Semi-supervised neighborhoods and localized patient outcome prediction.

Robust statistical methods that can provide patients and their healthcare providers with individual predictions are needed to help guide informed medical decisions. Ideally an individual prediction would display the full range of possible outcomes (full predictive distribution), would be obtained with a user-specified level of precision, and would be minimally reliant on statistical model assumptions. We propose a novel method that satisfies each of these criteria via the semi-supervised creation of an axis-parallel covariate neighborhood constructed around a given point defining the patient of interest. We then provide non-parametric estimates of the outcome distribution for the subset of subjects in this neighborhood, which we refer to as a localized prediction. We implement local prediction methods using dynamic graphical methods which allow the user to vary key options such as the choice of the variables defining the neighborhood, and the size of the neighborhood.

Preliminary aggregate safety and immunogenicity results from three trials of a purified inactivated Zika virus vaccine candidate: phase 1, randomised, double-blind, placebo-controlled clinical trials.

BACKGROUND: A safe, effective, and rapidly scalable vaccine against Zika virus infection is needed. We developed a purified formalin-inactivated Zika virus vaccine (ZPIV) candidate that showed protection in mice and non-human primates against viraemia after Zika virus challenge. Here we present the preliminary results in human beings. METHODS: We did three phase 1, placebo-controlled, double-blind trials of ZPIV with aluminium hydroxide adjuvant. In all three studies, healthy adults were randomly assigned by a computer-generated list to receive 5 µg ZPIV or saline placebo, in a ratio of 4:1 at Walter Reed Army Institute of Research, Silver Spring, MD, USA, or of 5:1 at Saint Louis University, Saint Louis, MO, USA, and Beth Israel Deaconess Medical Center, Boston, MA, USA. Vaccinations were given intramuscularly on days 1 and 29. The primary objective was safety and immunogenicity of the ZPIV candidate. We recorded adverse events and Zika virus envelope microneutralisation titres up to day 57. These trials are registered at ClinicalTrials.gov, numbers NCT02963909, NCT02952833, and NCT02937233. FINDINGS: We enrolled 68 participants between Nov 7, 2016, and Jan 25, 2017. One was excluded and 67 participants received two injections of Zika vaccine (n=55) or placebo (n=12). The vaccine caused only mild to moderate adverse events. The most frequent local effects were pain (n=40 [60%]) or tenderness (n=32 [47%]) at the injection site, and the most frequent systemic reactogenic events were fatigue (29 [43%]), headache (26 [39%]), and malaise (15 [22%]). By day 57, 52 (92%) of vaccine recipients had seroconverted (microneutralisation titre ≥1:10), with peak geometric mean titres seen at day 43 and exceeding protective thresholds seen in animal studies. INTERPRETATION: The ZPIV candidate was well tolerated and elicited robust neutralising antibody titres in healthy adults. FUNDING: Departments of the Army and Defense and National Institute of Allergy and Infectious Diseases.