First HIV-positive to HIV-negative liver transplant raises intriguing questions for cure research

An HIV-negative child received a liver transplant from its HIV-positive mother and now appears to have no evidence of HIV infection apart from a very weak antibody response, despite extensive testing for the virus, South African researchers report in the journal AIDS. Researchers cannot decide whether the child has cleared HIV or if the virus is deeply hidden, perhaps only in the liver.

The case is the first reported transplant from an HIV-positive donor to an HIV-negative recipient and raises the question of whether transmission through the donor organ might be preventable if the recipient begins antiretroviral therapy as prophylaxis – preventive therapy – prior to the transplant.

The case

The infant’s mother was diagnosed with HIV approximately six months before conceiving, and began antiretroviral treatment with a regimen of efavirenz, tenofovir and lamivudine six weeks prior to conceiving. Her infant received nevirapine prophylaxis for six weeks after birth and subsequently tested HIV negative.

The infant was referred for care at Wits Donald Gordon Medical Centre in Johannesburg at seven months of age. The infant required a liver transplant due to end-stage liver disease resulting from biliary atresia (a childhood disease in which the bile ducts are narrowed or blocked, leading to liver damage when bile cannot drain from the liver).

The average wait time for a paediatric liver transplant at Wits Donald Gordon Medical Centre is 49 days, but after 181 days a liver was still not available for transplant and the infant was admitted to hospital with life-threatening bleeding from varices (enlarged veins supplying the liver). 

At this point the clinical team decided to seek ethical approval for the mother’s request to become a living donor. This procedure involves taking a portion of the donor liver rather than the whole liver.

The transplant was approved and took place when the infant was 13 months old. Methylprednisone was given as an immunosuppressant during the operation, and oral corticosteroids for six months after the transplant to prevent organ rejection. Tacrolimus was also prescribed and is still being taken by the infant, again to prevent rejection of the transplanted liver. To prevent HIV transmission, a regimen of raltegravir, lamivudine and abacavir was commenced the day before transplantation.

To monitor the infant’s HIV infection status, testing took place for HIV antibodies, HIV DNA and HIV RNA. The child tested negative in all respects prior to transplantation.

The child tested positive for HIV antibodies 43 days post-transplant on the Architect HIV antigen/antibody test, and continued to test positive to day 379, although antibody titres declined rapidly to almost undetectable levels, indicating a lack of HIV replication to stimulate antibody responses.

No HIV DNA or RNA was detected in plasma at any point. A Western Blot test performed 225 days after transplantation was indeterminate. The assay showed that antibodies to HIV core proteins p24, p40, p55 and p65 were detectable, but no antibodies to envelope proteins could be detected. 

No proviral DNA could be detected in peripheral blood mononuclear cells or leukocytes sampled at day 225, suggesting either that no HIV-infected cells persist, or that so few cells had been infected, they were undetectable in the blood samples taken.

Analysis

The study investigators say that there are several possible explanations for these results.

One explanation is that although an HIV reservoir may have been established after the transplant, it is undetectable by current tests.

Another possibility is that infection is confined to cells in the transplanted donor liver and that no recipient cells have become infected. In this scenario, antibody responses are being produced by maternal B-cells in the donor liver tissue to antigens produced by the donor liver cells.

A third possibility is that the child is mounting an antibody response to HIV antigen produced by maternal liver cells. In both scenarios where the maternal liver is the source of antigen, the waning antibody response suggests that antigen production is insufficient to maintain a high level of antibodies.

The study investigators say that the only way to determine whether the child is still infected with HIV may be to stop treatment and monitor for viral load rebound, but they say that further discussion will be needed on the ethics of this approach and that the child will stay on continuous antiretroviral therapy for two years. The investigators say that they hope to undertake further transplants, but with more intensive sampling and monitoring.

In an accompanying editorial comment, liver specialists Professor Jurgen Rockstroh and Francisco Gonzalez-Scarano say that it is probably ethical to stop antiretroviral treatment at some point, “as one could not justify a lifetime of antiretroviral therapy absent proof of its necessity.” But they also say that the case does not provide sufficient evidence to adopt a new approach to the use of HIV-positive transplant donors, and that the question of whether and how HIV transmission is preventable in these circumstances needs to be answered “reassuringly, and in more than one instance.”

In particular, they note several cases where antiretroviral therapy was commenced within hours or days of transmission but did not prevent the establishment of infection. When antiretroviral treatment was stopped viral load eventually rebounded even though HIV DNA was undetectable for long periods prior to discontinuing treatment.