Immune Checkpoint Blockade (ICB) Data

Abstract

The development of immune checkpoint blockade (ICB) has changed the way we treat various cancers and led to the development of several new compounds that are now under clinical investigation. Despite the large number of Immune checkpoint inhibitors (ICB) clinical trials with available molecular data published every year, it is still complicated to assess the reproducibility of immune-oncology (IO) predictive and prognostic biomarkers. Recently, multiple biomarkers have been reported to be associated with immunotherapy response, however tumor complexity and heterogeneity limit the use of research context biomarkers in clinical practice. We have recently performed a meta-analysis of public clinical genomic datasets of patients treated with immunotherapies to assess the predictive value of published biomarkers and mRNA signatures. The collection of datasets contain a broad number of ICB clinical trials, with both molecular and clinical data available.

Background

The advent of immune checkpoint blockade (ICB) therapy, especially with monoclonal antibodies targeting PD-1, PD-L1, or CTLA-4, has transformed cancer treatment by providing durable survival benefits. Yet, 60% to 80% of patients do not achieve clinical benefits from ICB. This resistance is linked to tumor-intrinsic factors, such as genomic alterations, oncogenic signaling, and impaired antigen presentation, as well as tumor-extrinsic factors, including host immunity and the tumor microenvironment.

Methods

Whole genome and transcriptome sequencing data were processed using standard pipelines and reference genomes. The GATK software suite was used for whole genome data, and Kallisto version 0.46.1 was used for transcriptome data. Gene annotation was performed using Gencode V40, with Ensembl Gene ID as the primary identifier, linked to Entrez Gene ID, HUGO Gene Symbols, and chromosomal locations. All pipelines were fully transparent and reproducible on the ORCESTRA platform (Mammoliti et al. 2021). For RNA profiles, log2-transformed TPM (Transcripts Per Million) data were used. If raw FASTQ files were not available, raw count or TPM data were downloaded from publications. For studies that included FPKM (Fragments Per Kilobase of transcript per Million mapped reads) data, FPKM was converted to TPM. Somatic mutation data included gene-level alterations with mutation context information (e.g., synonymous or non-synonymous).

Data Description

Publicly available ICB data includes at least one type of genomic data and clinical outcome information from advanced solid tumor patients treated with anti-PD-1/L1 and/or anti-CTLA-4 ICB therapy. Genomic data is defined as RNA sequencing and/or tumor exome or targeted DNA sequencing. Clinical outcome data includes response (according to RECIST or other response criteria), progression-free survival (PFS), or overall survival (OS).

Response (R) was defined as either a complete or partial response according to RECIST (v1.1) or stable disease (SD) without a progression-free survival (PFS) event within 6 months. Non-response (NR) was defined as either progressive disease according to RECIST or stable disease (SD) with a PFS event occurring within 6 months. In cases where RECIST information was not available, patients without a PFS event at 6 months were classified as R, while those with a PFS event within 6 months were classified as NR. If PFS information was unavailable, patients with stable disease (SD) as the best response were classified as not assessable.

For a list of all included ICB datasets, see this spreadsheet.

Usage Notes

ICB data can be used to identify immuno-oncology biomarkers by analyzing genomic and clinical outcome data from patients treated with ICB therapies. These biomarkers can help predict or correlate with treatment response, provide insights into resistance mechanisms, and guide patient stratification and the development of more personalized cancer treatments (Bareche et al. 2022).

You'll need to install the multi assay experiment (MAE) R package through the Bioconductor channel:

https://www.bioconductor.org/packages/release/bioc/html/MultiAssayExperiment.html

# Install Bioconductor Manager
if (!require("BiocManager", quietly = TRUE))
install.packages("BiocManager")

# Install MAE Package
BiocManager::install("MultiAssayExperiment")

# Load MAE Package
library("MultiAssayExperiment")

You can traverse an ICB dataset as follows (using ICB_Limagne2 as an example):

immu = readRDS("ICB_Limagne2.RDS")
str(immu, 4)

Formal class 'MultiAssayExperiment' [package "MultiAssayExperiment"] with 5 slots
  ..@ ExperimentList:Formal class 'ExperimentList' [package "MultiAssayExperiment"] with 4 slots
  .. .. ..@ listData       :List of 4
  .. .. ..@ elementType    : chr "ANY"
  .. .. ..@ elementMetadata: NULL
  .. .. ..@ metadata       : list()
  ..@ colData       :Formal class 'DFrame' [package "S4Vectors"] with 6 slots
  .. .. ..@ rownames       : chr [1:26] "N15_158_RNA_116124" "N15_173_RNA_116117" "N15_178_RNA_116122" "N15_205_RNA_116129" ...
  .. .. ..@ nrows          : int 26
  .. .. ..@ elementType    : chr "ANY"
  .. .. ..@ elementMetadata: NULL
  .. .. ..@ metadata       : list()
  .. .. ..@ listData       :List of 66
  ..@ sampleMap     :Formal class 'DFrame' [package "S4Vectors"] with 6 slots
  .. .. ..@ rownames       : NULL
  .. .. ..@ nrows          : int 104
  .. .. ..@ elementType    : chr "ANY"
  .. .. ..@ elementMetadata: NULL
  .. .. ..@ metadata       : list()
  .. .. ..@ listData       :List of 3

Ex.1 Listing experiments:

str(immu@ExperimentList@listData, 2)

List of 4 
 $ expr_gene_tpm      :Formal class 'RangedSummarizedExperiment' [package "SummarizedExperiment"] with 6 slots
 $ expr_gene_counts   :Formal class 'RangedSummarizedExperiment' [package "SummarizedExperiment"] with 6 slots
 $ expr_isoform_tpm   :Formal class 'RangedSummarizedExperiment' [package "SummarizedExperiment"] with 6 slots
 $ expr_isoform_counts:Formal class 'RangedSummarizedExperiment' [package "SummarizedExperiment"] with 6 slots

Ex.2 Structure of one gene expression experiment (log TPM2):

 
str(immu@ExperimentList@listData$expr_gene_tpm@colData@listData, 2)

List of 66
 $ patientid              : chr [1:26] "N16_77_RNA_116120" "N16_45_RNA_116132" "N16_309_RNA_121159" "N15_205_RNA_116129" ...
 $ cancer_type            : chr [1:26] "Lung" "Lung" "Lung" "Lung" ...
 $ tissueid               : chr [1:26] "Lung" "Lung" "Lung" "Lung" ...
 $ treatmentid            : chr [1:26] "" "" "" "" ...
 $ response               : chr [1:26] "NR" "NR" "NR" "R" ...
 $ treatment              : chr [1:26] "anti-PD-1/anti-PD-L1" "anti-PD-1/anti-PD-L1" "anti-PD-1/anti-PD-L1" "anti-PD-1/anti-PD-L1" ...
 $ rna                    : chr [1:26] "tpm" "tpm" "tpm" "tpm" ...
 $ survival_time_pfs      : num [1:26] 3.5 2.8 1.6 16.1 2.3 1.8 12.3 2 1.6 0.4 ...
 $ event_occurred_pfs     : int [1:26] 1 1 1 0 1 1 1 1 1 1 ...
.
.
.

Ethics

The authors declare no ethics concerns.

Conflicts of Interest

The author(s) have no conflicts of interest to declare.

References

1. Bareche, Yacine, Deirdre Kelly, Farnoosh Abbas-Aghababazadeh, Minoru Nakano, Parinaz Nasr Esfahani, Denis Tkachuk, Hassan Mohammad, et al. 2022. “Leveraging Big Data of Immune Checkpoint Blockade Response Identifies Novel Potential Targets.” Annals of Oncology: Official Journal of the European Society for Medical Oncology / ESMO, August.
2. Mammoliti, Anthony, Petr Smirnov, Minoru Nakano, Zhaleh Safikhani, Christopher Eeles, Heewon Seo, Sisira Kadambat Nair, et al. 2021. “Orchestrating and Sharing Large Multimodal Data for Transparent and Reproducible Research.” Nature Communications 12 (1): 5797.