In the intention-to-treat analysis, 62% of tafenoquine recipients were free from P. vivax recurrence (95% confidence interval [CI], 55%-69%) at 6 months, as were 70% of primaquine recipients (95% CI, 60%-77%) and 28% of placebo recipients (95% CI, 20%-37%). Compared with placebo, the reduction in risk of recurrence was 70% with tafenoquine (hazard ratio [HR], 0.30; 95% CI, 0.22-0.40; P less than .001) and 74% with primaquine.
Declines in hemoglobin levels were greatest in the tafenoquine group but were not associated with symptomatic anemia and resolved without intervention, the investigators wrote.
In addition to the quantitative G6PD test, the investigators also evaluated a qualitative test, which “failed to identify 16 patients most at risk for hemolysis,” they reported. “If tafenoquine use is expanded, adoption of reliable quantitative point-of-care G6PD tests will be needed; such tests are not currently available but are in development.”
In the second study, Global Assessment of Tafenoquine Hemolytic Risk (), Alejandro Llanos-Cuentas, MD, of Universidad Peruana Cayetano Heredia, Lima, Peru, and his colleagues enrolled 251 patients with confirmed P. vivax infection from Peru, Brazil, Columbia, Vietnam, and Thailand. They attempted to recruit women with moderate G6PD levels, but only one participant met this criterion – all others had normal G6PD activity.
Patients received 3-day course of chloroquine and were randomly assigned on a 2:1 basis to either tafenoquine or primaquine at the same doses as in the DETECTIVE trial.
At 6 months, 2% (95% CI, 1%-6%) of tafenoquine recipients and 1% (95% CI, 0.2%-6%) of primaquine recipients had decreased hemoglobin levels, but none consequently needed treatment. The medications also caused a similar degree and time course of hemoglobin decrease, the investigators noted.
A meta-analysis of GATHER and DETECTIVE confirmed that tafenoquine more often led to a decreased hemoglobin level (4% versus 1.5% with primaquine). Tafenoquine also did not meet prespecified criteria for noninferiority compared with primaquine, with respective 6-month recurrence rates of 67% versus 73%.
However, GATHER deployed “extensive” support to help patients adhere to the 15-day primaquine course, Dr. Llanos-Cuentas and his colleagues wrote. “Without such interventions, adherence to primaquine has been reported to be as low as 24% in Southeast Asia, with a corresponding attenuation of efficacy.”
GlaxoSmithKline and Medicines for Malaria Venture funded both studies, and GSK funded and conducted the meta-analysis. Dr. Llanos-Cuentas and several coinvestigators reported ties to GSK, Medicines for Malaria Venture, the Gambia, and LSTMH. Dr. Lacerda reported having no conflicts of interest.