Kamau's Nula-cel Sickle Cell Gene Therapy: Three Patients Transfusion-Free After Precise Gene Correction
On November 4, Kamau Therapeutics announced the first multi-patient results from a gene therapy that does something no approved sickle cell treatment has done: correct the mutation itself. Three patients with severe sickle cell disease received nula-cel, a therapy made with the company's revised V2.0 manufacturing process. All three left the hospital transfusion-independent, and none has had a vaso-occlusive event — a sickle cell pain crisis — since the infusion.
The numbers behind that sentence are what make it worth reading twice. In the two patients with more than six months of follow-up, total hemoglobin rose above 10 g/dL (10.2 and 11.1 g/dL), sickle hemoglobin fell to 17.6% at 12 months and 19.1% at 8 months, and normal adult hemoglobin became the predominant type, at 76.8% and 66.8%. The corrected gene held: bone marrow HDR allele frequencies stabilized at about 20%, and single-cell analyses showed more than 90% of circulating red blood cells carried protective levels of anti-sickling hemoglobin. The very first patient ever treated with nula-cel, using the earlier V1.0 process, is now four years post-infusion. No serious adverse events related to nula-cel or the editing procedure were observed, and off-target editing stayed low and unchanged.
Sickle cell disease has a single root cause: one letter in the beta-globin gene. That letter produces sickle hemoglobin, which stiffens red cells, drives chronic hemolysis, and shortens life expectancy in the United States by about 30 years. Roughly 100,000 Americans live with the disease, and millions more worldwide. The gene therapies approved so far work around the mutation by reactivating fetal hemoglobin. Nula-cel takes the other path: it uses homology-directed repair to convert the sickle mutation back to the normal adult beta-globin sequence, in the patient's own blood stem cells, edited outside the body and returned under the body's own regulation. If it holds, the patient makes normal adult hemoglobin for life from a single treatment.
What I find most interesting is how the company's chief executive framed the result. "Results and a process built to repeat them are not the same thing," Vali Barsan said. "We built the process; patient after patient is how a corrected gene becomes ordinary medicine." That is the honest way to read three patients: a signal, not yet a therapy. The RESTORE trial enrolls about 15 participants, and the uncertainties are real. Three patients is a thin base, durability beyond a year is unproven, and the busulfan conditioning required before infusion is grueling. Ex vivo autologous therapy is also among the most complex manufacturing in medicine — every dose is a bespoke product, and the company reports its drug-product correction rates ran at 32% to 38% against a target above 20%, with a median of one apheresis cycle to collect enough cells. Those are good numbers, but they have to repeat fifteen times, then hundreds.
There is also the question the press release does not ask. Most people with sickle cell disease live in sub-Saharan Africa, where a therapy that costs millions and requires a transplant unit cannot reach them as currently built. The science is aimed at the mutation; the delivery system is aimed at wealthy health systems. Both facts matter, and the second one will decide whether this becomes a cure for a disease or a cure for a few.
The next check is already dated. Kamau will present the RESTORE data in an oral session at the American Society of Hematology meeting in New Orleans on December 14, 2026, presented by John DiPersio of Washington University School of Medicine. My prediction is that the update will include at least one additional patient treated with the V2.0 process beyond the three reported on November 4, and that every V2.0 patient reported will still be transfusion-independent, with no vaso-occlusive events and no treatment-related serious adverse events. I put it at 55%: the abstract was probably locked weeks ago, so a patient dosed since could plausibly appear, but autologous manufacturing is slow and the company may present exactly the abstract data. A miss would be informative either way — it would tell us the process is not yet repeating at pace, which is precisely the question Barsan raised.
If the December update holds, this will be the first time precise correction of a disease-causing mutation has worked in patients — not a workaround, a fix. That is a sentence worth watching for.