An HIV-positive man in China is doing well after receiving infusions of donor cells whose genes were edited with CRISPR 19 months ago, scientists reported on Wednesday in the New England Journal of Medicine. That makes him the longest-followed individual ever to be treated with CRISPR, a genetic technology that has inspired sky-high hopes for disease cures and spawned a multibillion-industry in just four years.
The patient’s mere survival, apparently without side effects from a treatment that a few studies suggested might trigger cancer or other disastrous genetic damage, provides some assurance that CRISPR-based therapies might be safe — and overshadows the fact that the intervention fell well short of its goal: eliminating HIV from the man’s cells.
“They attempted a moonshot, and while they did not land on the moon, they got back home safely,” said Fyodor Urnov of the Innovative Genomics Institute at the University of California, Berkeley, who previously helped lead a study for Sangamo Therapeutics (SGMO) of whether an older genetic technology, zinc fingers, could cure HIV/AIDS. Crucially, Urnov said of the Chinese study, “they highlighted how to get to the moon.”
It is rare but not unheard of for the New England Journal to publish papers describing a single patient. They do so, said spokeswoman Julia Morin, “when the findings warrant it.”
The 27-year-old patient was diagnosed with HIV/AIDS in May 2016 and, through additional testing, with acute lymphoblastic leukemia two weeks later, scientists led by Hongkui Deng of Peking University reported. The man received the standard AIDS treatment, antiretroviral therapy (ART), as well as chemotherapy for his form of leukemia, a cancer of the immune system’s T cells.
But the scientists saw an opportunity for a moonshot: a one-and-done cure for AIDS. Although ART keeps the virus in check, it does not eradicate HIV hiding in cells, so patients must take ART medications for the rest of their lives.
Serendipitously, however, scientists discovered a dozen years ago that transplants of cells that give rise to blood and immune cells, called hematopoietic stem and progenitor cells (HSPCs), can cure AIDS if their genomes contain a mutation in the CCR5 gene that blocks a common form of HIV from infecting cells.
The “serendipity” arose because that 2007 “Berlin patient,” Timothy Ray Brown, who had both AIDS and leukemia, received a bone marrow transplant for the latter. It just so happened that the donor’s cells had the HIV-blocking CCR5 mutation. Brown, who was treated in Berlin, was thereby cured of both his leukemia and AIDS, and has not taken ART since his cell transplant.
Deng, who received his Ph.D. at the University of California, Los Angeles, and is one of China’s most prominent biologists, reasoned that if hematopoietic stem cells that naturally had the CCR5 mutation can cure AIDS, maybe those whose CCR5 gene is disrupted by CRISPR could do the same.
Because HSPC transplantation from a healthy donor is a standard treatment for acute lymphoblastic leukemia, the risk-benefit balance was much clearer than if the patient had AIDS alone: He was receiving the cell transplant anyway, so only the cells’ CCR5 status would be different.
Deng and his colleagues therefore edited hematopoietic stem cells from a healthy bone marrow donor. They delivered the CRISPR molecules — a target-finding guide RNA and a DNA-cutting enzyme — not via a virus, as other experimental CRISPR therapies are doing, but by “electroporation,” using current to poke a minuscule opening into cells. The edit was designed to disable the CCR5 gene in a way that would keep HIV out of cells like a locked room keeps out visitors lacking a key.
As is typical with CRISPR, only a fraction of the cells were successfully edited. When Deng and his colleagues tested the hematopoietic stem cells before infusing them into the patient, they found that 17.8% were without a functioning CCR5, as intended.
Once the edited stem cells were infused into the patient and settled into his bone marrow, a series of measurements over time showed that only 5.2% to 8.28% of his bone marrow cells, progeny of the stem cells, had the intended CCR5 edit.
In an email interview, Deng acknowledged that was a “low efficiency of gene editing in the patient.” The likely explanation is that many of the edited, infused cells do not survive, and so are vastly outnumbered by the unedited donor cells or even the patient’s own.
His leukemia went into remission after the cell transplant. But the low fraction of CCR5-edited cells was insufficient to control his viral load. When the researchers took the patient off antiretroviral therapy seven months after the stem cell transplant, with his consent, HIV numbers rose and healthy T cell numbers fell. He went back on ART.
One encouraging finding, however, was that the percent of CCR5-edited cells rose during the drug holiday. Before then, the level of CCR5 disruption in T cells in the blood was 2.96%, Deng told STAT, and it “peaked at 4.39% during the interruption.” Unfortunately, he continued, “the ideal gene editing efficiency is 100%, as suggested by the case of Berlin patient.”
Two scientists not involved in the study questioned the decision to transplant stem cells when so few had been successfully edited.
It’s been known for years “that more efficient CCR5 knockout tracks with response,” said Berkeley’s Urnov. “They chose to treat a subject with cells that had less than 20% editing. This is not a decision I would have made.” That they did not achieve HIV control is not “surprising,” he said, but does not “reflect negatively on the approach.”
In fact, low editing rates might be meaningful. In the Sangamo study using zinc fingers to edit CCR5, 12 HIV-positive patients received a single dose of 10 billion T cells. Between 11% and 28% of the cells were edited as intended, no longer producing normal CCR5. Patients’ levels of a kind of T cell that HIV kills, called CD4 T cells, rose, suggesting the gene-edited T cells were not merely holding their own but actually proliferating, Sangamo reported in 2014. Although Sangamo concluded that the treatment was safe, and the company conducted a larger clinical trial, it is no longer pursuing that program, focusing instead on genome editing for sickle cell disease and other blood disorders.
Deng nevertheless called the editing levels he and his colleagues achieved “a promising approach for gene therapy” of HIV/AIDS. “The cells with the modified CCR5 gene persisted” for 19 months and counting, albeit at low levels, he said, and CRISPR hit the intended target and no others, as far as the researchers could determine by genome sequencing. That suggests CRISPR did not go rogue, editing out cancer-suppressing genes, for instance.
So far, there is only one patient in the study, Deng said. But “we are planning to improve editing efficiency in hematopoietic stem cells and optimize the transplantation protocol based on this patient,” he said, guided both by his results and those from mouse studies he and his colleagues reported in 2017. That experiment, they wrote, “provides evidence for translating CCR5 gene-edited [hematopoietic] transplantation for an HIV cure to the clinic.”
Deng was born in China and graduated from college there. After receiving his Ph.D. at UCLA in 1995, he worked as a postdoctoral fellow at New York University, focusing on HIV, and in 1998 became research director of the Boston-based stem cell biotech ViaCell. He was enticed back to China in 2001 by a prestigious professorship at Peking University, where he initially worked on the use of human embryonic stem cells to cure diabetes. In 2006 he won $1.9 million from the Bill and Melinda Gates Foundation’s Grand Challenges in Global Health, for research on HIV and hepatitis C vaccines.
Not everyone agrees that CRISPR is ready for use in HIV/AIDS patients. “In my view, it is too early to go to the bedside,” said biologist Shuliang Chen of Ohio State University and Wuhan University, who co-authored a recent review of gene editing for HIV/AIDS. For one thing, disabling CCR5 keeps out only one strain of HIV, not all of them. More generally, Chen said, “We need more evidence about CRISPR-related safety, off-target effects, and on-target efficiency.”
But there is a larger issue. “In most cases, HIV patients live well with antiviral drug treatment,” Chen said; Deng’s patient was doing just that. The only rationale for “high-risk gene editing,” he said, is if they also have leukemia, in which case one “can use gene-edited hematopoietic stem cells to kill two birds with one stone.”
Caveats notwithstanding, there will be more experiments trying to cure HIV/AIDS with CRISPR. “This is not an all-out success,” Urnov said, “but what to do next and how to get to that success is now even more clear.”