Vahé Asvatourian1,2, Clélia Coutzac3,4, Nathalie Chaput3,5, Caroline Robert4,6, Stefan Michiels7,8, Emilie Lanoy7,8. 1. Université Paris-Saclay, Univ. Paris-Sud, UVSQ, CESP, INSERM, Villejuif, France. vahe.asvatourian@gustaveroussy.fr. 2. Service de Biostatistique et d'Epidémiologie, Gustave-Roussy, Villejuif, France. vahe.asvatourian@gustaveroussy.fr. 3. Gustave Roussy, Laboratoire d'Immunomonitoring en Oncologie, and CNRS-UMS 3655 and INSERM-US23, F-94805, Villejuif, France. 4. Université Paris-Sud, Faculté de Médecine, F-94276, Le Kremlin Bicêtre, France. 5. Université Paris-Sud, Faculté de Pharmacie, F-92296, Chatenay-Malabry, France. 6. Department of Medicine, Dermatology Unit, Gustave Roussy Cancer Campus, Villejuif, France. 7. Université Paris-Saclay, Univ. Paris-Sud, UVSQ, CESP, INSERM, Villejuif, France. 8. Service de Biostatistique et d'Epidémiologie, Gustave-Roussy, Villejuif, France.
Abstract
BACKGROUND: Recently, the intervention calculus when the DAG is absent (IDA) method was developed to estimate lower bounds of causal effects from observational high-dimensional data. Originally it was introduced to assess the effect of baseline biomarkers which do not vary over time. However, in many clinical settings, measurements of biomarkers are repeated at fixed time points during treatment and, therefore, this method needs to be extended. The purpose of this paper is to extend the first step of the IDA, the Peter Clarks (PC)-algorithm, to a time-dependent exposure in the context of a binary outcome. METHODS: We generalised the so-called "PC-algorithm" to take into account the chronological order of repeated measurements of the exposure and proposed to apply the IDA with our new version, the chronologically ordered PC-algorithm (COPC-algorithm). The extension includes Firth's correction. A simulation study has been performed before applying the method for estimating causal effects of time-dependent immunological biomarkers on toxicity, death and progression in patients with metastatic melanoma. RESULTS: The simulation study showed that the completed partially directed acyclic graphs (CPDAGs) obtained using COPC-algorithm were structurally closer to the true CPDAG than CPDAGs obtained using PC-algorithm. Also, causal effects were more accurate when they were estimated based on CPDAGs obtained using COPC-algorithm. Moreover, CPDAGs obtained by COPC-algorithm allowed removing non-chronological arrows with a variable measured at a time t pointing to a variable measured at a time t´ where t´ < t. Bidirected edges were less present in CPDAGs obtained with the COPC-algorithm, supporting the fact that there was less variability in causal effects estimated from these CPDAGs. In the example, a threshold of the per-comparison error rate of 0.5% led to the selection of an interpretable set of biomarkers. CONCLUSIONS: The COPC-algorithm provided CPDAGs that keep the chronological structure present in the data and thus allowed to estimate lower bounds of the causal effect of time-dependent immunological biomarkers on early toxicity, premature death and progression.
BACKGROUND: Recently, the intervention calculus when the DAG is absent (IDA) method was developed to estimate lower bounds of causal effects from observational high-dimensional data. Originally it was introduced to assess the effect of baseline biomarkers which do not vary over time. However, in many clinical settings, measurements of biomarkers are repeated at fixed time points during treatment and, therefore, this method needs to be extended. The purpose of this paper is to extend the first step of the IDA, the Peter Clarks (PC)-algorithm, to a time-dependent exposure in the context of a binary outcome. METHODS: We generalised the so-called "PC-algorithm" to take into account the chronological order of repeated measurements of the exposure and proposed to apply the IDA with our new version, the chronologically ordered PC-algorithm (COPC-algorithm). The extension includes Firth's correction. A simulation study has been performed before applying the method for estimating causal effects of time-dependent immunological biomarkers on toxicity, death and progression in patients with metastatic melanoma. RESULTS: The simulation study showed that the completed partially directed acyclic graphs (CPDAGs) obtained using COPC-algorithm were structurally closer to the true CPDAG than CPDAGs obtained using PC-algorithm. Also, causal effects were more accurate when they were estimated based on CPDAGs obtained using COPC-algorithm. Moreover, CPDAGs obtained by COPC-algorithm allowed removing non-chronological arrows with a variable measured at a time t pointing to a variable measured at a time t´ where t´ < t. Bidirected edges were less present in CPDAGs obtained with the COPC-algorithm, supporting the fact that there was less variability in causal effects estimated from these CPDAGs. In the example, a threshold of the per-comparison error rate of 0.5% led to the selection of an interpretable set of biomarkers. CONCLUSIONS: The COPC-algorithm provided CPDAGs that keep the chronological structure present in the data and thus allowed to estimate lower bounds of the causal effect of time-dependent immunological biomarkers on early toxicity, premature death and progression.
Authors: Caroline Robert; Luc Thomas; Igor Bondarenko; Steven O'Day; Jeffrey Weber; Claus Garbe; Celeste Lebbe; Jean-François Baurain; Alessandro Testori; Jean-Jacques Grob; Neville Davidson; Jon Richards; Michele Maio; Axel Hauschild; Wilson H Miller; Pere Gascon; Michal Lotem; Kaan Harmankaya; Ramy Ibrahim; Stephen Francis; Tai-Tsang Chen; Rachel Humphrey; Axel Hoos; Jedd D Wolchok Journal: N Engl J Med Date: 2011-06-05 Impact factor: 91.245
Authors: Dirk Schadendorf; F Stephen Hodi; Caroline Robert; Jeffrey S Weber; Kim Margolin; Omid Hamid; Debra Patt; Tai-Tsang Chen; David M Berman; Jedd D Wolchok Journal: J Clin Oncol Date: 2015-02-09 Impact factor: 44.544
Authors: F Stephen Hodi; Steven J O'Day; David F McDermott; Robert W Weber; Jeffrey A Sosman; John B Haanen; Rene Gonzalez; Caroline Robert; Dirk Schadendorf; Jessica C Hassel; Wallace Akerley; Alfons J M van den Eertwegh; Jose Lutzky; Paul Lorigan; Julia M Vaubel; Gerald P Linette; David Hogg; Christian H Ottensmeier; Celeste Lebbé; Christian Peschel; Ian Quirt; Joseph I Clark; Jedd D Wolchok; Jeffrey S Weber; Jason Tian; Michael J Yellin; Geoffrey M Nichol; Axel Hoos; Walter J Urba Journal: N Engl J Med Date: 2010-06-05 Impact factor: 91.245
Authors: Marc Buyse; Stefan Michiels; Daniel J Sargent; Axel Grothey; Alastair Matheson; Aimery de Gramont Journal: Expert Rev Mol Diagn Date: 2011-03 Impact factor: 5.225
Authors: Markus Kalisch; Bernd A G Fellinghauer; Eva Grill; Marloes H Maathuis; Ulrich Mansmann; Peter Bühlmann; Gerold Stucki Journal: BMC Med Res Methodol Date: 2010-02-11 Impact factor: 4.615
Authors: Shrabanti Chowdhury; Ru Wang; Qing Yu; Catherine J Huntoon; Larry M Karnitz; Scott H Kaufmann; Steven P Gygi; Michael J Birrer; Amanda G Paulovich; Jie Peng; Pei Wang Journal: BMC Bioinformatics Date: 2022-08-05 Impact factor: 3.307