Having a hidden infection with hepatitis B may undermine viral suppression after a switch to dolutegravir / lamivudine (Dovato), a study of people with HIV who simplified treatment in Italian and French clinics has concluded.
The study investigators recommend careful monitoring of hepatitis B activity and HIV suppression in anyone with markers suggesting a hidden hepatitis B infection – antibodies to hepatitis B core antigen without corresponding antibodies to the hepatitis B surface antigen.
A switch from a regimen that contains tenofovir to nucleoside-sparing treatment or dual therapy containing lamivudine is not recommended for people who are positive for hepatitis B surface antigen (active hepatitis B infection). This is because people with HIV may experience hepatitis B reactivation and flares in hepatitis B replication that can lead to severe liver damage. Although lamivudine is active against hepatitis B, resistance to the drug can emerge rapidly.
People can have an ‘occult’ or hidden hepatitis B infection when hepatitis B DNA is detectable in liver tissue or blood without the presence of hepatitis B surface antigen (HBsAg) or antibodies to surface antigen. People exposed to hepatitis B who subsequently clear the infection retain antibodies to hepatitis B core antigen (HBcAg). Antibodies to hepatitis B core antigen can also be a sign of a new infection. The implications of occult hepatitis B are uncertain; those studies that have looked at liver disease in people with occult hepatitis B have reported contradictory findings.
The prevalence of occult hepatitis B varies between regions and appears higher where hepatitis B transmission is more common. A meta-analysis of prevalence studies carried out in people with HIV in Africa showed an overall prevalence of occult infection of 11%, rising to 26% in southern Africa. And one Italian study found that 30% of people with HIV who had antibodies to hepatitis B core antigen had very low-level hepatitis B replication, while another found that having antibodies to hepatitis B core antigen was correlated with poorer virological control on antiretroviral treatment.
Current European guidelines say that people with antibodies to hepatitis B core antigen should be monitored after switching to dolutegravir / lamivudine but do not advise against the use of the combination for this group.
Italian researchers were concerned by the risk that people switching from a three-drug regimen to two-drug treatment with dolutegravir / lamivudine might experience poorer virological control if they switched to two-drug treatment, due to less effective suppression of hepatitis B. They carried out a retrospective multicentre study with Karine Lacombe at the Hôpital Saint Antoine in Paris to investigate the issue.
The study population consisted of 267 people with HIV at participating clinics who had switched from three-drug treatment to dolutegravir / lamivudine and who had a viral load below 50 copies/ml at the time of switching and follow-up data. Twenty-eight percent had antibodies to hepatitis B core antigen and of these, three-quarters had antibodies to hepatitis B surface antigen.
Participants had a median age of 41 years and 32% were female. Before switching, 46% were taking two nucleoside reverse transcriptase inhibitors (NRTIs) and an integrase inhibitor, 33% two NRTIs and a non-nucleoside reverse transcriptase inhibitor and 15% a protease inhibitor with two NRTIs. The median CD4 count at the time of switching was 754.
The study examined changes in viral load from 24 months before switching until 36 months after switching.
Most participants (74%) had a history of prolonged viral load suppression (at least two years below 20 copies/ml and target not detected, indicating no detectable virus whatsoever). At the time of switching, 85% had viral load below 20 copies and target not detected, just under 10% had viral load below 20 copies/ml but target detected (indicating detectable virus but below 20 copies), and the remainder had viral loads between 20 copies and 50 copies/ml.
At six and 12 months after switching, 82% had viral load below 20 copies/ml and target not detected, while 8% and 9% respectively had viral load below 20 copies/ml but target detected. At month 6, 8% had viral load between 20 copies and 50 copies/ml. By month 12, 5% had viral load between 20 copies and 50 copies/ml and 3% had viral load above 50 copies but below 200 copies/ml.
At 24 and 36 months after switching, 84% had viral load below 20 copies/ml and target not detected, while 9% and 12% respectively had viral load below 20 copies/ml but target detected. At month 24, 4% had viral load between 20 copies and 50 copies/ml. By month 36, 2% had viral load between 20 copies and 50 copies/ml and 2% had viral load above 50 copies but below 200 copies/ml. At month 36, 3% had viral load between 20 copies and 50 copies/ml and less than 1% had viral load above 50 copies but below 200 copies/ml.
These figures disguise a degree of fluctuation in viral detectability. Only half of participants followed for 36 months maintained a viral load below 20 copies with target not detected, indicating complete viral suppression at all measurement points during the follow-up period.
Before switching, there was no difference in viral suppression between people who had antibodies to hepatitis B core antigen and those who did not. But from 12 months after switching, a lower proportion of those with antibodies to hepatitis B core antigen had viral load below 20 copies with target not detected (69% vs 87%, p=0.004). They were more likely to have target detected viral load measurements below 20 copies and to have viral loads between 20 and 50 copies/ml. The same pattern held true at subsequent time points, so that, by month 36, 26% of those with antibodies to hepatitis B core antigen had viral load below 20 copies with target detected and only 18% had maintained a viral load below 20 copies with target not detected at all time points.
"Low-level hepatitis B replication could be stimulating HIV replication, raising the risk of viral rebound."
A multivariable analysis that considered age, year of HIV acquisition, CD4 count, history of virological failure and hepatitis B core antigen status showed that people with antibody to hepatitis B core antigen were three times more likely to have detectable HIV at month 36 and this was the only factor significantly associated with viral detectability (p=0.005).
The study investigators speculate that low-level hepatitis B replication could be stimulating HIV replication, eventually raising the risk of viral rebound in people taking dolutegravir / lamivudine. This risk needs to be taken into account when people with antibodies to hepatitis B core antigen switch to tenofovir-sparing regimens, including long-acting regimens that do not contain agents that suppress hepatitis B.
However, the study investigators acknowledge that because most clinics did not measure hepatitis B DNA routinely in people with antibodies to hepatitis B core antigen, they do not know what proportion of participants already had detectable HBV DNA at the time of switching – or what proportion developed detectable HBV DNA during follow-up and how that correlated with the emergence of incomplete HIV suppression.
What’s needed, they conclude, is a prospective study which monitors hepatitis B DNA levels before and after treatment simplification and discontinuation of tenofovir-based treatment. A study should also investigate whether hepatitis B vaccination response in people with antibodies to hepatitis B core antigen has any impact on HIV or HBV viral detectability after switching, they suggest.
Malagnino V et al. HBcAb positivity as a risk factor for missing HIV RNA undetectability after the 3TC+DTG switch. Viruses, published online, 23 February 2024.