Novel imprint cytological classification is correlated with tumor spread through air spaces in lung adenocarcinoma.

Kimura, Toru; Nakamura, Harumi; Omura, Akiisa; Ike, Akihiro; Hiroshima, Takashi; Maniwa, Tomohiro; Honma, Keiichiro; Higashiyama, Masahiko; Okami, Jiro

OBJECTIVES Spread through air spaces (STAS) is a risk factor for local recurrence after sublobar resection in lung cancer patients. We recently proposed the novel Nakayama-Higashiyama imprint cytological classification (N-H classification) based on small-sized lung adenocarcinoma surgical specimens, which correlated with histological patterns and nodal involvement. This study aimed to evaluate the correlation between STAS and the N-H classification and to validate the N-H classification as an intraoperative predictor of the presence of STAS. MATERIALS AND METHODS We retrospectively analyzed 164 intraoperative imprint cytologies and their paired histologic specimens from patients undergoing surgical resection for lung adenocarcinoma in our institute in 2017-2019. Using the NH classification, imprint cytological findings were classified into 5 groups (Groups I to V) based on cell cluster shape, cell and nucleus size, and the existence of necrosis. We examined the characteristics of imprint cytology and STAS in the resected tissues and analyzed the relationship between them. RESULTS Tumor STAS was observed in 29 (17.7 %) cases. The presence of STAS was significantly associated with the NH classification (P < 0.0001). STAS was present in 6 of 57 cases (10.5 %) in NH classification Group II, 11 of 42 cases (26.2 %) in Group III, and 12 of 28 cases (42.9 %) in Group IV/V; STAS was not observed in any case in Group I. Logistic regression analysis revealed that tumors with a ground glass opacity rate of <50 % on computed tomography (P = 0.00867) and Groups III-V of the NH classification (P = 0.00201) were significant independent predictors for STAS. CONCLUSION Intraoperative imprint cytology with the N-H classification for lung adenocarcinoma is well correlated with the STAS status of the tumor and might have applications as an intraoperative predictive marker of STAS. This classification may be useful for intraoperative detection of STAS and in the decision-making process for the surgical procedure.