Proposal of a revised International Association for the Study of Lung Cancer grading system in pulmonary non-mucinous adenocarcinoma: The importance of the lepidic proportion

Objectives: We aimed to measure the validity of the International Association for the Study of Lung Cancer (IASLC) grading system in Korean patients and propose a modification for an increase of its predictability, especially in grade 2 patients.Materials and Methods: From 2012 to 2017, histopathologic characteristics of 1358 patients with invasive pul-monary adenocarcinoma (stage I-III) from two institutions were retrospectively reviewed and re-classified ac-cording to the IASLC grading system. Considering the amount of the lepidic proportion, the validity of the revised model (Lepidic-10), derived from the training cohort (hospital A), was measured using the validation cohort (hospital B). Its predictability was compared to that of the IASLC system.Results: Of the 1358 patients, 259 had a recurrence, and 189 died during follow-up. The Harrell's concordance index and area under the curve of the IASLC system were 0.685 and 0.699 for recurrence-free survival (RFS) and 0.669 and 0.679 for death, respectively. From the training cohort, the IASLC grade 2 patients were divided into grades 2a and 2b (Lepidic-10 model) with a 10 % lepidic pattern. This new model further distinguished patients in both institutions that had better performance than the IASLC grading (Hospital A, p < 0.001 for RFS and death; Hospital B, p = 0.0215 for RFS, p = 0.0429 for death).Conclusion: The IASLC grading system was easily applicable; its clinical use in predicting the prognosis of Korean patients with pulmonary adenocarcinoma was validated. Furthermore, the introduction of the lepidic proportion as an additional criterion to differentiate grade 2 patients improved its predictability.

Automated summary

The study aimed to validate the effectiveness of the IASLC grading system in Korean patients and propose a modification to improve its predictive ability, especially for grade 2 patients. By reviewing the histopathologic characteristics of 1358 patients with invasive pulmonary adenocarcinoma, the validity of the revised model (Lepidic-10) was evaluated and compared to the IASLC system. The results showed that the Lepidic-10 model had better performance in predicting survival and recurrence time in patients from two hospitals compared to the IASLC grading.