Jorge M Arevalillo1, Marcelo B Sztein2, Karen L Kotloff3, Myron M Levine4, Jakub K Simon5. 1. Department of Statistics and Operational Research, University Nacional Educación a Distancia, Paseo Senda del Rey 9, 28040 Madrid, Spain. Electronic address: jmartin@ccia.uned.es. 2. Center for Vaccine Development, Departments of Pediatrics and Medicine, University of Maryland School of Medicine, Baltimore, MD 21201, USA. Electronic address: msztein@medicine.umaryland.edu. 3. Center for Vaccine Development, Departments of Pediatrics and Medicine, University of Maryland School of Medicine, Baltimore, MD 21201, USA. Electronic address: KKOTLOFF@medicine.umaryland.edu. 4. Center for Vaccine Development, Departments of Pediatrics and Medicine, University of Maryland School of Medicine, Baltimore, MD 21201, USA. Electronic address: MLevine@medicine.umaryland.edu. 5. Merck & Co., Inc., Kenilworth, NJ, USA. Electronic address: jakub.simon@merck.com.
Abstract
BACKGROUND: Immunologic correlates of protection are important in vaccine development because they give insight into mechanisms of protection, assist in the identification of promising vaccine candidates, and serve as endpoints in bridging clinical vaccine studies. Our goal is the development of a methodology to identify immunologic correlates of protection using the Shigella challenge as a model. METHODS: The proposed methodology utilizes the Random Forests (RF) machine learning algorithm as well as Classification and Regression Trees (CART) to detect immune markers that predict protection, identify interactions between variables, and define optimal cutoffs. Logistic regression modeling is applied to estimate the probability of protection and the confidence interval (CI) for such a probability is computed by bootstrapping the logistic regression models. RESULTS: The results demonstrate that the combination of Classification and Regression Trees and Random Forests complements the standard logistic regression and uncovers subtle immune interactions. Specific levels of immunoglobulin IgG antibody in blood on the day of challenge predicted protection in 75% (95% CI 67-86). Of those subjects that did not have blood IgG at or above a defined threshold, 100% were protected if they had IgA antibody secreting cells above a defined threshold. Comparison with the results obtained by applying only logistic regression modeling with standard Akaike Information Criterion for model selection shows the usefulness of the proposed method. CONCLUSION: Given the complexity of the immune system, the use of machine learning methods may enhance traditional statistical approaches. When applied together, they offer a novel way to quantify important immune correlates of protection that may help the development of vaccines.
BACKGROUND: Immunologic correlates of protection are important in vaccine development because they give insight into mechanisms of protection, assist in the identification of promising vaccine candidates, and serve as endpoints in bridging clinical vaccine studies. Our goal is the development of a methodology to identify immunologic correlates of protection using the Shigella challenge as a model. METHODS: The proposed methodology utilizes the Random Forests (RF) machine learning algorithm as well as Classification and Regression Trees (CART) to detect immune markers that predict protection, identify interactions between variables, and define optimal cutoffs. Logistic regression modeling is applied to estimate the probability of protection and the confidence interval (CI) for such a probability is computed by bootstrapping the logistic regression models. RESULTS: The results demonstrate that the combination of Classification and Regression Trees and Random Forests complements the standard logistic regression and uncovers subtle immune interactions. Specific levels of immunoglobulin IgG antibody in blood on the day of challenge predicted protection in 75% (95% CI 67-86). Of those subjects that did not have blood IgG at or above a defined threshold, 100% were protected if they had IgA antibody secreting cells above a defined threshold. Comparison with the results obtained by applying only logistic regression modeling with standard Akaike Information Criterion for model selection shows the usefulness of the proposed method. CONCLUSION: Given the complexity of the immune system, the use of machine learning methods may enhance traditional statistical approaches. When applied together, they offer a novel way to quantify important immune correlates of protection that may help the development of vaccines.
Authors: Y Allory; C Bazille; A Vieillefond; V Molinié; B Cochand-Priollet; O Cussenot; P Callard; M Sibony Journal: Histopathology Date: 2007-11-22 Impact factor: 5.087
Authors: K L Kotloff; G A Losonsky; J P Nataro; S S Wasserman; T L Hale; D N Taylor; J W Newland; J C Sadoff; S B Formal; M M Levine Journal: Vaccine Date: 1995-04 Impact factor: 3.641
Authors: Thembi Mdluli; Ningbo Jian; Bonnie Slike; Dominic Paquin-Proulx; Gina Donofrio; Aljawharah Alrubayyi; Syna Gift; Rebecca Grande; Mary Bryson; Anna Lee; Vincent Dussupt; Letzibeth Mendez-Riveria; Eric Sanders-Buell; Agnès-Laurence Chenine; Ursula Tran; Yifan Li; Eric Brown; Paul T Edlefsen; Robert O'Connell; Peter Gilbert; Sorachai Nitayaphan; Punnee Pitisuttihum; Supachai Rerks-Ngarm; Merlin L Robb; Robert Gramzinski; Galit Alter; Sodsai Tovanabutra; Ivelin S Georgiev; Margaret E Ackerman; Victoria R Polonis; Sandhya Vasan; Nelson L Michael; Jerome H Kim; Michael A Eller; Shelly J Krebs; Morgane Rolland Journal: PLoS Pathog Date: 2020-12-08 Impact factor: 6.823
Authors: Biana Bernshtein; Esther Ndungo; Deniz Cizmeci; Peng Xu; Pavol Kováč; Meagan Kelly; Dilara Islam; Edward T Ryan; Karen L Kotloff; Marcela F Pasetti; Galit Alter Journal: Cell Rep Date: 2022-08-16 Impact factor: 9.995
Authors: Kristen A Clarkson; Robert W Frenck; Michelle Dickey; Akamol E Suvarnapunya; Lakshmi Chandrasekaran; Hailey P Weerts; Christopher D Heaney; Monica McNeal; Kate Detizio; Susan Parker; Amy Hoeper; August L Bourgeois; Chad K Porter; Malabi M Venkatesan; Robert W Kaminski Journal: mSphere Date: 2020-09-23 Impact factor: 4.389
Authors: Kristen A Clarkson; Chad K Porter; Kawsar R Talaat; Robert W Frenck; Cristina Alaimo; Patricia Martin; A Louis Bourgeois; Robert W Kaminski Journal: mSphere Date: 2021-07-14 Impact factor: 4.389