Trainee Enews April 2026

Medical Oncology

CHRYSALIS-2 Cohort C: amivantamab plus lazertinib for advanced non-small-cell lung carcinoma harbouring atypical EGFR mutations

EGFR is the second most frequently mutated oncogene in non-small-cell lung carcinoma (NSCLC) globally. Oncogenic mutations can be classified according to the region of the EGFR protein affected and the resulting effect on tyrosine kinase inhibitor (TKI) sensitivity. The common sensitising mutations, including exon 19 deletions and the L858R substitution in exon 21, confer the highest TKI sensitivity. In contrast, exon 20 insertions lead to resistance to conventional EGFR TKIs, and atypical sensitising EGFR mutations, including a wide variety of point mutations spanning exons 18 to 21, are associated with variable and intermediate EGFR TKI sensitivity.1 These are also referred to as rare or uncommon EGFR mutations. The MD Anderson Cancer Centre protein structure-based classification divides EGFR atypical mutations into (i) classical-like mutations, where the affected amino acid residue is distant from the drug-binding pocket and is predicted to have little impact on the overall protein structure, (ii) T790M-like mutations in the EGFR protein hydrophobic core, leading to increased affinity for ATP compared to classical mutations, and (iii) P-loop and αC-helix compressing (PACC) mutations, which narrow the drug-binding pocket.1 2

Afatinib demonstrated an objective response rate (ORR) of 71% and a median progression-free survival (mPFS) of 10.7 months in a pooled analysis of 38 participants from the LUX-Lung trials,3 resulting in its FDA label being broadened to include G719X, L861Q, and S768I point mutations. Osimertinib is commonly used globally for this indication based on the Japanese UNICORN and Korean KCSG-LU15-09 phase II studies, which showed ORR of 55 and 50% and mPFS of 9.4 and 8.2 months, respectively;4 5 however, this use remains off-label in Europe and the USA.

Borgeaud and colleagues examined the impact of EGFR mutational subtypes, as defined by the MDACC classification, on EGFR TKI activity. Their findings indicate that (i) classical-like mutations exhibit response rates to second- and third-generation TKIs comparable to those of common sensitising mutations; (ii) T790M-like mutations are resistant to second-generation TKIs and display variable sensitivity to third-generation TKIs; and (iii) for PACC mutations, second-generation TKIs demonstrate superior anti-cancer activity compared to third-generation TKIs.1

Cohort C of the CHRYSALIS-2 study evaluated the combination of the EGFR-MET bispecific antibody amivantamab and the third-generation EGFR TKI lazertinib in this population. This open-label, phase I/Ib multicenter study enrolled both untreated and pretreated patients (up to two prior lines of treatment). Key exclusion criteria were prior therapy with a third-generation EGFR TKI and the presence of compound common EGFR mutations, as these were predicted to positively influence response to EGFR TKIs.

The primary endpoint was investigator-assessed ORR. The cohort size was determined by a power calculation that estimated 92 evaluable patients were required to detect an increase in ORR from 50% (null hypothesis, i.e. ORR rate is equivalent with amivantamab plus lazertinib vs osimertinib) to 68% (alternative hypothesis) with 90% statistical power.

105 patients were included. Race was reported as Asian in 68%, white in 30%. 47% had received no prior systemic anti-cancer therapy; 35% had baseline CNS metastases, and 63% (37/59 participants with detectable pathogenic alterations in baseline circulating tumour DNA) had TP53 co-mutations. The most prevalent atypical EGFR mutation was G719X at 56%, followed by L861X at 26%, and S768X at 23%. Median follow-up was 16.1 months.

In the overall population, investigator-assessed ORR was 52% (95% CI: 42-62; all partial responses). Median progression-free survival (PFS) was 11.1 months (95% CI: 7.8-17.8), and the median overall survival (OS) was not estimable (95% CI: 22.8 to NE). In the untreated subgroup, investigator-assessed ORR was 57% and median PFS was 19.5 months. In pretreated patients, investigator-assessed ORR was 48% and median PFS was 7.8 months. Subgroup analysis by MDACC classification revealed ORRs of 64% (9/14) for classical-like mutations, 67% (2/3) for T790M-like mutations, and 45% (17/38 patients) for PACC mutations.

To compare the efficacy of amivantamab plus lazertinib vs currently available first-line EGFR-targeting therapies, the CHRYSALIS-2 investigators performed a propensity-score weighted analysis. This compared survival in the treatment-naïve CHRYSALIS-2 cohort C patient subgroup (n = 49) with real-world trial-matched patients from the NSCLC Flatiron Health/Foundation Medicine Clinico-Genomic Database (FH/FMI CGDB) who had ECOG PS 0-1 and received a physician-selected first-line EGFR TKI (n = 69; afatinib 41%, osimertinib 49%). There were no statistically significant differences in baseline characteristics between the two cohorts (median age 60 vs 71, smoking history 55% vs 74%, baseline CNS metastases 27% vs. 36%, mutational subtypes in FH/FMI CGDB cohort G719X 57%, L861X 38%, S768X 20%). Median OS in the FH/FMI CGDB cohort was 19.6 months vs not estimable for treatment-naive patients in CHRYSALIS-2 cohort C (HR, 0.29; 95% CI, 0.12-0.71; log-rank p =0.015). In the real-world cohort, 41% of patients died without receiving a second-line therapy; at 24 months, 77% of treatment-naïve CHRYSALIS-2 cohort C participants were alive vs. 47% in the FH/FMI CGDB cohort.

Adverse events were consistent with previous trials of amivantamab plus lazertinib, including toxicities directly related to EGFR inhibition (rash, paronychia, diarrhoea), MET inhibition (oedema, hypoalbuminaemia), and others (infusion-related reactions, venous thromboembolism). Treatment-emergent toxicity was not inconsiderable: dose reductions were required in 50% of patients and toxicity-related study agent discontinuation in 28%. Of note, since commencement of this trial, several prophylactic strategies have been developed to reduce the risk of infusion-related reactions, venous thromboembolism, and dermatological toxicities.6-8

An alternative way to improve outcomes for patients with atypical EGFR mutations is using EGFR TKIs with broader activity across EGFR mutations. Furmonertinib is a third-generation, brain-penetrant EGFR TKI with efficacy data reported specifically for patients with PACC mutations (FURTHER trial; full cohort, n = 60; treatment-naïve subgroup, n = 47).9 ORR was 68% in treatment-naïve patients and median PFS was 16.0 months. A global phase III study (ALPACCA; furmonertinib vs treatment of physician’s choice) began recruitment in late 2025. There are currently no actively recruiting studies investigating EGFR inhibition plus chemotherapy as a relevant treatment intensification strategy in this patient population.

CHRYSALIS-2 cohort C is the largest prospective cohort of patients with advanced NSCLC harbouring atypical EGFR mutations to date. Although the predefined threshold for the primary endpoint was not technically met, the investigators argue that survival outcomes support amivantamab plus lazertinib delivering a clinically meaningful benefit for this underserved molecular subset of patients with EGFR-mutant NSCLC. The efficacy of amivantamab plus lazertinib in treatment-naïve patients appears impressive, with median PFS almost doubled compared with historical data for afatinib and osimertinib. The propensity score matching model used appears a methodologically sound way to compare efficacy between the trial cohort and contemporaneous controls treated with standard-of-care options in the context of an early-phase non-randomised study. While treatment-emergent toxicity was not negligible, several effective risk reduction strategies have since been established. The CHRYSALIS-2 investigators conclude that amivantamab plus lazertinib shows clinically meaningful and durable antitumor activity while maintaining safety in patients with advanced NSCLC harbouring atypical EGFR mutations; certainly, this is an active regimen in this population and deserves further exploration.