The continuous re-emergence of yellow fever virus (YFV) poses a serious public health threat, with no approved specific antiviral drugs for treatment. Sofosbuvir and uprifosbuvir, developed for treatment of chronic hepatitis C, are nucleotide analogs with anti-YFV activity in vitro, and sofosbuvir has been used off-label in YFV-infected individuals. To study the barrier to resistance of these drugs, escape experiments were performed with the YFV 17D vaccine strain in human hepatoma Huh7.5cells. Compared to the original virus, sofosbuvir and uprifosbuvir escape variants exhibited 8.6- to 10.7-fold increased 50% effective concentration (EC50) values. Escape viruses developed putative resistance-associated-substitutions (RAS) M478K and A482G in the NS5 RNA-dependent RNA polymerase domain. Reverse-engineered YFV 17D mutants demonstrated that A482G was the primary determinant of drug resistance, resulting in 5.3- to 5.7-fold increased EC50; however, the substitutions acted synergistically when combined, leading to 12.3- and 11.1-fold increased EC50. A482G conferred slight cross-resistance to bemnifosbuvir, while no cross-resistance was observed for remdesivir, galidesivir, or NITD008. Escape and mutant viruses had slight delays in growth kinetics in Huh7.5cells and were viable in C6/36 and Aag2-AF5 mosquito cells. Structural analysis of a predicted AlphaFold3 YFV NS5-RNA structure suggests that M478K and A482G induce conformational changes in RdRp motif F, which may reduce sofosbuvir and uprifosbuvir affinities. Sofosbuvir RAS described for other orthoflaviviruses did not substantially affect drug susceptibility in reverse-engineered YFV 17D. This study demonstrates that YFV can develop resistance to sofosbuvir and uprifosbuvir with only two NS5 substitutions that have slight impact on viral fitness.

Letermovir (LMV), a novel antiviral agent targeting the cytomegalovirus (CMV) terminase complex, has significantly changed the management of adult allogenic hematopoietic stem cell transplant (allo-HSCT) recipients. In this retrospective, single-center study, we assessed our real-life experience in using an implemented diagnostic protocol to identify patients needing pre-emptive therapy (PET) or at high risk of developing CMV-related disease. The study included allo-HSCT recipients who received LMV prophylaxis (LMV group), compared with a historical control group of CMV seropositive patients undergoing allo-HSCT. These patients were managed using a pre-emptive strategy. Both study groups were followed from the day of transplant for up to 200 days. Among the 110 patients receiving LMV prophylaxis, CMV DNAemia was detected in 43 cases (39.1%). However, infectious virions associated with active CMV infection were identified only in four patients (4/43, 9.3%), who required the interruption of LMV and the initiation of PET (CS-CMVi cases). We evaluated the incidence of CS-CMVi at 100 and 200 days post-transplant, comparing patients undergoing LMV prophylaxis with a historical group managed by PET. At 100 days post-transplant, the cumulative incidence of CS-CMVi was significantly lower in the LMV group compared to the control group (3.6% versus 39.1%; p<0.001). The same result was observed at 200 days post-transplant. These findings confirm that LMV has significantly decreased the incidence of CS-CMVi in CMV seropositive patients and underline the clinical utility of an implemented diagnostic protocol including DNase or RNAemia tests to discriminate between active and abortive CMV infection.