In mid 2000s, key clinical trials reported the benefit of adjuvant chemotherapy in NSCLC. Platinum-based chemotherapy after curative resection has demonstrated to provide an absolute 5-year survival gain of 5% compared to surgery alone. Low-dose CT screening has increased the number of lung cancers detected at early stages which are potentially curable by surgical resection. Unlike advanced NSCLC, however, little progress has been made over the past decades in the treatment of early-stage patients for whom cure is most realizable. In recent years, most clinical trials in early-stage NSCLC focus on targeted agents and immune-checkpoint inhibitors which are effective therapies currently in use for metastatic disease. Adjuvant chemotherapy is generally recommended for patients with resected stage II-IIIA disease, but not recommended for patients with stage IA because of no proven benefit. Controversy exists for stage IB although guidelines suggest that adjuvant chemotherapy be considered for a subgroup of patients with high-risk features. Available evidence for stage IB that patients with larger (≥ 4cm) tumor may benefit from adjuvant chemotherapy came from a randomized trial with a relatively small sample size (CALGB 9633) and from subgroup analysis of other trials, which were underpowered to detect the small differences expected in survival. However, tumors ≥ 4cm were upstaged to stage IIA in the 8th edition of TNM staging system. Other high-risk factors in stage IB include vascular invasion, visceral pleural invasion, poor differentiation among others, mostly from retrospective studies. For adenocarcinoma, high-grade subtype (micropapillary and solid) emerged as a high-risk factor for the decision of adjuvant chemotherapy. While TNM staging remains the most important predictor of benefit from adjuvant therapy, clinical outcomes vary widely even among patients with the same stage. Evidence indicates that even stage I NSCLC comprised of heterogeneous diseases, which may lead to heterogenous prognosis. Therefore, predictive biomarkers are of vital importance to help identify patients who may benefit from adjuvant therapy, especially in stage I where the role of adjuvant chemotherapy is not apparent. Molecular biomarkers or signatures have been extensively investigated, but most were not validated. The most recent frontiers in this field have leveraged leading-edge technologies such as NGS and machinelearning, looking forward to the development of risk-stratification or prognosis-prediction models that integrate genomic and clinicopathologic factors. To ensure the cure of patients with the earliest stage NSCLC, future studies are required to investigate which subgroup of patients in stage I NSCLC should be selected for adjuvant therapy using state-of-the-art agents in welldesigned clinical trials powered with a large number of stage I patients.