HIV has been cured. Five times.

Physicians in Berlin, London, Düsseldorf, New York and Geneva cured five people of HIV through stem cell transplants given for blood cancers; the virus has not returned in any of them without medication.

A mutation that removes the door

HIV-1 enters a cell through two proteins on its surface: the CD4 receptor and a co-receptor, most often CCR5. A small minority of people, one to two percent of the general population according to City of Hope, carry two copies of a 32-base-pair deletion in the CCR5 gene, written CCR5 delta 32, and their cells make no functional CCR5. Homozygosity for the deletion confers resistance to acquiring HIV-1. The idea behind every case in this story is to replace a patient's immune system with one built from such cells, so that the virus has nowhere to go.

The replacement is a stem-cell transplant, a procedure with substantial mortality that is performed to treat blood cancers, not HIV. Every person counted as cured was first a cancer patient.

Berlin, 2007

Hütter and colleagues in Berlin reported the first case in the New England Journal of Medicine in February 2009. A man with acute myeloid leukemia and HIV-1 infection received stem cells from a donor homozygous for CCR5 delta 32. Antiretroviral therapy was discontinued, and 20 months after transplantation there had been no viral rebound. The authors wrote that the outcome demonstrated the critical role CCR5 plays in maintaining HIV-1 infection. The patient, later identified as Timothy Ray Brown, underwent transplants in 2007 and 2008. According to a memorial published by the IAS-USA, he died on September 29, 2020, at the age of 54, of a recurrence of leukemia, and remained free of HIV until his death.

London, 2019

For a decade Brown was the only case. In March 2019 Gupta and colleagues reported in Nature a second: a man treated for Hodgkin's lymphoma with a single transplant from a CCR5 delta 32 homozygous donor, using reduced-intensity conditioning and no total-body irradiation, a less aggressive regimen than Brown's. He had mild gut graft-versus-host disease. Antiretroviral therapy was interrupted 16 months after transplant, and remission had lasted a further 18 months at publication. Plasma HIV-1 RNA was undetectable at less than one copy per millilitre; viral outgrowth assays on 24 million resting CD4 T cells found no reactivatable virus. The authors wrote that it was premature to conclude that the patient was cured.

A follow-up in Lancet HIV in 2020 extended the observation to 30 months off therapy, with donor chimerism of 99 percent in peripheral T cells and no replication-competent virus in blood, cerebrospinal fluid, gut or lymph node tissue. Mathematical modelling put the probability of lifelong remission at 98 percent given 80 percent donor chimerism in HIV target cells. The authors proposed that these findings represented cure.

Düsseldorf, New York, California

Jensen and colleagues described the third case in Nature Medicine in 2023. A man, 53 years old at the time of writing, received a CCR5 delta 32 homozygous transplant for acute myeloid leukemia in February 2013 and stopped antiretroviral therapy in November 2018, 69 months later. Over more than nine years of monitoring, droplet digital PCR and in situ hybridization found sporadic traces of HIV-1 DNA in T-cell subsets and tissue, but repeated outgrowth assays, including transfer into humanized mice, recovered no replication-competent virus. Four years after stopping treatment, the authors called the absence of rebound strong evidence for cure.

Hsu and colleagues at Weill Cornell reported the fourth in Cell in 2023, and it differed in two ways. The patient was a woman of mixed race, and the CCR5 delta 32 homozygous cells came from umbilical cord blood, combined with adult stem cells from a half-matched relative, a method intended to widen the donor pool for people whose ancestry makes a matched adult delta 32 donor unlikely. She developed acute myeloid leukemia four years after her HIV diagnosis. Blood chimerism was 100 percent cord-derived by week 14. Antiretroviral therapy was stopped 37 months after transplant, and she had remained aviremic for 18 months at publication, with 4.8 years of total follow-up.

In February 2024, Jana Dickter and colleagues at City of Hope published the fifth in the New England Journal of Medicine. According to the hospital, Paul Edmonds, then 68, had lived with HIV for more than 31 years, received a reduced-intensity transplant on February 6, 2019, and had been off antiretroviral therapy for nearly three years.

Geneva: a donor without the mutation

The Geneva case, reported by Sáez-Cirión and colleagues in Nature Medicine in 2024, broke the pattern. A 53-year-old man diagnosed with HIV-1 in May 1990 received a transplant in July 2018 for an extramedullary myeloid tumour. The donor was unrelated, matched at nine of ten HLA alleles, and carried normal CCR5. Antiretroviral therapy was stopped on November 17, 2021. At 32 months off therapy, plasma viral load remained undetectable. His CD4 T cells remained fully susceptible to HIV-1 infection in the laboratory, yet no virus could be grown from them; only defective, not intact, proviral DNA was detected sporadically. He continued on ruxolitinib for chronic graft-versus-host disease throughout. The authors describe durable remission rather than cure, and note that other recipients of CCR5-normal transplants have rebounded.

Counting the cases

The Geneva paper's own opening sentence counts five individuals cured after transplants from CCR5 delta 32 homozygous donors before it: Berlin, London, Düsseldorf, New York and City of Hope. Geneva made six. In July 2024, at the 25th International AIDS Conference, Christian Gaebler of the Charité presented a seventh, described by the International AIDS Society as the next Berlin patient: a man transplanted for leukemia in late 2015 from a donor with only one copy of the delta 32 deletion, off antiretroviral therapy since late 2018 and still in remission about five and a half years later. That case was the first in which the donor was not homozygous.

What the story does and does not show

Each case is a single patient. All seven had a blood cancer that independently required a stem-cell transplant, a procedure whose risks make it unavailable as a treatment for HIV alone. Durations off therapy at publication range from 18 months to more than four years, and several of the authors describe remission rather than cure. Low-level HIV DNA, defective or otherwise, was detected in more than one patient; what has not been detected is virus capable of replicating.

The story's count of five is low. By mid-2024 six cases had been published in peer-reviewed journals and a seventh had been presented at a scientific conference. The story's cities also omit California. The slide stating that in most cases the donor carried the delta 32 mutation is accurate; the Geneva donor did not, and the seventh donor carried only one copy, which means the mechanism is not fully explained by the missing receptor alone. The sequence does show that HIV-1 can be removed from a human body and stay removed for years without medication. It does not yet show a way to do that for anyone who is not also being treated for cancer.

The virus can leave the body for good. It has, five times.