Phase 3 trials of neoadjuvant, perioperative, and adjuvant immune checkpoint inhibitors combined with chemotherapy (ICI-CT) in resectable early-stage NSCLC (eNSCLC) have reported that all three approaches confer an event-free or disease-free survival benefit over CT alone, with acceptable safety profiles. All three strategies are approved standards of care for eNSCLC. This review provides a detailed analysis of these phase 3 ICI-CT trials and addresses the considerations regarding the selection of each approach, including protocol schema and baseline patient and tumor differences, preoperative staging, surgical outcomes, efficacy end points, safety, treatment disposition, and the programmed death-ligand 1 (PD-L1) efficacy biomarker. The differences between regimens and study populations among these ICI-CT trials hamper cross-trial comparisons and highlight the need for head-to-head trials. Patients achieving pathologic complete response with neoadjuvant ICI-CT have better survival outcomes irrespective of subsequent treatment, but the optimal number of preoperative ICI-CT cycles needed to achieve pathologic complete response has not been defined. The choice between a neoadjuvant or perioperative versus adjuvant treatment approach involves a risk-benefit assessment of the potential for preoperative attrition to surgery, postoperative attrition to ICI-CT, and the anticipated toxicity profile. Current limitations of invasive lymph node staging mean that adjuvant ICI remains an important treatment strategy, but preoperative node staging is imperative. Future studies that identify the safety and toxicity contributions of each treatment phase in perioperative trials will confirm whether a pre- or postoperative ICI approach is superior, whether there is added benefit to adjuvant after neoadjuvant ICI-CT, and which patients will benefit the most from each approach.