Investigators conducted a nonrandomized, noncomparative phase 2 study of perioperative ivonescimab in patients with resectable NSCLC. The protocol administered neoadjuvant ivonescimab either as monotherapy (cohort 1) or combined with chemotherapy (cohort 2), followed by surgical resection and adjuvant ivonescimab. The article presents cohort-level pathological response and safety outcomes. The source did not provide details on individual dosing schedules or on randomization procedures, because the study was explicitly nonrandomized; those specifics were not reported in the text.
Enrollment produced small and unequal cohort sizes. Cohort 1 comprised a limited number of patients; cohort 2 was larger. Investigators predefined the primary efficacy endpoint as major pathological response (MPR). For cohort 2 the statistical assessment compared the observed MPR against a historical control rate, using an 80% two-sided confidence interval to test whether the lower bound exceeded that benchmark.
Cohort 2 met the study's primary endpoint. Data indicated an MPR rate of 70.1% (47/67) in the combination cohort, with the lower bound of the 80% two-sided confidence interval above the historical 41% rate used as the comparator. The study team observed an MPR of 60.0% (6/10) in the ivonescimab monotherapy cohort, although the confidence interval around that estimate was wide.
The authors reported residual viable tumor (RVT) percentages among patients who achieved MPR. In cohort 1, 5 of 6 patients (83.3%) with MPR had RVT ≤5%. In cohort 2, 46 of 47 patients (97.9%) with MPR had RVT ≤5%. These numbers describe the depth of pathological response in responders, and the trial report provides them directly as outcome measures linked to neoadjuvant treatment effect.
The source did not provide additional long-term efficacy outcomes such as event-free survival (EFS) or overall survival (OS) within the presented text. Readers should note the absence of reported survival endpoints in the published abstract and main text excerpt provided here.
Safety data appear in the report with treatment-related adverse events (AEs) summarized by cohort. Grade 3 or higher treatment-related AEs occurred in 36.4% (4/11) of patients in cohort 1 and in 30.4% (24/79) in cohort 2. The authors noted there were no grade 3 or higher hemorrhages and no treatment-related deaths.
A single patient (1.3%) experienced a treatment-related delay to surgery. The study team also identified surgery-related AEs judged to be related to ivonescimab; these occurred in 10.0% (1/10) of cohort 1 and 10.4% (7/67) of cohort 2 patients, and no grade 3 or higher events were reported among those surgery-related events. The article therefore documents a tolerability profile that includes a measurable rate of higher-grade systemic toxicities but no catastrophic bleeding or treatment-attributed mortality in the cohorts described.
The authors placed their results in the context of established neoadjuvant immunochemotherapy trials for resectable NSCLC. They referenced prior randomized phase 3 studies in which perioperative PD-1 or PD-L1 inhibitors combined with chemotherapy improved pathological response rates compared with chemotherapy alone; those earlier trials reported MPR and pathological complete response (pCR) ranges that informed the historical control value used in this phase 2 analysis. The report states that cohort 2’s MPR exceeded the historical 41% comparator when evaluated by the prespecified confidence interval method.
This trial is not a head-to-head randomized comparison. The study team therefore relied on historical benchmarks to contextualize efficacy. The source did not present a direct statistical comparison with contemporaneous control arms within the same protocol.
The published report is clear about study design: nonrandomized and noncomparative. Small cohort sizes, particularly in cohort 1, produced wide confidence intervals for some estimates; the source reports these limits transparently. The paper did not include long-term survival results (EFS or OS) in the provided excerpt, so durability of the high MPR rates remains unreported in this text. Details about individual dosing regimens, the chemotherapy backbone used in cohort 2, and perioperative timing parameters were not reported in the source excerpt presented here and therefore were not summarized.
Outstanding questions left by the dataset include whether the high pathological response rates will translate into improved EFS or OS, how ivonescimab compares directly with approved PD-1/PD-L1–based perioperative regimens in randomized settings, and whether specific subgroups derive differential benefit. The trial registration number is NCT05247684 for reference to the full protocol and any subsequent updates that may address these unanswered clinical endpoints.