Baseline and on-treatment circulating tumor DNA (ctDNA) features, including TP53 mutations and methylation-based tumor fraction, were associated with early progression or prolonged progression-free survival (PFS) with first-line lorlatinib (Lorbrena) in patients with
Compared with long-term responders, who were progression free at 84 months (n = 30), early progressors, who had a PFS event within 12 months (n = 22), had a higher median number of altered genes at baseline (10 vs 6), a higher median blood-based tumor mutation burden (7.7 vs 5.1), and more frequent TP53 mutations (50% vs 17%). In the lorlatinib arm, the 7-year PFS rate was 65.0% (95% CI, 54.5%-77.4%) in patients with TP53 wild-type disease (n = 73) vs 40.8% (95% CI, 27.3%-61.1%) in those with TP53-mutated disease (n = 43; HR, 0.48; 95% CI, 0.268-0.857); median PFS was not reached (NR) vs 51.6 months (95% CI, 16.4-NR), respectively.
Among lorlatinib-treated patients, those with undetectable methylation-based tumor fraction at baseline (n = 18) had a median PFS of NR vs 80.3 months (95% CI, 34.6-NR) for those with detectable tumor fraction (n = 100; HR, 0.17; 95% CI, 0.04-0.69). Among patients with detectable baseline tumor fraction, clearance at week 4 (n = 54) was associated with longer PFS vs persistent tumor fraction (n = 45; median, NR vs 33.3 months; HR, 0.48; 95% CI, 0.26-0.86), and the association held at week 24 (n = 58 vs n = 15; median, NR vs 32.9 months; HR, 0.45; 95% CI, 0.21-0.96).
"Despite these excellent results, 30% of patients develop disease progression within 2 years, and it remains important to understand mechanisms of resistance and identify patients at risk of early progression to further improve outcomes for this patient group," Benjamin J. Solomon, MBBS, PhD, of Peter MacCallum Cancer Centre, said in a presentation of the data.
CROWN randomly assigned 296 patients with treatment-naive, advanced ALK-positive NSCLC 1:1 to lorlatinib at 100 mg once daily (n = 149) or crizotinib (Xalkori) at 250 mg twice daily (n = 147).2 In the 7-year update, published in Annals of Oncology, median PFS was NR (95% CI, 68.5-NR) with lorlatinib vs 9.1 months (95% CI, 7.4-10.9) with crizotinib (HR, 0.19; 95% CI, 0.13-0.26), with 7-year PFS rates of 55% vs 3%. An earlier 5-year ctDNA analysis detected no emerging ALK resistance mutations at the end of lorlatinib treatment.3
For the current analysis, investigators used the Guardant Infinity assay, which assesses genomic alterations across 755 genes and quantifies tumor fraction using methylation-based technology.1 Samples were collected at baseline (n = 120) and at the end of treatment after progression (n = 18); patients enrolled in China were excluded. The data cutoff was October 31, 2025.
Baseline ALK rearrangements were detected by ctDNA in 59% of patients with baseline samples (n = 120). Paired baseline and post-progression samples (n = 17) revealed no new ALK resistance mutations at the end of treatment. Instead, changes in genomic alteration status suggested that resistance may be driven by bypass mechanisms and other cancer-related pathways, with no dominant pathway identified.
Among patients with paired samples, alterations were observed in DNA damage repair genes (35%), inflammatory-response genes such as AXL, MET, MYC, and ROS1 (35%), RTK/RAS/MAPK pathway genes (35%), oncogene amplification of MET, MYC, or NOTCH2 (29%), tumor suppressor genes (29%), FLT1/VEGFR1 (18%), and HDAC6 (12%).
Inflammatory response was also the predominant pathway associated with early progression at baseline, observed in 32% of early progressors vs 7% of long-term responders. Ahead of the meeting, Dylan Fortman, MD, a hematology/oncology fellow at the University of Michigan Medical School, noted that resistance to first-line lorlatinib
Lorlatinib prolonged PFS vs crizotinib regardless of baseline TP53 status; in the crizotinib arm, median PFS was 10.9 months (95% CI, 9.0-12.8) in patients with TP53 wild-type disease (n = 65) and 7.2 months (95% CI, 5.6-9.3) in those with TP53-mutated disease (n = 44; HR, 0.78; 95% CI, 0.513-1.186).1
PFS with lorlatinib was durable across EML4::ALK variants. Median PFS was NR (95% CI, 26.0-NR) with variants 1/2 (n = 25) and 60.0 months (95% CI, 33.3-NR) with variant 3 (n = 22), with 7-year PFS rates of 52.7% (95% CI, 35.7%-77.7%) and 42.6% (95% CI, 24.1%-75.4%), respectively (HR, 1.09; 95% CI, 0.463-2.579). With crizotinib, median PFS was 7.4 months (95% CI, 3.7-9.3) and 7.2 months (95% CI, 5.4-9.3), respectively. Among lorlatinib-treated patients with variant 2 (n = 6), the objective response rate was 100% and median PFS was NR.
The investigators concluded that baseline genomic complexity and inflammatory-response alterations may help identify patients at risk of early progression, and that methylation-based tumor fraction was prognostic both at baseline and on treatment.
Solomon noted that further work is needed to more precisely identify patients at early risk of progression to inform earlier therapeutic interventions, including treatment intensification. Overall survival follow-up in CROWN is ongoing.2
Disclosures: Solomon reported serving in consulting or advisory roles with Amgen, AstraZeneca, BeiGene, Bristol Myers Squibb, D3 Bio, GlaxoSmithKline, Johnson & Johnson/Janssen, Lilly, MSD, Nuvalent, Pfizer, Revolution Medicine, Roche/Genentech, and Takeda; and receiving honoraria from Amgen, AstraZeneca, MSD, Pfizer, and Roche/Genentech.
