Elsevier

The Lancet Oncology

Volume 19, Issue 9, September 2018, Pages 1180-1191
The Lancet Oncology

Articles
A radiomics approach to assess tumour-infiltrating CD8 cells and response to anti-PD-1 or anti-PD-L1 immunotherapy: an imaging biomarker, retrospective multicohort study

https://doi.org/10.1016/S1470-2045(18)30413-3Get rights and content

Summary

Background

Because responses of patients with cancer to immunotherapy can vary in success, innovative predictors of response to treatment are urgently needed to improve treatment outcomes. We aimed to develop and independently validate a radiomics-based biomarker of tumour-infiltrating CD8 cells in patients included in phase 1 trials of anti-programmed cell death protein (PD)-1 or anti-programmed cell death ligand 1 (PD-L1) monotherapy. We also aimed to evaluate the association between the biomarker, and tumour immune phenotype and clinical outcomes of these patients.

Methods

In this retrospective multicohort study, we used four independent cohorts of patients with advanced solid tumours to develop and validate a radiomic signature predictive of immunotherapy response by combining contrast-enhanced CT images and RNA-seq genomic data from tumour biopsies to assess CD8 cell tumour infiltration. To develop the radiomic signature of CD8 cells, we used the CT images and RNA sequencing data of 135 patients with advanced solid malignant tumours who had been enrolled into the MOSCATO trial between May 1, 2012, and March 31, 2016, in France (training set). The genomic data, which are based on the CD8B gene, were used to estimate the abundance of CD8 cells in the samples and data were then aligned with the images to generate the radiomic signatures. The concordance of the radiomic signature (primary endpoint) was validated in a Cancer Genome Atlas [TGCA] database dataset including 119 patients who had available baseline preoperative imaging data and corresponding transcriptomic data on June 30, 2017. From 84 input variables used for the machine-learning method (78 radiomic features, five location variables, and one technical variable), a radiomics-based predictor of the CD8 cell expression signature was built by use of machine learning (elastic-net regularised regression method). Two other independent cohorts of patients with advanced solid tumours were used to evaluate this predictor. The immune phenotype internal cohort (n=100), were randomly selected from the Gustave Roussy Cancer Campus database of patient medical records based on previously described, extreme tumour-immune phenotypes: immune-inflamed (with dense CD8 cell infiltration) or immune-desert (with low CD8 cell infiltration), irrespective of treatment delivered; these data were used to analyse the correlation of the immune phenotype with this biomarker. Finally, the immunotherapy-treated dataset (n=137) of patients recruited from Dec 1, 2011, to Jan 31, 2014, at the Gustave Roussy Cancer Campus, who had been treated with anti-PD-1 and anti-PD-L1 monotherapy in phase 1 trials, was used to assess the predictive value of this biomarker in terms of clinical outcome.

Findings

We developed a radiomic signature for CD8 cells that included eight variables, which was validated with the gene expression signature of CD8 cells in the TCGA dataset (area under the curve [AUC]=0·67; 95% CI 0·57–0·77; p=0·0019). In the cohort with assumed immune phenotypes, the signature was also able to discriminate inflamed tumours from immune-desert tumours (0·76; 0·66–0·86; p<0·0001). In patients treated with anti-PD-1 and PD-L1, a high baseline radiomic score (relative to the median) was associated with a higher proportion of patients who achieved an objective response at 3 months (vs those with progressive disease or stable disease; p=0·049) and a higher proportion of patients who had an objective response (vs those with progressive disease or stable disease; p=0·025) or stable disease (vs those with progressive disease; p=0·013) at 6 months. A high baseline radiomic score was also associated with improved overall survival in univariate (median overall survival 24·3 months in the high radiomic score group, 95% CI 18·63–42·1; vs 11·5 months in the low radiomic score group, 7·98–15·6; hazard ratio 0·58, 95% CI 0·39–0·87; p=0·0081) and multivariate analyses (0·52, 0·35–0·79; p=0·0022).

Interpretation

The radiomic signature of CD8 cells was validated in three independent cohorts. This imaging predictor provided a promising way to predict the immune phenotype of tumours and to infer clinical outcomes for patients with cancer who had been treated with anti-PD-1 and PD-L1. Our imaging biomarker could be useful in estimating CD8 cell count and predicting clinical outcomes of patients treated with immunotherapy, when validated by further prospective randomised trials.

Funding

Fondation pour la Recherche Médicale, and SIRIC-SOCRATE 2.0, French Society of Radiation Oncology.

Introduction

Computational medical imaging, known as radiomics, involves the analysis and translation of medical images into quantitative data.1, 2, 3 High-dimensional imaging data allow an in-depth characterisation of tumour phenotypes, with the underlying hypothesis that imaging reflects not only macroscopic but also the cellular and molecular properties of tissues. The objective of radiomics is to generate image-driven biomarkers that serve as instruments that provide a deeper understanding of cancer biology to better aid clinical decisions.3, 4 Radiomic features are complementary to biopsies and have the advantage of being non-invasive, which allows evaluation of a tumour and its microenvironment, characterisation of spatial heterogeneity, and longitudinal assessment of disease evolution.

Immunotherapy has substantially changed the therapeutic strategies for cancers such as melanomas,5 lymphomas,6 and lung tumours.7 Unfortunately, only 20–50% of patients with advanced solid tumours respond to treatment.8 There is therefore a need for the development of methods to identify patients who are most likely to respond to immunotherapy. Several studies8, 9, 10, 11 have shown that pre-existing tumoral and peritumoral immune infiltration correlates with patient response to anti-programmed cell death protein (PD)-1 and anti-programmed cell death ligand 1 (PD-L1) immunotherapy. Three distinct immune phenotypes have been described: immune-inflamed, immune-excluded, and immune-desert.12 Immune-inflamed tumours are characterised by dense, functional CD8 cell infiltration, increased interferon-γ signalling, expression of cell checkpoint markers (such as PD-L1), and a high mutational burden. These tumours tend to respond to immunotherapy.9, 12, 13 In immune-excluded tumours, several biological signals (such as signalling by transforming growth factor-β, activation of myeloid-derived suppressor cells, and angiogenesis) prevent infiltration by T cells into the tumour. The immune-desert phenotype is characterised by low infiltration by CD8 cells and highly proliferating tumour cells. Both immune-excluded and immune-desert phenotypes are considered not to be inflamed.

Research in context

Evidence before this study

We searched PubMed and Google Scholar for papers published before Aug 10, 2018, with the terms (“texture analysis” OR “radiomics” OR “radiomic” OR “computational imaging”) AND (“immunotherapy” OR “immune”), with no language restrictions. To our knowledge, there are no published studies that evaluate the use of radiomics to predict response to immunotherapy. Four studies have evaluated the associations between radiomics and immune infiltration or molecular pathways involved in the immune response; however, the level of evidence of these studies is still low: only one study has an external validation cohort.

Added value of this study

To our knowledge, our study is the first to develop and validate a radiomics-based biomarker of tumour-infiltrating CD8 cells to show a correlation of the number of tumour-infiltrating lymphocytes (as estimated by a pathologist), tumour immune phenotypes, and clinical responses to anti-programmed cell death protein-1 or anti-programmed cell death ligand 1 immunotherapy in three independent cohorts of patients with advanced solid tumours.

Implications of all the available evidence

Our study suggests that there is potential for non-invasive biomarker development in immunotherapy.

Radiomics could allow evaluation of immune infiltration of tumours and, thus, lead to the identification of novel predictors of the efficacy of immunotherapy. We aimed to develop a radiomic signature of immune infiltration of tumours and to assess the ability of this signature to predict clinical outcomes in patients treated with anti-PD-1 or anti-PD-L1 immunotherapy. The adopted computational imaging vocabulary is defined in the appendix (p 1).

Section snippets

Study design and data sources

In this multicohort study, radiomic analysis was applied retrospectively to four independent cohorts of patients older than 18 years who had solid tumours (figure 1).14, 15, 16 The MOSCATO dataset,14 the immune phenotype dataset, and the immunotherapy-treated dataset17, 18 included patients treated at Gustave Roussy Institute (Villejuif, France) and the other dataset was from the databases of The Cancer Genome Atlas (TCGA) and The Cancer Imaging Archive (TCIA). The MOSCATO dataset was used as a

Results

The MOSCATO training dataset that was used to build the radiomic signature consisted of 135 patients recruited between May 1, 2012, and March 31, 2016 (table 1). The TCGA validation dataset included 119 patients of the 435 patients who were available for screening and who had available RNA-seq data and CT scans at the time of inclusion on June 30, 2017 (appendix pp 2 and 10). The immune phenotype dataset consisted of 100 patients who had been randomly selected from the database of our institute

Discussion

With increasing use of immunotherapy in cancer, knowledge of an individual's immune status could help identify those who will respond to treatment.9, 13 Radiomics approaches, when combined with tumour biopsies and genomics, could improve treatment selection. In our study, we developed a radiomic signature of tumour immune infiltration from CT scans. Access to both RNA-seq data and images of the biopsied lesion in the MOSCATO trial14 and the clinical data of patients from phase 1 trials17, 18 of

References (51)

  • C Robert et al.

    Pembrolizumab versus ipilimumab in advanced melanoma

    N Engl J Med

    (2015)
  • SM Ansell

    Hodgkin lymphoma: MOPP chemotherapy to PD-1 blockade and beyond

    Am J Hematol

    (2015)
  • M Reck et al.

    Pembrolizumab versus chemotherapy for PD-L1-positive non-small-cell lung cancer

    N Engl J Med

    (2016)
  • DS Chen et al.

    Elements of cancer immunity and the cancer-immune set point

    Nature

    (2017)
  • RS Herbst et al.

    Predictive correlates of response to the anti-PD-L1 antibody MPDL3280A in cancer patients

    Nature

    (2014)
  • PS Hegde et al.

    The where, the when, and the how of immune monitoring for cancer immunotherapies in the era of checkpoint inhibition

    Clin Cancer Res

    (2016)
  • PC Tumeh et al.

    PD-1 blockade induces responses by inhibiting adaptive immune resistance

    Nature

    (2014)
  • C Massard et al.

    High-throughput genomics and clinical outcome in hard-to-treat advanced cancers: results of the MOSCATO 01 trial

    Cancer Discov

    (2017)
  • P Grabowski et al.

    Tumor infiltrating lymphocytes and PD-L1 expression differ in low and high grade neuroendocrine tumors (abstract)

    Neuroendocrinology

    (2015)
  • S Champiat et al.

    Hyperprogressive disease is a new pattern of progression in cancer patients treated by anti-PD-1/PD-L1

    Clin Cancer Res

    (2017)
  • E Becht et al.

    Estimating the population abundance of tissue-infiltrating immune and stromal cell populations using gene expression

    Genome Biol

    (2016)
  • K Clark et al.

    The Cancer Imaging Archive (TCIA): maintaining and operating a public information repository

    J Digit Imaging

    (2013)
  • JN Weinstein et al.

    The Cancer Genome Atlas Pan-Cancer analysis project

    Nat Genet

    (2013)
  • V Sridharan et al.

    Immune profiling of adenoid cystic carcinoma: PD-L2 expression and associations with tumor-infiltrating lymphocytes

    Cancer Immunol Res

    (2016)
  • M Hatt et al.

    18F-FDG PET uptake characterization through texture analysis: investigating the complementary nature of heterogeneity and functional tumor volume in a multi-cancer site patient cohort

    J Nucl Med

    (2015)
  • Cited by (811)

    • A Glimpse Into the Future for Unresectable Stage III Non-Small Cell Lung Cancer

      2024, International Journal of Radiation Oncology Biology Physics
    • In Response

      2024, Journal of Thoracic Oncology
    View all citing articles on Scopus
    *

    Contributed equally and are joint first authors

    Contributed equally and are joint senior authors

    View full text