Celine Le Bailly De Tilleghem

Statistics Belgium

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Trends

Author Order

Document Type

Co-Authors

Peter A. Fasching

Comprehensive Cancer Center Erlangen

Bernadette Govaerts

UCLouvain

Amin Haiderali

Merck & Co., Inc., Rahway, NJ, USA (United States)

Albert Leung

Mallinckrodt (United States)

Nicolas Cappoen

Statistics Belgium

Carolyn DaSilva

Merck & Co., Inc., Rahway, NJ, USA (United States)

Joyce O’Shaughnessy

Texas Oncology

P. Hu

Merck & Co., Inc., Rahway, NJ, USA (United States)

Mitashri Chaudhuri

Merck & Co., Inc., Rahway, NJ, USA (United States)

Peter Hu

Merck & Co., Inc., Rahway, NJ, USA (United States)

Cooperative Institutions

Merck & Co., Inc., Rahway, NJ, USA (United States)

Merck (Japan)

United States Military Academy

Analysis Group (United States)

Institute of Group Analysis

University of Florida

KU Leuven

University of Liège

Statistics Belgium

Weatherford College

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

Uncertainty propagation in multiresponse optimization using a desirability index

Bernadette Govaerts Celine Le Bailly De Tilleghem

Optimizing the quality of a product is widespread in the industry. Products have to be manufactured such that they best fit some quality properties. Varying the product settings leads to different product qualities and the aim of the manufacturer is to find the factors settings that simultaneously optimize the quality properties. The classical approach to solve such optimization problem is based on response surface methodology. First, a designed experiment is used to collect data and to adjust models capturing the relationship between the responses of interest and the factors settings. Those fitted models can then predict the quality properties for any design point of the experimental domain. Secondly, a desirability index is built to combine the predicted properties into a value belonging to the [0; 1] interval. This index provides a ranking of possible factors settings in the solutions space and the optimum can be found by an adequate optimization algorithm. But, as model predictions are suiled with error, so is the desirability index and the optimal solution found. In practice, in the related literature and design of experiment software, this error is neglected. This paper proposes an optimization methodology based on the fact that a desirability index is a random variable. The expectation of this index is taken as the criteria to be optimized and, since it can only be estimated, condence and predicted intervals are constructed to take into account the propagation of the models error on the expected or predicted desirability index. The stochastic character of the index leads also to an uncertainty on the optimum and a methodology is proposed to build an equivalence zone containing no significantly different optimal solutions. This methodology is illustrated on a simulated example and compared to the classical optimization methodology.

Source

Cite

Citations (4)

A review of quantitative structure-activity relationship (QSAR) models

Celine Le Bailly De Tilleghem Bernadette Govaerts

Source

Cite

Citations (4)

159P Q-TWiST analysis of pembrolizumab combined with chemotherapy as first-line treatment of metastatic TNBC that expresses PD-L1

Annals of Oncology (2022)

Peter A. Fasching Mo Huang Amin Haiderali W. Pan P. Hu

Triple-negative breast cancer

10.1016/j.annonc.2022.03.177

Cite

Citations (0)

Cost–Effectiveness of Pembrolizumab Plus Chemotherapy as First-Line Treatment in PD-L1-Positive Metastatic Triple-Negative Breast Cancer

Immunotherapy (2022)

Min Huang Peter A. Fasching Amin Haiderali Wilbur Pan Emma Gray

Objective: This study evaluated the cost–effectiveness of pembrolizumab/chemotherapy combinations for previously untreated metastatic triple-negative breast cancer patients in the USA with PD-L1 combined positive score ≥10. Methods: A partitioned-survival model was developed to project health outcomes and direct medical costs over a 20-year time horizon. Efficacy and safety data were from randomized clinical trials. Comparative effectiveness of indirect comparators was assessed using network meta-analyses. A series of sensitivity analyses were performed to test the robustness of the results. Results: Pembrolizumab/chemotherapy resulted in total quality-adjusted life-year (QALY) gains of 0.70 years and incremental cost–effectiveness ratio of US$182,732/QALY compared with chemotherapy alone. The incremental cost–effectiveness ratio for pembrolizumab/nab-paclitaxel versus atezolizumab/nab-paclitaxel was US$44,157/QALY. Sensitivity analyses showed the results were robust over plausible values of model inputs. Conclusion: Pembrolizumab/chemotherapy is cost effective compared with chemotherapy as well as atezolizumab/nab-paclitaxel as first-line treatment for PD-L1-positive metastatic triple-negative breast cancer from a US payer perspective.

Atezolizumab

Triple-negative breast cancer

10.2217/imt-2022-0082

Cite

Citations (8)

Efficacy and safety of losartan in children with Alport syndrome--results from a subgroup analysis of a prospective, randomized, placebo- or amlodipine-controlled trial

Nephrology Dialysis Transplantation (2011)

Nicholas J.A. Webb Chun Sing Lam S. Shahinfar J. Strehlau Thomas G. Wells

No prospective, randomized, double-blind trials of angiotensin-converting enzyme inhibitors or angiotensin receptor blockers have previously been reported in adults or children with proteinuria secondary to Alport syndrome.This 12-week, double-blind multinational study investigated the effects of losartan 0.7-1.4 mg/kg/day compared with placebo (normotensive patients) or amlodipine 0.1-0.2 mg/kg/day up to 5 mg/day (hypertensive patients) on proteinuria [early morning-void urinary protein/creatinine ratio (UPr/Cr), baseline ≥ 34 mg/mmol] in 30 children of up to 17 years of age with Alport syndrome.Twelve weeks of treatment with losartan significantly reduced proteinuria compared with placebo/amlodipine: losartan -14.7 mg/mmol (interquartile range -49.7 to -5.7 mg/mmol) or 31.6% reduction using a mixed model approach versus placebo/amlodipine 2.3 mg/mmol (-26.0 to 18.1 mg/mmol), P = 0.01 or 2.3% increase using a mixed model approach. Adverse event incidence was low and comparable between losartan and placebo/amlodipine groups.Losartan significantly lowered proteinuria and was well tolerated after 12 weeks of treatment in children aged 1-17 years with proteinuria secondary to Alport syndrome with or without hypertension, a population that has not previously been rigorously studied.

Amlodipine

Interquartile range

10.1093/ndt/gfq797

Cite

Citations (59)

Odanacatib in the treatment of postmenopausal women with low bone mineral density: Five years of continued therapy in a phase 2 study

Journal of Bone and Mineral Research (2012)

Bente Langdahl Neil Binkley Henry G. Bone Nigel Gilchrist Heinrich Resch

Odanacatib (ODN) is a selective inhibitor of the collagenase cathepsin K that is highly expressed by osteoclasts. In this 2-year, phase 2, dose-ranging trial, postmenopausal women with bone mineral density (BMD) T-scores -2.0 to -3.5 at spine or hip were randomized to weekly placebo or ODN 3, 10, 25, or 50 mg plus vitamin D(3) and calcium. Prespecified trial-extensions continued through 5 years. In year 3, all women were re-randomized to ODN 50 mg or placebo. For years 4 and 5, women who received placebo or ODN 3 mg in years 1 and 2 and placebo in year 3 received ODN 50 mg; others continued year 3 treatments. Endpoints included lumbar spine (primary), hip, 1/3 radius, and total body BMD; markers of bone metabolism; and safety. Women in the year 4 to 5 extension receiving placebo (n = 41) or ODN 50 mg (n = 100) had similar baseline characteristics. For women who received ODN (10-50 mg) for 5 years, spine and hip BMD increased over time. With ODN 50 mg continually for 5 years (n = 13), mean lumbar spine BMD percent change from baseline (95% confidence interval [CI]) was 11.9% (7.2% to 16.5%) versus -0.4% (-3.1% to 2.3%) for women who were switched from ODN 50 mg to placebo after 2 years (n = 14). In pooled results of women receiving continuous ODN (10-50 mg, n = 26-29), year 5 geometric mean percent changes from baseline in bone resorption markers cross-linked N-telopeptide of type I collagen (NTX)/creatinine and cross-linked C-telopeptide (CTX) were approximately -55%, but near baseline for bone formation markers bone-specific alkaline phosphatase (BSAP) and amino-terminal propeptide of type I procollagen (P1NP). In women switched from ODN 10 to 50 mg to placebo after 2 years (n = 25), bone turnover markers were near baseline. In summary, women receiving combinations of ODN (10-50 mg) for 5 years had gains in spine and hip BMD and showed larger reductions in bone resorption than bone formation markers. Discontinuation of ODN resulted in reversal of treatment effects. Treatment with ODN for up to 5 years was generally well-tolerated.

10.1002/jbmr.1695

Cite

Citations (166)

Cost-Effectiveness of Neoadjuvant Pembrolizumab Plus Chemotherapy Followed by Adjuvant Single-Agent Pembrolizumab for High-Risk Early-Stage Triple-Negative Breast Cancer in the United States

Advances in Therapy (2023)

Min Huang Peter A. Fasching Amin Haiderali Weiguang Xue Chelsey Yang

The randomized phase III KEYNOTE-522 trial demonstrated that addition of pembrolizumab to neoadjuvant chemotherapy provided a significant improvement in event-free survival and a favorable trend in overall survival for high-risk early-stage triple-negative breast cancer (eTNBC). This analysis evaluated the cost-effectiveness of pembrolizumab in combination with chemotherapy as neoadjuvant treatment and continued as a single-agent adjuvant treatment after surgery vs. neoadjuvant chemotherapy for patients with high-risk eTNBC in the USA.The analysis was conducted from a US third-party public healthcare payer perspective. A multistate transition model was developed using efficacy and safety data from the KEYNOTE-522 trial. The model included four mutually exclusive health states: event-free, locoregional recurrence, distant metastasis, and death to simulate patients' lifetime disease course. Quality-adjusted life years (QALYs) were calculated on the basis of EuroQoL-5 Dimensions utility data collected in KEYNOTE-522. Costs for drug acquisition/administration, adverse events, disease management, and subsequent therapies were reported (2021 US dollars). Costs and outcomes were discounted at 3% annually. A series of sensitivity analyses were performed to test the robustness of the main results.In the base case scenario, pembrolizumab plus chemotherapy followed by pembrolizumab resulted in expected gains of 3.37 life years (LYs) and 2.90 QALYs, and an incremental cost of $79,046 versus chemotherapy. The incremental cost per QALY gained was $27,285, which is lower than all commonly cited US willingness-to-pay thresholds. Sensitivity analyses showed the results were robust over plausible values of key model inputs and assumptions.Compared with neoadjuvant chemotherapy, pembrolizumab in combination with chemotherapy as neoadjuvant treatment and continued as a single-agent adjuvant treatment after surgery is considered a cost-effective option for high-risk eTNBC in the USA.

Neoadjuvant Therapy

Triple-negative breast cancer

10.1007/s12325-022-02365-1

Cite

Citations (9)

99P Cost effectiveness of pembrolizumab in combination with chemotherapy as neoadjuvant therapy and continued as a single agent as adjuvant therapy for high-risk early-stage TNBC in the United States

Annals of Oncology (2022)

Peter A. Fasching Mo Huang Amin Haiderali Weiguang Xue W. Pan

Neoadjuvant Therapy

10.1016/j.annonc.2022.03.115

Cite

Citations (3)

Measuring the Impact of a Dependence Among Insured Lifelengths

Belgian Actuarial Bulletin (2001)

Michel Denuit Jan Dhaene Celine Le Bailly De Tilleghem S Teghem

Actuaries usually compute multiple life premiums based on the unrealistic assumption of independence of the lifelengths of insured persons. Many clinical studies, however, have demonstrated dependence of the lifetimes of paired lives such as husband and wife. In this respect, the present article tries to give an answer to the following question: does this simplifying hypothesis constitute a real financial danger for the insurance company? The answer turns out to be affirmative: this dependence materially affects the values of annuities and insurances involving multiple lives. In order to quantify the impact of a possible dependence on the amount of premium charged for annuities, insurances and widow's pension, we resort here on the Frechet bounds, Markov processes and some copula models. These techniques are applied to classical insurance contracts issued to married couples and illustrated on NIS data as well as on observations from Brussels city.

Annuity

Life insurance

Independence

Source

Cite

Citations (61)

Statistical contribution to the virtual multicriteria optimisation of combinatorial molecules libraries and to the validation and application of QSAR models

Celine Le Bailly De Tilleghem

This thesis develops an integrated methodology based on the desirability index and QSAR models to virtually optimise molecules. Statistical and algorithmic tools are proposed to search in huge collections of compounds obtained by combinatorial chemistry the most promising ones. First, once the drugability properties of interest have been precisely defined, QSAR models are developed to mimic the relationship between those optimised properties and chemical descriptors of molecules. The literature on QSAR models is reviewed and the statistical tools to validate the models, analyse their fit and their predictive power are detailed. Even if a QSAR model has been validated and sounds highly predictive, we emphasise the importance of measuring extrapolation by the definition of its applicability domain and quantifying the prediction error for a given molecule. Indeed, QSAR models are often massively applied to predict drugability properties for libraries of new compounds without taking care of the reliability of each individual prediction. Then, a desirability index measures the compromise between the multiple estimated drugability properties and allows to rank the molecules in the combinatorial library in preference order. The propagation of the models prediction error on the desirability index is quantified by a confidence interval that can be constructed under general conditions for linear regression, PLS regression or regression tree models. This fulfills an important lack of the desirability index literature that considers it as exact. Finally, a new efficient algorithm (WEALD) is proposed to virtually screen the combinatorial library and retain the molecule with the highest desirability indexes. For each explored molecule, it is checked if it belongs to the applicability domain of each QSAR models. In addition, the uncertainty of the desirability index of each explored molecule is taken into account by gathering molecules that can not be distinguished from the optimal one due to the propagation of QSAR models prediction error. Those molecules do not have a significantly smaller desirability than the optimal molecule found by WEALD. This constitutes another important improvement in the use of desirability index as a tool to compare solutions in a multicriteria optimisation problem. This integrated methodology has been developed in the context of lead optimisation and is illustrated on a real combinatorial library provided by Eli Lilly and Company. This is the main application of the thesis. Nevertheless, as the results on desirability index uncertainty are applicable under general conditions, they can be applied to any multicriteria optimisation problem, like it often occurs in industry.

Applicability domain

Molecular descriptor

Source

Cite

Citations (0)