Evaluating EFS and Surgical Outcomes in Resectable NSCLC: Insights From a Clinical Trial

IMFINZI^® (durvalumab) in combination with platinum-containing chemotherapy as neoadjuvant treatment, followed by IMFINZI continued as a single agent as adjuvant treatment after surgery, is indicated for the treatment of adult patients with resectable (tumors ≥4 cm and/or node positive) non-small cell lung cancer (NSCLC) and no known epidermal growth factor receptor (EGFR) mutations or anaplastic lymphoma kinase (ALK) rearrangements.

Immunotherapy (checkpoint inhibitors) have become part of the evolving perioperative approach to resectable non–small cell lung cancer (rNSCLC).^1,2 The phase 3 AEGEAN study evaluated a perioperative approach combining immune checkpoint inhibition with platinum-based chemotherapy, reporting event-free survival (EFS) and pathologic complete response (pCR) results.³ A pre-specified secondary analysis further explored surgical data.⁴

Trial Design: A Multiphase Approach to Immunotherapy Integration
AEGEAN was a randomized, double-blind, phase 3 study enrolling 802 adults with resectable Stage IIA to selected Stage IIIB NSCLC. Eligible tumors were ≥4 cm and/or node-positive (N1 or N2), and all patients were candidates for curative-intent thoracic surgery, including lobectomy, bilobectomy, sleeve resection, or other anatomically appropriate procedures. Following a protocol amendment, patients with known EGFR or ALK alterations were excluded.³

Patients were randomized 1:1 to receive 4 cycles of platinum-based chemotherapy with either durvalumab or placebo in the neoadjuvant phase, followed by surgery and then up to 12 cycles of adjuvant durvalumab or placebo monotherapy every 4 weeks. Treatment with durvalumab or placebo continued until completion of the treatment, disease progression that precluded definitive surgery, disease recurrence in the adjuvant phase, or unacceptable toxicity.³

Randomization was stratified by PD-L1 tumor cell expression (≥1% and <1%), clinical stage (II and III), and histology. The modified intention-to-treat (mITT) population excluded patients with EGFR or ALK alterations enrolled before the protocol amendment.³

Primary Endpoints: Event-Free Survival and Pathologic Complete Response
The trial’s co-primary endpoints were EFS evaluated by blinded independent central review and pCR evaluated by blinded central pathology review. Both endpoints were assessed in the mITT population using prespecified statistical methods.³

• In AEGEAN’s first interim analysis, with 31.9% data maturity, median EFS in the mITT population³:
  • Median EFS was not reached in the durvalumab + neoadjuvant chemotherapy arm (95% CI, 31.9-NR)
  • Median EFS was 25.9 months in the neoadjuvant chemotherapy + placebo arm (95% CI, 18.9-NR)
  • Had a stratified hazard ratio for disease progression, recurrence, or death of 0.68 (95% CI, 0.53–0.88; P=0.004 by stratified log-rank test)
• pCR was defined as the absence of a viable tumor in both the resected lung specimen and all sampled lymph nodes. At the final analysis (November 10, 2022), which occurred 46.3 months after study initiation, pCR in the mITT population³:
  • pCR was achieved by 17.2% in the durvalumab arm (63 of 366; 95% CI, 13.5-21.5)
  • pCR was achieved by 4.3% in the placebo arm (16 of 374, 95% CI, 2.5-6.8)
  • Had a difference of 13.0 percentage points between the two treatment arms (95% CI, 8.7-17.6; P<0.001 based on interim analysis [N=402]). The 2-sided P value for pCR was calculated based on a stratified Cochran-Mantel-Haenszel test. Stratification factors include PD-L1 and disease stage. 

Surgical Completion and Extent of Resection
Surgical outcomes were assessed as a pre-specified secondary endpoint. In the mITT population, surgery was initiated in 80.6% of patients in the durvalumab arm (295 of 366) and 80.7% in the placebo arm (302 of 374). Surgery referred to any curative-intent thoracic procedure.⁴

Completed resection, defined as removal of all macroscopic disease, was reported in 284 patients in the durvalumab group (77.6%) and 287 in the placebo group (76.7%). Among those who completed surgery, approximately 68% had Stage III disease and 71% had baseline nodal involvement.^3,4

R0 resection, indicating negative margins, was achieved in over 94% of patients who completed surgery in both arms.⁴

Treatment completion data and surgical outcome data were not tested for statistical significance.

Treatment Completion and Continuity Across Phases
Minimally invasive approaches were used in 49.2% in the durvalumab group and 47.0% with placebo. Surgical complications occurred in 59.1% of patients who received durvalumab and 60.1% of those in the placebo group.⁴

Complete resection (R0) was achieved in 94.7% of patients in the durvalumab arm versus 91.3% in the placebo group. The median duration of surgery was 3.5 hours in the durvalumab group and 3.3 hours with placebo, each ranging from 1 to 24 hours.⁴

Timing from neoadjuvant therapy to surgery demonstrated median intervals of 50.0 days (range, 22-136) in the durvalumab group and 52.0 days (range, 21-141) in the placebo group. Surgery delays, defined as occurring more than 40 days after the final neoadjuvant dose, were reported in 17.3% and 22.2% of patients, respectively. The AEGEAN study was conducted during the COVID-19 pandemic. Most delays in surgery were logistical in nature, potentially related to scheduling restrictions caused by coronavirus.⁴

Lobectomy remained the predominant approach, performed in 88.1% of patients in the durvalumab arm and 85.4% in the placebo arm. Pneumonectomy rates were 9.2% and 9.6%, respectively. Less common procedures such as wedge resections accounted for 2.7% in the durvalumab group and 5.0% with placebo.⁴

Treatment completion data and surgical outcome data were not tested for statistical significance.

Lymph Node Response Following Treatment
Nodal status at the time of surgery offered additional insight into regional disease control. Among patients with baseline N2 disease, 47.3% in the durvalumab group and 40.2% in the placebo group were downstaged to N0. In patients with baseline N1 disease, downstaging to N0 was reported in 53.6% (durvalumab) and 46.2% (placebo).⁴

Stable nodal status, such as persistent N2, was observed in 42.6% and 42.4% of patients in the durvalumab and placebo arms, respectively.⁴

These data were descriptive and based on investigator-assessed clinical staging at baseline and pathologic findings at surgery.

Radiologic Response and Pathologic Correlation
A separate exploratory analysis assessed whether radiographic response prior to surgery correlated with pathologic outcomes. Imaging was evaluated using RECIST v1.1 by site investigators.⁴

Among patients with radiographic complete or partial response⁴:

• pCR was achieved in 27.2% (durvalumab) and 8.5% (placebo)
• mPR (≤10% viable tumor) occurred in 50.5% and 24.6%, respectively

No patient with radiographic progression achieved either pCR or mPR.⁴

These findings were descriptive, and no formal correlation testing was performed.

Perioperative Safety Profile
Surgical safety was evaluated in patients who underwent resection. Any-grade adverse events (AEs) considered possibly related to surgery occurred in 40.2% of patients in the durvalumab arm and 39.2% in the placebo arm. Grade 3 or 4 AEs occurred in <10% of patients across groups.⁴

One patient in the durvalumab group and 2 in the placebo group experienced an AE with fatal outcome deemed possibly related to surgery.⁴

Treatment discontinuation was attributed to adverse events, disease progression, withdrawal, or physician decision.⁴

All safety findings were descriptive and not tested for statistical significance.

Limitations
Surgical outcomes, nodal assessments, and treatment exposure data in AEGEAN were part of a pre-specified secondary analysis and were not powered or tested for statistical comparison between treatment groups. Surgical decisions, techniques, and timing were determined by site investigators according to local practice and were not centrally standardized. Nodal status was based on investigator-assessed clinical staging at baseline and pathology at surgery; imaging and pathologic correlations were exploratory and descriptive. Additionally, delays in surgery were attributed in part to external factors, including the COVID-19 pandemic, which may limit generalizability.^3,4

Practice Considerations
While AEGEAN was primarily designed to assess oncologic efficacy, the secondary analysis contributes clinical context around surgical feasibility, continuity of treatment, and nodal outcomes.^3,4 These findings may inform surgical planning and multidisciplinary coordination as perioperative immunotherapy based regimens continue to be explored in resectable NSCLC.