Editorial Comment: Recent advances in our understanding of lung cancer visceral pleural invasion and other forms of minimal invasion: implications for the next TNM classification

Abstract

Involvement of the visceral pleura (VP) has long been recognized as an adverse prognostic factor in non-small-cell lung carcinoma (NSCLC). It was shown more recently with the aid of elastic stains that a tumour invading beyond the VP elastic layer has the same prognostic impact as a tumour extending to the VP surface [1]. Based on this and other data, the 7th edition TNM adopted a proposal to classify VP invasion (VPI) as PL0 if the tumour does not invade past the elastic layer, as PL1 if it invades past the elastic layer, PL2 if it invades to the pleural surface and PL3 if it invades to the parietal pleura with the recommendation to classify PL0 as T1, PL1 and PL2 as T2 and PL3 as T3 [1]. However, relatively little progress has been made towards understanding the prognostic significance of a tumour invading an adjacent lobe. Based on limited data, the TNM classification, 7th edition, recommended tumours with direct invasion of an adjacent lobe across the fissure or by direct extension at a point where the fissure is deficient be categorized as T2a unless other T factors result in a higher T category [2]. This recommendation left it unresolved whether the behaviour of tumours extending into an adjacent lobe along an incomplete fissure is truly akin to that of tumours that invade through the fissural pleura into an adjacent lobe, and for that matter, to that of tumours that involve the VP away from a fissure. In this issue of the journal, Ohtaki et al. [3] provide illumination on this topic. For T1 and T2a-sized (≤5 cm) tumours, invasion across a fissure into an adjacent lobe conferred a prognosis similar to T2b, whereas direct involvement of an adjacent lobe in an area of incomplete fissure had no effect on survival (Table 1). These data suggest that the invasion of the VP at the interlobar fissure, which in fact consists of two layers of VP, rather than actual involvement of the adjacent lobe, drives the poorer prognosis of tumours with adjacent lobe invasion, presumably as a result of gaining access to the rich network of lymphatics and capillaries that invest the VP. An important aspect of the Ohtaki et al. study is the careful attention given to macroscopical assessment and histological sampling of the pathology specimen for tumour invasion into the adjacent lobe and use of elastic stains [3]. These findings largely corroborate the results of two other recent reports on adjacent lobe invasion, one of which showed survival being slightly worse for T1 and T2-sized (≤7 cm) tumours that invade through the interlobar fissure than those with VPI away from a fissure (Table 1) [4]. The other demonstrated a 10–15% reduction in 5-year survival in Stage I tumours that extend across an interlobar fissure (P = 0.037) [5]. Discussion of the valuable study by Ohtaki et al. provides an opportunity to review the recently accumulated T factor data on VPI as well as invasion of the hilar region. Multiple studies have demonstrated the prognostic significance of VPI as defined by the 7th edition TNM, according to univariate analysis (Table 2) [6], but less often by multivariate analysis (Table 2) [6]. However, several recent reports question whether PL1 is prognostically equivalent to PL2. One found that survival among patients with PL2 was significantly worse than those with PL1 [7]. In another study, PL2 but not PL1 adversely affected patient survival, but only cases with pleural retraction were analysed (Table 2). Another study of Stage I cancers attempted to separate two VP elastic layers (internal—VPIL and external—VPEL), which is not recommended by the 7th edition TNM classification. This study failed to show any difference in survival according to VPI by this approach [8]. However, these authors used overall survival (OS), which may not be as meaningful as disease-free survival (DFS) or recurrence-free probability (RFP) in Stage I NSCLC since these patients often die of other causes. Currently, the T category of tumours >3 cm is not altered by the presence of VPI [2]. The results of several recent studies suggest that raising the T category based on the presence of VPI is appropriate not only for ≤3 cm tumours, but also those that are T2 by size (Table 1). In one of the studies, 3.1–5 cm (T2a by size) tumours with VPI had survival equivalent to T2b tumours [6], while two studies suggest that survival among 5.1–7 cm (T2b by size) tumours with VPI is comparable with that of T3 tumours [6, 9]. Another study showed comparably poor survival among T2-sized tumours with VPI and T3 tumours [10]. The hilar soft tissue represents the ’no man’s land’ of lung cancer staging. The pleural reflection is incomplete in the hilar