The addition of canakinumab to frontline pembrolizumab and chemotherapy did not improve survival in previously untreated patients with locally advanced or metastatic non–small cell lung cancer. However, signals of activity were seen in patients with a baseline inflammatory biomarker high sensitivity-C-reactive protein.
The addition of canakinumab to frontline pembrolizumab (Keytruda) and chemotherapy did not improve progression-free and overall survival (OS) in previously untreated patients with locally advanced or metastatic non–small cell lung cancer (NSCLC). However, signals of activity were seen in a subset of patients with a baseline inflammatory biomarker high sensitivity (hs)-C-reactive protein (CRP), according to data from the phase 3 CANOPY-1 trial (NCT03631199).1
Results, which were presented during the 2022 AACR Annual Meeting, showed that the median progression-free survival (PFS) was 6.8 months in both arms (HR, 0.85; 95% CI, 0.67-1.09; one-sided P = .1) at a median follow-up of 4.8 months (interquartile range [IQR], 2.8-5.5). The median OS was 20.8 months with canakinumab and 20.2 months with placebo at a median follow-up of 16.2 months (IQR, 6.6-20.2), leading to a 13% reduction in the risk of death (HR, 0.87; 95% CI, 0.70-1.10; one-sided P = .123).
However, additional assessments revealed reductions in hs-CRP and hs-interleukin-6 (IL-6) levels that were more pronounced in those who received the canakinumab combination based on a linear mixed model.
When divided by quartile, in patients with hs-CRP less than Q1, there was a 43% reduction in the risk of disease progression or death (HR, 0.57; 95% CI, 0.27-1.21). This PFS benefit was more pronounced compared with those who had hs-CP ranging from Q1 to less than Q2 (HR, 1.24; 95% CI, 0.63-2.42), Q2 to less than Q3 (HR, 0.74; 95% CI, 0.44-1.25), and Q3 or higher (HR, 0.97; 95% CI, 0.60-1.57).
Similar findings were reported regarding OS in patients whose hs-CRP levels were less than Q1 (HR, 0.45; 95% CI, 0.22-0.93). In those with with hs-CRP of Q1 to less than Q2, the HR was 1.15 (95% CI, 0.67-1.97) and the HR was 0.76 (95% CI, 0.48-1.23) in those with hs-CRP of Q2 to less than Q3. For patients whose hs-CRP levels were in Q3 or higher, the HR for OS was 1.09 (95% CI, 0.70-1.71).
“Baseline hs-CRP and hs-IL-6 levels were prognostic factors and their levels decreased with treatment in both study arms, with a more pronounced decrease in the canakinumab arm,” lead study author Daniel S. W. Tan, MD, a senior consultant of medical oncology at the National Cancer Centre Singapore, said in a presentation during the meeting. “A comprehensive exploratory biomarker analysis is ongoing to verify the immune cell composition of the tumor microenvironment and/or CRP/IL-6 modulations are associated with treatment outcomes.”
IL-1β is an inflammatory cytokine with pleiotropic functions and may have a role in tumorigenesis, tumor invasiveness, and metastases. Preclinical data of IL-1β and PD-1 inhibition demonstrated synergistic antitumor activity compared with either agent alone, supporting the rationale for IL-1β combined with PD-1 inhibition in NSCLC, Tan noted.
Although pembrolizumab plus chemotherapy has remained a standard frontline option for patients with advanced NSCLC, not all patients respond to this approach, and activity varies depending on PD-L1 expression levels. Canakinumab is a monoclonal anti-interleukin-1β antibody with pro-tumor inflammation inhibition and is thought to enhance antitumor immune responses and to have synergistic potential with PD-1 inhibitors and chemotherapy.
In the double-blind, placebo-controlled, phase 3 CANOPY-1 trial, investigators randomized 643 patients with previously untreated stage IIIB/C or IV NSCLC 1:1 to receive first-line pembrolizumab plus platinum-based doublet chemotherapy with the addition of canakinumab (n = 320) or placebo (n = 323). Pembrolizumab and platinum-based chemotherapy were given every 3 weeks for 4 cycles, followed by maintenance canakinumab or placebo with pembrolizumab plus or minus pemetrexed.
To be eligible for enrollment, patients had to be treatment naïve with stage IIIB to IV NSCLC, known PD-L1 status, and at least 1 measurable lesion via RECIST v1.1 criteria. Patients also needed to have an ECOG performance status of 0 or 1. Patients could not have known EGFR or ALK alterations; they were stratified based on PD-L1 status, geographic region, and histology.
The coprimary end points of the trial were investigator-assessed PFS via RECIST v1.1 criteria and OS. Secondary end points included objective response rate (ORR), disease control rate (DCR), duration of response (DOR), patient-reported outcomes (PROs), and safety. Exploratory biomarker analyses were also performed, specifically with regard to PD-L1, tumor mutational burden (TMB), hs-CRP, and hs-IL-5.
Baseline characteristics were well balanced between the 2 arms. The median age was 63 years (range, 56-69), and most patients (80.9%) were current or former smokers. Moreover, 47.3% of patients had a PD-L1 expression of less than 1%, and 30.6% had squamous histology. Ten percent of patients had the presence of brain metastases.
The data cutoff dates for PFS and OS were May 18, 2020, and August 9, 2021, respectively. Treatment was ongoing in 23.8% of those in the investigative arm and 22.0% of those in the control arm. In the canakinumab and placebo arms, 76.3% and 77.7% of patients, respectively, discontinued treatment due to progressive disease (46.6% vs 45.8%), death (10.9% vs 14.9%), an adverse effect (AE; 8.8% vs 9.0%), physician decision (5.9% vs 4.0%), or patient or guardian decision (4.1% vs 4.0%).
The median duration of treatment exposure was 8.7 months (IQR, 4.7-19.2) and 7.9 months (IQR, 4.1-18.6) for the canakinumab and placebo arms, respectively. The median number of treatment administrations in both arms was 11 (IQR, 6.0-25.0).
Additional results showed that across key patient subgroups, including age, ECOG performance status, smoking history, PD-L1 expression, and TMB, there was no difference in PFS nor OS between the 2 treatment arms. Although there did appear to be a benefit with canakinumab in females (HR, 0.59; 95% CI, 0.36-0.95), Tan said that these findings are mixed in various meta-analyses with the impact of gender and outcomes.
The ORR was similar in both arms, at 45.6% (95% CI, 40.1%-51.3%) and 45.5% (95% CI, 40.0%-51.1%) with canakinumab and placebo, respectively. In the canakinumab and placebo arms, respectively, the complete response rates were 0.3% vs 0.9%, partial response rates were 45.3% vs 44.6%, stable disease rates were 41.3% vs 39.3%, progressive disease was seen in 7.2% vs 7.7% of patients, and 5.9% vs 7.4% of patients were not evaluable.
The DCR was 86.9% (95% CI, 82.7%-90.4%) with canakinumab and 84.8% (95% CI, 80.4%-88.6%) with placebo. The median DOR was 14.3 months (95% CI, 10.4–not evaluable [NE]) and 13.6 months (95% CI, 10.3-NE), respectively.
Investigators also assessed hs-IL-6 levels, and a more pronounced reduction in IL-6 levels was observed in those within the canakinumab arm.
“There really is a challenge in determining what the right assay should be with the appropriate dynamic range in advanced NSCLC, and there are active ongoing efforts to refine this threshold of significance and will be perhaps defined and correlated with tumor transcriptomic profiles of clinical outcomes with longer follow-up,” Tan noted.
Regarding safety, grade 3/4 AEsa were reported in 64.1% and 59.3% of patients on the canakinumab and placebo arms, respectively. Any-grade AEs that led to treatment discontinuation occurred in 22.5% of patients on canakinumab and 18.9% of those on placebo. Serious AEs that led to death occurred in 11.6% of patients and 14.6% of canakinumab- and placebo-treated patients, respectively. Any-grade dose adjustments or interruptions from AEs occurred in 73.1% and 57.1% of patients, respectively.
Additionally, PRO compliance rates at baseline were 91% for canakinumab and 87% for placebo. The mean PRO scores at baseline were well balanced between arms. Time to deterioration with coughing (HR, 0.59; 95% CI, 0.43-0.82; 1-sided P = .0007), chest pain (HR, 0.62; 95% CI, 0.45-0.86; P = .002), and dyspnea (HR, 0.66; 95% CI, 0.51-0.84; 1-sided P = .0005) were longer with canakinumab as measured via the EORTC QLQ-C30, EORTC QLQ-LC13, and EQ-5D-5L questionnaires.