Combinations of immuno-checkpoint inhibitors predictive biomarkers only marginally improve their individual accuracy

Matteo Pallocca, Davide Angeli, Fabio Palombo, Francesca Sperati, Michele Milella, Frauke Goeman, Francesca De Nicola, Maurizio Fanciulli, Paola Nisticò, Concetta Quintarelli, Gennaro Ciliberto

Research output: Contribution to journalArticle

Abstract

BACKGROUND: There are no accepted universal biomarkers capable to accurately predict response to immuno-checkpoint inhibitors (ICI). Although recent literature has been flooded with studies on ICI predictive biomarkers, available data show that currently approved companion diagnostics either leave out many possible responders, as in the case of PD-L1 testing for first-line metastatic lung cancer, or apply to a small subset of patients, such as the recently approved treatment for microsatellite instability-high or mismatch repair deficiency tumors. In this study, we conducted a survey of the available data on ICI trials with matched genomic or transcriptomic datasets in order to cross-validate the proposed biomarkers, to assess whether their prediction power was confirmed and, mainly, to investigate if their combination was able to generate a better predictive tool.

METHODS: We extracted clinical information and sequencing data details from publicly available datasets, along with a list of possible biomarkers obtained from the recent literature. After an operation of data harmonization, we validated the performance of all the biomarkers taken individually. Furthermore, we tested two strategies to combine the best performing biomarkers in order to improve their predictive value.

RESULTS: When considered individually, some of the biomarkers, such as the ImmunoPhenoScore, and the IFN-γ signature, did not confirm their originally proposed predictive power. The best absolute scoring biomarkers are TIDE, one of the ICB resistance signatures and CTLA4 with a mean AUC > 0.66. Among the combinations tested, generalized linear models showed the best performance with an AUC of 0.78.

CONCLUSIONS: We confirmed that the available biomarkers, taken individually, fail to provide a satisfactory predictive value. Unfortunately, also combination of some of them only provides marginal improvements. Hence, in order to generate a more robust way to predict ICI efficacy it is necessary to analyze and combine additional biomarkers and interrogate a wider set of clinical data.

Original languageEnglish
Pages (from-to)131
JournalJournal of Translational Medicine
Volume17
Issue number1
DOIs
Publication statusPublished - Apr 23 2019

Fingerprint

Biomarkers
Area Under Curve
Microsatellite Instability
Microsatellite Repeats
Tumors
Linear Models
Lung Neoplasms
Repair
Testing

Cite this

Combinations of immuno-checkpoint inhibitors predictive biomarkers only marginally improve their individual accuracy. / Pallocca, Matteo; Angeli, Davide; Palombo, Fabio; Sperati, Francesca; Milella, Michele; Goeman, Frauke; De Nicola, Francesca; Fanciulli, Maurizio; Nisticò, Paola; Quintarelli, Concetta; Ciliberto, Gennaro.

In: Journal of Translational Medicine, Vol. 17, No. 1, 23.04.2019, p. 131.

Research output: Contribution to journalArticle

@article{7c98c529b5a1445d9049f5121fb0e3b9,
title = "Combinations of immuno-checkpoint inhibitors predictive biomarkers only marginally improve their individual accuracy",
abstract = "BACKGROUND: There are no accepted universal biomarkers capable to accurately predict response to immuno-checkpoint inhibitors (ICI). Although recent literature has been flooded with studies on ICI predictive biomarkers, available data show that currently approved companion diagnostics either leave out many possible responders, as in the case of PD-L1 testing for first-line metastatic lung cancer, or apply to a small subset of patients, such as the recently approved treatment for microsatellite instability-high or mismatch repair deficiency tumors. In this study, we conducted a survey of the available data on ICI trials with matched genomic or transcriptomic datasets in order to cross-validate the proposed biomarkers, to assess whether their prediction power was confirmed and, mainly, to investigate if their combination was able to generate a better predictive tool.METHODS: We extracted clinical information and sequencing data details from publicly available datasets, along with a list of possible biomarkers obtained from the recent literature. After an operation of data harmonization, we validated the performance of all the biomarkers taken individually. Furthermore, we tested two strategies to combine the best performing biomarkers in order to improve their predictive value.RESULTS: When considered individually, some of the biomarkers, such as the ImmunoPhenoScore, and the IFN-γ signature, did not confirm their originally proposed predictive power. The best absolute scoring biomarkers are TIDE, one of the ICB resistance signatures and CTLA4 with a mean AUC > 0.66. Among the combinations tested, generalized linear models showed the best performance with an AUC of 0.78.CONCLUSIONS: We confirmed that the available biomarkers, taken individually, fail to provide a satisfactory predictive value. Unfortunately, also combination of some of them only provides marginal improvements. Hence, in order to generate a more robust way to predict ICI efficacy it is necessary to analyze and combine additional biomarkers and interrogate a wider set of clinical data.",
author = "Matteo Pallocca and Davide Angeli and Fabio Palombo and Francesca Sperati and Michele Milella and Frauke Goeman and {De Nicola}, Francesca and Maurizio Fanciulli and Paola Nistic{\`o} and Concetta Quintarelli and Gennaro Ciliberto",
year = "2019",
month = "4",
day = "23",
doi = "10.1186/s12967-019-1865-8",
language = "English",
volume = "17",
pages = "131",
journal = "Journal of Translational Medicine",
issn = "1479-5876",
publisher = "BioMed Central Ltd.",
number = "1",

}

TY - JOUR

T1 - Combinations of immuno-checkpoint inhibitors predictive biomarkers only marginally improve their individual accuracy

AU - Pallocca, Matteo

AU - Angeli, Davide

AU - Palombo, Fabio

AU - Sperati, Francesca

AU - Milella, Michele

AU - Goeman, Frauke

AU - De Nicola, Francesca

AU - Fanciulli, Maurizio

AU - Nisticò, Paola

AU - Quintarelli, Concetta

AU - Ciliberto, Gennaro

PY - 2019/4/23

Y1 - 2019/4/23

N2 - BACKGROUND: There are no accepted universal biomarkers capable to accurately predict response to immuno-checkpoint inhibitors (ICI). Although recent literature has been flooded with studies on ICI predictive biomarkers, available data show that currently approved companion diagnostics either leave out many possible responders, as in the case of PD-L1 testing for first-line metastatic lung cancer, or apply to a small subset of patients, such as the recently approved treatment for microsatellite instability-high or mismatch repair deficiency tumors. In this study, we conducted a survey of the available data on ICI trials with matched genomic or transcriptomic datasets in order to cross-validate the proposed biomarkers, to assess whether their prediction power was confirmed and, mainly, to investigate if their combination was able to generate a better predictive tool.METHODS: We extracted clinical information and sequencing data details from publicly available datasets, along with a list of possible biomarkers obtained from the recent literature. After an operation of data harmonization, we validated the performance of all the biomarkers taken individually. Furthermore, we tested two strategies to combine the best performing biomarkers in order to improve their predictive value.RESULTS: When considered individually, some of the biomarkers, such as the ImmunoPhenoScore, and the IFN-γ signature, did not confirm their originally proposed predictive power. The best absolute scoring biomarkers are TIDE, one of the ICB resistance signatures and CTLA4 with a mean AUC > 0.66. Among the combinations tested, generalized linear models showed the best performance with an AUC of 0.78.CONCLUSIONS: We confirmed that the available biomarkers, taken individually, fail to provide a satisfactory predictive value. Unfortunately, also combination of some of them only provides marginal improvements. Hence, in order to generate a more robust way to predict ICI efficacy it is necessary to analyze and combine additional biomarkers and interrogate a wider set of clinical data.

AB - BACKGROUND: There are no accepted universal biomarkers capable to accurately predict response to immuno-checkpoint inhibitors (ICI). Although recent literature has been flooded with studies on ICI predictive biomarkers, available data show that currently approved companion diagnostics either leave out many possible responders, as in the case of PD-L1 testing for first-line metastatic lung cancer, or apply to a small subset of patients, such as the recently approved treatment for microsatellite instability-high or mismatch repair deficiency tumors. In this study, we conducted a survey of the available data on ICI trials with matched genomic or transcriptomic datasets in order to cross-validate the proposed biomarkers, to assess whether their prediction power was confirmed and, mainly, to investigate if their combination was able to generate a better predictive tool.METHODS: We extracted clinical information and sequencing data details from publicly available datasets, along with a list of possible biomarkers obtained from the recent literature. After an operation of data harmonization, we validated the performance of all the biomarkers taken individually. Furthermore, we tested two strategies to combine the best performing biomarkers in order to improve their predictive value.RESULTS: When considered individually, some of the biomarkers, such as the ImmunoPhenoScore, and the IFN-γ signature, did not confirm their originally proposed predictive power. The best absolute scoring biomarkers are TIDE, one of the ICB resistance signatures and CTLA4 with a mean AUC > 0.66. Among the combinations tested, generalized linear models showed the best performance with an AUC of 0.78.CONCLUSIONS: We confirmed that the available biomarkers, taken individually, fail to provide a satisfactory predictive value. Unfortunately, also combination of some of them only provides marginal improvements. Hence, in order to generate a more robust way to predict ICI efficacy it is necessary to analyze and combine additional biomarkers and interrogate a wider set of clinical data.

U2 - 10.1186/s12967-019-1865-8

DO - 10.1186/s12967-019-1865-8

M3 - Article

C2 - 31014354

VL - 17

SP - 131

JO - Journal of Translational Medicine

JF - Journal of Translational Medicine

SN - 1479-5876

IS - 1

ER -