Jong‐Seok Lee

PURPOSE Non–small-cell lung cancer (NSCLC) with epidermal growth factor receptor ( EGFR) exon 20 insertion (Exon20ins) mutations exhibits inherent resistance to approved tyrosine kinase inhibitors. Amivantamab, an EGFR-MET bispecific antibody with immune cell–directing activity, binds to each receptor's extracellular domain, bypassing resistance at the tyrosine kinase inhibitor binding site. METHODS CHRYSALIS is a phase I, open-label, dose-escalation, and dose-expansion study, which included a population with EGFR Exon20ins NSCLC. The primary end points were dose-limiting toxicity and overall response rate. We report findings from the postplatinum EGFR Exon20ins NSCLC population treated at the recommended phase II dose of 1,050 mg amivantamab (1,400 mg, ≥ 80 kg) given once weekly for the first 4 weeks and then once every 2 weeks starting at week 5. RESULTS In the efficacy population (n = 81), the median age was 62 years (range, 42-84 years); 40 patients (49%) were Asian, and the median number of previous lines of therapy was two (range, 1-7). The overall response rate was 40% (95% CI, 29 to 51), including three complete responses, with a median duration of response of 11.1 months (95% CI, 6.9 to not reached). The median progression-free survival was 8.3 months (95% CI, 6.5 to 10.9). In the safety population (n = 114), the most common adverse events were rash in 98 patients (86%), infusion-related reactions in 75 (66%), and paronychia in 51 (45%). The most common grade 3-4 adverse events were hypokalemia in six patients (5%) and rash, pulmonary embolism, diarrhea, and neutropenia in four (4%) each. Treatment-related dose reductions and discontinuations were reported in 13% and 4% of patients, respectively. CONCLUSION Amivantamab, via its novel mechanism of action, yielded robust and durable responses with tolerable safety in patients with EGFR Exon20ins mutations after progression on platinum-based chemotherapy.

Due to their increasing spread, confidence in neural network predictions became more and more important. However, basic neural networks do not deliver certainty estimates or suffer from over or under confidence. Many researchers have been working on understanding and quantifying uncertainty in a neural network's prediction. As a result, different types and sources of uncertainty have been identified and a variety of approaches to measure and quantify uncertainty in neural networks have been proposed. This work gives a comprehensive overview of uncertainty estimation in neural networks, reviews recent advances in the field, highlights current challenges, and identifies potential research opportunities. It is intended to give anyone interested in uncertainty estimation in neural networks a broad overview and introduction, without presupposing prior knowledge in this field. A comprehensive introduction to the most crucial sources of uncertainty is given and their separation into reducible model uncertainty and not reducible data uncertainty is presented. The modeling of these uncertainties based on deterministic neural networks, Bayesian neural networks, ensemble of neural networks, and test-time data augmentation approaches is introduced and different branches of these fields as well as the latest developments are discussed. For a practical application, we discuss different measures of uncertainty, approaches for the calibration of neural networks and give an overview of existing baselines and implementations. Different examples from the wide spectrum of challenges in different fields give an idea of the needs and challenges regarding uncertainties in practical applications. Additionally, the practical limitations of current methods for mission- and safety-critical real world applications are discussed and an outlook on the next steps towards a broader usage of such methods is given.

The literature on network effects has popularized a hypothesis that competition between incompatible technologies results in the “winner-take-all” outcome. For the survival of the firm in this sort of competition, the installed base has been emphasized. We argue that the validity of this hypothesis depends on how customers interact with one another (e.g., if they exchange advice or files). In some interaction networks, customers influenced by their acquaintances may adopt a lagging technology even when a lead technology has built a large installed base. The presence of such a local bias facilitates the persistence of incompatibilities. When local bias cannot be sustained in other interaction networks, one technology corners the market. Our study suggests that overemphasizing the installed base, while ignoring network structure, could mislead practitioners.

PURPOSE Epidermal growth factor receptor ( EGFR ) tyrosine kinase inhibitors (TKIs) are standard first-line therapy for EGFR -mutant, metastatic non–small cell lung cancer (NSCLC); however, most patients experience disease progression. We report results from the randomized, double-blind, phase III KEYNOTE-789 study of pemetrexed and platinum–based chemotherapy with or without pembrolizumab for TKI-resistant, EGFR -mutant, metastatic nonsquamous NSCLC (ClinicalTrials.gov identifier: NCT03515837 ). METHODS Adults with pathologically confirmed stage IV nonsquamous NSCLC, documented DEL19 or L858R EGFR mutation, and progression after EGFR-TKI treatment were randomly assigned 1:1 to 35 cycles of pembrolizumab 200 mg or placebo once every 3 weeks plus four cycles of pemetrexed and carboplatin or cisplatin once every 3 weeks and then maintenance pemetrexed. Dual primary end points were progression-free survival (PFS) and overall survival (OS). Final PFS testing was completed at the second interim analysis (IA2; data cutoff, December 3, 2021); OS was tested at final analysis (FA; data cutoff, January 17, 2023). Efficacy boundaries were one-sided P = .0117 for PFS and OS. RESULTS Four hundred ninety-two patients were randomly assigned to pembrolizumab plus chemotherapy (n = 245) or placebo plus chemotherapy (n = 247). At IA2, the median PFS was 5.6 months for pembrolizumab plus chemotherapy versus 5.5 months for placebo plus chemotherapy (hazard ratio [HR], 0.80 [95% CI, 0.65 to 0.97]; P = .0122). At FA, the median OS was 15.9 versus 14.7 months, respectively (HR, 0.84 [95% CI, 0.69 to 1.02]; P = .0362). Grade ≥3 treatment-related adverse events occurred in 43.7% of pembrolizumab plus chemotherapy recipients versus 38.6% of placebo plus chemotherapy recipients. CONCLUSION Addition of pembrolizumab to chemotherapy in patients with TKI-resistant, EGFR -mutant, metastatic nonsquamous NSCLC did not significantly prolong PFS or OS versus placebo plus chemotherapy in KEYNOTE-789.