Objective To elucidate the correlation between radiological tumor size (RTS) and pathological tumor size (PTS), and to evaluate the accuracy of clinical T staging. Methods Data on patients who underwent complete resection between September 2018 and June 2019 were retrospectively collected. The correlation between RTS and PTS was analyzed by and we assessed the agreement between clinical and pathologic T staging. Results Finally, 1 880 patients were included. There were 778 males and 1 102 females at average age of 57±11 years. In the entire cohort, the RTS and PTS was 19.1±13.5 mm and 17.7±14.0 mm, respectively (P<0.001). The RTS and PTS showed a strong linear correlation with the Pearson’s correlation coefficient calculated as 0.897. The mean RTS was significantly larger than PTS (P<0.001) in tumors≤3 cm, but significantly smaller in tumors>4 cm. The overall concordance rate between clinical and pathological T staging was 65.6%. Clinical staging failed to detect T4 disease in 29.4% (5/17) of patients. Male patients and the presence of cavities within nodules were independent significant factors leading to inaccurate clinical T staging. Conclusions The correlation between the tumor sizes measured on thin-section computed tomography and pathologic specimens varies with the real tumor size. Methods and techniques for improving clinical T staging accuracy is in urgent need.

Along with the popularity of low-dose computed tomography lung cancer screening, an increasing number of early-stage lung cancers are detected. Radical lobectomy with systematic nodal dissection (SND) remains the standard-of-care for operable lung cancer patients. However, whether SND should be performed on non-metastatic lymph nodes remains controversy. Unnecessary lymph node dissection can increase the difficulty of surgery while also causing additional surgical damage. In addition, non-metastatic lymph nodes have been recently reported to play a key role in immunotherapy. How to reduce the surgical damage of mediastinal lymph node dissection for early-stage lung cancer patients is pivotal for modern concept of "minimally invasive surgery for lung cancer 3.0". The selective mediastinal lymph node dissection strategy aims to dissect lymph nodes with tumor metastasis while preserving normal mediastinal lymph nodes. Previous studies have shown that combination of specific tumor segment site, radiology and intraoperative frozen pathology characteristics can accurately predict the pattern of mediastinal lymph node metastasis. The personalized selective mediastinal lymph node dissection strategy formed from this has been successfully validated in a recent prospective clinical trial, providing an important basis for early-stage lung cancer patients to receive more personalized selective lymph node dissection with "precision surgery" strategies.