No increased risk of IRIS in people with low CD4 counts receiving raltegravir in the REALITY trial

Despite more rapid viral load decline with a raltegravir-intensified regimen compared with a standard one, there was no evidence that this increased the risk of IRIS in people starting ART with advanced HIV. This important finding from the REALITY trial was presented at CROI 2018.

The REALITY trial, conducted in ART naive adults and children with less than 100 CD4 cells/mm3 in sub-Saharan Africa, showed that raltegravir (RAL)-intensified treatment led to faster viral load decline but no reduction in mortality or WHO 3/4 events.

As integrase inhibitors are steadily replacing NNRTIs first-line, there is concern that rapid viral load reduction might increase rates of immune reconstitution inflammatory syndrome (IRIS) in people starting ART with very low CD4 – which is still common in low- and middle-income countries.

Diana Gibb presented an evaluation of rates of IRIS events among people receiving treatment intensified with RAL versus standard ART on behalf of the REALITY investigators.

Participants were randomised to start ART of 2NRTI + NNRTI with 12 weeks RAL (standard of care; SoC + RAL) vs without (SoC). A blinded endpoint review committee adjudicated serious adverse and WHO grade 3/4 events, causes of death and compatibility with IRIS. Predictors of time to first fatal/non-fatal IRIS-compatible event were identified using backwards elimination (exit p=0.05) treating death from other causes as a competing risk.

Overall 1805 participants with median age of 36 years (4% were 5–17 years), CD4 37 cells/mm3 and viral load 249,770 copies/mL – almost three quarters of whom had 100,000 copies/mL or more – were randomised to SoC + RAL (n=902) vs SoC (n=903).

Approxmiately 90% of participants received an efavirenz-based SoC regimen and 80% an TDF/FTC NRTI backbone.

At week 4 from starting ART the percentage of participants with viral load <50 copies/mL was 41.0% vs 13.4% in the SoC + RAL vs SoC arms respectively; at week 12 these values were 71.9% vs 51.7% (both p<0.0001). By weeks 24 and 48 percentages were similar in both arms, approximately 75% and 80% respectively. Mean change in log10 viral load at week 4 was -3.4 in SoC + RAL and -2.7 in SoC (p<0.001).

All-cause mortality at 24 weeks was similar in both arms: 10.9% vs 10.2% in SoC + RAL vs SoC. Fatal IRIS was seen in 36 (4.0%) vs 31 (3.4%) in SoC + RAL vs SoC (p=0.54) occurring a median 4.4 (IQR 2.6– 9.4) weeks after starting ART. Overall fatal/non-fatal IRIS events occurred in 89 (9.9%) vs 86 (9.5%) in SoC + RAL vs SoC (p=0.79). Over half of the fatal and non-fatal IRIS events in both arms were TB IRIS.

Risks of fatal/non-fatal IRIS were independently higher in participants with lower pre-ART CD4 (p<0.001), older people (p=0.004) and those with TB when starting ART (p=0.01).

Enhanced prophylaxis of cotrimoxazole + 12 weeks isoniazid/B6 + 12 weeks fluconazole + 5 days azithromycin + single dose albendazole (also evaluated as a factorial in the REALITY trial) was protective of IRIS vs standard cotrimoxazole alone. Fatal/non-fatal IRIS compatible events occurred in 67 (7.4%) vs 108 (12.0%) in the enhanced vs standard prophylaxis groups (p=0.001).

comment

These raltegravir data can likely also be applied to dolutegravir as rapid viral load reductions are similar with all integrase inhibitors. The results provide reassurance that the current transition to ART with dolutegravir first-line will not result in increased IRIS.

Pre-ART CD4 is still needed to identify people with advanced HIV at high mortality and IRIS risk for whom enhanced prophylaxis would be helpful.