For viral load testing to prevent treatment failure, drug resistance and onward transmission, treatment programmes will need to invest in better record-keeping and clinic procedures, human resources, demand creation and decentralisation of second-line treatment provision, a large pilot study of viral load provision in southern Africa has found.
Findings from the study, funded by UNITAID, the international drug and diagnostic purchase fund and carried out in seven countries by Médecins Sans Frontières (MSF), were presented this week at the 21st International AIDS Conference (AIDS 2016) in Durban, South Africa.
Routine implementation of viral load testing in treatment programmes is essential for identifying viral load rebound and treatment failure. Early identification of viral load rebound minimises the risk of developing drug resistance, so preserves future treatment options.
Maintaining viral suppression also minimises the risk of onward transmission and the monitoring of viral suppression rates gives some idea of the potential for HIV incidence in the population. Measurement of viral load is essential at a programmatic level in order to measure facility, district and national performance in achieving the last of the 90-90-90 targets - 90% of people on treatment virally suppressed.
Until recently viral load testing was largely unavailable outside large urban hospitals or research settings in low- and middle-income countries, but recent reductions in the cost of viral load testing, coupled with World Health Organization guidance recommending the use of viral load in preference to CD4 counts as the primary laboratory monitoring test for HIV treatment, are leading to the rapid expansion of viral load testing in sub-Saharan Africa and other settings where it was once considered unsuitable for use.
PEPFAR (The US President’s Emergency Plan for AIDS Relief) has already funded rapid scale-up of viral load testing in six countries and plans to use the lessons learned to support `Test and Start` activities in other PEPFAR focus countries, using routine viral load testing. At the same time, the cost of viral load testing is falling: following negotiations between manufacturer Roche and the Diagnostic Access Initiative, the company announced a price reduction to $9.40 per test for lower- and middle-income countries.
Further price reductions are plausible: speaking at the UN 90-90-90 Target Workshop preceding the conference, Trevor Peter of the Diagnostic Access Initiative said that demand for viral load tests is forecast to rise from 5-10 million tests to at least 22 million by 2020. The 2015 price reduction enabled 1.5 million extra tests in the past year, he added.
Despite the rapid progress towards availability of viral load testing, there is still little information available on how viral load is being implemented in practice, and on what challenges implementers face as they adopt viral load testing.
MSF's field implementation study
At AIDS 2016, MSF presented the fruits of four years of field implementation of viral load testing, supported by a grant from UNITAID.
MSF introduced routine viral load monitoring into its treatment programmes in Lesotho, Malawi, Mozambique and Zimbabwe in 2012, and in all districts except Changara in Mozambique, viral load monitoring was being carried out in rural settings.
An analysis presented at the conference by Dr Munyaradzi Dhodho of MSF Zimbabwe examined how treatment programmes performed during 2015 when implementing the viral load testing algorithm. Viral load testing is not an end in itself; it is only useful if it results in actions in response to detected viral rebound. The viral load testing algorithm recommends the following sequence of actions at facility level in the event of detectable viral load above 1000 copies/ml:
- Enhanced adherence counselling
- Second viral load test 2-9 months after detectable viral load result
- Switch to local second-line regimen if viral load is not resuppressed on the second test.
The study found big variations between district programmes in the four countries, not only in the coverage of viral load testing but also in compliance with other steps in the viral load algorithm.
Whereas 91% of people receiving treatment in Buhera in Zimbabwe received a viral load test, only 32% of people receiving treatment in Nyasa, Malawi, received a viral load test. Rates of viral detectability varied widely, and did not correspond to the coverage of viral load testing; whereas only 9% of people who underwent viral load testing in Thyolo, Malawi, had detectable viral load, 40% of people in Changara, Mozambique had detectable viral load. A lack of staff to draw blood for viral load testing was cited as a major obstacle to systematic compliance with the viral load algorithm.
Provision of enhanced adherence counselling to people with detectable viral load results failed to take place in 43% of people in Buhera, Zambia, and almost a quarter of people failed to receive counselling in the best-performing programme (Gutu), also in Zimbabwe. Staff at these sites said that a shortage of dedicated staff to perform enhanced counselling was a major obstacle.
Switching to second-line treatment occurred in a little more than one-third of patients at the best performing sites, and at one site (Thyolo, Malawi) only 15% of people with a detectable viral load above the very high local threshold (5000 copies/ml) were switched to second-line therapy. When confronted with these findings all sites said that reluctance to task shift and decentralise second-line antiretroviral therapy (ART) prescribing formed the biggest barrier to complying with this stage of the algorithm. In Malawi, for example, recommendations for switches to second-line treatment had to be approved by a national-level committee before medication could be supplied.
Sites with a higher rate of switching reported that getting results on the same day as testing took place, having a clinician on site at least once a week who was authorised to switch people to second-line treatment, and having second-line drugs available on site for people who were switched, were the factors that facilitated acting on viral load results.
In a report on the pilot programme issued this week at AIDS 2016, MSF highlighted a number of lessons learned regarding implementation:
- Health care workers benefit from continuous training on the benefits of routine viral load testing, and are motivated by monthly clinic-wide targets for the number of tests carried out.
- A catch-up phase in which everyone is tested regardless of time on ART may help familiarise clinic staff with viral load, but may also lead to heavy workload. Ultimately good record-keeping and a system for flagging up patients due for viral load tests need to be implemented.
- Reminder systems on paper and electronic records will help.
- Consider automatic viral load testing whenever a CD4 count is requested if viral load uptake is poor.
- Viral load awareness campaigns by and for civil society should be part of national scale-up plans, and need to be properly funded.
- Patient education and counselling on viral load needs to be prioritised. Enhanced adherence counselling needs to be funded properly; too often it is being carried out by lay counsellors without proper recognition that they are playing an essential role.
- Undetectable viral load results are motivating for patients and offer an opportunity to channel patients to differentiated care (e.g. adherence clubs) that can reduce their need for clinic visits and waiting time.
MSF reports on the programmatic experience and the laboratory implications of making viral load testing routine can be downloaded from the MSF website.
MSF South Africa has also produced a An Implementer's Guide To Introducing HIV Viral Load Monitoring – a toolkit providing training materials for clinic and laboratory implementers, including exercises to examine how procedures and workflow can be adjusted to implement viral load monitoring.
Dhodho M et al. Implementation of routine viral load monitoring in Lesotho, Malawi, Mozambique and Zimbabwe: a cascade analysis. 21st International AIDS Conference, Durban, abstract WEAE0301, 2016.