Primary Dermal Melanoma (PDM): Histological and Clinical Interdependence to Guide Therapy.

BACKGROUND/PURPOSE: This study examined clinical and histological parameters of primary dermal melanoma (PDM) to aid in its distinction from dermal metastasis. METHODS: Retrospective analysis of a prospective cohort of PDM patients. Includes patients fulfilling the strict histologic criteria for PDM (N = 9) and patients who did not, but clinically, unequivocally had an intradermal melanoma-clinical PDM (cPDM; N = 17).Histopathology slides were re-examined. Prognosticators and outcome measures were compared between groups. Sentinel nodes' retrieval and wide local excision (WLE) were offered to all patients as primary treatment. RESULTS: 26 patients identified, 15 females with a median age of 69 years (range 3.5-85). Mean Breslow was 7.9 ± 5.7 mm (median 5.8, range 1.8-25.0), and the mean mitotic rate was 4.9 ± 3.8/mm2 (median 4.0, range 0-17). Initial treatment and follow-up were as for cutaneous melanoma. One patient in each group with a palpable stage III underwent primary radical dissection. Sentinel nodes were retrieved in all 20 lymphatic mappings performed and found to be metastatic in 5 (25%) patients. Treatment consisted of completion lymph-node dissection. At a median postoperative follow-up of 62 months (range 8-132), 20 patients were disease-free, including 6 of 7 patients with stage III disease at presentation. Six patients died all of cPDM; 5 of 6 patients had primary ulcerated or epidermal-abutting melanomas. CONCLUSIONS: This is the first study to highlight cPDM. Diagnosis requires expert pathology review and a tight correlation to the clinical parameters. Patients seem to benefit from WLE with sentinel node retrieval and complete dissection when appropriate. However, clinical guidelines for dissection have changed since the time period of this retrospective review. Based on this series, complete nodal dissection in these melanomas is associated with better than expected outcome, for stage III disease.

Different patterns of expression of the erbB family of receptor tyrosine kinases in common nevi, dysplastic nevi, and primary malignant melanomas: an immunohistochemical study.

erbB receptors contribute to tumor formation and progression. Variable expression of erbB1, erbB2, and erbB3 has been reported in nevi and melanomas; erbB4 has hardly been investigated. We examined the expression of all 4 erbB receptors in common and dysplastic nevi and melanomas. Formalin-fixed, paraffin-embedded tissues of 100 melanomas, 27 common nevi, and 23 dysplastic nevi were immunostained with antibodies against the 4 erbB receptors. erbB3 and erbB4 showed stronger positivity in nevi than in melanomas, and in common than in dysplastic nevi. Staining pattern was more orderly in nevi than in melanomas. Common nevi showed more prominent membranous staining for erbB3 than dysplastic nevi followed by melanomas. In melanomas, greater thickness was associated with more widespread erbB2 and erbB3 staining in the vertical than in the radial growth phase, and in the dermal than in the epidermal component. Higher mitotic counts were associated with more widespread and intense erbB2 expression in the vertical growth phase than in the radial growth phase and in the dermal than in the epidermal component. Melanomas with more widespread erbB2 staining had heavier lymphocytic infiltrates. erbB1 expression was negligible in all groups. erbB2, erbB3, and erbB4 are expressed in all subtypes of melanocytic lesions, but with quantitative and qualitative differences. Receptor expression seems to decrease and to become less mature and orderly with tumor progression. The complex patterns of erbB receptor expression in melanocytic lesions warrant further investigation.

Patterns of failure in patients with cutaneous head and neck melanoma.

INTRODUCTION: The incidence of head and neck melanoma is increasing. Various factors influence prognosis. OBJECTIVE: We sought to investigate the subgroup of patients with head and neck melanoma who fail primary treatment and to define the patterns of failure. METHODS: The database of a tertiary medical center was reviewed for patients diagnosed and surgically treated for cutaneous head and neck melanoma in 1995-2014. Regional disease failure was defined as disease confirmed in positive SLNB at first assessment or at recurrence. RESULTS: The cohort included 141 patients followed for a median duration of 6.8 years (range 1-20 years). Median tumor thickness was 2.1 mm (range 0.5-12 mm). Ulceration was documented in 38 patients (26.9%). Sentinel lymph node biopsy (SLNB) was positive in 18 patients (12.8%). Total disease failure rate was 32.6% with similar rates of regional (n = 26, 18.4%) and distal (n = 22, 15.6%) failure. Most patients (86.3%) with systemic recurrence had a negative SNLB as did 6/26 patients (23%) with regional failure. Forty-three patients (30.4%) died during follow-up, half of them (23 patients, 16.3%) of melanoma. On multivariate analysis, Breslow thickness was the only significant predictor of outcome. CONCLUSIONS: The pattern of treatment failure in patients with head and neck melanoma relate predominantly to Breslow thickness. The high false-negative rate of SNLB and the relatively high rate of systemic failures in patients with negative SNLB indicate a low predictive value of this procedure. Efforts to detect systemic disease during follow-up need to be intensified.

Melanoma-Secreted Lysosomes Trigger Monocyte-Derived Dendritic Cell Apoptosis and Limit Cancer Immunotherapy.

The recent success of checkpoint blockade therapies has established immunotherapy as one of the most promising treatments for melanoma. Nonetheless, a complete curative response following immunotherapy is observed only in a fraction of patients. To identify what factors limit the efficacy of immunotherapies, we established mouse models that cease to respond to immunotherapies once their tumors exceed a certain stage. Analysis of the immune systems of the organisms revealed that the numbers of tumor-infiltrating dendritic cells (TIDC) drastically decreased with time. Further, in contrast to the current paradigm, once melanoma was established, TIDC did not migrate into sentinel lymph nodes. Instead, they underwent local cell death due to excessive phagocytosis of lysosomes. Importantly, TIDC were required to license the cytotoxic activity of tumor CD8+ T cells, and in their absence, T cells did not lyse melanoma cells. Our results offer a paradigm shift regarding the role of TIDC and a framework to increase the efficacy of immunotherapies. SIGNIFICANCE: This work redefines the role of monocyte-derived dendritic cells in melanoma and provides a novel strategy to increase the efficacy of T-cell-based immunotherapies in nonresponding individuals. GRAPHICAL ABSTRACT: http://cancerres.aacrjournals.org/content/canres/80/10/1942/F1.large.jpg.

Failure to identify sentinel lymph nodes for malignant melanoma - Outcome after over 10 years median follow up.

BACKGROUND: Sentinel lymph node biopsy (SLNB) is routinely performed during surgery for malignant melanoma, using double mapping. Still, in some cases, a sentinel lymph node identified pre-operatively by lymphoscintigraphy is not identified during surgery. We hypothesized that disease specific survival would not be significantly impacted by intra-operative lymph node mapping (IOLM) failure. METHODS: The patient population study included 1300 malignant melanomas operated on by a single surgical oncologist (H.G.) after sentinel lymph node scintigraphy. Patients were included in the analysis if intra-operative lymph node (IOLM) mapping failed. RESULTS: Among 1300 patients who underwent surgery for malignant melanoma during the study period and after median follow up of >10 years, 33/36 lymphatic drainage basins with failed sentinel node identification were free of disease. Disease specific survival for the entire group of 33 patients with IOLM failure was 91.0%, which is comparable to previously published disease specific survival for all melanoma patients. CONCLUSION: We conclude that failure to identify a pre-operatively marked sentinel lymph node by an experienced melanoma surgeon has, generally, no impact on disease specific survival, as demonstrated in this review of a series of surgical melanoma patients.

MRI diagnosis and follow-up of chest wall and breast desmoid tumours in patients with a history of oncologic breast surgery and silicone implants: A pictorial report.

INTRODUCTION: Breast and chest wall desmoid tumours can cause debilitating symptoms and deformity. The mutilating effects of surgical treatment have prompted a shift to medical treatments and even to a wait-and-see approach. This study sought to highlight specific characteristics of breast and chest wall desmoid tumours on long-term follow-up by sequential MRI scans. METHODS: Thirty-two breast MRI scans from six patients with chest wall or breast desmoid tumours followed up for up to 6 years were retrospectively reviewed. RESULTS: All patients underwent breast surgery prior to the development of the desmoid tumour. Five of the patients had reconstruction or augmentation using silicone implants. Two desmoids were treated primarily with surgery, three with medical means and one is under wait-and-see approach. On MRI, tumours appeared either oval and lobulated (chest wall) or spiculated with architectural distortion (breast). Chest wall desmoids demonstrated both an enhancing high-T2-signal component and a non-enhancing low-T2- signal component. The histologically defined phases during the course of desmoid tumours (progression, regression, residual disease) could be demonstrated by corresponding MRI changes in each of the components. CONCLUSIONS: Magnetic resonance imaging delineates the complex infiltrative features of chest wall and breast desmoid tumours. In tumours with a bright cellular enhancing and dark collagenous non-enhancing component, treatment response may be predicted by changes on serial T2-weighted sequences, beyond the tumour-dimension-based RECIST assessment alone.

A distinct subset of FcγRI-expressing Th1 cells exert antibody-mediated cytotoxic activity.

While a high frequency of Th1 cells in tumors is associated with improved cancer prognosis, this benefit has been attributed mainly to support of cytotoxic activity of CD8+ T cells. By attempting to potentiate antibody-driven immunity, we found a remarkable synergy between CD4+ T cells and tumor-binding antibodies. This surprising synergy was mediated by a small subset of tumor-infiltrating CD4+ T cells that express the high-affinity Fcγ receptor for IgG (FcγRI) in both mouse and human patients. These cells efficiently lyse tumor cells coated with antibodies through concomitant crosslinking of their T cell receptor (TCR) and FcγRI. By expressing FcγRI and its signaling chain in conventional CD4+ T cells, we successfully employed this mechanism to treat established solid cancers. Overall, this discovery sheds new light on the biology of this T cell subset, their function during tumor immunity, and the means to utilize their unique killing signals in immunotherapy.

Surgical management of gastrointestinal stromal tumors: analysis of outcome with respect to surgical margins and technique.

This report reviews the methods and goals of treatment of gastrointestinal stromal tumor (GIST), the most common mesenchymal tumor of the gastrointestinal tract. GISTs express CD117, which serves as an immunohistochemical diagnostic marker. Surgical excision is the definitive treatment for all primary GISTs greater than 2 cm without evidence of peritoneal seeding or metastasis. Preoperative or intraoperative biopsy is not indicated except when the differential diagnosis includes another type of malignancy. Resection may be performed by traditional open surgery or by laparoscopic or laparoscopy-assisted procedures. Regardless of the approach, oncological precautions must be strictly observed. Tumor disruption is to be avoided at all costs; tumor enucleation leaves a tumor-seeded pseudocapsule behind and is considered insufficient. Because GISTs rarely metastasize through the lymphatics, routine lymphadenectomy is not indicated. The importance of achieving negative microscopic margins is controversial, although patients who undergo incomplete microscopic resection may be at greater risk of locoregional recurrence. Other factors, such as tumor grade and size, may play a more significant role in predicting recurrence. Cases of advanced disease or involvement of adjacent structures should be evaluated on an individual basis by a multidisciplinary team.

Resection margins in modern rectal cancer surgery.

At present, the preferred treatment for rectal cancer is low anterior resection with total mesorectal excision and sphincter preservation. Complete removal of the tumor's lymphatic and vascular pad with free resection margins has led to a reduction in rates of local recurrence and improved disease-specific survival. In addition to the distal and proximal margins from the tumor edge, for an optimal outcome, it is essential to consider distal mesorectal spread and the circumferential mesorectal margin.

Concurrent chemobiotherapy with cisplatin, dacarbazine, decrescendo interleukin-2 and interferon alpha2b in patients with metastatic melanoma.

We aimed to evaluate a concurrent chemobiotherapy (CBT) regimen consisting of cisplatin (CDDP), dacarbazine (DTIC), decrescendo interleukin-2 (IL-2), and interferon alpha2b (INF-alpha2b), in metastatic melanoma patients. A total of 60 patients with biopsy proven, metastatic melanoma were treated between October 2000 and November 2005 at the Oncology Institutes of RMC and CSMC. Patients received concurrent CBT for 5 days, consisting of CDDP, DTIC, decrescendo IL-2, and subcutaneous INF-alpha2b. GM-CSF was given subcutaneously on days 8 to 12 of each cycle, to the first 26 patients. Treatment was administered q21d for a total of six cycles or until severe toxicity or progression; 57 patients who received at least two cycles, followed for at least 24 months, were included in response analysis. The overall response rate (RR) reached 44% (28/57 patients); 14 patients had a complete response (CR, 25%); 11 (19%) reached a partial response. The median progression-free survival was 7 months. Median overall survival (OS) was 11.7 months. At a median follow-up of 29 months, 8 of 14 complete responders remain alive for more than two years, with no clinical evidence of disease. Median OS of patients with CR has not been reached; 17% of the courses were modified due to toxicity, and 20% of the patients were removed from the protocol due to toxicity or refusal to continue. The data from this study indicate that this protocol of concomitant CBT is feasible with a fraction of the patients achieving a durable CR.

"Gloves-and-socks" melanoma: does histology make a difference?

BACKGROUND: Acral lentiginous melanoma (ALM) is associated with low survival. OBJECTIVE: The aim of the study was to compare the clinical course of ALM, non-ALM hand and foot melanoma, and melanoma of the extremities in nonacral locations. METHODS: Data on 168 patients operated on for cutaneous melanoma of the extremities from 1993 to 2005 were examined. Twenty-nine had ALM, 16 non-ALM, and 123 other-extremity melanoma. All known melanoma prognosticators were analyzed for their impact on survival at a median of 53 months' follow-up. RESULTS: The ALM group was significantly older (p=.015). No differences between the ALM and non-ALM groups were noted in tumor characteristics, lymph node status, and survival. However, the other-extremity melanoma group presented with significantly thinner lesions, fewer positive sentinel lymph nodes, and lower tumor stage and, consequently, had significantly better disease-specific and disease-free survival (p=.006, p=.0001). The acral lesions were nearly free of peritumoral lymphocytic infiltration. Multivariate analysis identified only tumor thickness (p=.0127), stage (p=.00001), and patient age (p=.012) as independent prognosticators of disease-specific survival. CONCLUSION: Cutaneous melanomas in acral sites, regardless of histology, tend to be diagnosed at an advanced stage probably owing to older patient age, difficult-to-see sites, and biologic factors, leading to reduced patient survival.

Post-traumatic soft tissue tumors: case report and review of the literature a propos a post-traumatic paraspinal desmoid tumor.

BACKGROUND: Antecedent trauma has been implicated in the causation of soft tissue tumors. Several criteria have been established to define a cause-and-effect relationship. We postulate possible mechanisms in the genesis of soft tissue tumors following antecedent traumatic injury. CASE PRESENTATION: We present a 27-year-old woman with a paraspinal desmoid tumor, diagnosed 3-years following a motor vehicle accident. Literature is reviewed. CONCLUSION: Soft tissue tumors arising at the site of previous trauma may be desmoids, pseudolipomas or rarely, other soft tissue growths. The cause-and-effect issue of desmoid or other soft tissue tumors goes beyond their diagnosis and treatment. Surgeons should be acquainted with this diagnostic entity as it may also involve questions of longer follow-up and compensation and disability privileges.

Sentinel node-guided evaluation of drainage patterns for melanoma of the helix of the ear.

The head and neck region, and especially the ear and its helix, is notorious for its ambiguous pattern of lymphatic drainage. Therefore, the primary nodal drainage basins in melanoma of the helix of the ear are often unpredictable. The aim of the study was to examine the value of sentinel lymph node biopsy in melanoma of the helix of the ear and to describe the natural history of the disease. Fifteen consecutive patients (14 men) with primary melanoma of the helix of the ear (median thickness, 1.2 mm; range, 0.7-10.0) underwent preoperative lymphoscintigraphy, followed by intraoperative lymphatic mapping, using blue dye in combination with a hand-held gamma probe and sentinel lymphadenectomy. The melanomas were characterized by low mitotic rate, low lymphocytic infiltrate, low spontaneous-regression rate, and mostly epitheloid cell type. In one patient, preoperative lymphoscintigraphy failed to demonstrate the draining nodes. The sentinel lymph nodes were identified and retrieved in all patients during surgery. In 13 patients (87%), they were found in the upper jugular lymphatic basin (level IIA); none were found in the retroauricular region. All sentinel lymph nodes were tumor-negative. At a median follow-up of 39 months (range, 12-73), all 15 patients were disease-free. In conclusion, sentinel lymph node biopsy for helix melanoma is an excellent alternative to elective lymph node neck dissection and superficial parotidectomy, with a high success rate and low morbidity. Melanoma of the helix of the ear has an indolent natural history.

Lymphatic drainage of calf melanoma skipping the superficial groin.

BACKGROUND: Metastases of melanoma often follow predictable patterns of lymphatic drainage. However, some cases demonstrate first-echelon drainage to an unexpected basin. We describe a patient with drainage from melanoma on the calf to a sentinel lymph node in the iliac basin. METHODS AND RESULTS: Biopsy of the sentinel lymph node was guided by preoperative lymphoscintigraphy and intraoperative use of a gamma probe and blue dye. The node excised from the iliac basin showed evidence of metastasis. CONCLUSION: The failure to detect aberrant sentinel lymph nodes and bypassed basins may lead to improper assessment of disease stage and deficient patient management.

Patterns of recurrence in patients with melanoma after radical lymph node dissection.

HYPOTHESIS: Previous interventions (excisional biopsy, incomplete dissection) in the regional basin that drain a melanoma site prior to definitive surgical procedures significantly increase the risk of melanoma recurrence in the surgical field. DESIGN: Retrospective analysis. SETTING: Tertiary care referral center. PATIENTS: One hundred forty-one consecutive patients who underwent radical lymph node dissection (RLND) either in the groin or the axilla owing to malignant melanoma were followed up for a median period of 41 months. INTERVENTIONS: All of the 141 patients received either elective or therapeutic RLND. Their medical records were analyzed for demographic data, disease history, previous treatments, recurrence patterns, and survival. MAIN OUTCOME MEASURES: Patterns of first recurrence after RLND and survival. RESULTS: Radical lymph node dissection was performed on 148 lymph node basins (141 patients; 86 axillae and 62 groins). Nineteen patients (13%) received previous open interventions in the lymph node basin (tampering) other than radical dissection. Radical lymph node dissection was performed prophylactically in 38 basins (26%), for palpable disease in 75 (51%), and for a positive sentinel node in 35 (24%). There were 74 failures (52%) of RLND: 51 patients (70%) with systemic disease, 12 (16%) with recurrence in the surgical field, 9 (11%) with in-transit metastases, and 2 (3%) with local recurrence. On multivariate analysis, the only significant predictors of recurrence after RLND were Breslow thickness of greater than 4 mm (P = .02), tampering (P = .01), and lymph node capsular invasion (P = .001). Tampering was the only independent prognosticator of failure in the surgical field, as tampering was noted in 10 (83%) of 12 patients with failure in the surgical field as compared with 6 (10%) of 62 patients with other types of first failures (P<.001). This effect did not translate into a survival difference (P = .54). Failure in the surgical field was not detected in any of the patients who underwent sentinel lymph node biopsy. CONCLUSIONS: Previous interventions (excisional biopsy, incomplete dissection) in the regional basin that drain a melanoma site prior to definitive surgical procedures significantly increase the risk of melanoma recurrence in the surgical field, and they should be avoided. Fine-needle aspiration and sentinel node biopsy, performed with strict surgical oncologic techniques, are safe with regard to failure in the surgical field.

Are solitary breast papillomas entirely benign?

HYPOTHESIS: Solitary breast papillomas are potentially malignant and are associated with a higher risk of breast cancer. DESIGN: Retrospective review of all pathological reports containing breast papilloma (1983-2000) and review of selected specimens. SETTING: Tertiary, referral, university-affiliated medical center. PARTICIPANTS: Ninety-five women with a breast specimen containing a papilloma or papillomatosis. Patients with overt papillary carcinoma without papilloma were excluded. INTERVENTION: All histopathological characteristics of the papilloma and the surrounding tissue were noted. The incidence of malignant and other proliferative histopathological findings were analyzed, comparing solitary ductal papilloma cases to multiple papilloma cases. The Fisher exact test and chi2 test were applied for statistical analysis. MAIN OUTCOME MEASURES: Surgical removal of solitary ductal papillomas should include margins wide enough to secure removal of any proliferative tissue within or around the papilloma and to enable thorough evaluation of the risk for future breast cancer. RESULTS: Solitary papillomas were associated with breast carcinoma in 7 patients (10%) in this series. An additional 9% (n = 6) presented with invasive or noninvasive carcinoma within the papilloma. Atypical papilloma was noted in 6% of patients (n = 4). The risk of associated malignancy was not significantly different between solitary ductal papilloma and multiple papilloma. CONCLUSION: Increased risk of breast cancer is associated with all forms of papilloma.

Lymphatic drainage to the popliteal basin in distal lower extremity malignant melanoma.

UNLABELLED: Hypotheses Melanoma of the distal lower extremity may drain to the popliteal basin. Drainage pathways and retrieval of the popliteal sentinel nodes may affect patient outcome. DESIGN: Retrospective analysis of popliteal involvement in patients with stage IB or higher melanoma, operated on from August 1, 1993, to July 31, 2003. SETTING: Tertiary referral, university-affiliated medical center. PATIENTS: One hundred six melanoma patients who underwent combined lymphoscintigraphy and blue dye-guided sentinel node biopsy, radical popliteal dissection, or both. MAIN OUTCOME MEASURES: Incidence and patterns of drainage to popliteal nodes; effect on staging and outcome. RESULTS: Lymphoscintigraphy (n = 8) and physical examination (n = 2) identified 10 cases (9%) of draining to the popliteal basin, with concurrent drainage to the groin. Three distinct drainage patterns were identified, with different popliteal node locations. Seven of 8 popliteal sentinel nodes were retrieved, 1 of which was metastatic with no groin metastasis. Two patients had synchronous palpable popliteal and groin metastases and underwent radical groin and popliteal dissection. All 3 patients with popliteal metastases relapsed early with synchronous systemic and in-transit disease. One of 7 patients with negative sentinel nodes is alive with in-transit disease; all others are disease free. CONCLUSIONS: According to this series, the popliteal basin is the site of first drainage in about 9% of patients, with concurrent drainage to the groin. The 3 distinct patterns of drainage to the popliteal region and the presence of isolated popliteal metastases may affect the surgical treatment. Therefore, drainage to popliteal sentinel nodes and the pattern of this drainage should be noted in all distal lower extremity melanomas.

Risk factors for lymph node metastases in breast ductal carcinoma in situ with minimal invasive component.

HYPOTHESIS: Clinical and pathological variables may be predictors of axillary dissemination in T1mic and T1a breast carcinoma. DESIGN: Retrospective medical chart review. SETTING: University-affiliated tertiary referral center. PATIENTS: All patients diagnosed as having ductal carcinoma in situ (DCIS) with microinvasion between January 1, 1988, and December 30, 1998. MAIN OUTCOME MEASURES: Pathology slides were reviewed according to the 1997 Cancer Staging Manual put forth by the American Joint Committee on Cancer. The number of involved ducts was noted. Patients with no invasive component or invasive components larger than 5 mm were excluded. Pathological and clinical variables were analyzed for their effect on axillary lymph node metastases. RESULTS: The study group included 57 women aged 37 to 71 years (median, 60 years), 37 with T1mic disease and 20 with T1a. Modified radical mastectomy was performed in 29 patients (18 with T1mic and 11 with T1a) and breast-preserving surgery in 28 (19 with T1mic and 9 with T1a). Forty-three patients (28 with T1mic and 15 with T1a) underwent axillary lymph node dissection. Axillary involvement was detected in 3 patients in each group. Forty-seven patients received adjuvant therapy (radiotherapy alone, or with hormones or chemotherapy). Follow-up was 3 to 120 months (median, 40 months). One patient was unavailable for follow-up, another died of disseminated disease, and a third developed contralateral primary carcinoma. Comedo DCIS (P<.03) and the number of DCIS-involved ducts (P<.002) in the T1mic group, and nuclear grade 3 (P<.001) in both groups, were independent significant predictors of axillary metastases. CONCLUSIONS: The significant rate of axillary metastases in T1a and T1mic breast tumors makes axillary staging a must. High nuclear grade, comedo DCIS, and high number of DCIS-involved ducts may predict axillary metastasis and should be considered when axillary dissection is done selectively.

Older age and second skin cancer as prognostic factors in localized malignant melanomas.

Of the numerous prognostic factors for patients with localized malignant melanoma (LMM), none is superior to the simple parameter of tumor thickness. The aim of the present study was to better define prognostic factors for this disease. Between January 1992 and December 1994, 188 consecutive patients with LMM were treated at the Rabin Medical Center. Patient and tumor characteristics were retrospectively examined as potential prognostic factors. Patients (n=173) who had had at least two-year follow-up were included in the overall survival (OS) analysis, and 159 patients for whom accurate data on recurrent disease were available were included in the disease-free survival (DFS) analysis. At a median follow-up of 85 months (range 24-114), 48 patients (30%) had recurrent disease which resulted in death in 35 (20%). The five-year OS and DFS rates for the entire group were 82 and 72%, respectively. On univariate analysis, female gender, age younger than 75 years, metachronous or synchronous second skin cancer (including melanoma), light skin color, tumor thickness and TNM stage were predictive of both OS and DFS. Tumor location and ulceration, correlated with only one endpoint, OS or DFS, respectively. On multivariate analysis, three factors retained statistical significance with regard to both OS and DFS: tumor thickness (p=0.000 for both), second skin cancer (p=0.02 for both), and age (p=0.04 for both). Alongside the well-established predictive factor of tumor thickness in LMM, older age and the presence of a second skin cancer also have prognostic significance. The prognostic importance of the latter is reported here for the first time.