NEW YORK – First-line treatment with Janssen's bispecific antibody Rybrevant (amivantamab) plus chemotherapy reduced the risk of disease progression or death by 60 percent in patients with advanced non-small cell lung cancer bearing EGFR exon 20 insertion mutations compared to chemotherapy alone in a Phase III trial.
Researchers presented data from the PAPILLON trial at the European Society for Medical Oncology Congress on Saturday and simultaneously published the results in the New England Journal of Medicine. Janssen said in July that it would submit this data to the US Food and Drug Administration, seeking to not only expand the use of Rybrevant to the first-line setting for advanced EGFR exon 20-positive NSCLC but also satisfy the regulatory requirements for single-agent Rybrevant's accelerated approval.
In 2021, the FDA granted accelerated approval to Rybrevant as a treatment for patients with locally advanced or metastatic NSCLC with EGFR exon 20 insertion mutations who have progressed on platinum-based chemotherapy. Following accelerated approval, sponsors must submit data confirming the therapy's efficacy and safety in the intended-use population.
"The PAPILLON study is the first Phase III study to demonstrate a clinical benefit in patients with EGFR exon 20 insertion mutated non-small cell lung cancer, and based on these results, [Rybrevant] plus chemotherapy represents the new standard of care for the first-line treatment of these patients," lead study author Nicolas Girard, a medical oncologist at Institut Curie in Paris, said in a presentation at the conference.
EGFR exon 20 insertion mutations show up in 4 percent to 10 percent of NSCLC cases, and patients have poor outcomes. In this subset, median overall survival ranges from 16 months to 24 months and less than 10 percent of these patients survive five years after diagnosis. EGFR exon 20-positive tumors don't respond to tyrosine kinase inhibitors designed to target more common EGFR mutations, and immunotherapy-based regimens don't improve overall survival in newly diagnosed NSCLC patients better than standard-of-care chemotherapy.
Rybrevant is a bispecific antibody that binds the extracellular domains of EGFR and MET and is designed to degrade tumor cell receptors and cause cell death via trogocytosis and antibody-dependent cellular toxicity. Chemotherapy has been found to complement the activity of Rybrevant, leading to deeper and more durable responses.
In granting accelerated approval to Rybrevant monotherapy, the FDA considered results from the single-arm CHRYSALIS Phase I trial in which patients on the treatment had an overall response rate of 40 percent and a median response duration of 11.1 months.
The PAPILLON trial, by comparison, is a randomized trial in which 306 treatment-naïve patients with metastatic EGFR exon 20-positive NSCLC received Rybrevant plus chemotherapy or just chemotherapy. Patients randomized to the chemotherapy arm were able to receive Rybrevant monotherapy upon disease progression as a second-line treatment. The primary endpoint of the trial is progression-free survival as assessed by blinded independent central review and secondary endpoints included objective response rate, overall survival, duration of response, and progression-free survival on first subsequent therapy (PFS2), which includes Rybrevant or another therapy.
In the 153-patient Rybrevant-chemotherapy arm, 81 patients discontinued treatment either due to disease progression or adverse events, while 70 patients are still receiving treatment. In the chemotherapy arm, 131 patients, or 85 percent, discontinued treatment. Seventy-one patients in the chemotherapy arm crossed over to Rybrevant monotherapy and 24 are still receiving chemotherapy.
After a median follow-up of 14.9 months, patients on Rybrevant-chemotherapy had a median progression-free survival of 11.4 months compared to 6.7 months for those on just chemotherapy. The 18-month progression-free survival rates were 31 percent and 3 percent, respectively, and the objective response rates were 73 percent and 47 percent, respectively. Median PFS2 was not reached for patients receiving Rybrevant-chemotherapy but was 17.2 months for patients on chemo alone.
"Among patients who had the first subsequent systemic treatment, most patients randomized to [Rybrevant] plus chemotherapy went on second-line chemotherapy, and in contrast, in the chemotherapy-randomized patients, the most preferred second-line treatment was [Rybrevant] monotherapy," said Girard.
At the time of the interim analysis, median overall survival was not reached in the Rybrevant-chemotherapy arm and was 24.4 months in the chemotherapy arm. Girard said the data were not mature enough to draw a "robust conclusion" on overall survival.
In a discussion of the results, Benjamin Besse, a medical oncologist at Gustave Roussy in Paris, called the study "highly, highly positive" but flagged that 42 percent of patients had dose interruptions, 36 percent required dose reductions, and 15 percent discontinued Rybrevant-chemotherapy. "This is not a completely easy drug, and we have to manage aggressively the side effects," Besse said.
He compared the results from the earlier CHRYSALIS trial of single-agent Rybrevant with the PAPILLON study to try to contextualize the added value of pairing the bispecific antibody with chemo. In CHRYSALIS, patients on Rybrevant had a median progression-free survival of 8.3 months, while in PAPILLON, patients on just chemo had median progression-free survival of 6.7 months. "It was wise to combine chemotherapy and [Rybrevant]," Besse concluded, drawing attention to the 11.4 months of median progression-free survival in the Rybrevant-chemo arm. But he asserted that the field can still do better for this subset of patients "if we find predictive factors."
The location of the exon 20 insertion mutation, which a prior analysis of data from CHRYSALIS suggested could affect patients' ability to respond to Rybrevant, was not an issue in PAPILLON. Besse said based on prior studies, the absence of baseline RAS/RAF/MEK alterations could be predictive, but those data are "exploratory."