MATERIALS AND METHODS: A literature search was carried out to gather eligible studies from the following widely sourced electronic databases such as Scopus, PubMed and Google Scholar using the combination of the following keywords: AD, MRS, brain metabolites, deep learning (DL), machine learning (ML) and artificial intelligence (AI); having the aim of taking the readers through the advancements in the usage of MRS analysis and related AI applications for the detection of AD.

RESULTS: We elaborate on the MRS data acquisition, processing, analysis, and interpretation techniques. Recommendation is made for MRS parameters that can obtain the best quality spectrum for fingerprinting the brain metabolomics composition in AD. Furthermore, we summarise ML and DL techniques that have been utilised to estimate the uncertainty in the machine-predicted metabolite content, as well as streamline the process of displaying results of metabolites derangement that occurs as part of ageing.

METHODS: Treatment-naive patients with EGFR-mutated advanced NSCLC were randomized one-to-one to lazertinib (240 mg/d) or gefitinib (250 mg/d). Patients with asymptomatic or stable CNS metastases were included if any planned radiation, surgery, or steroids were completed more than 2 weeks before randomization. For patients with CNS metastases confirmed at screening or subsequently suspected, CNS imaging was performed every 6 weeks for 18 months, then every 12 weeks. End points assessed by blinded independent central review and Response Evaluation Criteria in Solid Tumors version 1.1 included intracranial progression-free survival, intracranial objective response rate, and intracranial duration of response.

RESULTS: Of the 393 patients enrolled in LASER301, 86 (lazertinib, n = 45; gefitinib, n = 41) had measurable and or non-measurable baseline CNS metastases. The median intracranial progression-free survival in the lazertinib group was 28.2 months (95% confidence interval [CI]: 14.8-28.2) versus 8.4 months (95% CI: 6.7-not reached [NR]) in the gefitinib group (hazard ratio = 0.42, 95% CI: 0.20-0.89, p = 0.02). Among patients with measurable CNS lesions, the intracranial objective response rate was numerically higher with lazertinib (94%; n = 17) versus gefitinib (73%; n = 11, p = 0.124). The median intracranial duration of response with lazertinib was NR (8.3-NR) versus 6.3 months (2.8-NR) with gefitinib. Tolerability was similar to the overall LASER301 population.

PATIENTS AND METHODS: Patients were 18 years and older with no previous systemic anticancer therapy. Neurologically stable patients with CNS metastases were allowed. Patients were randomly assigned 1:1 to lazertinib 240 mg once daily orally or gefitinib 250 mg once daily orally, stratified by mutation status and race. The primary end point was investigator-assessed progression-free survival (PFS) by RECIST v1.1.

RESULTS: Overall, 393 patients received double-blind study treatment across 96 sites in 13 countries. Median PFS was significantly longer with lazertinib than with gefitinib (20.6 v 9.7 months; hazard ratio [HR], 0.45; 95% CI, 0.34 to 0.58; P < .001). The PFS benefit of lazertinib over gefitinib was consistent across all predefined subgroups. The objective response rate was 76% in both groups (odds ratio, 0.99; 95% CI, 0.62 to 1.59). Median duration of response was 19.4 months (95% CI, 16.6 to 24.9) with lazertinib versus 8.3 months (95% CI, 6.9 to 10.9) with gefitinib. Overall survival data were immature at the interim analysis (29% maturity). The 18-month survival rate was 80% with lazertinib and 72% with gefitinib (HR, 0.74; 95% CI, 0.51 to 1.08; P = .116). Observed safety of both treatments was consistent with their previously reported safety profiles.