Enlarge /. Richard Watkins, 49, (in bed) has complications from sickle cell disease.
Gene therapy has a long and sometimes difficult history. Many human genetic disorders can be traced back to problems with a single gene, which makes them an attractive target for correction. But someone died in a very early gene therapy attempt that put the whole field back on hold. And despite a far more cautious approach, the risks are still substantial, with only two deaths occurring during a trial this year.
But for researchers in the field and those suffering from genetic diseases, this week gives hope that the field's long-belated promise will eventually be fulfilled. At a virtual scientific conference, a group presented the results of a large safety study in which 50 out of 52 patients were able to discontinue treatment for hemophilia. A separate article describes the use of CRISPR gene editing and blood stem cell transplantation to successfully treat patients with sickle cell anemia or a related disorder.
The hemophilia study typified most of the early attempts at gene therapy. In this case, the disease is caused by a defect in a single gene, so if the cells are given a new copy, the problem will be resolved. And because the protein that is encoded by this gene circulates in the blood, you don't have to target a small and potentially inaccessible population of cells to correct things – a new copy of the gene can target cells that can put proteins in exporting the bloodstream will work.
In this case, the defective gene produces a protein called factor IX, which is part of the cascade that enables clotting. People with a defective factor IX gene can receive infusions of a purified version, but they must be repeated every two weeks and are quite expensive. Gene therapy promises to eliminate the need for it. The method used in the new study involves using a virus that carries the factor IX gene to infect liver cells. The virus integrates into the genome of the infected cells and provides them with a permanent copy of a functional factor IX gene.
According to a summary of the presentation of the meeting, 52 patients were enrolled in the Phase III study, of which 50 had completed therapy. Six months after an infusion of the virus, these patients had increased factor IX levels to such an extent that they were at risk of bleeding similar to the lower end of the general population. This level is classified as very mild hemophilia and does not require treatment. The side effects mainly related to an immune response to the virus, which was being treated with steroids.
Both patients who did not respond to treatment also had immune problems from previous exposure to the virus that was used to insert the replacement gene. In one case, while the virus was being transfused, the virus triggered a strong immune response that had to be stopped. The second had very high levels of antibodies that neutralized the virus. But 40 percent of the other patients were also previously exposed to the virus, which was not a problem during the study.
The second attempt at gene therapy was much smaller and the therapy far more complex. It focused on two types of anemia in which the underlying mutations provide some protection against malaria: sickle cell anemia and some form of β-thalassemia. These change how the red blood cells work and cause serious health problems. β-thalassemia damages one of the genes for hemoglobin, causing severe underproduction; People with sickle cell disease produce hemoglobin that forms polymers, resulting in misshapen red blood cells.
While it is possible to treat β-thalassemia with gene therapy similar to that used for hemophilia – just provide a replacement copy of the defective gene – it may not work for sickle cell anemia, which creates an altered form of hemoglobin. One of the ideas that have been considered for treating these anemias is to reactivate the fetal hemoglobin gene. This has a higher affinity for oxygen, which allows it to take up oxygen from the adult form in the placenta. But it turns off within a few years of giving birth.
This shutdown is mediated in part by a protein called BCL11A. So if you eliminate BCL11A, you can theoretically reactivate the fetal hemoglobin genes. Unfortunately, it is not easy to get rid of because it performs essential functions in other cells.
To get around this problem, the researchers behind the new work obtained blood stem cells from patients with β-thalassemia and sickle cell disease. These were then subjected to CRISPR gene editing, which deleted a piece of DNA that was essential for the activation of BCL11A in red blood cells. It wasn't perfectly efficient as expected, but it reached levels where about 80 percent of the copies of the BCL11A were processed. And when these processed stem cells produced red blood cells in culture, they produced increased levels of fetal hemoglobin.
The actual clinical trial involved a risky process: the blood stem cells of two patients, one with β-thalassemia and one with sickle cell disease, were eliminated. The gene-edited stem cells were then infused so that patients could use them to develop a new blood supply. This is a very aggressive procedure and requires extensive medical assistance. Both had severe events that needed to be addressed while recovering. A profile of a participant in the study can help explain why someone would risk it.
Not perfect, but good
It worked for both patients. For the ß-thalassemia patient, the fetal hemoglobin level before the procedure was around 30 milligrams per liter and gradually increased until it reached 1,300 grams per liter a year later. In the sickle cell patient, fetal hemoglobin made up half of their total number one year after the procedure. The amount of sickle cell hemoglobin decreased accordingly. The latter patient had an average of seven major vascular events per year, over three of which typically required hospitalization. She has only had three since the procedure, all of which appear to have been recovering.
While only two patients had passed the annual limit since the start of the study, another eight patients had passed the three-month time point since receiving the same treatment. While the paper does not go into details, it does indicate that the results "are broadly in agreement with the results of the two patients described here".
There are a number of reasons why this procedure does not completely switch patients to fetal hemoglobin. Gene editing wasn't 100 percent efficient at first. In addition, its activity is regulated by factors other than the protein discussed here, and activation does not turn off other forms of hemoglobin. This is not a problem with ß-thalassemia since the mutation only underproduces normal hemoglobin. However, this means that sickle cell patients will continue to produce an altered shape. The key factor is that there is enough of the normal shape to disrupt the formation of hemoglobin polymers.
One of CRISPR's major concerns is that its editing is not always in the correct order. There is sometimes "off-target" editing. In this case, the researchers ran tests to identify off-target changes using cultured cells and found none. That doesn't mean a rare instance hasn't occurred in patients, but it does make them a lot less likely.
None of this means that this particular approach to gene therapy will be widespread (although you can be the two companies behind it, hope it is). For one, a number of other approaches to testing have already been approved, some of which are already being used in patients. And all of them require lots of additional safety and effectiveness tests. And both procedures require enough medical attention that they never become routine.
However, the results suggest that gene therapy is making some headway again after an extensive and necessary pause to define and resolve safety issues. And over the past several years, biologists have developed a number of additional tools that have the potential to make it far more effective.
NEJM, 2020. DOI: 10.1056 / NEJMoa2031054 (About DOIs).