Target Product Profile for a Diagnostic Assay to Differentiate between Bacterial and Non-Bacterial Infections and Reduce Antimicrobial Overuse in Resource-Limited Settings: An Expert Consensus

Abstract

Acute fever is one of the most common presenting symptoms globally. In order to reduce the empiric use of antimicrobial drugs and improve outcomes, it is essential to improve diagnostic capabilities. In the absence of microbiology facilities in low-income settings, an assay to distinguish bacterial from non-bacterial causes would be a critical first step. To ensure that patient and market needs are met, the requirements of such a test should be specified in a target product profile (TPP). To identify minimal/optimal characteristics for a bacterial vs. non-bacterial fever test, experts from academia and international organizations with expertise in infectious diseases, diagnostic test development, laboratory medicine, global health, and health economics were convened. Proposed TPPs were reviewed by this working group, and consensus characteristics were defined. The working group defined non-severely ill, non-malaria infected children as the target population for the desired assay. To provide access to the most patients, the test should be deployable to community health centers and informal health settings, and staff should require <2 days of training to perform the assay. Further, given that the aim is to reduce inappropriate antimicrobial use as well as to deliver appropriate treatment for patients with bacterial infections, the group agreed on minimal diagnostic performance requirements of >90% and >80% for sensitivity and specificity, respectively. Other key characteristics, to account for the challenging environment at which the test is targeted, included: i) time-to-result <10 min (but maximally <2 hrs); ii) storage conditions at 0–40°C, ≤90% non-condensing humidity with a minimal shelf life of 12 months; iii) operational conditions of 5–40°C, ≤90% non-condensing humidity; and iv) minimal sample collection needs (50–100μL, capillary blood). This expert approach to define assay requirements for a bacterial vs. non-bacterial assay should guide product development, and enable targeted and timely efforts by industry partners and academic institutions.

Biomarkers to distinguish bacterial from viral pediatric clinical pneumonia in a malaria endemic setting

“…These models meet or exceed the performance guidelines proposed by a FIND-sponsored expert consensus document for a diagnostic test of bacterial pneumonia. 14 Because the BvV model was highly predictive, we investigated the underlying marker proteins for information about the inflammatory processes that typify bacterial and viral infections. Gene enrichment and pathway analyses showed that the processes in bacterial infections were dominated by neutrophils.…”

“…Thus, host response signatures have been explored as possible diagnostic indicators. To date, however, these approaches have not proven to be sufficiently reliable [9][10][11][12][13] , based on a recommended benchmark 14 of acceptable and desirable thresholds for sensitivity (desirable ≥95%, acceptable ≥90%) and specificity (≥90% and ≥80%) for diagnostic tests. We hypothesized that the distinctive cellular host responses could be detected at the protein level.…”

Diagnostic tools for tackling febrile illness and enhancing patient management

“…The ASSURED criteria, together with additional specific requirements, are summarized in Target Product Profiles (TPPs), which technology developers should meet during their development stages. Examples of TPPs exist for tuberculosis, Ebola, and differentiation between viral and bacterial infections, and are driven by the WHO, the Médecins Sans Frontières (MSF), the Foundation for Innovative New Diagnostics (FIND), and other organizations [36,37].…”

Point-of-care and point-of-‘can’: leveraging reference-laboratory capacity for integrated diagnosis of fever syndromes in the tropics

“…Healthcare-or community-associated invasive bacterial infection Gram stain, where expertise exists; however not of use for bloodstream infections and prone to many errors [12] Bacterial culture and antimicrobial susceptibility testing [12,19] Implementation of quality-assured basic techniques is lacking in most LRS reference laboratories and should be prioritized [12,19] At the primary level: biomarker and digitally assisted decision tools to distinguish bacterial from non-bacterial fevers [8] and/or multi-disease detection platforms [64] At the secondary levelþ: simplified approaches to classic bacteriology techniques [12], new tools for rapid phenotypic determination of AST NAAT and MALDI-TOF-based AST face significant challenges because of the wide array and changeability of mechanisms of resistance Dengue…”

“…Combination RDTs that detect both IgM/IgG and the NS1 antigen [65]. Able to diagnose dengue at different points in time after initiation of symptoms Commercial availability and production history of RDT has been inconsistent since its introduction At the primary level: biomarker and digitally assisted decision tools to distinguish bacterial from non-bacterial fevers [8] and/or multi-disease detection platforms [64] At the secondary levelþ: simplified approaches to classic bacteriology techniques [12,50] resource settings may play in the near future, and how they can interact with end-users and NMFI patients ( Fig. 1).…”