Neoadjuvant durvalumab plus radiation versus durvalumab alone in stages I-III non-small cell lung cancer: survival outcomes and molecular correlates of a randomized phase II trial

Abstract

We previously reported the results of a randomized phase II trial (NCT02904954) in patients with early-stage non-small cell lung cancer (NSCLC) who were treated with either two preoperative cycles of the anti-PD-L1 antibody durvalumab alone or combined with immunomodulatory doses of stereotactic radiation (DRT). The trial met its primary endpoint of major pathological response, which was significantly higher following DRT with no new safety signals. Here, we report on the prespecified secondary endpoint of disease-free survival (DFS) regardless of treatment assignment and the prespecified exploratory analysis of DFS in each arm of the trial. DFS at 2 and 3 years across patients in both arms of the trial were 73% (95% CI: 62.1-84.5) and 65% (95% CI: 52.5-76.9) respectively. For the exploratory endpoint of DFS in each arm of the trial, three-year DFS was 63% (95% CI: 46.0-80.4) in the durvalumab monotherapy arm compared to 67% (95% CI: 49.6-83.4) in the dual therapy arm. In addition, we report post hoc exploratory analysis of progression-free survival as well as molecular correlates of response and recurrence through high-plex immunophenotyping of sequentially collected peripheral blood and gene expression profiles from resected tumors in both treatment arms. Together, our results contribute to the evolving landscape of neoadjuvant treatment regimens for NSCLC and identify easily measurable potential biomarkers of response and recurrence.

Fig. 1. Disease-free survival in all randomized and surgically resected patients.

Disease-free survival for all randomized patients (n = 60) (a) and in each arm of the trial (b). DFS in patients in all surgically resected patients (n = 52) (c) and each arm (d). All panels, Kaplan-Meier survival method. Panels (b) and (d), log rank test p-values. Source data for all panels are provided as a Source Data file.

Fig. 2. Progression-free survival in all resected patients and stratified by treatment arm, stage and major pathological response.

Progression-free survival for all surgically resected patients (n = 52) (a) and in each arm of the trial (b). PFS in patients with clinical stages I/II (c) and clinical stage III (d) in each arm. PFS in all 52 patients with and without MPR (e). PFS in 34 patients without MPR in each arm of the trial (f). All panels, Kaplan-Meier survival method. Panels (b) through (f), log rank test p-values. Source data for all panels are provided as a Source Data file.

Fig. 3. Association of pretreatment circulating T cell populations with MPR (≤90% residual viable tumor cells).

A PaCMAP projection and FlowSOM clustering of CD4+ T cells from all PBMC samples. Heatmap overlays indicates relative expression for each phenotypic marker. Tnv = naïve T cell phenotype, T_reg = regulatory T cell phenotype, T_em = effector memory T cell phenotype, T_cm = central memory T cell phenotype. In all panels, patients are stratified by percent tumor regression: those with ≥90% tumor regression (defined clinically as major pathological response) and those with <90% tumor regression. Comparison between treatment arms shown in panels (B, D, G, I, K, M). Comparison between MPR and no MPR regardless of treatment arm shown in panels (C, E, H, J, L, N). B, C Frequency of CD4_P1 (CD103-expressing) T cell population in PBMC samples. D, E Frequency of CD103+/CD4+ T cells in PBMC samples measured by manual gating. F PaCMAP projection and FlowSOM clustering of CD8+ T cells from all PBMC samples. Heatmap overlays indicates relative expression for each phenotypic marker. Tnv = naïve T cell phenotype, T_emra = terminal effector T cell phenotype, T_em = effectory memory T cell phenotype, T_cm = central memory T cell phenotype. G, H Frequency of CD8_P7 (CD103-expressing) T cell population in PBMC samples. I, J Frequency of CD103+/CD8+ T cells in PBMC samples measured by manual gating. K, L Combined per-patient frequency of CD8_P7 and CD4_P1 T cell populations in PBMC samples. M, N Frequency of PD1+ T cells in PBMC samples measured by manual gating. All panels, two-tailed Mann-Whitney test. Panels (B, D, G, I, K, M) Durva <90% (n = 13); Durva + SBRT < 90% (n = 8); Durva ≥90% (n = 0); Durva + SBRT, ≥90% (n = 8). Panels (C, E, H, J, L, N): <90% (n = 21); ≥90% (n = 8). Source data for all panels are provided as a Source Data file.

Fig. 4. Association of circulating T cell populations with long-term disease-free survival.

Study participants were stratified into two groups based on their disease status at the conclusion of the trial: those with no evidence of disease (NED) and those with progression (died from disease or alive with disease). These groups were compared for frequency of circulating: A, B PD1⁺ T cells, C, D CD8_P7 and CD4_P1 (CD103-expressing) T cell populations. A–D Two-tailed Mann-Whitney test. A Progressed (n = 9), NED (n = 18); B Progressed (n = 5), NED (n = 21); C all panels, Progressed (n = 9), NED (n = 18). Source data for all panels are provided as a Source Data file.

Fig. 5. CD103 gene (ITGAE) expression.

a CD103 (ITGAE) gene expression (FPKM) from bulk RNAseq of pretreatment (tumor biopsy) and at surgery (resected tumor) for monotherapy cases by MPR (n = 15 matched samples). Two-sided, paired Mann-Whitney test. b CD103 (ITGAE) gene expression (FPKM) from bulk RNAseq of pretreatment (tumor biopsy) and at surgery (resected tumor) for dual therapy cases that did not achieve MPR. Two-sided, paired Mann-Whitney test (n = 6 matched samples). c CD103 (ITGAE) gene expression (FPKM) from bulk RNAseq of pretreatment (tumor biopsy) and at surgery (resected tumor) for dual therapy cases that achieved MPR. Two-sided, paired Mann-Whitney test (n = 10 matches samples). d CD103 (ITGAE) gene expression (FPKM) from bulk RNAseq of resected tumors (at surgery), excluding MPR samples, with (n = 9) and without (n = 20) disease recurrence. Solid symbols are from the dual therapy arm. Mean ± SEM. Two-sided, unpaired Mann-Whitney test. Source data for all panels are provided as a Source Data file.

Fig. 6. Characteristics of tumors that do not recur.

a Pathology response, as percent of cancer cell killing, for tumors without MPR grouped by disease recurrence. Mean ± SEM. No recur (n = 23) and Recur (n = 9). Two-sided, unpaired Mann-Whitney test. b PD-L1 positive cancer cells as percent of total cancer cells determined by IHC for tumors without MPR grouped by disease recurrence. Mean ± SEM. No recur (n = 22) and recur (n = 9) Two-sided, unpaired Mann-Whitney test. c Tumor mutational burden for tumors without MPR grouped by disease recurrence. Mean ± SEM. No recur (n = 11) and recur (n = 6). Two-sided, unpaired Mann-Whitney test. d Heatmap of differentially expressed genes between tumors with or without disease recurrence, excluding tumors that achieved MPR. Solid symbols are from the dual therapy arm. e Some GO pathways enriched among genes upregulated in tumors that did not recur. Percentages of the genes in the gene set that were differentially expressed in the data. Source data for panels (a, b, c, e) are provided as a Source Data file.