The Gleason grading system, currently the most powerful prognostic predictor of prostate cancer, is based solely on the tumor’s histological architecture and has high inter-observer variability. We propose an automated Gleason scoring system based on deep neural networks for diagnosis of prostate core needle biopsy samples. To verify its efficacy, the system was trained using 1133 cases of prostate core needle biopsy samples and validated on 700 cases. Further, system-based diagnosis results were compared with reference standards derived from three certified pathologists. In addition, the system’s ability to quantify cancer in terms of tumor length was also evaluated via comparison with pathologist-based measurements. The results showed a substantial diagnostic concordance between the system-grade group classification and the reference standard (0.907 quadratic-weighted Cohen’s kappa coefficient). The system tumor length measurements were also notably closer to the reference standard (correlation coefficient, R = 0.97) than the original hospital diagnoses (R = 0.90). We expect this system to assist pathologists to reduce the probability of over- or under-diagnosis by providing pathologist-level second opinions on the Gleason score when diagnosing prostate biopsy, and to support research on prostate cancer treatment and prognosis by providing reproducible diagnosis based on the consistent standards.

Objectives Ground-glass opacity (GGO) on computed tomography is associated with prognosis in early-stage non-small cell lung cancer (NSCLC) patients. However, the stratification of the prognostic value of GGO is controversial. We aimed to evaluate clinicopathologic characteristics of early-stage NSCLC based on the consolidation-to-tumor ratio (CTR), conduct multi-pronged analysis, and stratify prognosis accordingly. Methods We retrospectively investigated 944 patients with clinical stage IA NSCLC, who underwent curative-intent lung resection between August 2018 and January 2020. The CTR was measured and used to categorize patients into six groups (1, 0%; 2, 0–25%; 3, 25–50%; 4, 50–75%; 5, 75–100%; and 6, 100%). Results Pathologic nodal upstaging was found in 1.8% (group 4), 9.0% (group 5), and 17.4% (group 6), respectively. The proportion of patients with a high grade of tumor-infiltrating lymphocytes tended to decrease as the CTR increased. In a subtype analysis of patients with adenocarcinoma, all of the patients with predominant micro-papillary patterns were in the CTR > 50% groups, and most of the patients with predominant solid patterns were in group 6 (47/50, 94%). The multivariate analysis demonstrated that CTR 75–100% (hazard ratio [HR], 3.85; 95% confidence interval [CI], 1.58–9.36) and CTR 100% (HR, 5.58; 95% CI, 2.45–12.72) were independent prognostic factors for DFS, regardless of tumor size. Conclusion We demonstrated that the CTR could provide various noninvasive clinicopathological information. A CTR of more than 75% is the factor associated with a poor prognosis and should be considered when making therapeutic plans for patients with early-stage NSCLC.

Aims The prognostic role of EGFR mutations remains controversial. We aimed to evaluate the prognostic role of EGFR mutation in consideration of the IASLC histological grade in patients with resected early‐stage lung adenocarcinoma. Methods and results A total of 3297 patients with stages I–IIA resected lung adenocarcinoma who had had EGFR mutation tests between January 2014 and December 2019 at the Samsung Medical Center, Seoul, Korea were included. Recurrence‐free survival (RFS) was compared by EGFR mutation status (EGFR‐M+ versus EGFR‐WT) and IASLC histological grade (G1, G2 and G3). Cox proportional hazards models were used to estimate the adjusted HRs (aHRs) and 95% confidence intervals (CIs). Results Compared to the EGFR‐WT group, the EGFR‐M+ group had a significantly lower proportion of G3 tumour (16 versus 33%, P < 0.001). During a median follow‐up of 41.4 months, 376 patients experienced recurrence. After adjusting for histological grade, the aHR for recurrence comparing the EGFR‐M+ to the EGFR‐WT was 1.30 (95% CI = 1.04–1.62, P = 0.022). The EGFR‐M+ group had a significantly lower 5‐year RFS than the EGFR‐WT group among G3 patients (58.4 versus 71.5%, P < 0.001), but not among G1 and G2 patients. Conclusions EGFR mutation status was associated with a risk of recurrence after consideration of the IASLC histological grading, especially in G3 tumours. The results of this study would be useful for developing a new staging system and identifying a subset of patients who may benefit from adjuvant targeted therapy.