HIV – a new horizon
HIV mutates 4 000 times faster than most viruses, evolving and reproducing itself at a relentless pace. As a result, treatments that may have been effective against the original virus become ineffective, creating drug resistance. RISKSA looks at a ground-
According to the World Health Organisation ( WHO), nearly one in every 20 adults in sub- Saharan Africa lives with HIV. There are between five and six million HIV- positive individuals in South Africa alone. Currently, patients are given a first line of ARV treatment. If this fails, the patient is moved on to secondline treatment. If this treatment fails, there are certain third line treatments possible, but this is where government funding and a lot of medical aid funding will stop. WHO protocol requires that every person receiving antiretroviral treatment should be tested for drug resistance. However, in the majority of developing, resource- limited countries, including South Africa, this is not the case. Such testing is expensive and logistically intensive. Researchers from the University of the Western Cape’s ( UWC) South African Bioinformatics Institute ( SANBI) have spent two and a half years developing a software tool to dramatically speed this process, reduce costs and simplify the logistics required.
The drug resistance problem
“There are multiple genetic variants of HIV. Each one can be resistant to different drugs and each one responds better or worse to different treatments,” explains Imogen Wright, PhD candidate and researcher on the project. For this reason, over the course of a human life, almost every patient receiving ARV treatment will develop some form of resistance, but the amount of time taken to build up resistance varies between patients. Once resistance develops, patients progress from HIV positive to full- blown Aids cases. “For some, the strain they are initially infected with, is already resistant and their treatment will immediately fail as a result.” Wright’s background is computer science and physics, but the interdisciplinary field of bioinformatics, which uses a combination of computer science, mathematics and engineering to generate useful information from biological data, has allowed a fresh application of her skills. “If you give someone a drug and they are resistant to that drug, research has shown that the resistant variant will explode in their body and become dominant. If they infect someone else, this will be the variant passed on. So, you are actually propagating drug resistance when you give someone a drug and they are resistant to it,” she explains. According to the WHO's global strategy for prevention and assessment of HIV drug resistance, the consequences of drug resistance include treatment failure, increased direct and indirect health costs associated with the need
to start more costly second- line treatment for patients, the spread of resistant strains of HIV and the need to develop new HIV drugs. Since the rollout of antiretroviral treatment in South Africa began in 2004, HIV mortality rates have decreased substantially. Once seen as an immediate death sentence, the discourse around HIV in the insurance and medical management context is that, if managed properly, it is no more high- risk than diabetes or any other chronic condition. Studies indicate that by not controlling the resistance aspect, South Africa is potentially creating a situation that we will not be able to control in the future.
HIV drug resistance can be tested using two different methodologies; phenotypic testing, a highly accurate but expensive culture test; or genotypic testing, which sequences strands of RNA from the virus to determine resistance. For genotypic testing, sequencing can be achieved in two ways. Traditional Sanger sequencing requires the manually guided sequencing of individual strands of RNA. Again, while highly accurate, this is extremely expensive and not practical for large- scale roll out. The second approach to genotypic testing is called next- generation sequencing. By using computer technology to sequence many strands of RNA, potentially from multiple patients, at high speed, the cost of the process is reduced substantially. For this reason, there is a strong drive to use this method for wide- scale drug resistance testing. The limitation with next generation sequencing is that sequencing machines, while fast, are error prone and frequently sequence with inaccuracies, making reliable results difficult. To counter this, it is necessary to sequence the same RNA strand thousands of times. It is then necessary to cross reference the repeated sequencing to eliminate inconsistencies, identify the correct sequence and obtain the accurate result. Wright and her team have built a tool to tackle this aspect of the process. Currently, the cross referencing process takes days and adds considerably to the cost of the test. By creating a means for the rapid cross referencing through rapid mathematical probability modelling, software enables consistent accuracy and can draw a result from thousands of different readings of the same sequence in seconds. The tool is now capable of completing this process for 48 patients in less than 15 minutes. The results are returned in a report detailing if a person is drug resistant or not and what course of treatment should be implemented.
A reason that HIV is less of a problem in First World countries is that any person with HIV is tested for resistance.
Previously, each individual sequence read was sent to the HIV Drug Resistance Database at Stanford University, where the algorithm used to determine drug resistance was first developed, for analysis. As there are thousands of sequence reads required for one patient, this represented a significant bottleneck in the process.
How it will operate
The tool will run as a web service, accessible from anywhere in the country via a log- in. “The ultimate pipe- dream is to have a sequencing machine available to every doctor anywhere in the country. Right now access to a well- equipped lab is necessary, but researchers at Oxford Nanopore have created an experimental sequencing machine on a USB flash drive, so we hope to see improvement in access soon." " The data file would then be compressed so that it can be sent from a low bandwidth connection and uploaded to the web service at UWC. A report is then generated at UWC and returned with recommendations for each patient’s treatment programme,” Wright explains.
The current phase
Funded by the Bill and Melinda Gates Foundation, the Department of Science and Technology, the Atlantic Philanthropies and the German Academic Exchange Programme, the pipeline, software and front- end development for the tool is complete. The project will now move into a wide- scale research and testing phase, to be implemented by hospitals and research teams across the country, before it can be used to determine patient treatment. Data from across the continent will be used for the study. “We’re hoping to eventually black box the technology and ship it out to every lab in the country and let them all run it. Right now it runs on my laptop just fine. Hopefully that means that it can be run on a laptop in rural Malawi,” says Wright.
When a person is on the right ARV treatment, their viral load is fully suppressed and when they have no viral load, they are not contagious. “A reason that HIV is less of a problem in First World countries is that any person with HIV is tested for resistance and then put on, and adhere to, the correct course of treatment for the HIV variants in their body, keeping their
viral load suppressed and thus containing the spread of infection,” says Wright. Because there is no testing, there is little data to determine the resistance rates in developing countries. Wright estimates that two to three per cent are infected with a resistant strain and that within five years of the inception of treatment, 20 to 30 per cent of infected patients develop resistance. “Ultimately, resistance will develop in every case. It is only a matter of how long it takes,” she adds. According to the WHO, the extent of HIV drug resistance resulting from recent ART scale- up in resource- limited countries has not systematically been quantified due to the lack of reliable data and information. “Ideally, this sequencing is not something that would happen once, it would happen once every one to two years and ideally as soon as a patient experiences a problem such as a persistent illness or infection.” Given the dramatic direct and indirect costs to the economy and the impact on the labour force and productivity, the potential benefits for South Africa and the continent as a whole are significant. For the financial industry, the principal advantages will be in cost reductions for the testing itself, disability claims associated with secondary infections and associated with second- line treatments and even third- line treatments where this is covered by medical aid schemes. The cost benefits will also be felt by HIVpositive consumers, notes Dr Pieter Coetzer, chief medical adviser at Sanlam Life. “It will be beneficial to people living with HIV to have access to regular and affordable resistance testing. It should reduce transmission rates of the disease and increase the life expectancy of the client. This should ultimately translate into cheaper premiums for life insurance.”
Ultimately, resistance will develop in every case, it is only a matter of how long it takes.
A home- grown solution
Given the road South Africa has travelled in treating and managing HIV and Aids, it seems remarkable that a South African innovation has the potential to enable a transformed approach to ART protocol across the developing world. It is a stark and welcome contrast with the denialism, misinformation and inaction of the past. “I think that hopefully the upshot of owning the solution is that the solution will be free forever for people in this country. We will never charge a government lab to use this technology. It is also significant that we are the nation with the problem." " We’re in a position where our economy is suffering so we need to make this an economic priority and put in the funding to develop the protocol. In the West it’s hard to make the argument for funding because the problem is so much smaller and more contained,” Wright concludes.