INTRODUCTION: Immunotherapy targeting the programmed cell death protein–1 (PD-1) axis elicits durable antitumor responses in multiple cancer types. However, clinical responses vary, and biomarkers predictive of response may help to identify patients who will derive the greatest therapeutic benefit. Clinically validated biomarkers predictive of response to the anti–PD-1 monoclonal antibody pembrolizumab include PD-1 ligand 1 (PD-L1) expression in specific cancers and high microsatellite instability (MSI-H) regardless of tumor type. Tumor mutational burden (TMB) and T cell–inflamed gene expression profile (GEP) are emerging predictive biomarkers for pembrolizumab. Both PD-L1 and GEP are inflammatory biomarkers indicative of a T cell–inflamed tumor microenvironment (TME), whereas TMB and MSI-H are indirect measures of tumor antigenicity generated by somatic tumor mutations. However, the relationship between these two categories of biomarkers is not well characterized. RATIONALE: This study assessed the potential for TMB and a T cell–inflamed GEP to jointly predict clinical response to pembrolizumab in >300 patient samples with advanced solid tumors and melanoma across 22 tumor types from four KEYNOTE clinical trials. To assess the individual and joint clinical utility of TMB and GEP, patients were stratified in four biomarker–defined clinical response groups [GEP low and TMB low (GEPlo TMBlo), GEP low and TMB high (GEPlo TMBhi), GEPhi TMBlo, and GEPhi TMBhi] based on predefined cutoffs for TMB and GEP. These patient–defined biomarker groups were further used to guide transcriptome and exome analyses of tumors in a large molecular database [The Cancer Genome Atlas (TCGA)] (n = 6384 tumors) to identify targetable patterns of biology that may modulate response and resistance. RESULTS: TMB and GEP exhibited only modest correlation and were independently predictive of response across the KEYNOTE clinical datasets. We found that objective response rates were strongest in patients with GEPhi TMBhi (37 to 57%), moderate in those with GEPhi TMBlo (12 to 35%) and GEPlo TMBhi (11 to 42%), and reduced or absent in those with GEPlo TMBlo (0 to 9%) (see the figure). Additionally, longer progression–free survival times were seen in patients with higher levels of both TMB and GEP. Findings were comparable when TMB and PD-L1 expression were jointly assessed. Within TCGA database, GEP and TMB again had a low correlation, demonstrating the potential to jointly stratify transcriptomic and genomic features across cancer types. Specific gene expression patterns reflective of TME biology showed significant associations with TMB, GEP, or both. In particular, gene set enrichment analysis identified proliferative and stromal, myeloid, and vascular biology corresponding to specific TMB-defined subgroups within GEPhi tumors. In TMBhi tumors, indication-dependent somatic DNA alterations in key cancer driver genes showed a strong negative association with GEP. CONCLUSION: This analysis shows that TMB and inflammatory biomarkers (T cell–in...

Clinical studies support the efficacy of programmed cell death 1 (PD-1) targeted therapy in a subset of patients with metastatic gastric cancer (mGC). With the goal of identifying determinants of response, we performed molecular characterization of tissues and circulating tumor DNA (ctDNA) from 61 patients with mGC who were treated with pembrolizumab as salvage treatment in a prospective phase 2 clinical trial. In patients with microsatellite instability-high and Epstein-Barr virus-positive tumors, which are mutually exclusive, dramatic responses to pembrolizumab were observed (overall response rate (ORR) 85.7% in microsatellite instability-high mGC and ORR 100% in Epstein-Barr virus-positive mGC). For the 55 patients for whom programmed death-ligand 1 (PD-L1) combined positive score positivity was available (combined positive score cut-off value ≥1%), ORR was significantly higher in PD-L1(+) gastric cancer when compared to PD-L1(-) tumors (50.0% versus 0.0%, P value <0.001). Changes in ctDNA levels at six weeks post-treatment predicted response and progression-free survival, and decreased ctDNA was associated with improved outcomes. Our findings provide insight into the molecular features associated with response to pembrolizumab in patients with mGC and provide biomarkers potentially relevant for the selection of patients who may derive greater benefit from PD-1 inhibition.

The widespread emergence of methicillin-resistant Staphylococcus aureus (MRSA) has dramatically eroded the efficacy of current β-lactam antibiotics and created an urgent need for new treatment options. We report an S. aureus phenotypic screening strategy involving chemical suppression of the growth inhibitory consequences of depleting late-stage wall teichoic acid biosynthesis. This enabled us to identify early-stage pathway-specific inhibitors of wall teichoic acid biosynthesis predicted to be chemically synergistic with β-lactams. We demonstrated by genetic and biochemical means that each of the new chemical series discovered, herein named tarocin A and tarocin B, inhibited the first step in wall teichoic acid biosynthesis (TarO). Tarocins do not have intrinsic bioactivity but rather demonstrated potent bactericidal synergy in combination with broad-spectrum β-lactam antibiotics against diverse clinical isolates of methicillin-resistant staphylococci as well as robust efficacy in a murine infection model of MRSA. Tarocins and other inhibitors of wall teichoic acid biosynthesis may provide a rational strategy to develop Gram-positive bactericidal β-lactam combination agents active against methicillin-resistant staphylococci.

A number of machine learning-based predictors have been developed for identifying immunogenic T-cell epitopes based on major histocompatibility complex (MHC) class I and II binding affinities. Rationally selecting the most appropriate tool has been complicated by the evolving training data and machine learning methods. Despite the recent advances made in generating high-quality MHC-eluted, naturally processed ligandome, the reliability of new predictors on these epitopes has yet to be evaluated. This study reports the latest benchmarking on an extensive set of MHC-binding predictors by using newly available, untested data of both synthetic and naturally processed epitopes. 32 human leukocyte antigen (HLA) class I and 24 HLA class II alleles are included in the blind test set. Artificial neural network (ANN)-based approaches demonstrated better performance than regression-based machine learning and structural modeling. Among the 18 predictors benchmarked, ANN-based mhcflurry and nn_align perform the best for MHC class I 9-mer and class II 15-mer predictions, respectively, on binding/non-binding classification (Area Under Curves = 0.911). NetMHCpan4 also demonstrated comparable predictive power. Our customization of mhcflurry to a pan-HLA predictor has achieved similar accuracy to NetMHCpan. The overall accuracy of these methods are comparable between 9-mer and 10-mer testing data. However, the top methods deliver low correlations between the predicted versus the experimental affinities for strong MHC binders. When used on naturally processed MHC-ligands, tools that have been trained on elution data (NetMHCpan4 and MixMHCpred) shows better accuracy than pure binding affinity predictor. The variability of false prediction rate is considerable among HLA types and datasets. Finally, structure-based predictor of Rosetta FlexPepDock is less optimal compared to the machine learning approaches. With our benchmarking of MHC-binding and MHC-elution predictors using a comprehensive metrics, a unbiased view for establishing best practice of T-cell epitope predictions is presented, facilitating future development of methods in immunogenomics.

Many important cellular processes are performed by molecular machines, composed of multiple proteins that physically interact to execute biological functions. An example is the bacterial peptidoglycan (PG) synthesis machine, responsible for the synthesis of the main component of the cell wall and the target of many contemporary antibiotics. One approach for the identification of essential components of a cellular machine involves the determination of its minimal protein composition. Staphylococcus aureus is a Gram-positive pathogen, renowned for its resistance to many commonly used antibiotics and prevalence in hospitals. Its genome encodes a low number of proteins with PG synthesis activity (9 proteins), when compared to other model organisms, and is therefore a good model for the study of a minimal PG synthesis machine. We deleted seven of the nine genes encoding PG synthesis enzymes from the S. aureus genome without affecting normal growth or cell morphology, generating a strain capable of PG biosynthesis catalyzed only by two penicillin-binding proteins, PBP1 and the bi-functional PBP2. However, multiple PBPs are important in clinically relevant environments, as bacteria with a minimal PG synthesis machinery became highly susceptible to cell wall-targeting antibiotics, host lytic enzymes and displayed impaired virulence in a Drosophila infection model which is dependent on the presence of specific peptidoglycan receptor proteins, namely PGRP-SA. The fact that S. aureus can grow and divide with only two active PG synthesizing enzymes shows that most of these enzymes are redundant in vitro and identifies the minimal PG synthesis machinery of S. aureus. However a complex molecular machine is important in environments other than in vitro growth as the expendable PG synthesis enzymes play an important role in the pathogenicity and antibiotic resistance of S. aureus.