Expanded Approaches Focus on Balancing Efficacy and Toxicity in NSCLC

James Chih-Hsin Yang, MD, PhD

Director

Graduate Institute of Oncology

Director

Department of Oncology

National Taiwan University Hospital

Taipei, Taiwan

Expanded approaches in non–small lung cancer (NSCLC) highlight much of the data presented during the European Lung Cancer Congress (ELCC 2025), which took place in Paris, France, from March 26 to 29, 2025. Although using osimertinib (Tagrisso) remains a frontline stalwart, data presented during ELCC 2025 can spur additional conversations with clinicians and their patients regarding optional approaches that balance efficacy with toxicity. In particular, findings from the phase 3 MARIPOSA trial (NCT04487080) and phase 2 KRYSTAL-7 trial (NCT04613596) explored treatment options in patients with EGFR-mutant and KRAS G12C–mutated NSCLC.

MARIPOSA Trial

Updated data from MARIPOSA,¹ which evaluated the combination of amivantamab-vmjw (Rybrevant) plus lazertinib (Lazcluze) vs osimertinib in the frontline setting, demonstrated long-term benefit of the combination in this patient population. After a median follow-up of 37.8 months, the median overall survival (OS) with the doublet (n = 429) was not reached (NR; 95% CI, 42.9-NR) vs 36.7 months (95% CI, 33.4-41.0) with osimertinib (n = 429). According to investigators, this resulted in a 25% reduction in the risk of death (HR, 0.75; 95% CI, 0.61-0.92; P < .005). The OS curves continued to widen, and based on an exponential assumption of OS in both arms, the improvement in median OS is projected to exceed 1 year. The 36-month OS rates with the combination vs osimertinib were 60% and 51%, respectively; at 42 months, these rates were 56% and 44%.

“Patients live longer with first-line amivantamab plus lazertinib, with MARIPOSA demonstrating practice-changing superior OS vs osimertinib and potentially extending median survival beyond 4 years,” James Chih-Hsin Yang, MD, PhD, director, Graduate Institute of Oncology and the Department of Oncology at the National Taiwan University Hospital, in Taipei, Taiwan, said in a presentation of the data.

The pivotal study enrolled patients with treatment-naive, locally advanced, or metastatic NSCLC who had documented EGFR exon 19 deletion or L858R and an ECOG performance status no higher than 1. Study participants (n = 1074) were randomly assigned 2:2:1 to 1 of 3 arms: amivantamab plus lazertinib (open-label), osimertinib (blinded), or lazertinib monotherapy (blinded). The third arm was included to evaluate the contribution of components. Stratification factors included EGFR mutation type (exon 19 deletion vs L858R), being of Asian race (yes vs no), and history of brain metastases (yes vs no).

The study’s primary end point was progression-free survival (PFS) by blinded independent central review and RECIST v1.1 criteria. Earlier data from the study showed that the doublet reduced the risk of disease progression or death by 30% vs osimertinib (HR, 0.70; 95% CI, 0.58-0.85; P <.001).² The median PFS with the combination was 23.7 months vs 16.6 months with osimertinib. These findings supported the FDA’s decision to approve amivantamab plus lazertinib for this population in August 2024.³

Image by Jennifer Kralyevich / MJH Life Sciences using AI

A key secondary end point of the study was protocol-specified final OS, which is what the current presentation focused on.¹ Other end points reported on at the meeting included intracranial PFS (icPFS), intracranial overall response rate (icORR), intracranial duration of response (icDOR), time to symptomatic progression (TTSP), and safety. “OS was tested with a 2-sided alpha of 0.05, determined by O’Brien-Fleming alpha spending approach as implemented by the Lan-DeMets method,” Yang noted.

Of the 1074 eligible patients, 858 were randomly assigned to the combination arm or osimertinib monotherapy. In the combination arm, 421 patients were treated, and 260 discontinued therapy. The most common reason for discontinuation was progression (33%). In the osimertinib arm, 428 patients received therapy, and 310 discontinued, with progression (55%) being the most common reason for discontinuation.

Baseline Characteristics

Yang explained that baseline characteristics were well balanced across the treatment arms. The median patient age was 63.5 years (range, 25-88); more than half were female (doublet, 64%; osimertinib, 59%) and Asian (58%; 59%), and most had an ECOG performance status of 1 (67% vs 65%). About one-third of patients in both arms had a history of smoking, and about 40.5% of patients had a history of brain metastases. Sixty percent of patients in both arms had EGFR exon 19 deletion, and the remaining 40% had EGFR L858R. Almost all (97%) had an adenocarcinoma subtype.

A generally consistent OS benefit was observed with the doublet vs osimertinib in predefined subgroups, except those aged 65 years or older (HR, 1.11; 95% CI, 0.84-1.48). Seventy-four percent of patients received second-line treatment, suggesting that a long-term treatment plan after frontline treatment with amivantamab/lazertinib is feasible, according to Yang. The most common subsequent therapy class was chemotherapy-based regimens for both arms. Notably, the trial did not permit crossover.

Intracranial PFS

Investigators reported that patients who received the combination therapy had a longer icPFS, 25.4 months vs 22.2 months (95% CI, 18.4-26.9), compared with the monotherapy (HR, 0.79; 95% CI, 0.61-1.02; P =.07). The 3-year landmark icPFS rates in the respective arms were 36% and 18%. The doublet also showcased greater durability of response than osimertinib, with improved icDOR. The icORR was 78% (95% CI, 71%84%) with amivantamab plus lazertinib vs 77% (95% CI, 71%-83%) with osimertinib, and the median icDOR in the respective arms was 35.7 months (95% CI, 25.8-NR) and 29.6 months (95% CI, 23.9-34.1).

“The doublet continues to significantly delay the time to onset or worsening of lung cancer symptoms by a median of greater than 14 months,” Yang said. The median TTSP with the combination was 43.6 months (95% CI, 36.0-NR) vs 29.3 months (95% CI, 26.4-33.4) with osimertinib (HR, 0.69; 95% CI, 0.57-0.83; P <.001).

Regarding safety, the median duration of treatment for the doublet was 27.0 months vs 22.4 months with osimertinib. The toxicity profile of the combination was similar to what was reported previously.

Most of the toxicities experienced with the doublet were related to EGFR or MET inhibition and were grade 1 or 2 in severity. Twenty-one percent of patients received antibiotics for rash at the start of the study.

At baseline, 5% of patients were taking anticoagulation therapy; venous thromboembolic events (VTEs) occurred in 40% of those who received the doublet and 11% of those who were given osimertinib monotherapy. Notably, most first-onset adverse events (AEs) occurred early, between 0 and 4 months. Longer-term follow-up did not reveal any new safety signals, suggesting that long-term treatment with the combination is feasible.

Yang noted that early-onset AEs can be significantly reduced by leveraging prophylactic strategies. For example, the findings presented in the COCOON trial (NCT06120140)⁴ led to an approximate 2-fold reduction in grade 2 or higher dermatologic toxicities, the SKIPPirr trial (NCT05663866 )⁵ led to an approximate 3-fold decrease in infusion-related reactions, and prophylactic anticoagulation led to an approximate 2-fold reduction in VTEs.

KRYSTAL-7

In patients with KRAS G12C–mutated NSCLC and a PD-L1 tumor proportion score (TPS) of 50% or higher, the combination of adagrasib (Krazati) and pembrolizumab (Keytruda) continued to demonstrate efficacy, according to 20-year follow-up findings from the phase 2 portion of KRYSTAL-7.⁶

At a data cutoff date of August 23, 2024, the doublet elicited an objective response rate (ORR) of 59% (95% CI, 45.0%-72.4%) in this subset of patients (n = 54). The best overall responses included complete response (2%), partial response (57%), stable disease (22%), and progressive disease (4%); no postbaseline imaging assessment was done for 15% of patients. The disease control rate was 81% (95% CI, 68.6%-90.7%). The median duration of response (DOR) was 26.3 months (95% CI, 26.3-not evaluable [NE]), and the 12-month DOR rate was 83%.

The PFS was 27.7 months (95% CI, 8.1NE); the 12-month PFS rate with the combination was 60%, and the 18-month PFS rate was 51%. The median OS was not reached (95% CI, 15.4-NE); the 12- and 18-month OS rates were 70% and 62%, respectively.

“ORR and PFS compared favorably with treatment expectations from a PD-L1 inhibitor alone in patients with [a] PD-L1 [TPS of] 50% or higher,” Marina C. Garassino, MD, MBBS, professor of medicine at the University of Chicago, in Illinois, said in a presentation of data. “Treatment-related adverse effects [TRAEs] were consistent with the known safety profiles of adagrasib and pembrolizumab and were effectively managed with established algorithms.”

The phase 2 study included patients with advanced, unresectable, or metastatic NSCLC harboring KRAS G12C mutations who had not previously received systemic therapy for locally advanced or metastatic disease. To participate, patients needed to have a known PD-L1 TPS score. Notably, those with treated, neurologically stable brain metastases were permitted.

Cohorts 1a and 2 received 400 mg of oral adagrasib twice daily and 200 mg of intravenous pembrolizumab every 3 weeks. The primary end point of the study was investigator-assessed ORR by RECIST 1.1, and key secondary end points included investigator-assessed DOR and PFS, OS, and safety.

Earlier data from the study showed that at a median follow-up of 10.1 months, the doublet elicited an ORR of 63%, and the median PFS was NR in this subgroup.⁷ At ELCC, investigators reported updated efficacy in the subset of patients with a PD-L1 TPS of at least 50% and updated safety in all participants.⁶

TPS

In the subset of patients with a PD-L1 TPS of 50% or higher, the median age was 66 years (range, 40-80), and 52% were female. Most patients were White (78%), non-Hispanic or Latino (94%), had adenocarcinoma (94%), and had an ECOG performance status of 1 (67%). All patients were current or former smokers. At baseline, patients had adrenal (17%), bone (30%), central nervous system (20%), and liver (20%) metastases.

Safety

Any-grade TRAEs occurred in 95% of patients who received the combination; 8% of events were grade 1, 16% were grade 2, 58% were grade 3, and 11% were grade 4. The most common TRAEs included diarrhea (grade 1, 35%; grade 2, 9%; grade 3, 3%; grade 4, 0%), nausea (33%; 20%; 3%; 0%, respectively), vomiting (20%; 12%; <1%; 0%), increased alanine aminotransferase (ALT) levels (15%; 13%; 11%; <1%), decreased appetite (15%; 9%, <1%; 0%), fatigue (14%; 11%; 4%; 0%), increased aspartate aminotransferase (AST) levels (13%; 8%; 13%; 1%), and increased lipase (3%; 9%; 11%; 1%). Immune-related AEs occurred in 22% (grade 1, 4%; grade 2, 13%; grade 3, 3%; grade 4, 1%).

TRAEs led to dose reduction or interruption of adagrasib for 48% and 67% of patients, respectively. Most adagrasib dose reductions were because of grade 1 or 2 TRAEs, and the most common TRAEs leading to these reductions were increased ALT levels (11%), increased AST levels (7%), nausea (7%), and increased lipase levels (5%). TRAEs resulted in discontinuation of adagrasib only or pembrolizumab for 7% and 17% of patients, respectively. Forty-six percent of pembrolizumab discontinuations were because of grade 1 or 2 TRAEs. Seven percent of patients experienced TRAEs, which led to the discontinuation of both agents. Other trials focusing on KRAS G12C–mutated NSCLC include the KROCUS trial (NCT05756153).⁸

KROCUS

The phase 2 efficacy and safety KROCUS trial evaluated fulzerasib (GFH925) and cetuximab (Erbitux) as a first-line treatment in patients with advanced NSCLC. At the data cutoff of January 14, 2025, the combination elicited an ORR of 80.0% in evaluable patients (n = 47); the confirmed ORR was 68.9%. More than half of patients (57.8%) experienced at least 50% tumor shrinkage. The disease control rate (DCR) was 100%. Ten patients with brain metastases (n = 14) achieved systemic response by RECIST 1.1 (ORR, 71.4%). All nontarget lesions either disappeared or became stable. In 5 patients who had brain target lesions, all shrank during the treatment period.

At a median follow-up of 12.8 months (range, 0.16-18.4), the median duration of response (n = 31) was not reached (NR; 95% CI, 10.7-not applicable [NA]). The median progression-free survival (n = 47) was 12.5 months (95% CI, 7.39-NA). The median overall survival was also NR.

The single-arm, open-label, phase 2 trial included patients with pathologically confirmed advanced NSCLC with KRAS G12C mutations who had not previously received systemic treatment for advanced or metastatic lesions and had an ECOG performance status no higher than 1. Patients could not have other known actionable driver mutations or alterations, but those with stable and asymptomatic brain metastases were permitted.

Participants were treated with 600 mg of fulzerasib twice daily plus 500 mg/m² of cetuximab every 2 weeks. Treatment continued until disease progression, intolerable toxicity, or other withdrawal criteria were met. ORR by RECIST 1.1 served as the trial’s primary end point, and key secondary end points included PFS by RECIST 1.1 and safety. Exploratory end points included analysis of PD-L1 (TPS) and EGFR (H score) expression levels as well as the examination of comutations at baseline and end of therapy. Prior findings from the study indicated that the doublet induced an ORR of 81.8% and a DCR of 100%.⁹

In the 47 evaluable patients, the median age was 68 years (range, 46-87), and 53.2% were male. Most patients were White (95.7%), were former smokers (72.3%), had an ECOG performance status of 1 (55.3%), and had adenocarcinoma (97.9%). Patients had TNM stage IIIB (4.3%), IVA (53.2%), or VB (42.6%) disease.1 Moreover, 34.0% of patients had brain metastasis, 23.4% had bone metastasis, and 19.1% had adrenal gland metastasis. One patient received prior neoadjuvant therapy, and 2 patients previously received chemoradiation. As of the data cutoff date, 24 patients were still receiving treatment. The median duration of treatment was 10.1 months (range, 0.2-18.4).

When investigators looked at PD-L1 and EGFR expression, they noted comparable response rates across PD-L1 subsets and did not find any correlation between response to the doublet and EGFR expression levels. The confirmed ORRs (cORRs) in those with baseline PD-L1 TPS expression of less than 1% (n = 14), 1% to 49% (n = 8), and 50% or higher (n = 12) were 57.1%, 62.5%, and 75.0%, respectively. The cORRs in those with baseline EGFR H scores under 200 (n = 19) and 200 or higher (n = 13) were 68.4% and 76.9%, respectively.

Responses were observed in those with STK11 or KEAP1 comutations, as well. Those with PD-L1 TPS of less than 1% and a STK11 comutation (n = 8) had a cORR of 62.5%; those with a PD-L1 TPS of 1% to 49% and STK11 comutation (n = 36) had a cORR of 69.4%. Those with an EGFR H score under 200 and a KEAP1 comutation (n = 3) had a cORR of 66.7%; those with an EGFR H score of 200 or higher and a KEAP1 comutation (n = 41) had a cORR of 68.3%.

At least 1 TRAE occurred in 87.2% of those who received the doublet, and these effects were grade 3 for 14.9% of patients. At least 1 serious TRAE was experienced by 4.3% of patients, but none were related to fulzerasib. At least 1 TRAE led to dose reduction, interruption, or discontinuation for 10.6%, 27.7%, and 6.4% of patients, respectively; they were related to fulzerasib for 8.5%, 14.9%, and 0% of cases, respectively.

The most common TRAEs experienced with the combination were rash (any grade, 53%; grade 3, 2%), pruritus (32%; 0%), asthenia (26%; 2%), and nausea (26%; 0%).

RMC-6236-001 In patients with RAS G12X mutations, Salman R. Punekar, MD, and colleagues reported on the first-in-human phase 1 RMC-6236-001 trial (NCT05379985) that evaluated the multiselective, tricomplex RAS(ON) inhibitor daraxonrasib (RMC6236) in patients with NSCLC.¹⁰ Patients were treated with daraxonrasib at either the 120-mg or 220-mg dose (n = 40) and investigators reported an ORR of 38%. The median time to response was 1.5 months (range, 1.26.2), and the median duration of response (DOR) was 15.1 months (95% CI, 11.1-NE).

TRAEs

Regarding safety, TRAEs led to dose interruptions, dose reductions, and treatment discontinuation in 48%, 27%, and 6% of patients treated with daraxonrasib at a dose of 120 mg to 300 mg per day (n = 124), respectively; a total of 52% of these patients required any dose modifications due to TRAEs. Additionally, 41% of patients treated at doses ranging from 120 mg to 220 mg per day (n = 73) required dose modifications due to TRAEs; the respective rates of dose interruption, dose reduction, and treatment discontinuation were 34%, 21%, and 4%.

Among patients treated at 120 mg to 220 mg, grade 3 or higher treatment- related rash occurred in 9% of patients who did not receive prophylaxis (n = 58), and 7% of these patients required dose modifications due to treatment-related rash. The 200-mg dose was selected for further evaluation in phase 3 testing.

“Based on these data, a randomized phase 3 study called RASolve 301 [NCT06881784] of daraxonrasib vs docetaxel in patients with previously treated, locally advanced or metastatic RAS-mutant NSCLC has been initiated,” Punekar, an assistant professor in the Department of Medicine, New York University Grossman School of Medicine of NYU Langone Health, New York, said in a presentation of the data.

During dose-escalation, KRAS G12D mutations were required; RAS mutations confirmed via DNA sequencing were needed for dose optimization/expansion. Prior standard therapy appropriate for a patient’s tumor type and stage was required. Other key inclusion criteria consisted of an ECOG performance status of 0 or 1 and adequate organ function. The study included patients with KRAS, HRAS, or NRAS mutations at codon 12, codon 13, or codon 61; patients with KRAS G12C mutations were enrolled in the NSCLC cohort.¹⁰ During dose escalation, daraxonrasib was evaluated at once-daily doses ranging from 10 mg to 400 mg. The 120-, 160-, 220-, and 300-mg doses were selected for the dose-optimization phase. Safety and the incidence of dose-limiting toxicities are the trial’s primary end points.² Secondary end points include pharmacokinetics, ORR, DOR, time to response, disease control rate, and PFS.

In all patients with NSCLC treated with daraxonrasib at 120 mg to 300 mg per day (n = 124), the median age was 67 years (range, 31-89), and 60% were female.¹ Eighty-one percent of patients had an ECOG performance status of 1, and 76% of patients were current or past smokers. Fifty-seven percent of patients had metastatic disease at diagnosis, and 29% had baseline brain metastases. Patients received a median of 2 prior lines of therapy (range, 1-6).

Additional Findings

Additional findings also demonstrated that patients treated with daraxonrasib at 120 mg to 220 mg per day achieved a median PFS of 9.8 months (95% CI, 6.0-12.3) and a median OS of 17.7 months (95% CI, 13.7-NE). The most common TRAEs reported in at least 10% of patients treated with daraxonrasib at 120 mg to 300 mg per day included rash (anygrade, 89%; grade ≥3, 7%), diarrhea (70%; 8%, respectively), nausea (55%; 0%), vomiting (44%; 2%), stomatitis (38%; 2%), paronychia (21%; 0%), fatigue (16%; 0%), dry skin (15%; 0%), increased AST levels (14%; 1.6%), and increased ALT levels (12%; 2.4%).

Overall findings presented during ELCC 2025 demonstrate the benefit of using targeted combinations earlier in the disease progression in this heterogeneous population. Clinician and patient discussions will focus on the various toxicities associated with each approach, but clinicians are excited about multiple therapy options.

REFERENCES:

1. Yang JC-H, Kim YJ, Lee S-H, et al. Amivantamab plus lazertinib vs osimertinib in first-line EGFR-mutant advanced NSCLC. Presented at: 2025 European Lung Cancer Congress; March 26-29, 2025; Paris, France. Abstract 4O.

2. Cho BC, Lu S, Felip E, et al; MARIPOSA Investigators. Amivantamab plus lazertinib in previously untreated EGFR-mutated advanced NSCLC. N Engl J Med. 2024;391(16):1486-1498. doi:10.1056/NEJMoa2403614

3. Rybrevant (amivantamab-vmjw) plus Lazcluze (lazertinib) approved in the U.S. as a first-line chemotherapy-free treatment for patients with EGFR-mutated advanced lung cancer. News release. Johnson & Johnson. August 20, 2024. Accessed March 26, 2025. https://tinyurl.com/yzfrtv47

4. Cho BC, Girard N, Sauder MB, et al. Enhanced vs standard dermatologic management with amivantamab-lazertinib in advanced NSCLC: phase 2 COCOON Study. J Thor Oncol. 2024;19(10):S347. doi:10.1016/j.jtho.2024.09.626

5. Paz-Ares LG, Spira AI, Han J-Y, et al. Preventing infusion-related reactions with intravenous amivantamab: updated results from SKIPPirr, a phase II study. Ann Oncol. 2024;35(suppl 2):S812. doi:10.1016/j.annonc.2024.08.1326

6. Garassino MC, Theelen WSME, Jänne PA, et al. First-line adagrasib (ADA) with pembrolizumab (PEMBRO) in patients with advanced/metastatic KRAS G12C-mutated non-small cell lung cancer (NSCLC) and PD-L1 ≥ 50% from the phase 2 portion of KRYSTAL-7. Presented at: 2025 European Lung Cancer Congress; March 26-29, 2025; Paris, France. Abstract 5MO.

7. Garassino MC, Theelan WSME, Jotte R, et al. LBA65 KRYSTAL-7: Efficacy and safety of adagrasib with pembrolizumab in patients with treatment-naive, advanced non-small cell lung cancer (NSCLC) harboring a KRAS G12C mutation. Ann Oncol. 2023;34(suppl 2):S1309-S1310. doi:10.1016/j.annonc.2023.10.066

8. Majem M, Gregorc V, Lo Russo G, et al. First-line (1L) fulzerasib + cetuximab in KRAS G12Cm advanced NSCLC: Updated efficacy and safety from KROCUS study. Presented at: 2025 European Lung Cancer Congress; March 26-29, 2025; Paris, France. Abstract LBA1.

9. Gregorc V, González-Cao M, Salvagni S, et al. KROCUS: A phase II study investigating the efficacy and safety of fulzerasib (GFH925) in combination with cetuximab in patients with previously untreated advanced KRAS G12C mutated NSCLC. J Clin Oncol. 2024;42(suppl 17):LBA8511. doi:10.1200/JCO.2024.42.17_suppl.LBA8511

10. Punekar S, Hong DS, Luo J, et al. Safety and clinical activity of daraxonrasib (RMC-6236) in RAS mutant non-small cell lung cancer (NSCLC). J Thorac Oncol. 2025;20(3):S10-S11. doi:10.1016/S1556-0864(25)00201-1