2555. Predicting Risk of Breakthrough Invasive Pneumococcal Disease in Children After 13-Valent Pneumococcal Conjugate Vaccination. (26th November 2018)

• Record Type: Journal Article
• Authors: Yildirim, Melike; Pelton, Stephen I; Keskinocak, Pinar; Yildirim, Inci
• Is Part Of: Open forum infectious diseases. Volume 5(2018)Supplement 1
• Page Start: S68
• Page End: S69
• Publication Date: 2018-11-26
• Subjects: Communicable diseases -- Periodicals; Medical microbiology -- Periodicals; Infection -- Periodicals; 616.9
• DOI: 10.1093/ofid/ofy209.163
• Languages: English
• ISSNs: 2328-8957

Abstract

Abstract: Background: Thirteen-valent-pneumococcal conjugate vaccine (PCV13) replaced PCV7 in the childhood immunization schedule in Massachusetts beginning in April 2010. We evaluated the predictors of vaccine-type (VT) invasive pneumococcal infection (IPD) occurrence despite vaccination. Methods: Cases of IPD in children <18 years of age were detected through an enhanced surveillance system in MA since 2001. All cases and Streptococcus pneumoniae (SP) isolates are submitted to Department of Public Health (MDPH) and parents/physicians are interviewed for confirmation of demographic and clinical data. All available isolates are serotyped by Quellung reaction. Children who received any dose of PCV7 were excluded from this study. We used 4-layer, feed-forward, neural network with backpropagation learning algorithm, random forest algorithm with 150 classification trees, and extreme gradient boosting (XGBoost) algorithm based on boosted trees with over than 200 iterations to make prediction about risk of nonvaccine serotype (NVST) causing IPD. Results: Overall, 144 IPD cases have been identified between April 1, 2010, and March 31, 2017, and 27 (19%) were VT IPD. Compared with children with complete PCV13 vaccination, IPD among those with incomplete immunization was more likely to be due to VT (9%, 95% CI 4–18% vs. 19%, 95% CI 10–31%, respectively). Despite complete immunization, 80 of 144 (55.6%) of all IPD in >60 months was breakthrough IPD (Figure 1). Among children with ≥1 comorbid condition and incomplete PCV13, 4 of 18 (22%) IPD were due to VT. Children with incomplete vaccination and pneumonia were most likely (11/17, 65%) to have VT; however, bacteremia without focus cases with incomplete vaccination were most likely (44/45, 3%) to have NVST (Figure 2). Our algorithm performs with 85% accuracy and 92% precision scores. Conclusion: IPD due to VT after PCV13 vaccination mostly occurs in older children with incomplete PCV13 immunization, among those with underlying comorbidity, and among those who present with pneumococcal pneumonia. Evaluating the immune response following PCV13 vaccination in children with comorbidity could increase our understanding of breakthrough pneumococcal infections despite vaccination with PCV13. Disclosures: All authors: No reported disclosures.