CNS Efficacy of Osimertinib With or Without Chemotherapy in Epidermal Growth Factor Receptor-Mutated Advanced Non-Small-Cell Lung Cancer

Abstract

Purpose: We report CNS efficacy of first-line osimertinib plus chemotherapy versus osimertinib monotherapy in patients with epidermal growth factor receptor (EGFR)-mutated advanced non-small-cell lung cancer (NSCLC) from the phase III FLAURA2 study according to baseline CNS metastasis status.

Methods: Patients were randomly assigned to osimertinib plus platinum-pemetrexed (combination) or osimertinib monotherapy until disease progression or discontinuation. Brain scans were performed in all patients at baseline and progression and at scheduled assessments until progression for patients with baseline CNS metastases; scans were assessed by neuroradiologist CNS blinded independent central review (BICR).

Results: On the basis of baseline CNS BICR, 118 of 279 (combination) and 104 of 278 (monotherapy) randomly assigned patients had ≥one measurable and/or nonmeasurable CNS lesion and were included in the CNS full analysis set (cFAS); 40 of 118 and 38 of 104 had ≥one measurable target CNS lesion and were included in the post hoc CNS evaluable-for-response set (cEFR). In the cFAS, the hazard ratio (HR) for CNS progression or death was 0.58 (95% CI, 0.33 to 1.01). In patients without baseline CNS metastases, the HR for CNS progression or death was 0.67 (95% CI, 0.43 to 1.04). In the cFAS, CNS objective response rates (ORRs; 95% CI) were 73% (combination; 64 to 81) versus 69% (monotherapy; 59 to 78); 59% versus 43% had CNS complete response (CR). In the cEFR, CNS ORRs (95% CI) were 88% (73 to 96) versus 87% (72 to 96); 48% versus 16% had CNS CR.

Conclusion: Osimertinib plus platinum-pemetrexed demonstrated improved CNS efficacy compared with osimertinib monotherapy, including delaying CNS progression, irrespective of baseline CNS metastasis status. These data support this combination as a new first-line treatment for patients with EGFR-mutated advanced NSCLC, including those with CNS metastases.

Trial registration: ClinicalTrials.gov NCT04035486.

FIG 1.

CONSORT diagram, patient disposition. Data cutoff: April 3, 2023. Neuroradiologist CNS BICR was evaluated on the basis of data derived from the brain scans (MRI or computed tomography) of all patients in the FAS. The cFAS included patients with ≥one CNS lesion (measurable and/or nonmeasurable) on CNS scans assessed by neuroradiologist BICR. The cEFR included only patients with ≥one measurable CNS lesion at baseline, on CNS scans assessed by neuroradiologist BICR. BICR, blinded independent central review; cEFR, CNS evaluable-for-response set; cFAS, CNS full analysis set; FAS, full analysis set; MRI, magnetic resonance imaging.

FIG 2.

CNS PFS in (A) cFAS and (B) cEFR. (C) Cumulative incidence of CNS progression, taking into account competing risks of non-CNS progression and death resulting from any cause, in cFAS. (D) Analysis of CNS lesion status at data cutoff by study BICR in the overall FLAURA2 study population. (A): The median CNS PFS was 30.2 months (95% CI, 28.4 months to NC) with osimertinib plus platinum-pemetrexed and 27.6 months (95% CI, 22.1 months to NC) with osimertinib monotherapy. Censored data are indicated by tick marks. (C): CNS progression included patients who experienced progression in the CNS (by CNS BICR) and in other anatomies at the same overall visit; non-CNS progression included those with radiological documentation of progression outside the CNS only (overall study BICR); death included patients who died in the absence of radiological confirmed CNS or non-CNS progression. On the basis of this competing risk analysis, the estimated 24-month cumulative incidence rates with osimertinib plus platinum-pemetrexed versus osimertinib monotherapy were CNS progression, 9% versus 23% and non-CNS progression, 17% versus 50%. The 24-month cumulative incidence rates for death were NC (as seen in the figure, no death events at 24 months or later contributed to the competing risk analysis for the CNS PFS). ^aCR indicates those patients with a CR that was maintained at the time of data cutoff. ^bAmong patients included in the cFAS (ie, those with ≥one BICR CNS lesion at baseline), 10 patients in the osimertinib plus platinum-pemetrexed arm and five patients in the osimertinib monotherapy arm had no postbaseline assessment by BICR. BICR, blinded independent central review; cEFR, CNS evaluable-for-response set; cFAS, CNS full analysis set; CR, complete response; HR, hazard ratio; NC, not calculable; PD, progressive disease; PFS, progression-free survival; PR, partial response; SD, stable disease.

FIG 3.

Best percentage change from baseline in CNS target lesion size with (A) osimertinib plus platinum-pemetrexed and (B) osimertinib monotherapy (cFAS). ^aTwo patients had ≥one measurable CNS lesion at baseline by CNS BICR but died before the follow-up CNS BICR scan. (A) and (B): This evaluation only includes patients with measurable (≥10 mm) baseline CNS lesions and ≥one follow-up assessment. BICR, blinded independent central review; cFAS, CNS full analysis set; CR, complete response; PD, progressive disease; PR, partial response; SD, stable disease.

FIG A1.

FLAURA2 study schema and scan schedule. The FLAURA2 study schema for the overall population (full analysis set) is shown, including key inclusion criteria, randomly assigned treatment, and primary and key secondary end points. Additional detail on inclusion criteria for patients with CNS metastases is provided, alongside details on follow-up. The tumor assessment schedule per RECIST 1.1 for systemic assessments in all patients is shown. The brain scan schedule is shown for patients with and without CNS metastases at baseline (as assessed by investigator). Brain scans were subsequently assessed by neuroradiologist BICR per modified RECIST 1.1. ^aPemetrexed maintenance continued until a discontinuation criterion was met. ^bRECIST 1.1-defined radiological PD was by investigator assessment. AE, adverse event; BICR, blinded independent central review; CNS, central nervous system; CTCAE, Common Terminology Criteria for AEs; DCR, disease control rate; DoR, duration of response; EGFR, epidermal growth factor receptor; EGFRm, EGFR-mutated; Ex19del, exon 19 deletion; HRQoL, health-related quality of life; MRI, magnetic resonance imaging; NSCLC, non–small-cell lung cancer; ORR, objective response rate; OS, overall survival; PD, progressive disease; PFS, progression-free survival; PFS2, second progression-free survival; PS, performance status; pts, patients.

FIG A2.

Overlap between patients with CNS metastases at baseline per investigator and those included in the cFAS (≥one BICR CNS lesion at baseline) in the overall FLAURA2 population. Most patients with CNS metastases at baseline per investigator were also included in the cFAS. Thirty-seven patients were not included in the cFAS but did have CNS metastases at baseline per investigator. Thirty-three patients were included in the cFAS but did not have CNS metastases at baseline per investigator. BICR, blinded independent central review; cFAS, CNS full analysis set.

FIG A3.

CNS PFS in patients without CNS metastases at baseline by BICR. Censored data are indicated by tick marks. BICR, blinded independent central review; HR, hazard ratio; PFS, progression-free survival.

FIG A4.

Brain MRI scans from two patients receiving osimertinib plus carboplatin-pemetrexed who had durable CNS CRs in the brain. (A) Baseline images show three lesions in the right precentral gyrus with associated vasogenic edema. Leptomeningeal disease was most prominent in the posterior fossa evidenced by enhancement in the bilateral cerebellar subarachnoid spaces. At week 6, a >90% reduction in tumor size, resolution of vasogenic edema, and reduction in leptomeningeal enhancement was observed. By week 12, CR was recorded in the brain (not shown). Images at week 108 (>2 years) showed CR was maintained until data cutoff. Additional details for this patient are provided in Appendix 2. (B) The baseline image shows an enhancing nodule (<10 mm) in the right cerebellar subarachnoid space, indicating leptomeningeal disease; this was considered a nontarget lesion. The image at 3.0 months/13 weeks shows that the lesion had not increased in size; as a nontarget lesion, this was classed as a response of non-CR/non-PD. The image at 5.3 months/23 weeks shows that the lesion had resolved because there was no residual enhancing lesion; this was classed as a CR. The images at 8.5 months/37 weeks and 27.6 months/120 weeks (>2 years; last follow-up) show that the lesion remained resolved as there was no recurrent enhancing lesion; the response was classed as a CR. CR, complete response; MRI, magnetic resonance imaging; PD, progressive disease.