Materials Genes of CO2 Hydrogenation on Supported Cobalt Catalysts: An Artificial Intelligence Approach Integrating Theoretical and Experimental Data.

Designing materials for catalysis is challenging because the performance is governed by an intricate interplay of various multiscale phenomena, such as the chemical reactions on surfaces and the materials' restructuring during the catalytic process. In the case of supported catalysts, the role of the support material can be also crucial. Here, we address this intricacy challenge by a symbolic-regression artificial intelligence (AI) approach. We identify the key physicochemical parameters correlated with the measured performance, out of many offered candidate parameters characterizing the materials, reaction environment, and possibly relevant underlying phenomena. Importantly, these parameters are obtained by both experiments and ab initio simulations. The identified key parameters might be called "materials genes", in analogy to genes in biology: they correlate with the property or function of interest, but the explicit physical relationship is not (necessarily) known. To demonstrate the approach, we investigate the CO2 hydrogenation catalyzed by cobalt nanoparticles supported on silica. Crucially, the silica support is modified with the additive metals magnesium, calcium, titanium, aluminum, or zirconium, which results in six materials with significantly different performances. These systems mimic hydrothermal vents, which might have produced the first organic molecules on Earth. The key parameters correlated with the CH3OH selectivity reflect the reducibility of cobalt species, the adsorption strength of reaction intermediates, and the chemical nature of the additive metal. By using an AI model trained on basic elemental properties of the additive metals (e.g., ionization potential) as physicochemical parameters, new additives are suggested. The predicted CH3OH selectivity of cobalt catalysts supported on silica modified with vanadium and zinc is confirmed by new experiments.

Self-interaction corrected SCAN functional for molecules and solids in the numeric atom-center orbital framework.

Semilocal density-functional approximations (DFAs), including the state-of-the-art SCAN functional, are plagued by the self-interaction error (SIE). While this error is explicitly defined only for one-electron systems, it has inspired the self-interaction correction method proposed by Perdew and Zunger (PZ-SIC), which has shown promise in mitigating the many-electron SIE. However, the PZ-SIC method is known for its significant numerical instability. In this study, we introduce a novel constraint that facilitates self-consistent localization of the SIC orbitals in the spirit of Edmiston-Ruedenberg orbitals [Rev. Mod. Phys. 35, 457 (1963)]. Our practical implementation within the all-electron numeric atom-centered orbitals code FHI-aims guarantees efficient and stable convergence of the self-consistent PZ-SIC equations for both molecules and solids. We further demonstrate that our PZ-SIC approach effectively mitigates the SIE in the meta-generalized gradient approximation SCAN functional, significantly improving the accuracy for ionization potentials, charge-transfer energies, and bandgaps for a diverse selection of molecules and solids. However, our PZ-SIC method does have its limitations. It cannot improve the already accurate SCAN results for properties such as cohesive energies, lattice constants, and bulk modulus in our test sets. This highlights the need for new-generation DFAs with more comprehensive applicability.

MET Fusions in NSCLC: Clinicopathologic Features and Response to MET Inhibition.

INTRODUCTION: MET fusions have been described only rarely in NSCLC. Thus, data on patient characteristics and treatment response are limited. We here report histopathologic data, patient demographics, and treatment outcome including response to MET tyrosine kinase inhibitor (TKI) therapy in MET fusion-positive NSCLC. METHODS: Patients with NSCLC and MET fusions were identified mostly by RNA sequencing within the routine molecular screening program of the national Network Genomic Medicine, Germany. RESULTS: We describe a cohort of nine patients harboring MET fusions. Among these nine patients, two patients had been reported earlier. The overall frequency was 0.29% (95% confidence interval: 0.15-0.55). The tumors were exclusively adenocarcinoma. The cohort was heterogeneous in terms of age, sex, or smoking status. We saw five different fusion partner genes (KIF5B, TRIM4, ST7, PRKAR2B, and CAPZA2) and several different breakpoints. Four patients were treated with a MET TKI leading to two partial responses, one stable disease, and one progressive disease. One patient had a BRAF V600E mutation as acquired resistance mechanism. CONCLUSIONS: MET fusions are very rare oncogenic driver events in NSCLC and predominantly seem in adenocarcinomas. They are heterogeneous in terms of fusion partners and breakpoints. Patients with MET fusion can benefit from MET TKI therapy.

Association Between Lung Immune Prognostic Index and Durvalumab Consolidation Outcomes in Patients With Locally Advanced Non-Small-Cell Lung Cancer.

INTRODUCTION: The LIPI, based on pretreatment derived neutrophils/[leukocytes-neutrophils] ratio (dNLR) and LDH, is associated with immune checkpoint inhibitors (ICI) outcomes in advanced non-small-cell lung cancer (NSCLC). We aimed to assess baseline LIPI correlation with durvalumab consolidation outcomes in the locally advanced setting. MATERIAL AND METHODS: Multicentre retrospective study (330 patients) with stage III unresectable NSCLC treated with durvalumab after chemo-radiotherapy between April 2015 and December 2020; 65 patients treated with chemo-radiotherapy only. Baseline LIPI characterized 3 groups: good (dNLR≤3+LDH≤ULN), intermediate (dNLR>3/LDH>ULN) and poor (dNLR>3+LDH>ULN). Primary endpoint was overall survival (OS). RESULTS: In the durvalumab cohort, median age was 67 years, 95% smokers, 98% with a performance status of 0-1; 60% had nonsquamous histology and 16% a PD-L1 expression <1%. Radiotherapy was delivered concurrently in 81%. LIPI was evaluable in 216 patients: 66% good, 31% intermediate, 3% poor. LIPI significantly correlated with median OS (median follow-up: 19 months): 18.1 months vs. 47.0 months vs. not reached in poor, intermediate and good LIPI groups, respectively (P = .03). A trend between objective response rate and LIPI groups was observed: 0% vs. 41% vs. 45%, respectively (P = .05). The pooled intermediate/poor LIPI group was associated with shorter OS (HR 1.97; P = .03) and higher risk of progressive disease (OR 2.68; P = .047). Survivals and response were not influenced in the control cohort. CONCLUSION: Baseline LIPI correlated with outcomes in patients with locally advanced NSCLC treated with durvalumab consolidation, but not in those who only received chemo-radiotherapy, providing further evidence of its prognostic and potential predictive role of ICI benefit in NSCLC.

Recent advances in the SISSO method and their implementation in the SISSO++ code.

Accurate and explainable artificial-intelligence (AI) models are promising tools for accelerating the discovery of new materials. Recently, symbolic regression has become an increasingly popular tool for explainable AI because it yields models that are relatively simple analytical descriptions of target properties. Due to its deterministic nature, the sure-independence screening and sparsifying operator (SISSO) method is a particularly promising approach for this application. Here, we describe the new advancements of the SISSO algorithm, as implemented into SISSO++, a C++ code with Python bindings. We introduce a new representation of the mathematical expressions found by SISSO. This is a first step toward introducing "grammar" rules into the feature creation step. Importantly, by introducing a controlled nonlinear optimization to the feature creation step, we expand the range of possible descriptors found by the methodology. Finally, we introduce refinements to the solver algorithms for both regression and classification, which drastically increase the reliability and efficiency of SISSO. For all these improvements to the basic SISSO algorithm, we not only illustrate their potential impact but also fully detail how they operate both mathematically and computationally.

Clinicopathologic and molecular characteristics of small-scale ROS1-mutant non-small cell lung cancer (NSCLC) patients.

BACKGROUND: ROS1 fusions are well treatable aberrations in NSCLC. Besides solvent-front mutations (SFM) in resistance to targeted therapy, small-scale ROS1 mutations are largely unknown. We exploratively analyzed the clinical and molecular characteristics of small-scale ROS1 mutations in NSCLC patients without activating ROS1 fusions or SFMs. METHODS: Next-generation sequencing was performed on tissue samples from NSCLC patients within the Network Genomic Medicine. Patients with ROS1 fusions and SFMs were excluded. We analyzed clinical characteristics of patients harboring small-scale ROS1-mutations, ROS1- and co-occurring mutations, and their response to systemic therapy. RESULTS: Of 10,396 patients analyzed, 101 (1.0%) patients harbored small-scale ROS1 mutations. Most patients were male (73.3%) and smokers (96.6%). Nearly half of the patients presented with squamous-cell carcinoma (SqCC, 40.4%). Most mutations were transversions (50.5%), and 66% were in the kinase domain. Besides TP53 mutations (65.3%), KRAS (22.8%), EGFR (5.9%), PIK3CA (9.9%) and FGFR1-4 mutations (8.9%) co-occurred. In 10 (9.9%) patients, ROS1 mutation was the only aberration detected. Median overall survival (mOS) differed significantly in patients with or without KRAS co-mutations (9.7 vs 21.5 months, p = 0.02) and in patients treated with or without immune-checkpoint blockade (ICB) during treatment (21.5 vs 4.4 months, p = 0.003). CONCLUSION: The cohort's clinical characteristics contrasted ROS1-fused cohorts. Co-occurrence of KRAS mutations led to shortened survival and patients benefited from ICB. Our data does not support the idea of ROS1 small-scale mutations as strong oncogenic drivers in NSCLC, but rather as relevant bystanders altering the efficacy of treatment approaches.

KEAP1/NFE2L2 Pathway Signature Outperforms KEAP1/NFE2L2 Mutation Status and Reveals Alternative Pathway-Activating Mutations in NSCLC.

INTRODUCTION: Activation of the antioxidant KEAP1/NFE2L2 (NRF2) pathway leads to increased glutamine dependence and an aggressive phenotype in NSCLC. Because this pathway has been explored as a clinical target, we developed a transcriptomic signature for identifying KEAP1/NFE2L2-activated tumors. METHODS: A total of 971 NSCLC samples were used to train an expression signature (K1N2-score) to predict KEAP1/NFE2L2 mutations. There were 348 in-house NSCLCs that were analyzed using a NanoString expression panel for validation. RESULTS: The 46-gene K1N2 score robustly predicted KEAP1/NFE2L2 mutations in the validation set irrespective of histology and mutation (area under the curve: 89.5, sensitivity: 90.2%), suggesting that approximately 90% of KEAP1/NFE2L2 mutations are pathway-activating. The K1N2-score outperformed KEAP1/NFE2L2 mutational status when predicting patient survival (score p = 0.047; mutation p = 0.215). In K1N2 score-positive but KEAP1/NFE2L2 wild-type samples, enrichment testing identified SMARCA4/BRG1 and CUL3 mutations as mimics of KEAP1/NFE2L2 mutations. CONCLUSIONS: The K1N2-score identified KEAP1/NFE2L2-activated NSCLC by robustly detecting KEAP1/NFE2L2mut cases and discovering alternative genomic activators. It is a potential means for selecting patients with a constitutively active KEAP1/NFE2L2 pathway.

Anharmonicity in Thermal Insulators: An Analysis from First Principles.

The anharmonicity of atomic motion limits the thermal conductivity in crystalline solids. However, a microscopic understanding of the mechanisms active in strong thermal insulators is lacking. In this Letter, we classify 465 experimentally known materials with respect to their anharmonicity and perform fully anharmonic ab initio Green-Kubo calculations for 58 of them, finding 28 thermal insulators with κ<10 W/mK including 6 with ultralow κ≲1 W/mK. Our analysis reveals that the underlying strong anharmonic dynamics is driven by the exploration of metastable intrinsic defect geometries. This is at variance with the frequently applied perturbative approach, in which the dynamics is assumed to evolve around a single stable geometry.

Comutations and KRASG12C Inhibitor Efficacy in Advanced NSCLC.

Molecular modifiers of KRASG12C inhibitor (KRASG12Ci) efficacy in advanced KRASG12C-mutant NSCLC are poorly defined. In a large unbiased clinicogenomic analysis of 424 patients with non-small cell lung cancer (NSCLC), we identified and validated coalterations in KEAP1, SMARCA4, and CDKN2A as major independent determinants of inferior clinical outcomes with KRASG12Ci monotherapy. Collectively, comutations in these three tumor suppressor genes segregated patients into distinct prognostic subgroups and captured ∼50% of those with early disease progression (progression-free survival ≤3 months) with KRASG12Ci. Pathway-level integration of less prevalent coalterations in functionally related genes nominated PI3K/AKT/MTOR pathway and additional baseline RAS gene alterations, including amplifications, as candidate drivers of inferior outcomes with KRASG12Ci, and revealed a possible association between defective DNA damage response/repair and improved KRASG12Ci efficacy. Our findings propose a framework for patient stratification and clinical outcome prediction in KRASG12C-mutant NSCLC that can inform rational selection and appropriate tailoring of emerging combination therapies. SIGNIFICANCE: In this work, we identify co-occurring genomic alterations in KEAP1, SMARCA4, and CDKN2A as independent determinants of poor clinical outcomes with KRASG12Ci monotherapy in advanced NSCLC, and we propose a framework for patient stratification and treatment personalization based on the comutational status of individual tumors. See related commentary by Heng et al., p. 1513. This article is highlighted in the In This Issue feature, p. 1501.

Role of circulating-tumor DNA in the early-stage non-small cell lung carcinoma as a predictive biomarker.

Lung cancer is one of the most common solid malignancies. Tissue biopsy is the standard method for accurately diagnosing lung and many other malignancies over decades. However, molecular profiling of tumors leads to establishing a new horizon in the field of precision medicine, which has now entered the mainstream in clinical practice. In this context, a minimally invasive complementary method has been proposed as a liquid biopsy (LB) which is a blood-based test that is gaining popularity as it provides the opportunity to test genotypes in a unique, less invasive manner. Circulating tumor cells (CTC) captivating the Circulating-tumor DNA (Ct-DNA) are often present in the blood of lung cancer patients and are the fundamental concept behind LB. There are multiple clinical uses of Ct-DNA, including its role in prognostic and therapeutic purposes. The treatment of lung cancer has drastically evolved over time. Therefore, this review article mainly focuses on the current literature on circulating tumor DNA and its clinical implications and future goals in non-small cell lung cancer.

Osimertinib and Selpercatinib Efficacy, Safety, and Resistance in a Multicenter, Prospectively Treated Cohort of EGFR-Mutant and RET Fusion-Positive Lung Cancers.

PURPOSE: Acquired RET fusions have been reported at resistance to treatment with EGFR inhibitors in EGFR-mutant non-small cell lung cancer (NSCLC); however, a multicenter cohort of patients with EGFR-mutant lung cancers treated with osimertinib and selpercatinib for RET fusion-mediated osimertinib resistance has not previously been published. PATIENTS AND METHODS: Patients who received selpercatinib in combination with osimertinib on a prospective expanded access clinical trial (NCT03906331) and single-patient compassionate use programs across five countries were centrally analyzed. All patients had advanced EGFR-mutant NSCLC with a RET fusion detected from tissue or plasma following osimertinib therapy. Clinicopathologic and outcomes data were collected. RESULTS: Fourteen patients with EGFR-mutant and RET fusion-positive lung cancers who experienced prior progression on osimertinib received osimertinib and selpercatinib. EGFR exon 19 deletions (±T790M, 86%) and non-KIF5B fusions (CCDC6-RET 50%, NCOA4-RET 36%) predominated. Osimertinib 80 mg daily and selpercatinib 80 mg twice daily were the most commonly administered dosages. The response rate, disease control rate, and median treatment duration were 50% [95% confidence interval (CI), 25%-75%, n = 12], 83% (95% CI, 55%-95%), and 7.9 months (range, 0.8-25+), respectively. Resistance was complex, involving EGFR on-target (EGFR C797S), RET on-target (RET G810S), and off-target (EML4-ALK/STRN-ALK, KRAS G12S, BRAF V600E) mechanisms; RET fusion loss; or polyclonal mechanisms. CONCLUSIONS: For patients with EGFR-mutant NSCLC with an acquired RET fusion as a mechanism of EGFR inhibitor resistance, the addition of selpercatinib to osimertinib was feasible and safe and offered clinical benefit, supporting the prospective evaluation of this combination. See related commentary by Krebs and Popat, p. 2951.

Data-Centric Heterogeneous Catalysis: Identifying Rules and Materials Genes of Alkane Selective Oxidation.

Artificial intelligence (AI) can accelerate catalyst design by identifying key physicochemical descriptive parameters correlated with the underlying processes triggering, favoring, or hindering the performance. In analogy to genes in biology, these parameters might be called "materials genes" of heterogeneous catalysis. However, widely used AI methods require big data, and only the smallest part of the available data meets the quality requirement for data-efficient AI. Here, we use rigorous experimental procedures, designed to consistently take into account the kinetics of the catalyst active states formation, to measure 55 physicochemical parameters as well as the reactivity of 12 catalysts toward ethane, propane, and n-butane oxidation reactions. These materials are based on vanadium or manganese redox-active elements and present diverse phase compositions, crystallinities, and catalytic behaviors. By applying the sure-independence-screening-and-sparsifying-operator symbolic-regression approach to the consistent data set, we identify nonlinear property-function relationships depending on several key parameters and reflecting the intricate interplay of processes that govern the formation of olefins and oxygenates: local transport, site isolation, surface redox activity, adsorption, and the material dynamical restructuring under reaction conditions. These processes are captured by parameters derived from N2 adsorption, X-ray photoelectron spectroscopy (XPS), and near-ambient-pressure in situ XPS. The data-centric approach indicates the most relevant characterization techniques to be used for catalyst design and provides "rules" on how the catalyst properties may be tuned in order to achieve the desired performance.

An Adjusted Treatment Comparison Comparing Amivantamab Versus Real-World Clinical Practice in Europe and the United States for Patients with Advanced Non-Small Cell Lung Cancer with Activating Epidermal Growth Factor Receptor Exon 20 Insertion Mutations.

INTRODUCTION: Patients with advanced, epidermal growth factor receptor (EGFR)-mutated, non-small cell lung cancer (NSCLC) with Exon 20 insertion mutations (Exon20ins) have poor prognoses, exacerbated by a previous lack of specific treatment guidelines and unmet need for targeted therapies. Amivantamab, an EGFR and MET bispecific antibody, demonstrated efficacy and tolerability in patients with advanced EGFR-mutated NSCLC with Exon20ins following platinum-based therapy in CHRYSALIS (NCT02609776; Cohort D+). Since CHRYSALIS was single-arm, individual patient data (IPD)-based adjusted analyses versus similar patients in real-world clinical practice (RWCP) were conducted to generate comparative evidence. METHODS: RWCP cohorts were derived from seven European and US real-world sources, comprising patients fulfilling CHRYSALIS Cohort D+ eligibility criteria. Amivantamab was compared with a basket of RWCP treatments. Differences in prognostic characteristics were adjusted for using inverse probability weighting (IPW; average treatment effect among the treated [ATT]). Balance between cohorts was assessed using standardized mean differences (SMDs). Overall response rate (ORR; investigator- [INV] and independent review committee-assessed [IRC]), overall survival (OS), progression-free survival (PFS; INV and IRC) and time-to-next treatment (TTNT) were compared. Binary and time-to-event endpoints were analyzed using weighted logistic regression and proportional hazards regression, respectively. RESULTS: Pre-adjustment, baseline characteristics were comparable between cohorts. IPW ATT-adjustment improved comparability, giving closely matched characteristics. ORR (INV) was 36.8% for amivantamab versus 17.0% for the adjusted EU + US cohort (response rate ratio [RR]: 2.16). Median OS, PFS (INV) and TTNT were 22.77 versus 12.52 months (hazard ratio [HR]: 0.47; p < 0.0001), 6.93 versus 4.17 months (HR: 0.55; p < 0.0001) and 12.42 versus 5.36 months (HR: 0.44; p < 0.0001) for amivantamab versus the adjusted EU + US cohort, respectively. Results were consistent versus EU- and US-only cohorts, and when using IRC assessment. CONCLUSION: Adjusted comparisons demonstrated significantly improved outcomes for amivantamab versus RWCP, highlighting the value of amivantamab in addressing unmet need in patients with advanced EGFR Exon20ins NSCLC following platinum-based therapy. TRIAL REGISTRATION: CHRYSALIS: NCT02609776.

Resistance to MET inhibition in MET-dependent NSCLC and therapeutic activity after switching from type I to type II MET inhibitors.

OBJECTIVES: Resistance to MET inhibition occurs inevitably in MET-dependent non-small cell lung cancer and the underlying mechanisms are insufficiently understood. We describe resistance mechanisms in patients with MET exon 14 skipping mutation (METΔex14), MET amplification, and MET fusion and report treatment outcomes after switching therapy from type I to type II MET inhibitors. MATERIALS AND METHODS: Pre- and post-treatment biopsies were analysed by NGS (next generation sequencing), digital droplet PCR (polymerase chain reaction), and FISH (fluorescense in situ hybridization). A patient-derived xenograft model was generated in one case. RESULTS: Of 26 patients with MET tyrosine kinase inhibitor treatment, eight had paired pre- and post-treatment biopsies (Three with MET amplification, three with METΔex14, two with MET fusions (KIF5B-MET and PRKAR2B-MET).) In six patients, mechanisms of resistance were detected, whereas in two cases, the cause of resistance remained unclear. We found off-target resistance mechanisms in four cases with KRAS mutations and HER2 amplifications appearing. Two patients exhibited second-site MET mutations (p.D1246N and p. Y1248H). Three patients received type I and type II MET tyrosine kinase inhibitors sequentially. In two cases, further progressive disease was seen hereafter. The patient with KIF5B-MET fusion received three different MET inhibitors and showed long-lasting stable disease and a repeated response after switching therapy, respectively. CONCLUSION: Resistance to MET inhibition is heterogeneous with on- and off-target mechanisms occurring regardless of the initial MET aberration. Switching therapy between different types of kinase inhibitors can lead to repeated responses in cases with second-site mutations. Controlled clinical trials in this setting with larger patient numbers are needed, as evidence to date is limited to preclinical data and case series.

Resection of a Large Growing Mediastinal Germ Cell Tumor Using a Multidisciplinary Approach.

Mediastinal germ cell tumors (GCTs) are rare. Post-chemotherapy residual masses in patients with a nonseminomatous GCT require resection. A patient with a large mediastinal GCT involving the left subclavian artery, superior vena cava (SVC) and hilum of the right lung is presented. Despite a biochemical response to chemotherapy, the tumor enlarged on serial imaging. With guidance from medical oncology, a multidisciplinary surgical team, including cardiac anesthesia, cardiac surgery and thoracic surgery resected the tumor with a staged reconstruction of the SVC. The procedure was well tolerated and yielded clear margins. The final pathology showed a significant associated component of rhabdomyosarcoma.

Advancing Critical Chemical Processes for a Sustainable Future: Challenges for Industry and the Max Planck-Cardiff Centre on the Fundamentals of Heterogeneous Catalysis (FUNCAT).

Catalysis is involved in around 85 % of manufacturing industry and contributes an estimated 25 % to the global domestic product, with the majority of the processes relying on heterogeneous catalysis. Despite the importance in different global segments, the fundamental understanding of heterogeneously catalysed processes lags substantially behind that achieved in other fields. The newly established Max Planck-Cardiff Centre on the Fundamentals of Heterogeneous Catalysis (FUNCAT) targets innovative concepts that could contribute to the scientific developments needed in the research field to achieve net zero greenhouse gas emissions in the chemical industries. This Viewpoint Article presents some of our research activities and visions on the current and future challenges of heterogeneous catalysis regarding green industry and the circular economy by focusing explicitly on critical processes. Namely, hydrogen production, ammonia synthesis, and carbon dioxide reduction, along with new aspects of acetylene chemistry.

Hierarchical Symbolic Regression for Identifying Key Physical Parameters Correlated with Bulk Properties of Perovskites.

Symbolic regression identifies nonlinear, analytical expressions relating materials properties and key physical parameters. However, the pool of expressions grows rapidly with complexity, compromising its efficiency. We tackle this challenge hierarchically: identified expressions are used as inputs for further obtaining more complex expressions. Crucially, this framework can transfer knowledge among properties, as demonstrated using the sure-independence-screening-and-sparsifying-operator approach to identify expressions for lattice constant and cohesive energy, which are then used to model the bulk modulus of ABO_{3} perovskites.