Meta-analysis of tumor- and T cell-intrinsic mechanisms of sensitization to checkpoint inhibition.

Litchfield K, Reading JL, Puttick C, Thakkar K, Abbosh C, Bentham R, Watkins TBK, Rosenthal R, Biswas D, Rowan A, Lim E, Al Bakir M, Turati V, Guerra-Assunção JA, Conde L, Furness AJS, Saini SK, Hadrup SR, Herrero J, Lee SH, Van Loo P, Enver T, Larkin J, Hellmann MD, Turajlic S, Quezada SA, McGranahan N, Swanton C. · 2021

Read Meta-analysis of tumor- and T cell-intrinsic mechanisms of sensitization to checkpoint inhibition. by Litchfield K, Reading JL, Puttick C, Thakkar K, Abbosh C, Bentham R, Watkins TBK, Rosenthal R, Biswas D, Rowan A, Lim E, Al Bakir M, Turati V, Guerra-Assunção JA, Conde L, Furness AJS, Saini SK, Hadrup SR, Herrero J, Lee SH, Van Loo P, Enver T, Larkin J, Hellmann MD, Turajlic S, Quezada SA, McGranahan N, Swanton C. (2021) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc7933824.txt.

Introduction

Litchfield and co-authors report a meta-analysis of what makes tumours respond to checkpoint inhibitors. These drugs free the immune system to attack cancer. Some patients gain lasting benefit, while others gain none. The team pooled genetic and clinical data to weigh the rival explanations for that split, inside both the tumour and its T cells.

The data covered more than 1,000 treated patients across seven tumour types. Clonal mutation burden was the strongest predictor of response. Total mutation burden and one immune signal, CXCL9, followed. Several favoured markers failed to hold up across cancers. Two gene changes also stood out: one loss tied to response, one gain tied to resistance. Single-cell work named two further T-cell markers of sensitivity. The study gives trial designers a ranked list of what to measure.

Catalog record

CREATORLitchfield K, Reading JL, Puttick C, Thakkar K, Abbosh C, Bentham R, Watkins TBK, Rosenthal R, Biswas D, Rowan A, Lim E, Al Bakir M, Turati V, Guerra-Assunção JA, Conde L, Furness AJS, Saini SK, Hadrup SR, Herrero J, Lee SH, Van Loo P, Enver T, Larkin J, Hellmann MD, Turajlic S, Quezada SA, McGranahan N, Swanton C.
YEAR2021
COLLECTIONPAPERS
FORMATTXT
FILE SIZE132 KB
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC7933824

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Meta-analysis of tumor- and T cell-intrinsic mechanisms of sensitization to checkpoint inhibition By Kevin Litchfield, James L Reading, Clare Puttick, Krupa Thakkar, Chris Abbosh, Robert Bentham, Thomas BK Watkins, Rachel Rosenthal, Dhruva Biswas, Andrew Rowan, Emilia Lim, Maise Al Bakir Published in: Cell DOI: 10.1016/j.cell.2021.01.002 ABSTRACT Summary Checkpoint inhibitors (CPIs) augment adaptive immunity. Systematic pan-tumor analyses may reveal the relative importance of tumor-cell-intrinsic and microenvironmental features underpinning CPI sensitization. Here, we collated whole-exome and transcriptomic data for >1,000 CPI-treated patients across seven tumor types, utilizing standardized bioinformatics workflows and clinical outcome criteria to validate multivariable predictors of CPI sensitization. Clonal tumor mutation burden (TMB) was the strongest predictor of CPI response, followed by total TMB and CXCL9 expression. Subclonal TMB, somatic copy alteration burden, and histocompatibility leukocyte antigen (HLA) evolutionary divergence failed to attain pan-cancer significance. Dinucleotide variants were identified as a source of immunogenic epitopes associated with radical amino acid substitutions and enhanced peptide hydrophobicity/immunogenicity. Copy-number analysis revealed two additional determinants of CPI outcome supported by prior functional evidence: 9q34 (TRAF2) loss associated with response and CCND1 amplification associated with resistance. Finally, single-cell RNA sequencing (RNA-seq) of clonal neoantigen-reactive CD8 tumor-infiltrating lymphocytes (TILs), combined with bulk RNA-seq analysis of CPI-responding tumors, identified CCR5 and CXCL13 as T-cell-intrinsic markers of CPI sensitivity. Keywords: neoantigen, mutation, immunogenicity, checkpoint inhibitors, immunotherapy, biomarkers, CXCL9, clonal TMB, meta-analysis INTRODUCTION To date, multiple biomarkers have been associated with immune checkpoint inhibitor (CPI) response, which can be broadly grouped into four categories: (1) sources of antigen that elicit T cell responses, (2) mechanisms of immune evasion that drive resistance, (3) host factors, and (4) markers of immune infiltration. Despite these promising insights, large-scale studies of CPI response in patients with in-depth whole-exome and transcriptome data have been lacking. Furthermore, given that CPIs activate the immune system rather than target cancer-cell-intrinsic pathways, we hypothesized that a systematic pan-tumor analysis could help elucidate the critical features underpinning CPI response and enable appropriately powered biomarker discovery. Accordingly, we collated raw exome/transcriptome data across multiple studies and cancer types, totaling n = 1,008 CPI-treated patients (termed the “CPI1000+ cohort”; Figure 1) from 12 individual cohorts (see STAR methods), and reprocessed these data through a uniform bioinformatics pipeline to maximize comparability across cohorts. Furthermore, we harmonized radiological clinical response definitions across the 12 studies to ensure strict consistency in outcome measurement (“responder” is defined as a RECIST-criteria-based radiological response with complete response [CR] or partial response [PR], and “nonresponder” is defined as stable disease [SD] or progressive disease [PD]). We note this is a conservative definition of response, and patients with SD and extended survival can be considered as experiencing clinical benefit from treatment; however, the “CR/PR versus SD/PD” definition allows the clearest response interpretation and is consistent with the most recent literature (Cristescu et al., 2018; Mariathasan et al., 2018). Furthermore, in a subset of patients with both radiological response and overall survival data, we found a strong relationship in biomarker effect sizes for response and survival (Figure S1A; p = 0.001). The CPI1000+ cohort comprises data from seven tumor types (metastatic urothelial cancer [n = 387], malignant melanoma [n = 353], head and neck cancer [n = 107], non-small cell lung cancer [n = 76], renal cell carcinoma [RCC] [n = 51], colorectal cancer [n = 20], and breast cancer [n = 14]), treated with three classes of CPIs (anti-CTLA-4 [n = 155], anti-PD-1 [n = 432], and anti-PD-L1 [n = 421]) (Table S1). Samples predominantly represented baseline pretreatment specimens, treated with single agent CPI, with a small number of exceptions (n = 55, 5.5%) in which the patient had either undergone prior lines of anti-CTLA-4 treatment or the biopsy was taken on treatment (see Table S1). As a validation cohort, we obtained processed copy-number segment and overall survival data from n = 1,600 cases from CPI-treated patients profiled using the MSK-IMPACT panel (Consortium, 2017; Samstein et al., 2019) (referred to hereafter as the MSK1600 cohort; RECIST response outcome data were not available for this cohort). [Figure: Design of the meta-analysis studyInput studies to the meta-analysis (Figure 2) results (top) and validation cohorts for the multivariable predictive modeling (Figure 3) (bottom).] [Figure: Supplementary meta-analysis data, related to Figure 2Panel A shows the correlation in biomarker effect sizes for radiological response and overall survival clinical endpoints (Spearman's correlation). Panel B shows response rate by number of NMD-escape mutations for all available samples. Panel C shows results from previously published histology specific biomarkers, or metrics that could not be calculated in > 75% of the cohort samples.] RESULTS Benchmarking of previously reported biomarkers of CPI response Samples were processed from raw sequencing reads, and standardized processing/quality control procedures were executed as described in STAR methods. We began the analysis by benchmarking previously published predictors of CPI response using a literature search to systematically identify relevant biomarkers. We reviewed 723 articles that matched the search terms (see STAR methods), yielding a panel of 55 unique biomarkers (methods). To allow biomarkers with varying measurement scales (e.g., mutation counts versus gene expression values) to be compared equivalently based on effect size rather than p value (Wasserstein et al., 2019), all biomarker values were converted to standard Z scores. We note Z score conversion has been similarly applied in other large-scale tumor mutation burden (TMB) projects (Vokes et al., 2019), and as a control all analyses were repeated without Z score conversion, with the top-ranked biomarkers found to be the same (data not shown). Finally, to avoid data pooling (Bravata and Olkin, 2001), each biomarker in each study was analyzed individually, and then the effect sizes/standard errors were combined through meta-analysis (Figure 2A). [Figure: The biomarker landscape of CPI response(A) Previously published biomarkers are shown as rows and individual cohorts within the CPI1000+ cohort as columns. The heatmap indicates the effect size of each biomarker in each cohort, measured as the log2 odds ratio (OR) for response “CR/PR” versus no response “SD/PD/NE” derived from logistic regression. Blue denotes association with response, red association with no response. Drug class and cohort sizes are annotated, and the right-hand forest plot shows the overall effect size and significance of each biomarker in meta-analysis across all studies, based on effect sizes and standard errors from each individual cohort. p values are shown from meta-analysis (random effects, on account of the different tumor types), with the first set of p-values including all samples (p-meta all cohorts) and last set (p-meta validation cohorts) including validation cohorts only (i.e., when a biomarker was originally discovered in a cohort, this cohort was excluded from the meta-analysis). For clarity of plotting, outlier OR values were capped between OR = 0.1 and OR = 10 (all outlier values were nonsignificant results skewed by rare event counts, and raw (uncapped) values were still used in the meta-analysis).(B) The CPI1000+ cohort broken into cancer/drug subgroups for combinations with two or more independent cohorts. OR effect sizes are shown on the y axis, and biomarkers that are either significant in the pan-cancer 2A analysis or within an individual subgroup are shown. Colors are arbitrary and are used only to distinguish the groups.(C) Correlation between biomarkers that are measured on a continuous scale.(D) Proportion of variance explained for each

Read the complete text