Pathologic heterogeneity of lung adenocarcinomas: A novel pathologic index predicts survival

Abstract

Although the most predominant subtype of invasive lung adenocarcinoma has been reported to have clinical significance, a major limitation of this concept is that most tumors are mixed-subtype. Therefore, we aimed to determine the individual prognostic significance of each subtype and also attempted to establish a pathologic index that reflects the pathologic subtypes and overall heterogeneity of lung adenocarcinomas and evaluated its prognostic significance. The individual prognostic impact of each subtype was assessed from the development cohort using the disease-free survival (DFS) curve of a previous large-scale study. Hazard ratios (HRs) from the development cohort were 1, 1.025, 1.059, 1.495, and 1.160 for the lepidic, acinar, papillary, micropapillary, and solid pattern subtype, respectively. Based on the calculated HR of each subtype, four indices representing pathologic heterogeneity were developed. The first and second indices were defined as the sum of the proportions of each subtype multiplied by their HRs, with the addition of either entropy or Gini coefficient, respectively. The third index was calculated as the sum of all subtype percentages multiplied by their HRs. To emphasize heterogeneity, the fourth index was defined as the simple arithmetic sum of the scores of the subtypes multiplied by their HRs. Each subtype was assigned a score of 0 if the subtype was absent and a score of 1 if the subtype was present in a binary fashion. We applied these four pathologic indices to a validation group of 148 patients with comprehensive histologic subtyping for completely resected lung adenocarcinomas. DFS curves were plotted and predictive ability of each pathologic index was evaluated. Among the four pathologic indices, only pathologic index 3 enabled significant patient stratification in the validation cohort according to DFS (P = 0.004) and showed the highest Harrell's C index of 0.691 of all four pathologic indices. In conclusion, we estimated the HR of each subtype and generated four pathologic indices that reflect heterogeneity. One of these, index 3, the pathologic heterogeneity index based on the sum of all subtype percentages multiplied by their HR, possesses good prognostic ability for predicting survival in patients with lung adenocarcinoma.

Prognostic impact of lung adenocarcinoma second predominant pattern from a large European database

“…The real impact of the second predominant pattern have been seldom discussed in the literature with inconsistent results. Before the latest revision of the adenocarcinoma classification, Sica et al, 4 evaluated the possible impact of second predominant pattern in mixed type adenocarcinomas; similarly, Lee et al, 15 elaborated four mathematical indexes based on the HR derived by a control cohort to calculate the different potential contribution to prognosis given by different adenocarcinoma subtypes. Among these four indexes, only one was verified to be a significant prognostic factor together with TNM at multivariate analysis.…”

Micropapillary Adenocarcinoma of the lung: Recent updates and literature review

“…One of their indices enabled signifi cant patient stratifi cation in the validation cohort according to DFS, and showed the highest Harrell's C index (a measure of how well binary outcomes in a logistic regression model fi t). This novel pathologic index refl ects tumor heterogeneity in lung adenocarcinomas and has good prognostic ability to predict survival [41]. In contrast, MMP9 levels were low in acinar and papillary subtypes, and even lower in the lepidic subtype.…”

“…Positive podoplanin was also associated with decreased DFS [39]. Based on the HR of each subtype, four indices were developed [41]. The validation group consisted of 148 patients with completely resected adenocarcinomas.…”

Progress in the Management of Early-Stage Non–Small Cell Lung Cancer in 2017

“…[40][41][42][43] The inclusion of a number of additional histopathologic features to further stratify risk of recurrence has been proposed such as secondary patterns, mitotic counts, nuclear grade, and spread through air spaces (STAS). [44][45][46][47] STAS is defined as discrete cell clusters, within air spaces in the surrounding lung beyond the tumor edge. STAS has been shown to be associated with recurrence in limited resection of early-stage lung cancer, and to be associated with high-grade histologic patterns such as solid and micropapillary.…”