Artificial intelligence estimates the importance of baseline factors in predicting response to anti-PD1 in metastatic melanoma

Alice Indini, Lorenza Di Guardo, Carolina Cimminiello, Filippo De Braud, Michele Del Vecchio

Research output: Contribution to journalArticle

Abstract

Objective: Prognosis of patients with metastatic melanoma has dramatically improved over recent years because of the advent of antibodies targeting programmed cell death protein-1 (PD1). However, the response rate is 40% and baseline biomarkers for the outcome are yet to be identified. Here, we aimed to determine whether artificial intelligence might be useful in weighting the importance of baseline variables in predicting response to anti-PD1. Methods: This is a retrospective study evaluating 173 patients receiving anti-PD1 for melanoma. Using an artificial neuronal network analysis, the importance of different variables was estimated and used in predicting response rate and overall survival. Results: After a mean follow-up of 12.8 (±11.9) months, disease control rate was 51%. Using artificial neuronal network, we observed that 3 factors predicted response to anti-PD1: neutrophil-to-lymphocyte ratio (NLR) (importance: 0.195), presence of ≥3 metastatic sites (importance: 0.156), and baseline lactate dehydrogenase (LDH) > upper limit of normal (importance: 0.154). Looking at connections between different covariates and overall survival, the most important variables influencing survival were: presence of ≥3 metastatic sites (importance: 0.202), age (importance: 0.189), NLR (importance: 0.164), site of primary melanoma (cutaneous vs. noncutaneous) (importance: 0.112), and LDH > upper limit of normal (importance: 0.108). Conclusions: NLR, presence of ≥3 metastatic sites, LDH levels, age, and site of primary melanoma are important baseline factors influencing response and survival. Further studies are warranted to estimate a model to drive the choice to administered anti-PD1 treatments in patients with melanoma.

Original languageEnglish
Pages (from-to)643-648
Number of pages6
JournalAmerican Journal of Clinical Oncology: Cancer Clinical Trials
Volume42
Issue number8
DOIs
Publication statusPublished - Aug 1 2019

Fingerprint

Artificial Intelligence
Melanoma
L-Lactate Dehydrogenase
Neutrophils
Lymphocytes
Survival
Programmed Cell Death 1 Receptor
Survival Rate
Retrospective Studies
Biomarkers
Skin
Antibodies

Keywords

  • anti-PD1
  • artificial intelligence
  • artificial neural network
  • metastatic melanoma
  • response

ASJC Scopus subject areas

  • Oncology
  • Cancer Research

Cite this

@article{47c5e9e6aa3546dd9872f4a66f5fa2f9,
title = "Artificial intelligence estimates the importance of baseline factors in predicting response to anti-PD1 in metastatic melanoma",
abstract = "Objective: Prognosis of patients with metastatic melanoma has dramatically improved over recent years because of the advent of antibodies targeting programmed cell death protein-1 (PD1). However, the response rate is 40{\%} and baseline biomarkers for the outcome are yet to be identified. Here, we aimed to determine whether artificial intelligence might be useful in weighting the importance of baseline variables in predicting response to anti-PD1. Methods: This is a retrospective study evaluating 173 patients receiving anti-PD1 for melanoma. Using an artificial neuronal network analysis, the importance of different variables was estimated and used in predicting response rate and overall survival. Results: After a mean follow-up of 12.8 (±11.9) months, disease control rate was 51{\%}. Using artificial neuronal network, we observed that 3 factors predicted response to anti-PD1: neutrophil-to-lymphocyte ratio (NLR) (importance: 0.195), presence of ≥3 metastatic sites (importance: 0.156), and baseline lactate dehydrogenase (LDH) > upper limit of normal (importance: 0.154). Looking at connections between different covariates and overall survival, the most important variables influencing survival were: presence of ≥3 metastatic sites (importance: 0.202), age (importance: 0.189), NLR (importance: 0.164), site of primary melanoma (cutaneous vs. noncutaneous) (importance: 0.112), and LDH > upper limit of normal (importance: 0.108). Conclusions: NLR, presence of ≥3 metastatic sites, LDH levels, age, and site of primary melanoma are important baseline factors influencing response and survival. Further studies are warranted to estimate a model to drive the choice to administered anti-PD1 treatments in patients with melanoma.",
keywords = "anti-PD1, artificial intelligence, artificial neural network, metastatic melanoma, response",
author = "Alice Indini and {Di Guardo}, Lorenza and Carolina Cimminiello and {De Braud}, Filippo and {Del Vecchio}, Michele",
year = "2019",
month = "8",
day = "1",
doi = "10.1097/COC.0000000000000566",
language = "English",
volume = "42",
pages = "643--648",
journal = "American Journal of Clinical Oncology",
issn = "0277-3732",
publisher = "Lippincott Williams and Wilkins",
number = "8",

}

TY - JOUR

T1 - Artificial intelligence estimates the importance of baseline factors in predicting response to anti-PD1 in metastatic melanoma

AU - Indini, Alice

AU - Di Guardo, Lorenza

AU - Cimminiello, Carolina

AU - De Braud, Filippo

AU - Del Vecchio, Michele

PY - 2019/8/1

Y1 - 2019/8/1

N2 - Objective: Prognosis of patients with metastatic melanoma has dramatically improved over recent years because of the advent of antibodies targeting programmed cell death protein-1 (PD1). However, the response rate is 40% and baseline biomarkers for the outcome are yet to be identified. Here, we aimed to determine whether artificial intelligence might be useful in weighting the importance of baseline variables in predicting response to anti-PD1. Methods: This is a retrospective study evaluating 173 patients receiving anti-PD1 for melanoma. Using an artificial neuronal network analysis, the importance of different variables was estimated and used in predicting response rate and overall survival. Results: After a mean follow-up of 12.8 (±11.9) months, disease control rate was 51%. Using artificial neuronal network, we observed that 3 factors predicted response to anti-PD1: neutrophil-to-lymphocyte ratio (NLR) (importance: 0.195), presence of ≥3 metastatic sites (importance: 0.156), and baseline lactate dehydrogenase (LDH) > upper limit of normal (importance: 0.154). Looking at connections between different covariates and overall survival, the most important variables influencing survival were: presence of ≥3 metastatic sites (importance: 0.202), age (importance: 0.189), NLR (importance: 0.164), site of primary melanoma (cutaneous vs. noncutaneous) (importance: 0.112), and LDH > upper limit of normal (importance: 0.108). Conclusions: NLR, presence of ≥3 metastatic sites, LDH levels, age, and site of primary melanoma are important baseline factors influencing response and survival. Further studies are warranted to estimate a model to drive the choice to administered anti-PD1 treatments in patients with melanoma.

AB - Objective: Prognosis of patients with metastatic melanoma has dramatically improved over recent years because of the advent of antibodies targeting programmed cell death protein-1 (PD1). However, the response rate is 40% and baseline biomarkers for the outcome are yet to be identified. Here, we aimed to determine whether artificial intelligence might be useful in weighting the importance of baseline variables in predicting response to anti-PD1. Methods: This is a retrospective study evaluating 173 patients receiving anti-PD1 for melanoma. Using an artificial neuronal network analysis, the importance of different variables was estimated and used in predicting response rate and overall survival. Results: After a mean follow-up of 12.8 (±11.9) months, disease control rate was 51%. Using artificial neuronal network, we observed that 3 factors predicted response to anti-PD1: neutrophil-to-lymphocyte ratio (NLR) (importance: 0.195), presence of ≥3 metastatic sites (importance: 0.156), and baseline lactate dehydrogenase (LDH) > upper limit of normal (importance: 0.154). Looking at connections between different covariates and overall survival, the most important variables influencing survival were: presence of ≥3 metastatic sites (importance: 0.202), age (importance: 0.189), NLR (importance: 0.164), site of primary melanoma (cutaneous vs. noncutaneous) (importance: 0.112), and LDH > upper limit of normal (importance: 0.108). Conclusions: NLR, presence of ≥3 metastatic sites, LDH levels, age, and site of primary melanoma are important baseline factors influencing response and survival. Further studies are warranted to estimate a model to drive the choice to administered anti-PD1 treatments in patients with melanoma.

KW - anti-PD1

KW - artificial intelligence

KW - artificial neural network

KW - metastatic melanoma

KW - response

UR - http://www.scopus.com/inward/record.url?scp=85068391816&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85068391816&partnerID=8YFLogxK

U2 - 10.1097/COC.0000000000000566

DO - 10.1097/COC.0000000000000566

M3 - Article

C2 - 31261257

AN - SCOPUS:85068391816

VL - 42

SP - 643

EP - 648

JO - American Journal of Clinical Oncology

JF - American Journal of Clinical Oncology

SN - 0277-3732

IS - 8

ER -