April 2011
Feature Story

Karen Lusky

What will it take to keep turning the tables on flu? Everything from faster molecular assays, which test makers continue to turn out, to blocking antiviral resistance.

Nanosphere’s RV+ received Food and Drug Administration clearance as a moderate-complexity test in January. It produces results in 2.5 hours for influenza A and B, H1, H3, 2009 H1N1, and respiratory syncytial virus A and B on a single cartridge, says William Moffitt, president and CEO of Nanosphere.

Those performing the RV+ pipette the sample into one of the test consumables, and the rest of the testing is automated, says Nanosphere product manager Zachary Crowther.

Idaho Technology is expecting to have FDA clearance this spring for its FilmArray respiratory panel, which provides results in one hour for more than a dozen viral targets, says Wade Stevenson, diagnostics marketing manager. The test includes flu A and B, H1N1 seasonal and 2009 strains, H3, parainfluenza 1, 2, 3, and 4, and numerous other respiratory viruses. Belinda Yen-Lieberman, PhD, director of clinical virology, serology, and cellular immunology at the Cleveland Clinic, says Idaho’s “PCR in a pouch” performs every step from extraction to amplification to detection in one hour. She says the walkaway test system is best used for hospitalized and immunocompromised patients.

Though only one test can be run at a time using the Film­Array, Stevenson says, a laboratory that purchases four of the relatively small instruments (for a total cost of less than $120,000) could run 40 to 60 tests daily over a 12- to 18-hour period.

AutoGenomics, which offers an expanded panel now of 25 viruses, has set its sights on providing a panel geared to identifying an emerging flu epidemic. The panel, says Jacques Corbeil, PhD, collaborator at Laval University in Canada, includes H1, H3, H5, N1, and N2, in addition to some of the resistance genes associated with antiviral treatment, with the gene linked to Tamiflu (oseltamivir) resistance being the primary one. AutoGenomics expects to have the panel available by the fourth quarter of this year, says senior VP Ram Vairavan.

What is the rationale for including H5N1 (Avian flu), which experts say hasn’t been detected in the United States? “Just like no one predicted [the 2009] H1N1,” Dr. Corbeil says, “we want to be ready if it turns out there is a circulating H5 strain.” He cautions that doing H5 testing routinely could lead to false-positives—and the inability to recognize H5 if it mutates.

Other commercial assays, including Luminex’s RVP assay and Gen-Probe’s ProFlu+, can detect H5N1, but they, unlike the AutoGenomics panel in development, do not subtype for it.

Gen-Probe’s “ProFlu+ test would tell you that a sample containing H5N1 is positive for a flu A, but it would not tell you that it’s H5,” says Steven Visuri, PhD, senior director of research and development for Prodesse products. Gen-Probe’s subtyping test, ProFAST+, however, would produce negative results for H1, H3, and 2009 H1N1. “Test users would send that sample to the health department to determine if it’s a novel virus or H5N1,” Dr. Visuri says.

Molecular tests do have the ability to get on the trail of a novel virus quickly, says Jay M. Lieberman, MD, medical director for infectious disease, Quest Diagnostics and subsidiary Focus Diagnostics. “Quest had a molecular test available within two weeks of the [CDC’s] publication of the novel H1N1 genetic sequence.”

The CDC, says Dan Jernigan, MD, deputy director of the agency’s Influenza Division in the National Center for Immunization and Respiratory Diseases, continues to make its PCR assay “better and more sensitive to the new circulating strains of seasonal flu—and to assure that the test can detect novel influenza strains.” In addition, the CDC is making the gene sequences and test protocols available for others to use in developing their own assays.

One downside to PCR testing is its inability to pick up seasonal variants.

Holly Alexander, PhD, clinical scientist at Via Christi Hospitals Wichita Inc. in Kansas, says her lab discovered that the Luminex xTAG RVP assay was not picking up an H3 variant in February when the staff began to see negative H1 and H3 samples. Initially they assumed the samples must be 2009 H1N1. The state health department laboratory determined that about half of them were novel H1N1 and the rest H3, Dr. Alexander says. (The H3 variant has not produced more serious illness than the seasonal H3, she adds.)

Jeremy Bridge-Cook, PhD, senior VP of the assay group at Luminex, says the company is aware that an H3 variant this flu season is testing negative on its assay in some cases. “The sensitivity of flu A remains unaffected, but the lab will receive an unsubtypable result for these H3 variants,” he says, just as it will for 2009 H1N1. Laboratories follow the guidelines they have in place in such situations, he notes. Meanwhile, Luminex is evaluating how best to address the H3 mutation, which may be to update the assay, Dr. Bridge-Cook says.

Karen Kaul, MD, PhD, board of directors chair of molecular pathology and director of the Molecular Pathology Division at NorthShore University HealthSystem, Evanston, Ill., says there have been a number of assays—commercial and laboratory-developed—that fail to detect mutated viruses. “A virus, particularly influenza A, can undergo changes where the virus exchanges large pieces of its genome with other viruses.” Seasonal drift also occurs, which she describes as accumulation of smaller nucleotide variations that have an impact on the genome at a lower level.

Viral mutations can also affect direct fluorescent antibody testing, but probably to a lesser degree, Dr. Kaul says, “unless the virus produces an altered protein or there is an amino acid change at a key site that is critical in the antibody interaction detected.” What about culture? That would still work because “a nucleotide change or swap of a larger genomic piece is not going to affect, in general, the ability of the virus to grow.” Yet culture is too slow and relatively insensitive compared with PCR, she notes.

Dr. Kaul says the laboratory community may need to address the problem of molecular testing being unable to pick up variants by “stepping up” use of multiple assays and multiple sets of primers within each assay. Then the assays won’t be subject to being “knocked out by a chance nucleotide change that hits primary binding sites.”

Rebecca Horvat, PhD, section director of clinical microbiology at the University of Kansas Medical Center, says laboratories need “a systematic plan to check multiplex assays in a financially appropriate way.” Having found that a reference lab this year was not detecting an RSV type A variant, UKMC plans to conduct surveillance by pooling negative specimens and doing viral cultures.

The laboratory at Texas Children’s Hospital in Houston uses the Gen-Probe ProFlu+ assay to detect influenza A, B, and RSV, and a homebrew assay to subtype influenza A. And the lab is rarely unable to subtype using the homebrew assay, says Paula Revell, PhD, director of clinical and molecular microbiology laboratories. “If we have an influenza A positive but cannot subtype the virus using our in-house methods, we can send those viruses to the CDC for further characterization,” she says.

While many see the value of molecular testing, questions remain about the value proposition of routinely using more expensive molecular testing for flu and other respiratory viruses. It’s a pro and con debate that Quest’s Dr. Lieberman argues should include more than simply looking at the cost of the test. “A falsely negative test can have significant consequences and be quite expensive,” he warns.

Dr. Lieberman points to a paper published last year in the New England Journal of Medicine, which he says shows how “a delay in the initiation of antiviral therapy in pregnant and peripartum women with [2009] H1N1 was associated with an increased likelihood of them becoming critically ill or dying. Of the eight patients who died from H1N1 infection, seven had false-negative rapid flu tests” (Louie JK. N Engl J Med. 2010;362:27–35).

The CDC’s Dr. Jernigan says “rapid antigen detection tests can be very useful for clinical management when the result is positive.” Making clinical decisions based on a negative result can be problematic, he cautions. That’s why the CDC has, in conjunction with input from professional groups, developed algorithms that guide clinicians in the use of rapid tests. (The 2010–2011 guidance is at www.cdc.gov/flu/professionals/diagnosis/clinician_guidance_ridt.htm.)

Subtyping flu A, which rapid antigen tests cannot do, provides information about the potential for resistance to Tamiflu, as the seasonal H1N1 is virtually 100 percent immune to it, experts say.

“If you look back three years ago, before swine flu,” Dr. Kaul says, “we had essentially seasonal H1N1 resistant to Tamiflu or seasonal H3 that was still sensitive. Our lab originally started to look at molecular subtyping on influenza A to allow clinicians to use that information to guide use of Tamiflu.” If the virus is resistant, “there are other antivirals they can use. Amantadine is the next course of therapy, and there are other things as well.”

What’s susceptible to antivirals now may not be so in the future, which has been a problem with the existing antivirals.

David Wurtman, MD, a former internist and now VP of corporate development at NexBio, a San Diego pharmaceutical company working on a novel anti-flu drug, cautions that seasonal H1N1 became 99 percent resistant to Tamiflu over a period of two years owing to the H274Y mutation. The pandemic H1N1, he says, was sensitive to Tamiflu only because it came from a novel (swine) source and didn’t have that specific mutation. But now it’s picking it up “with increasing speed,” he says.

Based on World Health Organization surveillance reports as of mid-March, 24 percent of the novel H1N1 virus’ resistance to Tamiflu had occurred within the last eight weeks, Dr. Wurtman says. “Overall, worldwide [resistance] is now seen in almost two percent of all surveillance samples WHO has tested.”

The Cleveland Clinic has been using Gen-Probe’s ProFlu+ assay for all patients the past three years. In a study this year, the flu A subtyping on samples from hospitalized patients was done with Idaho Technology’s FilmArray respiratory panel test. Thus far, with the dozen samples tested, about two-thirds of the flu A was H3; one-third was 2009 H1N1, says Dr. Yen-Lieberman. “That correlates to what the public health department was showing in this region” according to the CDC’s weekly “FluView,” she adds.

A new treatment approach for flu may be on the horizon. The President’s Council of Advisors on Science and Technology’s 2009 report on H1N1 preparedness recommends the federal government “stimulate the development of new influenza drugs that have novel mechanisms of action in order to reduce the potential for antiviral resistance” (www.whitehouse.gov/assets/documents/PCAST_H1N1_Report.pdf).

NexBio’s experimental novel drug (DAS181 or Fludase) is designed to do just that. Rather than target the flu virus, the drug targets the host (person). It renders receptors in the human lung and upper airway unable to bind virus, which means the virus cannot enter the cell and replicate, explains Dr. Wurtman. The receptors, which he says the drug clips off “like a lawnmower does to the tips of blades of grass,” grow back in a few days.

Dr. Wurtman says the drug appears to be active against any type of influenza, including H5N1, and for all parainfluenza viruses (PIV), which cause a flu-like syndrome. Currently, there’s no vaccine or FDA-approved drug for PIV. In Dr. Wurtman’s view, if the FDA approves DAS181, it would make a PIV rapid test “very useful,” in that there would now be a treatment for it.

NexBio has not yet published its human safety data for the drug. “But so far in all our trials, including the ongoing phase two trial, which we expect to wind up in a few weeks, we haven’t seen any drug-related significant adverse effects,” Dr. Wurtman says.

The drug has been shown to be much less susceptible to viral resistance than antivirals, he says. That’s because, as he puts it, “if you chase the virus, the virus will find a way to get around you. And DAS181 doesn’t chase the virus.”

As for the future role of molecular testing in the pursuit of flu, Nanosphere’s Moffitt predicts that as the turnaround time for molecular comes down to two to three hours, physicians will begin to consider relying on it more. That change will probably begin in the inpatient setting and then move to office-based practice where, Moffitt says, “at a minimum, clinicians will use molecular to reflex off a negative rapid test where a patient is very symptomatic.”

The CDC’s Dr. Jernigan, who notes that during the 2009 H1N1 pandemic, the community was able to get the molecular testing under CLIA rules at academic medical centers, believes the next step is for molecular to move to the “FDA-approved and possibly even CLIA-waived space.”

“That’s where we hope things are going,” he says.

Dr. Jernigan also advocates better partnerships between the CDC, medical centers, and professional societies like the CAP and the Association for Molecular Pathology. “That would be really fruitful for preparing the nation to respond more quickly to the next emerging infection,” he says.