O Straume1, L A Akslen. 1. Department of Pathology, Gade Institute, University of Bergen, Norway.
Abstract
BACKGROUND: The role of vascular invasion as an independent predictor of outcome in cutaneous melanoma is controversial, mostly because of a close relationship with tumor thickness. Therefore, the prevalence of vascular involvement and its prognostic impact has been studied in a series of patients with aggressive cutaneous melanoma. METHODS: One hundred and two patients with nodular malignant melanoma diagnosed between 1981 and 1989 were studied retrospectively. The vessels were stained with factor VIII antibody, and vascular invasion as well as other features of the primary tumor and essential clinical variables were related to time to recurrence and death due to melanoma using univariate survival analysis (product-limit method) and the multivariate Cox regression method. RESULTS: Definite vascular invasion was present in 15 cases (15%), and it was significantly associated with tumor thickness, histologic diameter, ulceration, and presence of lymph node metastases and distant spread at the time of diagnosis. In univariate analysis, vascular invasion was a significant predictor of patient survival (P = 0.006), and this variable also showed an independent prognostic impact in multivariate analysis (P = 0.02). Furthermore, anatomic site (P = 0.03), histologic diameter (P = 0.004), and Clark's level of invasion (P = 0.006) were independently associated with decreased patient survival. In the multivariate analysis of time to recurrence, anatomic site (P = 0.003), tumor thickness (P = 0.002), and histologic diameter (P = 0.003) were independent risk factors. CONCLUSIONS: Although vascular invasion was significantly associated with tumor thickness and histologic diameter of the primary tumor, it provided an independent prognostic information by multivariate analysis of patient survival, in addition to anatomic site and indicators of tumor size (histologic diameter and level of invasion). Our findings further indicate that the prognostic importance of vascular invasion is associated with both lymphatic and hematogeneous spread, although the latter is probably more important.
BACKGROUND: The role of vascular invasion as an independent predictor of outcome in cutaneous melanoma is controversial, mostly because of a close relationship with tumor thickness. Therefore, the prevalence of vascular involvement and its prognostic impact has been studied in a series of patients with aggressive cutaneous melanoma. METHODS: One hundred and two patients with nodular malignant melanoma diagnosed between 1981 and 1989 were studied retrospectively. The vessels were stained with factor VIII antibody, and vascular invasion as well as other features of the primary tumor and essential clinical variables were related to time to recurrence and death due to melanoma using univariate survival analysis (product-limit method) and the multivariate Cox regression method. RESULTS: Definite vascular invasion was present in 15 cases (15%), and it was significantly associated with tumor thickness, histologic diameter, ulceration, and presence of lymph node metastases and distant spread at the time of diagnosis. In univariate analysis, vascular invasion was a significant predictor of patient survival (P = 0.006), and this variable also showed an independent prognostic impact in multivariate analysis (P = 0.02). Furthermore, anatomic site (P = 0.03), histologic diameter (P = 0.004), and Clark's level of invasion (P = 0.006) were independently associated with decreased patient survival. In the multivariate analysis of time to recurrence, anatomic site (P = 0.003), tumor thickness (P = 0.002), and histologic diameter (P = 0.003) were independent risk factors. CONCLUSIONS: Although vascular invasion was significantly associated with tumor thickness and histologic diameter of the primary tumor, it provided an independent prognostic information by multivariate analysis of patient survival, in addition to anatomic site and indicators of tumor size (histologic diameter and level of invasion). Our findings further indicate that the prognostic importance of vascular invasion is associated with both lymphatic and hematogeneous spread, although the latter is probably more important.
Authors: Monica Mannelqvist; Ingunn M Stefansson; Geir Bredholt; Trond Hellem Bø; Anne M Oyan; Inge Jonassen; Karl-Henning Kalland; Helga B Salvesen; Lars A Akslen Journal: Am J Pathol Date: 2011-02 Impact factor: 4.307
Authors: G Viale; A Giobbie-Hurder; B A Gusterson; E Maiorano; M G Mastropasqua; A Sonzogni; E Mallon; M Colleoni; M Castiglione-Gertsch; M M Regan; K N Price; R W Brown; R Golouh; D Crivellari; P Karlsson; C Öhlschlegel; R D Gelber; A Goldhirsch; A S Coates Journal: Ann Oncol Date: 2009-07-24 Impact factor: 32.976