Adjuvant epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (TKIs) for the treatment of people with resected stage I to III non-small-cell lung cancer and EGFR mutation.

Background: Postoperative adjuvant epidermal growth factor receptor (EGFR) inhibitor osimertinib is the standard care for stage IB-IIIB non-small-cell lung cancer (NSCLC) with EGFR exon 19 deletions or exon 21 L858R mutation, following complete tumour resection, with or without prior platinum-based adjuvant chemotherapy. However, the role of EGFR tyrosine kinase inhibitors (TKIs) in this setting is debated, particularly concerning long-term curative effects versus recurrence delay. Uncertainties persist around treatment duration, harms, and effectiveness across disease stages, prior chemotherapy, or EGFR-sensitising mutation types.

Objective: To assess the effectiveness and harms of adjuvant EGFR tyrosine kinase inhibitors (TKIs) in people with resected stage I to III non-small-cell lung cancer (NSCLC) harbouring an activating EGFR mutation.

Methods: We searched major databases (CENTRAL, MEDLINE, Embase) to 9 December 2024, along with conference proceedings (from 2019) and clinical trial registries. Methods: We included randomised controlled trials (RCTs) reporting benefits or harms of adjuvant EGFR TKIs in adults with resected stage I-III NSCLC. Trials compared EGFR TKIs with platinum-based chemotherapy, placebo/best supportive care (BSC), or second-and/or third-generation EGFR TKIs versus first- and/or second-generation EGFR TKIs. Participants were adults with histologically confirmed stage I-III NSCLC. Methods: Three review authors independently assessed search results, resolving disagreements with a fourth author. Primary outcomes were overall survival (OS), disease-free survival (DFS), and adverse events (AEs); secondary outcomes included health-related quality of life (HRQoL), relapse risk during drug-off time, and brain relapse risk. We conducted meta-analyses using random-effects and fixed-effect models with hazard ratios (HRs) or risk ratios (RRs) and 95% confidence intervals (CIs). We assessed heterogeneity with the I² statistic.

Results: We included nine RCTs involving 2603 participants, and identified six ongoing trials. Five trials compared EGFR TKIs with placebo/BSC, and four compared them with chemotherapy. We found no trials comparing second-and/or third-generation to first- and/or second-generation EGFR TKIs. Six trials had low selection bias risk; most had unclear or high risk for detection or performance bias; and four were high risk for other biases. The certainty of the evidence (GRADE) ranged from moderate to very low, depending on the outcome. First-, second-, and/or third-generation EGFR TKIs versus placebo/BSC EGFR TKIs probably improve overall survival compared to placebo/BSC (HR 0.54, 95% CI 0.40 to 0.73; 3 studies, 864 participants; moderate-certainty evidence). TKIs may improve disease-free survival compared to placebo/BSC, but the evidence is very uncertain (HR 0.34, 95% CI 0.28 to 0.41; 5 studies, 1153 participants). We are uncertain if there is a difference between groups in serious adverse events (≥ grade 3) as the evidence is very uncertain, with wide confidence intervals spanning both potential harm and no effect (RR 2.52, 95% CI 0.44 to 14.37; 4 studies, 1134 participants). Mild-to-moderate adverse events (grades 1 and 2) may be more frequent with EGFR TKIs compared to placebo/BSC, but the evidence is very uncertain (RR 1.57, 95% CI 1.08 to 2.29; 4 studies, 1134 participants). One study assessed HRQoL, with no clinically meaningful decline compared to placebo/BSC (592 participants; moderate-certainty evidence). First-, second-, and/or third-generation EGFR TKIs versus chemotherapy Overall survival was similar between EGFR TKIs and chemotherapy (HR 0.79, 95% CI 0.52 to 1.18; 4 studies, 878 participants; moderate-certainty evidence). TKIs may have improved disease-free survival compared to chemotherapy (HR 0.54, 95% CI 0.35 to 0.83; 4 studies, 878 participants; low-certainty evidence). TKIs may have reduced serious adverse events (≥ grade 3) compared to chemotherapy (RR 0.31, 95% CI 0.18 to 0.52; 4 studies, 811 participants; low-certainty evidence). TKIs may have increased mild-to-moderate adverse events (grades 1 and 2) (RR 2.13, 95% CI 1.20 to 3.78; 4 studies, 811 participants; low-certainty evidence). Two studies assessed HRQoL, showing no clear difference compared to chemotherapy, as assessed with the Functional Assessment of Cancer Therapy-Lung instrument (2 studies, 399 participants) and the Lung Cancer Symptom Scale (2 studies, 400 participants), both with moderate-certainty evidence.

Conclusions: Adjuvant EGFR TKIs may improve disease-free survival compared to both placebo/BSC and chemotherapy. There is moderate-certainty evidence that EGFR TKIs increase overall survival compared to placebo/BSC. However, they likely result in little to no difference in overall survival compared to chemotherapy. We could not rule out a potential survival benefit of adjuvant chemotherapy in people with EGFR-mutant NSCLC. Approximately 50% of participants experienced relapse or death within one year of stopping TKI therapy, indicating that the disease-free survival benefit may wane after withdrawal. This raises the possibility that prolonged adjuvant TKI therapy could be associated with improved long-term outcomes, although further research is needed to clarify this.