乌鸦传媒

SAN DIEGO -- A bispecific immune checkpoint inhibitor significantly improved survival in advanced gastric/gastroesophageal junction (GEJ) cancer, irrespective of PD-L1 expression status, a large study from China showed.

Median overall survival (OS) increased from 10.8 months with chemotherapy to 15.0 months with the addition of cadonilimab, which targets both PD-1 and CTLA-4. Results significantly favored the cadonilimab arm in patients with a PD-L1 expression combined positive score (CPS) ≥5 and those with a CPS <5. Subgroup analysis showed a consistent benefit for the bispecific antibody.

The frequency of treatment-related adverse events (TRAEs) was similar between the cadonilimab arm and patients who received only chemotherapy, reported Jiafu Ji, MD, of the Beijing Cancer Hospital, at the American Association for Cancer Research annual meeting.

"Cadonilimab combined with chemotherapy showed promising results in treating patients with gastric or gastroesophageal junction adenocarcinoma," said Ji. "It leads to improved overall survival and PFS [progression-free survival], and overall response rate [ORR] and duration of response, compared to chemotherapy. For patients with lower CPS, it could become a new standard of management for late-stage gastric and gastroesophageal junction cancer."

A suboptimal control arm detracted from the trial's favorable results, said invited discussant Yelena Janjigian, MD, of Memorial Sloan Kettering Cancer Center in New York City. Chemotherapy stopped after 4.5 months in both arms, a shorter duration than used elsewhere in the world. Additionally, the control arm did not include a PD-1/L1 inhibitor, even though an anti-PD-1 agent plus chemotherapy has become standard treatment for advanced gastric/GEJ adenocarcinoma.

The data reported did not include microsatellite instability or claudin 18.2 status, and HER2 negativity was not mandated, she continued.

"Despite the suboptimal comparator arm, the data are important and offer a rationale for dual anti-PD-1/CTLA-4 blockade in gastroesophageal cancer," said Janjigian. "Additional efficacy and safety data in Western patients, including a comparison to chemotherapy plus anti-PD-1, might be needed before globalization of this regimen."

PD-1 inhibitors plus chemotherapy have demonstrated efficacy as first-line therapy for advanced gastric/GEJ adenocarcinoma. However, prior studies have shown limited survival benefit with the combination in patients with low PD-L1 expression.

Ji reported interim results from the phase III randomized trial conducted exclusively in China. A of cadonilimab in gastric/GEJ cancer showed a median PFS of 9.2 months and median OS of 17.4 months, and outcomes were similar in patients with a PD-L1 expression CPS ≥5 or <5.

COMPASSION-15 evaluated the bispecific antibody as first-line treatment in 600 patients with unresectable or metastatic gastric/GEJ adenocarcinoma. All patients received up to six cycles of capecitabine and oxaliplatin chemotherapy and were randomized to cadonilimab or placebo. Patients randomized to placebo received no further therapy, whereas those in the study arm received cadonilimab maintenance. The primary endpoint was OS in the intention-to-treat (ITT) population.

The study population had a median age of 64, 75-80% had gastric cancer, and 75% had primary metastatic disease. Analysis of PD-L1 expression was conducted with the . The 22C3 results showed that about a fourth of the patients had a CPS <1, a fourth had a CPS of 1-4, 15-20% had a CPS of 5-9, and a fourth had a CPS ≥10. According to the SP263 assay, 55% of the patients had a visualized CPS <5% and 40% had a visualized CPS ≥5%.

The ITT analysis showed that the addition of cadonilimab led to a 38% reduction in the survival hazard (95% CI 0.50-0.78, P<0.001). The bispecific antibody improved survival in patients with a CPS ≥5 (HR 0.56, 95% CI 0.39-0.80, P<0.001) and those with a CPS <5 (HR 0.70, 95% CI 0.51-0.95, P=0.011). All prespecified subgroups benefited from cadonilimab.

The PFS analysis showed a median of 7.0 months with cadonilimab versus 5.3 months with placebo in the ITT population (HR 0.53, 95% CI 0.44-0.65, P<0.001). Median PFS was 6.9 months versus 5.5 months for those with a CPS ≥5 (HR 0.51, 95% CI 0.37-0.70) and 6.9 versus 4.6 months for those with a CPS <5 (HR 0.60, 95% CI 0.45-0.79, P<0.001).

Overall response rate improved from 48.9% with chemotherapy alone to 65.2% with cadonilimab, and duration of response doubled from 4.4 months to 8.8 months.

Grade ≥3 TRAEs occurred in 66% of the cadonilimab arm versus 54% of the control arm. TRAE-associated discontinuation rates were 15.4% with cadonilimab and 5.3% with chemotherapy alone. Five (1.6%) deaths occurred with cadonilimab and seven (2.3%) with chemotherapy-placebo. The most common all-grade TRAEs in the cadonilimab group were decreased platelets (65-70%), decreased neutrophils (60%), decreased white blood cells (50%), anemia (50%), nausea (45%), and vomiting (45%).

Janjigian characterized the toxicity profile as "pretty good" and in line with what might be expected with the combination of a PD-1 inhibitor and chemotherapy.

Comment