Recently, the stromal invasion grading system was proposed for small adenocarcinomas of r2.0 cm. The system is based on the presence or absence of a fibrotic focus, and the extent of invasion into the fibrotic focus. Although stromal invasion grading system closely correlated with the prognosis, stromal invasion grade 3, representing stromal invasion into the center of a fibrotic focus, was the largest group of tumors and showed variable prognosis. In this study, we investigated whether stromal invasion grading system could be applied to and validated in pT1 adenocarcinomas as the TNM classification is the most universally used system. Furthermore, we investigated whether stromal invasion grade 3 cases could be subclassified according to the presence and absence of micropapillary pattern. The study included 120 cases of pT1 lung adenocarcinomas, of which 81 (68%) cases were stromal invasion grade 3. Micropapillary pattern was positive in 80% of grade 3 cases. For stromal invasion grade 3 cases, the 5-year survival rate of patients with micropapillary patternpositive carcinomas was 63%, which was significantly worse than 94% of those with micropapillary patternnegative carcinomas (P ¼ 0.0196). The latter was very close to that for patients with stromal invasion grade 0-2 (95%). Moreover, small cluster invasion was observed at sites of stromal invasion significantly more often in micropapillary pattern-positive cases than negative cases. Thus, the stromal invasion grading system is reproducible and correlates with prognosis even in pT1 lung adenocarcinomas. Moreover, among patients with stromal invasion grade 3 carcinomas, favorable prognosis is noted in micropapillary pattern-negative cases. The micropapillary pattern subclassification provides an advantage to the stromal invasion grading system and reconfirms the importance of micropapillary pattern as a prognostic marker. Our study is the first to point to the possible association of micropapillary pattern-positive carcinomas and small cluster invasion.