7.3 Monitoring response to ART and the diagnosis of treatment failure
Consolidated ARV guidelines, June 2013
7.3.2 Monitoring the response to ART and the diagnosis of treatment failure: Rationale and supporting evidence (part 1)
Although evidence from clinical trials for a survival benefit of viral load testing is limited, it can provide an early indication of treatment failure, and the 2013 guidelines strongly recommend using it for detecting virological failure and/or confirming treatment failure among people with evidence of clinical and/or immunological failure (Table 7.14). Since several clinical and epidemiological studies show that the risk of HIV transmission is very low when the viral load is lower than 1000 copies/ml (181), the Guidelines Development Group also recommended reducing the viral load threshold for treatment failure from 5000 copies/ml to 1000 copies/ml.
Virological monitoring (viral load) versus immunological (CD4) and clinical monitoring (WHO clinical staging)
The main rationale for recommending viral load monitoring as the preferred approach compared with immunological and clinical monitoring is to provide an early and more accurate indication of treatment failure and the need to switch to second-line drugs, reducing the accumulation of drug-resistance mutations and improving clinical outcomes. Measuring viral load can also help to discriminate between treatment failure and non-adherence (183) and can serve as a proxy for the risk of transmission at the population level (76).
There is still limited evidence to support any additional survival benefit of viral load monitoring over CD4 and/or clinical monitoring among individuals with HIV receiving ART. A systematic review identified three randomized clinical trials on virological versus immunological and clinical monitoring (184–186) (Web Annex). Compared with immunological and/or clinical monitoring, adding viral load monitoring has not been associated with reduced mortality. In one of these trials (185), no significant difference in the incidence of clinical failure, switching to second-line regimens and resistance mutations was found. One cohort modelling study among adults also found that adding virological monitoring to clinical and/or immunological criteria made no difference in mortality or new AIDS-defining illnesses (187).
Although randomized controlled trials have not yet shown that viral load monitoring translates into survival gains, follow-up has been limited (less than five years) and longer follow-up is required to examine the longer-term impact on survival, resistance profile and HIV transmission.
A systematic review provided moderate-quality evidence that current WHO guidelines on immunological and clinical monitoring for treatment failure have poor sensitivity and lower positive predictive value for identifying virological failure in adults (187–200) (Web Annex). This means that many of the people who are identified with immunological failure in fact have adequate virological suppression and risk being misclassified as having treatment failure and switched unnecessarily to second-line therapy. A further systematic review using data in children also provided moderate-quality evidence that immunological criteria (201–204) have low sensitivity and positive predictive value for identifying children with virological failure.
Immunological monitoring versus clinical monitoring
Where viral load monitoring is unavailable, clinical monitoring and CD4 monitoring are recommended (205). Although a systematic review of two randomized controlled trials (184,206) provide moderate-quality evidence of mortality and morbidity benefits with CD4 and clinical monitoring compared with routine clinical monitoring in adults receiving ART, these trials largely focused on CD4 and clinical monitoring in people who initiated ART at CD4 counts below 200 cells/mm3 (Web Annex). Existing immunological and clinical criteria may have decreased sensitivity and specificity to detect treatment failure in people who initiate ART at higher CD4 counts, and more accurate immunological criteria for these people remain to be identified.
Routine versus targeted viral load monitoring to detect treatment failure
Viral load should be monitored routinely (every 6–12 months) to enable treatment failure to be detected earlier and more accurately. In settings with limited access to viral load testing, a targeted viral load strategy to confirm failure suspected based on immunological or clinical criteria (Table 7.14) should be used to avoid unnecessary switching to second-line ART. Targeted viral load monitoring is less costly than routine viral load testing, but as with clinical and immunological monitoring, has the potential to delay switching to second-line ART and may subsequently increase the risk of disease progression, selection of ARV drug resistance and HIV transmission.