A 52-year-old man with HIV had contracted an acute HBV infection 15 years earlier. Antiretroviral therapy (ART) consisting of tenofovir disoproxil fumarate/emtricitabine (TDF/FTC) and lopinavir/ritonavir (LPV/RTV) was initiated at the time, and the acute hepatitis rapidly improved. The patient’s liver enzymes normalized 2 months after beginning ART, and he tested negative for HBV DNA and HBsAg 4 months after beginning ART (Table 1). Anti-HBs titers were detected at 2.03 IU/mL 9 months after beginning ART, and it was believed that his acute hepatitis B did not transition into a chronic HBV infection. His anti-HBs titers were again detected at 18.6 IU/mL 6 years after the initial diagnosis of the acute HBV infection. There was no reactivation of HBV while on stable ART containing TDF/FTC or TAF/FTC. Two years after the confirmation of seroconversion, his anti-HBs levels decreased below the cutoff value but remained around 8.0 mIU/mL. He tested positive for anti-HBc.

After 14 years of ART, the patient expressed a preference to change to a long-acting injectable ART. We considered 3 additional factors before making this medication change: the patient did not become an HBV carrier after his acute hepatitis B diagnosis, his HBsAg and his HBV DNA had remained undetectable for >10 years, and his anti-HBs had declined only after initial positive confirmation. His hepatitis B core-related antigen, which is shown to remain positive longer than HBsAg and HBV DNA in acute hepatitis B infections (5,6), was negative. After a 1-month lead-in with oral CAB/RPV, his ART was changed to an injectable CAB/RPV.

The patient’s liver enzyme levels increased to alanine aminotransferase (ALT) 103 U/L (reference range 4–44 U/L) 4 months after switching to CAB/RPV. Both his HBsAg and HBV DNA tested positive again, while the level of anti-HBs decreased to nearly zero (Table 2). The probability of reinfection with a novel HBV strain was considered low because his sexual partner was on ART containing tenofovir during this period and the patient denied any other potential exposures. The patient’s CD4 count was 512 cells/μL, indicating sufficient immune function. The patient’s liver enzymes returned to within reference ranges 4 months later, whereas the HBsAg and HBV DNA remained elevated. The patient’s ART regimen was switched to B/F/TAF 14 months after the change to CAB/RPV because of concerns of a concurrent chronic HBV infection.

We cloned 2 full-length HBV isolates and determined the genome sequences from serum samples obtained at 2 intervals: the first after 4 months of treatment with CAB/RPV, coinciding with the re-positivity of HBsAg, and the second 14 months after treatment with CAB/RPV (Appendix). Our genomic analysis revealed identical clones of genotype A2, and we observed no increases in mutations after reactivation (Table 3). Of note, E164V was identified within the S region and is recognized as a vaccine-escape mutation (Appendix Figure) (7). No drug resistance–associated mutations were observed in the polymerase region.

There are multiple explanations for the re-emergence of HBsAg in this case. The mostly likely explanation is our patient did not attain a functional cure of HBV. The reduction in HBsAg might have been because of ART containing TDF/FTC or TAF/FTC instead of acquired immunity against HBV, implying that anti-HBs levels were inadequate. After switching to CAB/RPV, the drugs no longer suppressed HBV. The exposure to reactivated HBV did not adequately boost the anti-HBs levels, which lead to breakthrough reactivation. Another potential explanation is a de novo reactivation of HBV independent of switching to CAB/RPV. However, we were unable to find documented cases of de novo HBV reactivation among persons on stable ART or with a stable immune status. Furthermore, whereas de novo HBV reactivation typically leads to severe hepatitis (8), our patient remained asymptomatic, with only a marginal elevation in transaminase levels.

E164V in the S region is known as a vaccine-escape mutation and is frequently identified in patients with occult HBV infection or de novo HBV reactivation (9,10). Cases of breakthrough infection are extremely rare worldwide, and it is not clear whether a single mutation is responsible for immune escape. In patients with occult HBV infection, escape mutations other than E164V are typically detected alongside other mutations (10,11).

HBV reactivation has been reported in patients with low CD4 counts (12,13). Cases have also been reported of HBV DNA detection at very low levels with negative HBsAg after switching to ART regimens without anti-HBV drugs (14). Our patient had a sufficient CD4 count for immunity and a history of anti-HBc and anti-HBs positivity, and still we found his HBV reactivated with re-emergence of HBsAg. We expect our patient will undergo the seroconversion of HBsAg again with ART containing F/TAF. Nonetheless, ART without tenofovir or lamivudine is not an option for this patient in the future.

Below the cutoff value, the preventive effect of anti-HBs may not be sufficient to prevent reactivation. Vaccination may be beneficial for patients with isolated anti-HBc positivity, even if their anti-HBs levels have declined since their initial anti-HBs positivity was confirmed (15). For many years, ART containing tenofovir or lamivudine has been widely used worldwide; however, providers are increasingly using ART regimens that do not include anti-HBV drug. Healthcare providers must consider the potential scenario where the detection of anti-HBs might be a consequence of HBV suppression through ART containing anti-HBV drugs, and not indicative of a functional cure. Switching to ART regimens without anti-HBV drugs should be approached with caution.