Under a $395,000 National Institutes of Health grant awarded last month, researchers at Seattle's PATH institute are developing a minimally instrumented, point-of-care nucleic acid amplification test to diagnose HIV-1 in infants in resource-poor areas of the world.
An initial prototype of the test, which uses an isothermal amplification technique called recombinase polymerase amplification and marketed by nucleic acid testing company TwistDx, has shown promising results, David Boyle, a senior research scientist in PATH's Diagnostics Group and the project's principle investigator, told PCR Insider.
According to Boyle, early internal tests suggest the assay should give results equivalent to commercially available PCR instruments and be able to eventually accurately diagnose all strains of HIV at the point of care, only much faster and at a much lower cost than centralized PCR testing. However, he also noted that the assay still needs to be refined and tested using a much larger sample size.
The PATH researchers in general are interested in adapting isothermal amplification technologies to detect a variety of infectious diseases. For infant HIV in particular, Boyle said, a point-of-care nucleic acid assay would improve AIDS-specific child mortality in areas where diagnosis and, ultimately, treatment is often greatly delayed by shipping samples to a centralized testing facility for traditional PCR.
"Although [a central PCR] lab can test the specimen quite quickly, it might take several weeks for the specimen to get to the lab or the result to get back to the health care worker that asked for the test," Boyle said. This is an especially vexing issue considering that more than half of all HIV-1-infected infants in sub-Saharan Africa die before age two unless treatment is begun early in life.
Boyle and colleagues are working to solve this by adapting TwistDx's RPA method. According to the company, RPA uses recombinase enzymes that are capable of pairing oligonucleotide primers with homologous sequences in duplex DNA, thus directing DNA synthesis to defined points in a DNA sample. If the desired target sequence is present, DNA amplification is initiated, requiring no other sample manipulation such as thermal or chemical melting.
Headquartered in Cambridge, Mass., and with operational facilities in Cambridge, UK, TwistDx has, over the past several years, been exploring applications for the RPA technique in several markets, including clinical diagnostics, biomedical research, food testing, animal health, agriculture, environmental testing, and biodefense.
According to Boyle, the method offers benefits in speed, operating temperature, and flexibility of read formats. For instance, the technique typically operates at a temperature below 42°C, and works optimally in the 34-40°C range, Boyle said.
"There are a couple things interesting about that," he said. "Since you don't need to obtain a high temperature, you could consider doing incubation at ambient temperature in a lot of places. So there's a reduced need for instrumentation to some degree. Of course, if it's above 42 degrees you can't perform the assay, so there is a slight problem there that we're aware of."
The method also allows for two different detection formats: real-time fluorescence and lateral flow strip detection. For the former, TwistDx sells a real-time fluorescence reader called Twista, he added, which is portable, battery powered, and "costs way less than a PCR machine." According to TwistDx's website, Twista retails for around £2,700 ($4,420).
"You can put in the thresholds for determining what's positive and negative. As the reaction continues, you can have a laptop hooked up, and it looks like a PCR curve, essentially… But on the instrument itself it gives you a plus or a minus. So there's actually no user interpretation required in the final device. A trained operator can set the thresholds for a particular run, and then anyone can use it," Boyle said.
However, Boyle noted that a lateral flow strip detection method is much more suitable for point-of-care applications because it virtually eliminates the need for instrumentation, although it would provide completely qualitative results as opposed to semi-qualitative results using the fluorescence reader.
Reagents for Twist's method are also very stable, according to Boyle. "We store them at minus 20 [degrees] but we've done some very preliminary experiments looking at the stability, and with the lateral flow detection format they appear to be stable at over a month at 45[°C] with no loss in performance compared to the ones we took out of the freezer," he said.
"That's really exciting," Boyle added. "We don't have to start investigating from scratch looking at stabilization because [TwistDx has] done a lot of this already."
Having completed an initial prototype of an RPA assay for HIV-1, the group is now refining the design to be able to detect all subtypes of HIV.
"We're developing that right now, starting with fluorescence, and then we can transfer to lateral flow strips, and make sure it works in all forms of HIV," Boyle said, adding that early results from this work are promising.
"But the next step — and this is really the elephant in the room for all clinical assays — is the specimen preparation." Boyle said the team is investigating a variety of methods to free up DNA from the mononuclear cells in infected peripheral blood.
Luckily, he said, RPA has a relatively high tolerance to whole blood. "It's not like PCR where you have a bit of hemoglobin and a lot of the assays just crap out because they can't tolerate the confounding factors," he said. "We've found that using a lateral flow detection method we can add whole blood. So we'll be looking at whole blood lysis methods."
More specifically, he said, the group is aiming for a rehydration buffer that contains a detergent and can be added directly to a certain volume of whole blood, perhaps on the order of 50 microliters, thus lysing all the cells, a fraction of which would be used to rehydrate and perform the assay.
So far, the group's test appears to be as sensitive as existing PCR methods for HIV-1 detection, Boyle said, though he stressed that the team has only performed about 25 assays and that the results are not yet statistically robust. Once they are finished redeveloping the lateral flow strip version, the researchers plan to test it on hundreds of samples to establish an accurate limit of detection, he said.
The final product will likely be contained in a thermos flask-type of container, similar to a non-instrumented nucleic acid assay being developed by another team of PATH researchers to detect Plasmodium falciparum and other pathogens, Boyle said. That test currently uses loop-mediated isothermal amplification and a heating system combining exothermic chemical reactions and phase change material (PCR Insider 6/9/2011)
The RPA-based test is housed similarly, with a lid containing a timer for the amplification and a holster for 200-µL tubes in which the RPA takes place. Underneath sits a heating and phase change material similar to that being employed in the LAMP-based test.
"You activate [the heating] material, put the tube holder over it, insert the tubes and seal the lid, turn the timer, wait until it goes ding, pick out the tubes, and perform testing on those lateral flow strips," Boyle said.
One remaining problem in the design comes in opening the reaction tubes to perform lateral flow strip detection. "If you open these tubes and it's been a positive reaction, the amplicons generated … contaminate everything," Boyle said.
To get around this, he said the group is considering adding a fully enclosed flow strip detection cassette product. Price terms will determine which technology the team chooses, and the group is also "still negotiating pricing terms with TwistDx to produce and supply the assay," Boyle said.
"Materials are fairly cheap, but we would be looking for TwistDx to have some sort of commitment to the markets [PATH sees]." In any case, he said "It's a lot cheaper than the commercially available PCR assays, and obviously less cost in terms of infrastructure because we're looking at either a small battery powered reactor [for real-time fluorescence], or lateral flow strips in conjunction with chemical heating or ambient temperatures if it's warm enough."
The team's funding runs through June 2012, and the project's reported end date is June 2014.
Have topics you'd like to see covered in PCR Insider? Contact the editor at mashford [at] genomeweb [.] com.