Primary End Point of Improved DFS Not Met With Adjuvant Canakinumab in NSCLC

Adjuvant treatment with canakinumab did not meet the disease-free survival primary end point in patients with completely resected non–small cell lung cancer versus placebo, nor did subgroup analyses show statistically significant differences in DFS, according to results from the phase 3 CANOPY-A trial presented at ESMO Congress 2022.

Adjuvant treatment with canakinumab (Ilaris) did not meet the disease-free survival (DFS) primary end point in patients with completely resected non–small cell lung cancer (NSCLC) versus placebo, nor did subgroup analyses show statistically significant differences in DFS, according to results from the phase 3 CANOPY-A trial (NCT03447769) presented at ESMO Congress 2022.1

“The role of inflammation in lung cancer has been long studied, but to date, it has not been effectively therapeutically harnessed. Canakinumab is a high-affinity antibody directed against interleukin (IL)-1β, a key mediator of inflammation,” Edward B. Garon, MD, MS, explained in his presentation of the data.

After an exploratory analysis of the phase 3 cardiovascular CANTOS trial (NCT01327846) showed reductions in NSCLC incidence and mortality, investigators launched the CANOPY-A trial. In the CANTOS study, a majority of patients who were diagnosed with lung cancer while enrolled had detectable circulating tumor DNA (ctDNA) at baseline. Among patients in the CANTOS trial, the cumulative incidence of fatal lung cancer was lowest with the highest dose of canakinumab (300 mg) and highest with placebo.2

In the CANOPY-A trial, 1382 adult patients with completely resected, stage IIA, IIA, or IIIB NSCLC were randomly assigned 1:1 to receive 200 mg of subcutaneous canakinumab every 3 weeks for 18 cycles versus placebo.1 Patients had to have received cisplatin-based chemotherapy unless they had tumors less than 5 cm and no lymph node involvement and/or radiation therapy to be enrolled on the study. Histology, region, and American Joint Committee on Cancer/Union for International Cancer Control staging were used to stratify patients.

Every 12 weeks for the first year, patients had radiologic evaluation. For years 2 and 3, it was every 26 weeks, then annually for years 4 and 5. Survival follow-up was performed every 12 weeks. The primary end point was DFS by investigator assessment, and secondary end points included overall survival (OS) if DFS was statistically significant, DFS and OS by biomarker subgroup, lung cancer–specific survival, safety, pharmacokinetics, immunogenicity, and patient-reported outcomes.

Patients on CANOPY-A had a median age of 63 years (range, 27-83) in the experimental group (n = 693) and 62 years (range, 21-82) in the placebo group (n = 689). For both arms, about 62% of patients were men, and 64% in the experimental arm and 65% in the placebo arm had ECOG performance status of 0. Forty percent of patients were from Western Europe and North America, 34% were from East Asia, and 25% were from the rest of the world. A majority of patients had non-squamous cell carcinoma and stage IIB or IIIA disease in both arms.

At a data cut-off of March 17, 2022, there was a median time between randomization and data cut-off of 18.9 months (range, 2.8-47.2). Risk factors and prior therapies were even between both arms, including median time from diagnosis to randomization (5.8 months), smoking history (61% former smoker), and prior type of therapy (chemotherapy received by 91.8% in the experimental arm and 90.0% in the placebo arm).

Biomarkers evaluated on this trial included PD-L1, with 30.4% and 29.5% of patients having less than 1% PD-L1 expression in the canakinumab arm and placebo arm, respectively, 14.3% and 17.3% with 1% to 49% PD-L1 expression, 12.4% and 13.9% having 50% or more PD-L1 expression, and 42.9% and 39.3% with missing data. High-sensitivity C-reactive protein of less than 2 mg/L was seen in 48.5% and 49.8% of the canakinumab and placebo arm, respectively, and 2 mg/L or greater was seen in 42.9% and 43.8%, respectively. The rest were missing this data (8.7% and 6.4%, respectively).

One patient in the canakinumab group was not treated after randomization. After the data cut-off, 21.4% of patients receiving canakinumab were still on treatment compared with 19.9% receiving placebo. Approximately half of patients completed treatment, and 30% of both arms discontinued treatment. The most common reason for discontinuation was progressive disease. Median treatment exposure duration was similar between these arms at 11.7 months and 11.8 months, respectively.

The median DFS for canakinumab was 35.0 months versus 29.7 months with placebo (HR, 0.94; 95% CI, 0.78-1.14; one-sided P = .258). This was not a statistically significant improvement for the CANOPY-A trial. In a subgroup analysis of demographics, baseline disease characteristics and biomarkers of interest, there were no meaningful differences in DFS.

OS was not formally tested since DFS was not proven to be statistically significant, however the data showed an OS hazard ratio of 0.722 (95% CI, 0.482-1.081). The lung cancer–specific survival was similar between the experimental and control arms, with a hazard ratio of 0.900 (95% CI, 0.534-1.517).

In terms of toxicity, adverse events (AEs), serious AEs, and fatal serious AEs had similar rates between arms. Canakinumab produced marginally higher treatment-related AEs than the placebo (all-grade 38.9% versus 31.5%, respectively). Fatal treatment-related serious AEs were experienced in 2 patients on the placebo arm and no patients on the canakinumab arm. AEs led to dose interruptions in 13.4% of patients who received canakinumab and 11.0% who received placebo.

The most common AEs, whether treatment-related or not, were cough and arthralgia. Both were reported more in the placebo arm more than the canakinumab arm (cough, 11.4% versus 14.8%; arthralgia, 10.0% versus 12.0%). Infections were the most common AE of special interest and the rate was comparable between the study arms.

Investigators will continue to analyze exploratory biomarker data, such as tumor molecular profiling and longitudinal analysis of circulating inflammatory markers.

“Evaluation of biomarker data from the entire CANOPY program, which in addition includes a neoadjuvant study as well as 2 studies in advanced disease—1 combined with chemoimmunotherapy in a frontline setting and 1 combined with chemotherapy in the second-line setting—will continue to be evaluated to try and further elucidate the role of IL-1β in NSCLC,” Garon concluded.


1. Garon EB, Lu S, Goto Y, et al. CANOPY-A: phase III study of canakinumab (CAN) as adjuvant therapy in patients (pts) with completely resected non-small cell lung cancer (NSCLC). Presented at: European Society for Medical Oncology Congress 2022; September 9-13, 2022; Paris, France. Abstract LBA49.

2. Ridker PM, MacFadyen JG, Thuren T, et al. Effect of interleukin-1β inhibition with canakinumab on incident lung cancer in patients with atherosclerosis: exploratory results from a randomised, double-blind, placebo-controlled trial. Lancet. 2017;390(10105):1833-1842. doi:10.1016/S0140-6736(17)32247-X