Prognostic impact of intratumoural microvascular invasion and microlymphatic permeation on node-negative non-small-cell lung cancer: which indicator is the stronger prognostic factor?

OBJECTIVES Microvascular invasion and microlymphatic permeation are indicators of microscopic tumour invasion into small vessels and have been considered to be powerful prognostic indicators for non-small-cell lung cancer (NSCLC). Several studies have suggested that these should be included in the TNM classification, but, there have been conflicting results regarding the prognostic impact of microvascular invasion and microlymphatic permeation. The aim of the current study was to clarify the prognostic impact of microvascular invasion and microlymphatic permeation on resected node-negative NSCLC by comparative analyses. METHODS We reviewed the data of 1039 consecutive patients with pathological size-based stage T1a-3N0M0 NSCLC who underwent lobectomy or greater resection between 1993 and 2005. The median follow-up period was 108 months. Microvascular invasion and microlymphatic permeation were identified by the Victoria blue-van Gieson staining. The overall survival was then analysed. RESULTS Microvascular invasion and microlymphatic permeation were observed in 358 (34.5%) and 205 (19.7%) of patients, respectively. Both microvascular invasion and microlymphatic permeation were more prevalent in non-adenocarcinoma and larger-sized tumours. The 5-year overall survival rate of the microvascular invasion-positive group and microlymphatic permeation-positive group were 69.2 and 84.6%, respectively, and the difference was statistically significant (P = 0.002). On multivariate analyses, microvascular invasion, but not microlymphatic permeation, was an independent prognostic factor (microvascular invasion, hazard ratio [HR] 1.648, P = 0.001; microlymphatic permeation, HR 1.138, P = 0.588). The 5-year overall survival rate of either the microvascular invasion- or microlymphatic permeation-positive T1a-b group was significantly lower than that of the corresponding double-negative (dn) T1a-b group (dnT1a-b, 93.7%; microvascular invasion-positive T1a-b, 85.2%, P < 0.001; microlymphatic permeation-positive T1a-b, 85.4%, P = 0.014), and overlapped to that of the dnT2a group (84.8%). However, in the T2a-b group, only microvascular invasion-positive T2a-b patients showed significantly lower overall survival than dnT2a-b patients, and their overall survival overlapped that of dnT3 patients (dn T2a-b, 83.5%; microvascular invasion-positive T2a-b, 60.6%, P < 0.001; dnT3, 53.8%; P = 0.316). The 5-year overall survival of microlymphatic permeation-positive T2a-b patients (86.2%) did not statistically differ from that of dnT2a-b patients (P = 0.856). CONCLUSIONS Microvascular invasion and microlymphatic permeation have different impact on survival, and microvascular invasion rather than microlymphatic permeation is a strong prognostic factor in resected node-negative NSCLC. Microvascular invasion and microlymphatic permeation should be examined separately by elastic staining.