This article is more than 5 years old. Information may no longer be current.
New method developed for monitoring fitness of antiviral-resistant influenza strains
In a mathematical modeling study published in Lancet Infectious Diseases, researchers introduced a method for providing reliable, real-time estimates of the fitness of antiviral-resistant influenza strains.
In the last 10 years, surveillance of antiviral-resistant (AVR) strains of influenza has been deemed essential for controlling influenza epidemics and pandemics. However, few advances in analytics and assessment of AVR surveillance data have been made in that time. Therefore, researchers aimed to develop a simple method for estimating AVR fitness from surveillance data.
“Characterization of the nonlinear epidemic dynamics underlying surveillance data typically requires inference of multiple parameters in transmission models,” Joseph T. Wu, PhD, associate professor in the division of epidemiology and biostatistics at the School of Public Health at the University of Hong Kong, and colleagues wrote. “Our method bypasses such complexity and is therefore easy to implement.”
Defining the fitness of AVR strains as their reproductive number relative to their co-circulating antiviral-sensitive (AVS) counterparts, researchers developed a model requiring information only on generation time with no other details regarding transmission dynamics.
To validate this method, they used simulations to show that the model yielded unbiased and robust fitness estimates in most epidemic scenarios.
After validating their method, they applied it to two retrospective case studies and one hypothetical case study.
In the first case study — a retrospective study of the Tamiflu (oseltamivir, Roche)-resistant influenza A(H1N1) virus in 2007-2008 — Wu and colleagues estimated that the resistant strain was 4% (95% credible interval [CrI], 3-5) more transmissible than the oseltamivir-sensitive strain.
In the second case study, they estimated that the oseltamivir-resistant pandemic A(H1N1) strain that emerged and circulated in Japan in 2013-2014 was 24% (95% CrI, 17-30) less transmissible than its drug-sensitive counterpart. They noted that, in this case, their method “could have correctly predicted that the AVR virus was less transmissible than its AVS counterpart … after both viruses had co-circulated for 2 weeks.”
The third case study was a hypothetical model of AVR fitness and drug pressure under large-scale antiviral interventions during a pandemic with co-circulation of AVS and AVR strains. The researchers concluded that, under these circumstances, their method could be used to inform optimal use of antivirals. This, they noted, could be essential for reducing mortality in the context of a large-scale antiviral intervention during a pandemic.
At the same time, they acknowledged that “a comprehensive evaluation of optimal antiviral use would require knowledge of additional parameters (eg, reproductive number and antiviral efficacy in reducing mortality).”
In a related editorial, Gerardo Chowell, PhD, of Georgia State University School of Public Health, and Cecile Viboud, PhD, of Fogarty International Center at NIH, wrote that the model presented by Wu and colleagues is an important step toward monitoring antiviral resistance in influenza. However, they argued, it is still unknown whether surveillance systems can provide sufficient data to guide policy in near real time.
“Accurate long-term predictions of the rise of antiviral resistance cannot proceed without a more detailed understanding of the mechanisms at play, which will require further experimental work combined with epidemiological and phylogenetic analyses,” they concluded. – by Sarah Kennedy
Disclosure: The researchers and editorial authors report no relevant financial disclosures.