Concerns about Viruses
Larry H. Bernstein, MD, FCAP, Curator
Sanjay Gupta, MD Medscape http://www.medpagetoday.com/InfectiousDisease/GeneralInfectiousDisease/55915
No one really saw Zika virus coming or cared much if it did.
In general, it has been regarded “clinically inconsequential,” Anthony Fauci, MD, director of the National Institute of Allergy and Infectious Diseases, told MedPage Today — so much so that it wasn’t even on a recent World Health Organization list of pathogens that need urgent research to prevent epidemics.
And — absent its apparent association with a spike in cases of microcephaly in Brazil — it probably still would be thought of as a minor nuisance, experts told MedPage Today.
But Zika virus illustrates a worrisome fact — the pace of emerging infectious diseases is both increasing and unpredictable.
Zika is a flavivirus, discovered in 1947, that is carried mainly by the mosquito Aedes aegypti. It causes a mild, self-limiting febrile illness in 20% to 25% of the people it infects; most people would never know they had it.
Until recently, it was pretty much confined to its ancestral home in Africa. Then in 2007 it was found in Micronesia and in 2013 ongoing transmission was documented in French Polynesia.
Early last year, it made its appearance in Brazil and it now appears to be established in 20 countries or territories in the Americas, including Puerto Rico.
Given that much of the region also has endemic dengue fever and chikungunya — with similar but more serious symptoms and also carried by A. aegypti — the appearance of Zika virus was originally just recorded with the notation that it would be nice to know more about these concurrent infections.
Then in September 2015, reports emerged of a spike in cases of microcephaly in the region of Brazil where the outbreak of Zika had been noted. It was an alarm bell, even though there’s still no definitive proof that Zika infection causes microcephaly.
“Microcephaly is obviously where the significant global public health concern is,” according to Michael Diamond, MD, PhD, of the Washington University School of Medicine in St. Louis.
But, he told MedPage Today, there have also been reports of a spike in Guillain-Barré syndrome during the Polynesian outbreak. Again, it’s an association with nothing to prove that Zika was responsible.
Still, there are now two clinical syndromes that have appeared at the same time as a Zika outbreak. It might be coincidence but health officials are urging precautions anyway.
And it’s yet another instance of a pathogen emerging from the shadows.
The term “emerging diseases” is widely used but it’s often bit of a misnomer. Many such pathogens are bugs that have moved into new places, while a few are actually novel. In the latter group, put HIV, SARS and MERS. But Zika — like dengue, chikungunya, and West Nile virus — is a traveller.
Previously known or not, the list of such pathogens has grown in recent years. Consider a partial list: West Nile in 1999, SARS in 2002, the H1N1 pandemic influenza in 2009, MERS in 2012, Ebola in 2013, chikungunya in 2013.
And now Zika.
Is the apparent increase real? If so, what’s causing it? And what can we do?
The answers, experts told MedPage Today, are:
- Yes, it’s probably a real phenomenon
- It has multiple causes
- And while there are steps we can take to minimize the effects — if we have the will and the cash — emerging diseases are going to be a continuing problem
“We’re definitely seeing more, there’s no question about it,” according to James LeDuc, PhD, director of the Galveston National Laboratory.
And it really shouldn’t be a surprise, he told MedPage Today: The National Academy of Sciences warned in 1992 that infectious disease had not been conquered and that — as a consequence of human activities — we were likely to see more and more pathogens spreading beyond their ancestral ranges.
The causes, that 1992 report said, include:
- Increasing human populations, often pushing into new places and coming in contact with new pathogens
- More and faster travel
- Growing urbanization
- Erosion of some traditional public health infrastructure, such as mosquito control programs
To those, we might have to add climate change, LeDuc said. For some of the mosquito-borne diseases especially, climate change might expand or move their ranges, as temperate regions become semi-tropical.
What propels Zika into the headlines is the link with birth defects. And however nuanced health officials try to be — it’s only an association, we still need more research, there might be other causes — just making the link creates fear.
“We’re still trying to figure out what’s going on with Zika and microcephaly,” commentedHeidi Brown, PhD, of the University of Arizona in Tucson. That’s going to take a lot of study and some time.
Put simply, an outbreak of disease needs three pro-conditions, Brown told MedPage Today: “You need the vector, you need the virus, and you need a human population that is susceptible.”
In the case of Zika, as well as dengue fever and chikungunya, the vector is A. aegypti, a mosquito that historically was implicated in the spread of yellow fever. In the early part of the last century, huge public health programs aimed to eradicate A. aegypti, with some success at reducing yellow fever.
But it’s the fate of successful public health programs to wither once they’ve achieved success and A. aegypti has made a comeback.
And, Brown said, A. aegypti is an “urban mosquito” — it likes to feed on people and to breed in the pools of standing water we all-too-often leave around our dwellings. The increasingly large cities of South and Central America, usually with slums where people can’t afford window screens or other protection against mosquito bites, offer a huge pool of targets.
Then an outbreak is just one plane ride away. “It’s very easy now for an infected person or an infected mosquito to move from one area to another,” Diamond said.
Of course, local conditions play an important role. Fauci told MedPage Today that it’s unlikely Zika will make huge inroads into the U.S. for two reasons. In the first place, most of the country has a severe enough winter to cause the mosquitoes to die off. And second, he said, “we can do vector control if we want to do vector control” — a capability that some other countries in the Americas don’t have.
Other experts noted that our cities are less densely populated than those in South and Central America and conditions are better — there’s air conditioning and household screens.
The same applies to other pathogens carried by A. aegypti, like dengue. But not every pathogen needs a mosquito. It’s still not clear what is the animal reservoir for Ebola, for instance, but in the recent epidemic, the vector was good old Homo sapiens. And other pathogens have intermediate hosts that don’t necessarily die off in the winter.
If the pathogens are likely to keep coming, what can we do? Slowing the pace and speed of travel is a nonstarter, we’re not going to stop living in cities, and our numbers, while growth is slowing, continue to rise.
In other words, the third of Brown’s triad — the pool of susceptible people — is going to remain.
That leaves the vector or the pathogen.
“Mosquitoes, in the end, don’t contribute much to society,” Diamond said, so A. aegypti is an obvious target if we want to prevent Zika, dengue, chikungunya, or yellow fever. And it’s something we know how to do, noted LeDuc, citing the mass eradication campaigns of the 20th century.
But that effort used “armies of people,” LeDuc noted. “That kind of commitment is just not economically feasible today,” he said.
On the other hand, the modern age has brought new tools. For instance, Australian researchers, focusing on dengue, think they can use Wolbachia, bacteria found in many insects, as a way to reduce the ability of A. aegypti to transmit viruses.
And the Brazilian city of Piracicaba is working with a British company, Oxitec, to release male mosquitoes genetically modified so their offspring don’t survive. The males don’t bite, so they can’t transmit disease, but if they outcompete normal males for mates, the net result would be a reduction in adult mosquitoes.
A similar program to prevent screwworm among livestock has been working in the U.S. since the 1950s, Brown said, so it’s not a pipe dream. But neither approach is a “silver bullet,” she said, and will need to be used in combination with other approaches.
Some approaches are decidedly low-tech. Eliminate sources of standing water. Wear insect repellent if you’re somewhere with mosquitoes. Ditto long sleeves and long pants. Put up bednets.
Those have the advantage that they work against all mosquito species, Diamond said, and therefore lots of pathogens.
A recurring theme in the story of emerging diseases is that there are no specific treatments and no vaccines. And if you think about it, that makes perfect sense — if we don’t know something is coming (because it’s emerging, after all), how can we have a vaccine or a therapy?
So consider the current Zika outbreak. Most people working in the field would not have predicted it for the next viral epidemic in the Americas and if they had would not have been especially worried.
“It took a lot of people by surprise and they were perhaps a little bit dismissive,” Brown said.
Other pathogens — Lassa fever, Rift Valley fever, Marburg, and MERS among them — might well have been higher on the priority list. Indeed, they are higher on the WHO’s blueprint for future research into epidemic prevention.
And who pays for the research? It’s not as if there is a huge commercial market for a vaccine or treatment for Zika, which in most cases causes mild or no illness. There might be a better market for other pathogens but how do you know where to focus?
The problem with vaccinology in this field, LeDuc said, is that vaccines generally have to be pathogen-specific and they are costly to develop. They’re also technically challenging; work on a dengue vaccine has been going on for years, he noted.
That said, Fauci commented, researchers on West Nile virus have developed a “platform” for a flavivirus vaccine that might be quickly adaptable to Zika. The issue then would be getting it through the regulatory hurdles and into the field — a long expensive process.
Even if a vaccine were available, how would it be used?
Writing with a colleague recently in the New England Journal of Medicine, Fauci noted that outbreaks are unpredictable, so vaccinating a population against a given pathogen would not be cost-effective, while stockpiling a vaccine for later deployment might be too slow to stop an epidemic.
And, of course, both approaches depend on knowing the pathogen is there or on its way.
The Ebola epidemic, which left thousands dead in West Africa, was missed for months because health officials in the region weren’t expecting it and didn’t recognize it when it arrived. In the case of Zika, the silent circulation of the virus in asymptomatic people makes it hard for surveillance systems to pick it up.
Then there’s treatment.
There is a specific therapy for just a handful of viruses, Diamond noted: hepatitis C, HIV, herpes simplex, and influenza. Such drugs are not easy to develop, especially in the throes of reacting to a crisis.
But LeDuc, for one, is “quite optimistic” that broad-spectrum antivirals can be developed. “The more we understand how pathogens cause disease,” he said, “the more we see common pathways” that might be avenues for intervention.
Once again, though, we run into the issue of getting drugs to people when they need it. Even if a Zika treatment were available, the vast majority of infected people would not take it because they would never know they were infected.
“One of the biggest challenges is diagnosis — and early diagnosis — so that we have a chance to intervene,” LeDuc said.
So what’s next?
Predicting the next outbreak is a mug’s game, as the case of Zika illustrates.
“There are many viruses that could emerge,” Diamond said, but whether they do or not depends on a host of variables, such things as the presence or absence of a vector and the titer needed to cause infection.
But the world could do better at being prepared, he said. “You can be reactionary or you can be proactive,” Diamond said.
The reactionary approach is to wait until something happens and then wheel out the fire trucks to put out the blaze. But we’d be better off, Diamond said, investing in “ways to make your house fireproof.”
Those investments would certainly include better surveillance, drugs, and vaccines, he said.
But first on the list, Diamond said, should be basic research on the nature of viruses so that we are “prepared to deal not just with the pandemic du jour but to really respond to any virus that comes up.”
HIV Growing Resistant to Common Treatment
Ryan Bushey, Associate Editor http://www.dddmag.com/news/2016/01/hiv-growing-resistant-common-treatment
Scientists from the University College London made a new discovery regarding the HIV virus.
The researchers learned the common HIV therapy tenofovir was less effective against certain strains of the pathogen after studying an estimated 2,000 patients, writes theBBC. Tenofovir is typically used in combination with other medication to suppress the growth of this infection.
A comparison was done between HIV patients in Africa versus those in Europe. Individuals in Africa were 60 percent more resistant to tenofovir whereas European patients experienced 20 percent more resistance.
Irregular dosing of the drug was partly to blame as well as sub-standard administration of the medication.
Lead author Dr. Ravi Gupta told the BBC, “If the right levels of the drug are not taken, as in they are too low or not regularly maintained, the virus can overcome the drug and become resistant.”
Gupta added that there should be a simultaneous global initiative and cash investment to improve facilities and disease monitoring in African countries.
The virus continues to spread as countries issue pregnancy advisories and drug firms pick up on vaccine development.
January 28, 2016|
As the mosquito-borne Zika virus has now spread to at least 23 countries in the Americas in recent months, the World Health Organization (WHO) is convening an emergency meeting on International Health Regulations Monday (February 1), Director-General Margaret Chan announced today (January 28).
Meanwhile, four countries—Ecuador, El Salvador, Jamaica, and Colombia—have asked women to delay getting pregnant for fear the virus can cause severe brain damage in fetuses. And some airlines have offered refund to pregnant travelers who booked trips to countries where Zika is circulating.
President Obama chimed in this week, calling for an acceleration of “research efforts to make available better diagnostic tests, to develop vaccines and therapeutics, and to ensure that all Americans have information about the Zika virus and steps they can take to better protect themselves from infection,” according to a White House statement.
There is currently no immunization or cure for Zika, but several pharmaceutical companies and academic labs have expressed interest in developing a vaccine. The University of Texas Medical Branch has already begun work on a Zika vaccine, which could be ready for testing in a year or two.
“What would take the longest time would be the process of passing it through the [US Food and Drug Administration] and other regulatory agencies to allow it for public use and that may take up to 10 to 12 years,” Nikos Vasilakis, who is working on the vaccine, told BBC News. His team is also advancing diagnostics that could help answer questions about the risks of fetal Zika exposure. (See “New Tests for Zika in the Works,” The Scientist, January 25, 2016.)
Sanofi, which has had a dengue shot recently approved in several countries, and GlaxoSmithKline have expressed interest in starting a vaccine program, while another firm, Inovio Pharmaceuticals, announcedMonday (January 25) it was beginning work on a Zika vaccine.
On Tuesday (January 26), the US Centers for Disease Control and Prevention (CDC) issued guidelines for screening babies whose mothers may have contracted Zika while pregnant. Infants should be tested for Zika if their mother tested positive or if they have microcephaly and their mothers were in a country with circulating virus while they were pregnant.
Microcephaly is not the only concern for exposed fetuses; these babies should also be screened for hearing and vision impairments, the CDC urged. “One rationale is we don’t know the spectrum of problems that perhaps are related to Zika virus, so we want to do a lot of screenings of infants out of an abundance of caution,” Cynthia Moore, the director of the CDC’s division of birth defects and developmental disabilities, told The New York Times. “We worry because other intrauterine infections may have some effects that last or show up after birth.”
Meanwhile, US health officials this week said a massive outbreak stateside is unlikely, given the geographic range of the mosquito species that transmit the virus and Americans’ housing conditions, with screens and air conditioning. “If you look at historically what we’ve seen, I think we can say that it’s a remote possibility and unlikely to happen,” Anthony Fauci, head of the National Institutes of Allergy and Infectious Disease, told NPR’s Shots.
Interim Guidelines for the Evaluation and Testing of Infants with Possible Congenital Zika Virus Infection — United States, 2016
Weekly / January 29, 2016 / 65(3);63–67 http://www.cdc.gov/mmwr/volumes/65/wr/mm6503e3.htm
On January 26, 2016, this report was posted as an MMWR Early Release on the MMWR website (http://www.cdc.gov/mmwr).
J. Erin Staples, MD, PhD1; Eric J. Dziuban, MD2; Marc Fischer, MD1; Janet D. Cragan, MD3; Sonja A. Rasmussen, MD4; Michael J. Cannon, PhD3; Meghan T. Frey, MPH3; Christina M. Renquist, MPH3; Robert S. Lanciotti, PhD1; Jorge L. Muñoz, PhD1; Ann M. Powers, PhD1; Margaret A. Honein, PhD3; Cynthia A. Moore, MD, PhD3 (View author affiliations)
CDC has developed interim guidelines for health care providers in the United States who are caring for infants born to mothers who traveled to or resided in an area with Zika virus transmission during pregnancy. These guidelines include recommendations for the testing and management of these infants. Guidance is subject to change as more information becomes available; the latest information, including answers to commonly asked questions, can be found online (http://www.cdc.gov/zika). Pediatric health care providers should work closely with obstetric providers to identify infants whose mothers were potentially infected with Zika virus during pregnancy (based on travel to or residence in an area with Zika virus transmission [http://wwwnc.cdc.gov/travel/notices]), and review fetal ultrasounds and maternal testing for Zika virus infection (see Interim Guidelines for Pregnant Women During a Zika Virus Outbreak*) (1). Zika virus testing is recommended for 1) infants with microcephaly or intracranial calcifications born to women who traveled to or resided in an area with Zika virus transmission while pregnant; or 2) infants born to mothers with positive or inconclusive test results for Zika virus infection. For infants with laboratory evidence of a possible congenital Zika virus infection, additional clinical evaluation and follow-up is recommended. Health care providers should contact their state or territorial health department to facilitate testing. As an arboviral disease, Zika virus disease is a nationally notifiable condition.
Zika virus is a mosquito-borne flavivirus primarily transmitted by Aedes aegypti mosquitoes (2,3). Aedes albopictus mosquitoes also might transmit the virus. Ae. aegypti and Ae. albopictusmosquitoes are found throughout much of the Region of the Americas, including parts of the United States, and also transmit dengue and chikungunya viruses (4). Zika virus infections have also been documented through both intrauterine transmission resulting in congenital infection and intrapartum transmission from a viremic mother to her newborn (5,6). Zika virus RNA has been detected in breast milk, but Zika virus transmission through breastfeeding has not been documented (5).
During outbreaks, humans are the primary host for Zika virus. An estimated 80% of persons infected with Zika virus are asymptomatic (2,7). Symptomatic disease generally is mild and characterized by acute onset of fever, maculopapular rash, arthralgia, or nonpurulent conjunctivitis. Symptoms typically last from several days to 1 week. Based on information from previous outbreaks, severe disease requiring hospitalization is uncommon and fatalities are rare (6,7). During the current outbreak in Brazil, Zika virus RNA has been identified in specimens (i.e., brain tissue, placenta, and amniotic fluid) from several infants with microcephaly and from fetal losses in women infected with Zika virus during pregnancy (6,8,9). The Brazil Ministry of Health has reported a marked increase from previous years in the number of infants born with microcephaly and intracranial calcifications in 2015, although it is not known how many of these cases are associated with Zika virus infection (6,8–11).
Zika Virus Testing Considerations and Classification
The diagnosis of Zika virus infection is made through molecular and serologic testing (2). This includes reverse transcription-polymerase chain reaction (RT-PCR) for viral RNA, and immunoglobulin (Ig) M ELISA and plaque reduction neutralization test (PRNT) for Zika virus antibodies. Because it is currently not known which type of testing most reliably establishes the diagnosis of congenital infection, CDC recommends both molecular and serologic testing of infants who are being evaluated for evidence of a congenital Zika virus infection (Box 1). No commercial tests for Zika virus are available; Zika virus testing is performed at CDC and some state and territorial health departments. Health care providers should contact their state or territorial health department to facilitate testing.
Zika virus RT-PCR testing should be performed on serum specimens collected from the umbilical cord or directly from the infant within 2 days of birth (12). In addition, cerebrospinal fluid (CSF) obtained for other studies, and frozen and fixed placenta obtained at delivery, should also be tested by RT-PCR. IgM ELISA for Zika virus and dengue virus should be performed on infant serum, infant CSF, and maternal serum; however, results of these assays can be falsely positive because of cross-reacting antibodies (9,12). PRNT can be performed to measure virus-specific neutralizing antibodies and to discriminate between cross-reacting antibodies from closely related flaviviruses (e.g., dengue or yellow fever viruses). Finally, immunohistochemical staining to detect Zika virus antigen on fixed placenta and umbilical cord tissues can be considered.
An infant is considered congenitally infected if Zika virus RNA or viral antigen is identified in any of the samples submitted, including testing of amniotic fluid and testing of the placenta or umbilical cord. In addition, Zika virus IgM antibodies with confirmatory neutralizing antibody titers that are ≥4-fold higher than dengue virus neutralizing antibody titers in the infant serum or CSF constitute evidence of a congenital Zika virus infection. If Zika virus neutralizing antibody titers are <4-fold higher than dengue, results are considered inconclusive.
The only way to prevent congenital Zika virus infection is to prevent maternal infection, either by avoiding areas where Zika virus transmission is ongoing or strictly following steps to avoid mosquito bites (15,18). Mosquito-bite prevention includes using air conditioning or window and door screens when indoors, wearing long sleeves and pants, using permethrin-treated clothing and gear, and using insect repellents. When used according to the product label, U.S. Environmental Protection Agency-registered insect repellents are safe for pregnant women (18).
The Lancet Infectious Diseases JANUARY 28, 2016
Global epidemiology of drug resistance after failure of WHO recommended first-line regimens for adult HIV-1 infection: a multicentre retrospective cohort study
Antiretroviral therapy (ART) is crucial for controlling HIV-1 infection through wide-scale treatment as prevention and pre-exposure prophylaxis (PrEP). Potent tenofovir disoproxil fumarate-containing regimens are increasingly used to treat and prevent HIV, although few data exist for frequency and risk factors of acquired drug resistance in regions hardest hit by the HIV pandemic. We aimed to do a global assessment of drug resistance after virological failure with first-line tenofovir-containing ART.
The TenoRes collaboration comprises adult HIV treatment cohorts and clinical trials of HIV drug resistance testing in Europe, Latin and North America, sub-Saharan Africa, and Asia. We extracted and harmonised data for patients undergoing genotypic resistance testing after virological failure with a first-line regimen containing tenofovir plus a cytosine analogue (lamivudine or emtricitabine) plus a non-nucleotide reverse-transcriptase inhibitor (NNRTI; efavirenz or nevirapine). We used an individual participant-level meta-analysis and multiple logistic regression to identify covariates associated with drug resistance. Our primary outcome was tenofovir resistance, defined as presence of K65R/N or K70E/G/Q mutations in the reverse transcriptase (RT) gene. Findings
We included 1926 patients from 36 countries with treatment failure between 1998 and 2015. Prevalence of tenofovir resistance was highest in sub-Saharan Africa (370/654 [57%]). Pre-ART CD4 cell count was the covariate most strongly associated with the development of tenofovir resistance (odds ratio [OR] 1·50, 95% CI 1·27–1·77 for CD4 cell count <100 cells per μL). Use of lamivudine versus emtricitabine increased the risk of tenofovir resistance across regions (OR 1·48, 95% CI 1·20–1·82). Of 700 individuals with tenofovir resistance, 578 (83%) had cytosine analogue resistance (M184V/I mutation), 543 (78%) had major NNRTI resistance, and 457 (65%) had both. The mean plasma viral load at virological failure was similar in individuals with and without tenofovir resistance (145 700 copies per mL [SE 12 480] versus 133 900 copies per mL [SE 16 650; p=0·626]).
We recorded drug resistance in a high proportion of patients after virological failure on a tenofovir-containing first-line regimen across low-income and middle-income regions. Effective surveillance for transmission of drug resistance is crucial.
More than 35 million people worldwide are living with HIV-1.1 There is no effective vaccine and therefore control of the HIV pandemic relies heavily on combination antiretroviral therapy (cART). WHO treatment guidelines for adult HIV-1 infection recommend the nucleotide reverse-transcriptase inhibitor (NRTI) tenofovir for first-line ART, in combination with lamivudine or emtricitabine and the non-nucleoside reverse-transcriptase inhibitor (NNRTI) efavirenz.2 Older NRTIs such as the thymidine analogue drugs are being replaced by tenofovir and the NNRTI nevirapine, although mentioned in WHO guidelines, is being phased out from first-line regimens.2
The global scale-up of cART has now reached 15 million treated individuals.1 The administration of cART at the time individuals with HIV-1 are initially diagnosed prevents immunological deterioration as early as possible and interrupts the spread of HIV-1 from newly diagnosed individuals.3 This strategy, referred to as treatment as prevention, is being studied especially in high-incidence regions and nearly always includes the use of first-line tenofovir-containing ART regimens. Likewise, the strategy of pre-exposure prophylaxis (PrEP) depends entirely on the administration of tenofovir or tenofovir and emtricitabine to uninfected individuals at high risk of HIV-1 infection.4
In individuals receiving tenofovir, HIV-1 develops phenotypically and clinically significant resistance usually as a result of one mutation at position 65 (lysine to arginine; K65R) in the reverse transcriptase (RT) gene.5 Data from clinical trials and cohorts in high-income settings using tenofovir combined with NNRTI have reported low prevalence of tenofovir resistance at viral failure,6, 7, 8 in stark contrast with reports from low-income and middle-income countries where prevalence seems to be much higher.9, 10 Similarly, high-level resistance to NNRTI and the cytosine analogue component (emtricitabine and lamivudine) arise through changes to one aminoacid, which suggests a low genetic barrier to resistance for these drugs as well. In view of the pivotal role of tenofovir-containing ART as both treatment and prophylaxis, and the striking potential for drug resistance, we did a global assessment of drug resistance after virological failure with first-line tenofovir-containing ART.
Research in context
Evidence before this study
We searched PubMed for studies of the prevalence of tenofovir resistance after failure of first-line antiretroviral therapy with efavirenz or nevirapine (non-nucleoside reverse-transcriptase inhibitors [NNRTIs]) in patients with HIV-1, published between January, 1999, and June, 2015, using the search terms “HIV” AND “tenofovir” AND “resistance”. We identified studies done in untreated adults (age >15 years) in which either efavirenz or nevirapine was combined with tenofovir and either emtricitabine or lamivudine as first line antiretroviral therapy. Several studies reported resistance data for tenofovir when the drug was started after initial use of stavudine or zidovudine; these studies were not reviewed further. We also excluded studies that reported tenofovir use without NNRTI because standard first-line antiretroviral therapy under a public health approach is based on NNRTI in adults.
We identified randomised controlled trials and a meta-analysis comparing NNRTI with protease inhibitors, in combination with tenofovir, which reported resistance data. Patients in high-income settings reported tenofovir resistance in 0–25% of virological failures and those in sub-Saharan Africa in 28–50%. The only other prospective study in sub-Saharan Africa was PASER-M, and was limited by few resistance data for patients given tenofovir plus NNRTI-based combination antiretroviral therapy (cART). The remaining studies were largely from South Africa and reported a wide range of prevalence (between 23% and 70%) of tenofovir resistance after virological failure. In west Africa, one study reported that 57% of virological failures were tenofovir resistant in a very small sample of 23 patients. Although aforementioned studies also reported NNRTI and cytosine analogue resistance, they were unable to quantify to what extent tenofovir resistance was a marker for high-level compromise of the regimen. We found no studies that specifically reported resistance data for patients given first-line tenofovir in east Africa. No study reported resistance data from more than one continent, and none seemed adequately powered to establish the effect of co-administered reverse-transcriptase inhibitors on the emergence of tenofovir resistance.
Added value of this study
This study reports the most comprehensive assessment of HIV-1 drug resistance after scale-up of first-line WHO recommended tenofovir-based antiretroviral regimens, showing that tenofovir resistance is surprisingly common in patients with treatment failure across many studies in all low-income regions. Importantly, these individuals also have notable resistance to other drugs in their regimen, leading to almost complete compromise of combination treatment. Challenging current perceptions in the specialty, our findings show that tenofovir resistant viruses have substantial transmission potential. Furthermore, our results show that viral strain affects tenofovir resistance in Europe but is not the main driver for resistance in viruses circulating in sub-Saharan Africa. Newly identified risk factors for resistance to tenofovir and NNRTI drugs include pre-treatment CD4 cell count (but not viral load) and co-administered antiretrovirals.
Implications of all the available evidence
Improvements in the quality of HIV care and viral load monitoring could mitigate the emergence and spread of tenofovir resistance, thereby prolonging the lifetime of tenofovir-containing regimens for both treatment and prophylaxis. Surveillance of tenofovir and NNRTI resistance should be a priority both in untreated and treated populations.
The TenoRes collaboration included 1926 individuals from 36 countries (figure 1 and appendix).Table 1 summarises the median size and year of ART initiation for the cohorts comprising the collaboration. Viral load monitoring was done in about 50% of the cohorts including nearly all of cohorts from upper-income regions and from a small proportion of the cohorts in low-income and middle-income countries (appendix shows income status for each cohort; table 1).
(A) Countries contributing data to resistance analysis and HIV-1 subtype distribution, (B) prevalence of drug resistance by mutation and by region
NNRTI=non-nucleotide reverse-transcriptase inhibitor. TDF=tenofovir disoproxil fumarate. *24% (n=462) of participants had tenofovir resistance when genotypes from viral load >1000 copies HIV-1 RNA per mL were considered.
Data are n, range, or n (%). cART=combination antiretroviral therapy.
The region-level pre-ART median CD4 cell count ranged from 44 to 104 cells per μL in sub-Saharan Africa, Asia, and Latin America (table 2). As expected, in north America pre-ART median CD4 cell count was 144 cells per μL and 190 cells per μL in Europe. The proportion of individuals using emtricitabine (vs lamivudine) and efavirenz (vs nevirapine) varied significantly by region. Emtricitabine was used significantly more than lamivudine in Europe, North America, and west and central Africa, and efavirenz was used significantly more than nevirapine in all regions apart from east and west and central Africa. The median duration of ART ranged from 11 to 26 months. Pre-treatment viral load ranged between 4·80 and 5·58 log copies per mL and was significantly higher in eastern and western and central Africa and Latin America than the other regions (table 2).
Data are n (%) or median (IQR). TDF=tenofovir disoproxil fumarate.
Crude prevalence of tenofovir resistance in patients with treatment failure was highest in low-income and middle-income regions (figure 1). Prevalence of cytosine analogue resistance (M184V/I) was highest in sub-Saharan Africa and Latin America and lowest in western Europe. By contrast, resistance to NNRTI did not show this pattern (figure 1). Furthermore, the M184V/I mutation was less common than NNRTI resistance across all regions except in eastern Africa. Of the 700 patients with tenofovir resistance in the dataset, 457 (65%) had resistance to both remaining drugs. Participants with tenofovir resistant viruses were likely to be resistant to one or both accompanying drugs and therefore have profound compromise of their regimen, as compared with those without tenofovir resistance (figure 1).
Low baseline CD4 cell count was consistently associated with a higher prevalence of tenofovir resistance across regions. The pooled OR for tenofovir in individuals with a CD4 cell count of less than 100 cells per μL versus 100 cells per μL was 1·50 (95% CI 1·27–1·77; figure 2). By contrast, a high baseline viral load was only associated with a small, not significant increase in tenofovir resistance (OR for viral load ≥100 000 copies per mL vs <100 000 copies per mL was 1·17, 95% CI 0·94–1·44;appendix). We compared tenofovir resistance by use of co-administered antiretrovirals with tenofovir as first-line therapy. Use of lamivudine rather than emtricitabine (NRTIs) was associated with a higher prevalence of tenofovir resistance (OR 1·48, 95% CI 1·20–1·82), as was use of the NNRTI nevirapine rather than efavirenz (OR 1·46, 1·28–1·67; appendix). Subgroup analysis showed that as well as associations being consistent across regions, they were also generally similar across a range of study settings (appendix), although there was some evidence of a greater effect size of baseline CD4 when efavirenz was co-administered with tenofovir, as compared with nevirapine.
Pooled odds ratios for tenofovir resistance after viral failure for baseline CD4 cell count <100 vs≥100 × 106 cells per μL
TDF+ denotes presence of tenofovir resistance. TDF=tenofovir disoproxil fumarate.
When considering the effect of baseline CD4, baseline viral load (figure 3), and co-administered antiretrovirals (appendix) on cytosine analogue and NNRTI resistance, we noted that the magnitude of associations were smaller than those recorded for tenofovir resistance.
Odds ratios for NNRTI resistance for (A) baseline CD4 cell count <100 vs ≥100 cells per μL, (B) viral load ≥100 000 vs <100 000 copies HIV-1 RNA per mL
NNRTI=non-nucleotide reverse-transcriptase inhibitor.
We also assessed the relation between viral subtype C on acquisition of tenofovir resistance. We restricted this analysis to western European studies in view of the consistent standard of care available in this region and relatively lower level of subtype diversity in other regions (figure 1A). We also limited the comparison to subtypes found in immigrant populations to minimise bias due to socioeconomic factors (thereby excluding subtype B infections mainly recorded in participants born in western Europe). Tenofovir resistance was higher in subtype C compared with non-C, non-B infections with a pooled OR of 2·44 (1·66–3·59).
As a sensitivity analysis we studied risk factors for tenofovir resistance using univariate (adjusted only for region) and multivariate logistic regression analyses (appendix). We noted a dose-response relationship for baseline CD4, which was not markedly altered by adjustment for baseline viral load, viral subtype, or type of co-administered drug used (appendix). Baseline viral load of 100 000 or more copies of HIV-1 RNA per mL was not significantly associated with tenofovir resistance (OR 1·31, 95% CI 0·91–1·91) and we noted no clear trend across increasing viral loads (appendix). Adjustment for several risk factors also had little effect on associations with tenofovir resistance of emtricitabine versus lamivudine and nevirapine versus efavirenz.
Finally, we compared the viral load at treatment failure in the presence and absence of tenofovir-associated mutations. The mean plasma viral load at treatment failure was not different in the presence or absence of tenofovir associated mutations (145 700 copies HIV RNA per mL [SE 12 480]vs 133 900 copies [SE 16 650]; p=0·626; figure 4 shows the within-study viral load by region). These results did not change when analysis was restricted to individuals who had evidence of the K65R mutation, either with or without M184V/I (appendix). Mutations at aminoacids K65 and M184 in the RT gene have been associated with suboptimum replication.13
Boxplot of log viral load by presence (TDF-positive) or absence (TDF-negative) of tenofovir resistance at viral failure in studies with at least ten patients with TDF resistance and a viral load measurement at treatment failure
We restricted to studies with at least ten TDF-resistant mutations to help with graphical clarity, although the pattern of similar distributions of failure viral load in the presence or absence of TDF resistance was true for all studies. TDF=tenofovir disoproxil fumarate. Blue dots represent outliers.
Our study has three main findings relating to the prevalence, risk factors for, and transmissibility of tenofovir resistance. First, we noted that levels of tenofovir resistance in individuals with viral failure ranged from 20% in Europe to more than 50% in sub-Saharan Africa. Second, a CD4 cell count of less than 100 cells per μL, treatment with nevirapine rather than efavirenz, and treatment with lamivudine rather than emtricitabine, were consistently associated with a 50% higher odds of tenofovir resistance in those with viral failure. Third, we noted that in patients with viral failure, viral loads were similar in the presence or absence of tenofovir resistance.
Our findings are important in view of the fact that following WHO recommendations,2 tenofovir is replacing thymidine analogues (zidovudine and stavudine) as part of the NRTI backbone in first-line regimens in resource-limited settings. Every drug in these regimens can be compromised by one aminoacid mutation, and the combination therapy is therefore potentially fragile. In view of the crucial role of tenofovir-containing ART in both treatment and prevention of new infections, restriction of drug resistance in high-burden settings is of paramount importance. Understanding how common tenofovir resistance is, and how and why it varies, is key to its prevention. Although our risk factors are only associated with a modest 50% increase in odds, this translates to a roughly 10% increase in resistance in those who fail when the overall tenofovir resistance prevalence is about 50% (as recorded in sub-Saharan Africa).
We hypothesise that the regional differences in tenofovir resistance are due to the frequency of viral load monitoring with close patient follow-up and feedback of results. For example, although viral load monitoring is not routinely done in most low-income and middle-income countries, in high-income countries viral load is tested three to four times per year with close patient follow-up and adherence support. Such an approach is likely to lead to earlier detection of viral failure, before selection of drug resistance mutations against tenofovir has occurred.14 This view is supported by the uncommon detection of drug resistance mutations in specimens with low viral load (400–1000 copies per mL) from patients given tenofovir in both high-income settings (figure 1; see higher prevalence of tenofovir resistance where viral load >1000 copies per mL is used as threshold in western Europe)15 and sub-Saharan Africa (Chunfu Yang, Centres for Disease Control, Atlanta, GA, USA, personal communication). Tenofovir resistance could be limited by viral load monitoring,16with rapid feedback to clinicians followed by adherence counselling to preserve first line, or switch to second line when this approach fails. Furthermore, pre-ART (baseline) resistance testing for key NNRTI mutations could potentially protect against tenofovir resistance by avoiding use of partly active treatment regimens. In our report, transmitted NNRTI resistance was low in the regions studied (<10%),17 and therefore not likely to be a major driver of wide variation in drug resistance across income settings.
Other factors that vary geographically could also affect success of ART and should be noted. Treatment failure is associated not only with drug resistance, but also side-effects. Efavirenz is associated with CNS side-effects such as sleep disturbance and is associated with treatment discontinuation.18 Furthermore, drug stock-outs and other indicators of quality of HIV services that have shown geographic variation would also predispose to treatment failure.19 The issue of regional variation in adherence levels has received considerable attention, with data from several studies suggesting that adherence is not worse in sub-Saharan Africa compared with North America.20, 21
With regards to increased tenofovir resistance in individuals with low baseline CD4 counts, this finding is consistent with results from the ACTG 5202 trial22 suggesting higher frequency of RT mutations in patients given ART with low CD4 cell counts, and offer a benefit of CD4 cell count measurement after diagnosis of HIV infection beyond establishing prophylaxis against opportunistic infections.23 Lamivudine warrants further study in first-line regimens in view of data presented in our study and the conflicting reports regarding virological efficacy of lamivudine versus emtricitabine.24,25, 26 Of note, the differences between lamivudine and emtricitabine might become less important in high-income regions where implementation of the second generation integrase inhibitor dolutegravir occurs, in view of the fact that this agent has not been associated with any cytosine analogue resistance at virological failure.27
Viral load has been associated with transmission risk.28 Despite evidence for diminished replication of tenofovir resistant viruses (containing the K65R mutation in the RT gene) in vitro, we noted similar viral loads in participants with and without tenofovir resistance. Therefore, there might be substantial potential for onward transmission to uninfected individuals,29 despite little evidence of K65R transmission up to now.30 This finding reinforces the need for drug resistance surveillance activities in both untreated and treated HIV-positive individuals.
There are several important limitations of our study. First, because we only included patients with virological failure related to existing study cohorts,1 our estimates of the prevalence of tenofovir resistance might not be representative in certain high-burden regions. Although this situation might have biased our findings on absolute prevalences of tenofovir resistance, it is unlikely to have affected associations with baseline CD4 or co-administered drugs. Second, we only included patients at failure so were unable to assess overall rates of tenofovir resistance in all patients starting first-line treatment. We used this method because many of the contributing studies had no clear denominator, especially those done in resource-limited settings. However, extensive WHO-led analysis reported that 15–35% (on treatment vs intention to treat) of patients in sub-Saharan Africa have virological failure by 12 months.31 Therefore, using a conservative 50% prevalence of tenofovir resistance at failure from our analysis, we suggest that it is likely that 7·5–17·5% of individuals given tenofovir plus cytosine analogue plus efavirenz will develop tenofovir resistance within 1 year of treatment initiation under present practices in sub-Saharan Africa.
Third, our findings on risk factors for tenofovir resistance were derived from an unadjusted meta-analysis involving very different study populations. Although this enhances the generalisability of results, it has the potential to lead to biased comparisons. However, we took measures to minimise biases. We exclusively used within-study and within-country comparisons for our primary analyses, thereby ensuring that comparisons were for participants undergoing similar treatment monitoring practices. We tested associations between risk factors and found that they were generally weak. For example, baseline CD4 cell count and viral load were only weakly associated with one another and neither was strongly associated with type of co-administered drug. Additionally, we undertook sensitivity analyses, which suggested that adjustment for other covariates had minimum effect on estimated associations. Lastly, our data tended to be consistent with previous studies—eg, our findings of higher resistance in subtype C patients are consistent with in-vitro data suggesting subtype C viruses are more susceptible to developing the K65R mutation.32
Fourth, despite our analysis being the largest drug resistance study ever undertaken after failure of first-line tenofovir-containing cART, patient numbers were somewhat limited by the slow uptake of tenofovir-based regimens in west and central Africa, eastern Europe, and Asia (in particular China and Russia), and information about baseline viral load in these settings was uncommon. As a result, European countries, Thailand, and South Africa contributed substantially to the analysis.
In summary, extensive drug resistance emerges in a high proportion of patients after virological failure on a tenofovir-containing first-line regimen across low-income and middle-income regions. Optimisation of treatment programmes and effective surveillance for transmission of drug resistance is therefore crucial.
Correspondence to: Dr Ravindra K Gupta, UCL, Department of Infection, London WC1E 6BT, UK firstname.lastname@example.org
Cross-Reactive Ebola Antibodies
Human monoclonal antibodies induced during Ebola infection are able to neutralize related viral species, scientists show.
January 21, 2016|
SCRIPPS RESEARCH INSTITUTE; CHARLES MURIN, ANDREW WARD
From blood samples of survivors of a 2007 Ebola outbreak in Uganda, researchers have isolated antibodies that are protective and can neutralize two other species of Ebolavirus, including Zaire ebolavirus—the one responsible for the massive 2014 outbreak in West Africa. The binding specificities of these human monoclonal antibodies and their activity in animal models of Ebolavirus appeared today (January 21) in Cell.
“What is exciting is that the authors demonstrated that cross-reactive antibodies exist in survivors, and these antibodies are protective in animal models,”Larry Zeitlin, president of Mapp Biopharmaceutical, who was not involved in the current study but is collaborating with its authors as part of a National Institutes of Health (NIH)-funded project to develop and test vaccines and therapeutics against Ebolaviruses, told The Scientist in an email. “This gives real hope that a single product could be developed for treating all the Ebolavirus species.”
“This is a very good paper,” said Lisa Hensley, who has worked on the pathogenesis of viruses including Ebola and is associate director for science at the NIH’s Integrated Research Facility in Maryland. “Looking in people who survived Bundibugyo ebolavirus,” the species behind the 2007 outbreak, “brings our understanding of these viruses a step further. The study moved what we only knew anecdotally into convincing and strong data,” Hensley said.
Within the genus, there are currently three identified viral species that have caused deadly human outbreaks: Z. ebolavirus, B. ebolavirus, and Sudan ebolavirus.
Isolating peripheral blood B cells from seven survivors of B. ebolavirus, James Crowe, Jr., a viral immunologist and director of the Vanderbilt University Vaccine Center in Nashville, Tennessee, and his colleagues identified 90 antibodies that bound to the virus’s outer glycoprotein. Characterizing these human antibodies, the researchers found that 63 percent of them bound the glycoprotein of at least two of the three Ebolavirus species in vitro. Thirty-one (34 percent) of these antibodies were able to neutralize B. ebolavirus in an assay that is commonly used as a surrogate of an in vivo infection, and seven of these neutralized all three viral species, binding to one of three highly conserved regions of the glycoprotein.
The researchers found two different neutralizing antibodies, each used separately to treat Ebola-infected mice, effective. A single treatment with one of these antibodies successfully rescued Ebola-infected guinea pigs, while a combination of two antibodies resulted in complete protection of guinea pigs infected with Z. ebolavirus, the scientists showed.
“Protection against Ebola virus achieved in the guinea pig model is quite predictive of what can happen in humans, and in this model we can achieve protection,” said study coauthor Alexander Bukreyev, a virologist at the University of Texas Medical Branch in Galveston. “The key [result] is that the human immune system does produce protective antibodies. We just need to choose the right ones and give them at a high concentration.”
Crowe agreed. “Some of [the human antibodies] are extremely potent, some of the most potent antivirus antibodies ever isolated. Some of these antibodies possess the two major qualities you want to see in a therapeutic treatment—potent neutralization and breadth of activity against multiple Ebolavirus species.”
A recent study showed that serum isolated from Ebola survivors did not improve patient prognosis. But Mapp Biopharmaceutical’s cocktail of three mouse-derived, humanized monoclonal antibodies against Z. ebolavirus, ZMapp, was last year shown to be an effective treatment in Ebola-infected macaques. “Our goal for the second generation ZMapp product is a pan-Ebola virus antibody cocktail,” Zeitlin told The Scientist.The results of the present study, he added, “provides proof-of-concept for our effort.”
“The previously developed mouse monoclonal antibodies that were part of ZMapp were a solid start,” said Crowe. “But now, this study with antibodies from human survivors shows that we may have been underestimating the ability of antibodies to kill Ebola.”
The Ebola epitopes to which the cross-reactive antibodies bind will not only be helpful to develop new antibody cocktails as treatments but also point to a vaccine that could be used for multiple Ebola species. “This is the deeper implication of this work,” said Crowe.
Crowe, Bukreyev, and their colleagues are now isolating and analyzing antibodies against Z. ebolavirusstrains from the most recent outbreak. The researchers are also looking to combine the two types of therapies that have so far been shown to be most potent against Ebola, for a one-two punch: antibodies, which can neutralize already formed viral particles, and small interfering RNAs (siRNAs) that block viral replication.
The ultimate test will be whether the antibodies might be used in a protective vaccine or to treat people infected in the next outbreak. “That is everyone’s goal,” said Hensley. “What we do in animal models is nice, but how does it reflect what will happen in humans?”
A.I. Flyak et al., “Cross-reactive and potent neutralizing antibody responses in human survivors of natural Ebolavirus infection,” Cell, doi:10.1016/j.cell.2015.12.022, 2016.
The term “HER2-positive lung cancer” may actually refer to two distinct entities, according to a new study. HER2 mutations and HER2 amplifications were found in similar numbers of lung adenocarcinoma cases, but they did not occur in the same samples, suggesting HER2-targeted agents should differentiate between mutation and amplification.