.8 percent in non-neurotropic DM cases.1 It has also been shown that DMs excised with margins <1 cm have a higher rate of local recurrence than those excised with margins >2 cm.1,17

Maurichi, et al28 compared outcomes with different surgical margins based on histologic type, in both thin (<2 mm) and thick (>2 mm) pDMs and mDMs. Thin pDMs excised with 1 cm margins had a higher rate of recurrence and worse 5-year OS (60 percent versus 85.2 percent, P=0.014) than those treated with 2 cm margins. Patients whose thick pDMs were excised with 2 cm margins had similar survival to patients whose thin pDMs were excised with 2cm margins (86.6 percent versus 85.2 percent). Finally, the mortality risk in patients with mDM increased according to their stage but was independent of width of excision margins, suggesting that the behavior of mDM is similar to that of non-desmoplastic melanomas. Hence, to minimize the risk of local recurrence and metastasis, WLE with clear margins is recommended for both subtypes of DM, and whenever feasible, a 2 cm margin appears to offer the highest cure rates.

However, there are cases where wide margins and deep excisions are not feasible, for example with DMs that are deeply infiltrating or located in cosmetically, structurally, or functionally sensitive areas such as the periorbital or periocular regions of the face.25,28 In these circumstances, adjuvant radiation therapy may improve local control of the disease.29 There is some evidence that adjuvant radiotherapy may provide added value for DMs at high risk for local recurrence, including tumors excised with narrow margins, those removed with positive excision margins, or those with neurotropism.30,31 However, given the limited data aboutadjuvant radiation therapy’s role in DM, the available evidence remains somewhat controversial, and the most appropriate dose, fractionation, and target volume of radiation remain to be elucidated.2,17,20,32 Ongoing prospective studies on adjuvant radiation therapy for DM will likely clarify its role in management of the disease.

Finally, since there are scarce followup data pertaining to the management of patients with distant metastatic DM, the role for systemic therapies with medications such as ipilimumab remains to be determined.33,34

  • Figure 1A. (left) Clinical image of an ill-defined pink nodule on the glabella of an 81-year-old male. Pathology demonstrated a 7.2mm pure DM. 

    Figure 1B. (right) Under dermoscopy, serpentine vessels and vascular blush (arrows) were observed.

  • Figure 2A. (left) Clinical image of a tan to slightly erythematous lesion with a focal bluish papule on the arm of an 81-year-old female. Palpation of the lesion revealed that the bluish papule was firm. Pathology demonstrated a 1.6mm mixed DM. 

    Figure 2B. (right) Under dermoscopy, off-center blotch, tan structureless areas, atypical dots/globules (solid square), polymorphous vessels (dashed square at left, dashed rectangle at right), and blue-white veil (black arrow) were observed.


Although DM can be a diagnostic challenge for clinicians and pathologists, certain clues may improve detection and aid in diagnosis of these tumors. For instance, diagnosis of DM should be considered when evaluating nonspecific, scar-like lesions displaying irregular vessels under dermoscopy, or when a dermal firm nodule is noted during palpation of a suspected LMM. Once a DM is diagnosed, the pathologic distinction between pDM and mDM should be documented, since this information enhances management decisions and provides prognostic information. While routine SLNB may not be necessary for pDMs, it should be discussed with patients and considered for those with additional high-risk features. To reduce local recurrence, WLE with clear margins is recommended, and whenever feasible, 2 cm margins are ideal. Adjuvant radiotherapy should be considered for tumors at high risk for local recurrence. We remain optimistic that future studies will improve our current knowledge and understanding of the biology of DMs. This in turn will likely lead to more accurate diagnosis and help to target therapies for this malignancy.


  1. Quinn MJ, Crotty KA, Thompson JF, Coates AS, O'Brien CJ, McCarthy WH. Desmoplastic and desmoplastic neurotropic melanoma: experience with 280 patients. Cancer 1998; 83:1128-35.
  2. Feng Z, Wu X, Chen V, Velie E, Zhang Z. Incidence and survival of desmoplastic melanoma in the United States, 1992-2007. J Cutan Pathol 2011; 38:616-24.
  3. Conley J, Lattes R , Orr W. Desmoplastic malignant melanoma (a rare variant of spindle cell melanoma). Cancer 1971; 28:914-36.
  4. Reed RJ, Leonard DD. Neurotropic melanoma. A variant of desmoplastic melanoma. Am J Surg Pathol 1979; 3:301-11.
  5. Busam KJ, Mujumdar U, Hummer AJ, Nobrega J, Hawkins WG, Coit DG, et al. Cutaneous desmoplastic melanoma: reappraisal of morphologic heterogeneity and prognostic factors. Am J Surg Pathol 2004; 28:1518-25.
  6. Busam KJ. Desmoplastic melanoma. Clin Lab Med 2011; 31:321-30.
  7. Jaimes N CL, Dusza SW, Carrera C, Puig S, Thomas L, et al.  Unpublished data, submitted to Arch Derm 2012.
  8. Debarbieux S, Ronger-Salve S, Dalle S, Balme B, Thomas L. Dermoscopy of desmoplastic melanoma: report of six cases. Br J Dermatol 2008; 159:360-3.
  9. de Almeida LS, Requena L, Rutten A, Kutzner H, Garbe C, Pestana D, et al. Desmoplastic malignant melanoma: a clinicopathologic analysis of 113 cases. Am J Dermatopathol 2008; 30:207-15.
  10. Busam KJ, Iversen K, Coplan KC, Jungbluth AA. Analysis of microphthalmia transcription factor expression in normal tissues and tumors, and comparison of its expression with S-100 protein, gp100, and tyrosinase in desmoplastic malignant melanoma. Am J Surg Pathol 2001; 25:197-204.
  11. Longacre TA, Egbert BM, Rouse RV. Desmoplastic and spindle-cell malignant melanoma. An immunohistochemical study. Am J Surg Pathol 1996; 20:1489-500.
  12. Ohsie SJ, Sarantopoulos GP, Cochran AJ, Binder SW. Immunohistochemical characteristics of melanoma. J Cutan Pathol 2008; 35:433-44.
  13. Lens MB, Newton-Bishop JA, Boon AP. Desmoplastic malignant melanoma: a systematic review. Br J Dermatol 2005; 152:673-8.
  14. Murali R, Shaw HM, Lai K, McCarthy SW, Quinn MJ, Stretch JR, et al. Prognostic factors in cutaneous desmoplastic melanoma: a study of 252 patients. Cancer 2010; 116:4130-8.
  15. Baer SC, Schultz D, Synnestvedt M, Elder DE. Desmoplasia and neurotropism. Prognostic variables in patients with stage I melanoma. Cancer 1995; 76:2242-7.
  16. Smithers BM, McLeod GR, Little JH. Desmoplastic melanoma: patterns of recurrence. World J Surg 1992; 16:186-90.
  17. Chen JY, Hruby G, Scolyer RA, Murali R, Hong A, Fitzgerald P, et al. Desmoplastic neurotropic melanoma: a clinicopathologic analysis of 128 cases. Cancer 2008; 113:2770-8.
  18. Busam KJ, Zhao H, Coit DG, Kucukgol D, Jungbluth AA, Nobrega J, et al. Distinction of desmoplastic melanoma from non-desmoplastic melanoma by gene expression profiling. J Investig Dermatol 2005; 124:412-9.
  19. Gerami P, Beilfuss B, Haghighat Z, Fang Y, Jhanwar S, Busam KJ. Fluorescence in situ hybridization as an ancillary method for the distinction of desmoplastic melanomas from sclerosing melanocytic nevi. J Cutan Pathol 2011; 38:329-34.
  20. Wasif N, Gray RJ, Pockaj BA. Desmoplastic melanoma—the step-child in the melanoma family? J Surg Oncol 2011; 103:158-62.
  21. Spatz A, Shaw HM, Crotty KA, Thompson JF, McCarthy SW. Analysis of histopathological factors associated with prolonged survival of 10 years or more for patients with thick melanomas (> 5 mm). Histopathology 1998; 33:406-13.
  22. Livestro DP, Muzikansky A, Kaine EM, Flotte TJ, Sober AJ, Mihm MC, Jr., et al. Biology of desmoplastic melanoma: a case-control comparison with other melanomas. J Clin Oncol 2005; 23:6739-46.
  23. Hawkins WG, Busam KJ, Ben-Porat L, Panageas KS, Coit DG, Gyorki DE, et al. Desmoplastic melanoma: a pathologically and clinically distinct form of cutaneous melanoma. Ann Surg Oncol 2005; 12:207-13.
  24. Smith VA, Lentsch EJ. Sentinel node biopsy in head and neck desmoplastic melanoma: an analysis of 244 cases. Laryngoscope 2012; 122(1):116-20.
  25. Pawlik TM, Ross MI, Prieto VG, Ballo MT, Johnson MM, Mansfield PF, et al. Assessment of the role of sentinel lymph node biopsy for primary cutaneous desmoplastic melanoma. Cancer 2006; 106:900-6.
  26. George E, McClain SE, Slingluff CL, Polissar NL, Patterson JW. Subclassification of desmoplastic melanoma: pure and mixed variants have significantly different capacities for lymph node metastasis. J Cutan Pathol 2009; 36:425-32.
  27. Mohebati A, Ganly I, Busam KJ, Coit D, Kraus DH, Shah JP, et al. The role of sentinel lymph node biopsy in the management of head and neck desmoplastic melanoma. Ann Surg Oncol 2012; July 6 [Epub ahead of print].
  28. Maurichi A, Miceli R, Camerini T, Contiero P, Patuzzo R, Tragni G, et al. Pure desmoplastic melanoma: a melanoma with distinctive clinical behavior. Ann Surg 2010; 252:1052-7.
  29. Vongtama R, Safa A, Gallardo D, Calcaterra T, Juillard G. Efficacy of radiation therapy in the local control of desmoplastic malignant melanoma. Head Neck 2003; 25:423-8.
  30. Mendenhall WM, Amdur RJ, Hinerman RW, Werning JW, Malyapa RS, Villaret DB, et al. Skin cancer of the head and neck with perineural invasion. Am J Clin Oncol 2007; 30:93-6.
  31. Newlin HE, Morris CG, Amdur RJ, Mendenhall WM. Neurotropic melanoma of the head and neck with clinical perineural invasion. Am J Clin Oncol  2005; 28:399-402.
  32. Foote MC, Burmeister B, Burmeister E, Bayley G , Smithers BM. Desmoplastic melanoma: the role of radiotherapy in improving local control. Aust NZ J Surg 2008; 78:273-6.
  33. Davison JM, Rosenbaum E, Barrett TL, Goldenberg D, Hoque MO, Sidransky D, et al. Absence of V599E BRAF mutations in desmoplastic melanomas. Cancer 2005; 103:788-92.
  34. Miller DD, Emley A, Yang S, Richards JE, Lee JE, Deng A, et al. Mixed versus pure variants of desmoplastic melanoma: a genetic and immunohistochemical appraisal. Modern Pathol 2012; 25(4):505-15.
  35. Kvaskoff M, Siskind V, Green AC. Risk factors for lentigo maligna melanoma compared with superficial spreading melanoma: a case-control study in Australia. Arch Dermatol 2012; 148(2):164-70.
  36. Chang AE, Karnell LH, Menck HR. The National Cancer Data Base report on cutaneous and noncutaneous melanoma: a summary of 84,836 cases from the past decade. The American College of Surgeons Commission on Cancer and the American Cancer Society. Cancer 1998; 83:1664-78.
  37. Chamberlain AJ, Fritschi L, Giles GG, Dowling JP, Kelly JW. Nodular type and older age as the most significant associations of thick melanoma in Victoria, Australia. Arch Dermatol 2002; 138:609-14.
  38. Gray RJ, Pockaj BA, Vega ML, Connolly SM, DiCaudo DJ, Kile TA, et al. Diagnosis and treatment of malignant melanoma of the foot. Foot Ankle Int 2006; 27:696-705.