A CRISPR sickle cell therapy looked highly effective in 28 patients, then stopped early

A gene-editing therapy for severe sickle cell disease produced striking results in a small trial published April 1 in the New England Journal of Medicine: 27 of 28 treated patients had no severe pain crises after infusion, and average hemoglobin levels rose from anemic to near-normal [s1]. The same paper also discloses something the topline numbers don't convey — the trial was terminated early by its sponsor, and the results reported are from an analysis that was not the one originally specified in the study's protocol [s1].

Both things can be true. The results are genuinely strong. They also come with a caveat serious enough that the authors state it plainly in their own methods section.

What was tested, and on whom

The therapy, renizgamglogene autogedtemcel (reni-cel), uses CRISPR-Cas12a gene editing to disrupt BCL11A binding sites in the HBG1 and HBG2 gene promoters — a modification designed to reactivate the body's production of fetal hemoglobin, which does not sickle the way adult hemoglobin does in people with sickle cell disease [s1]. The approach is similar in concept to other approved gene-editing therapies for sickle cell disease, though reni-cel targets a different mechanism.

The Phase 1-2 trial enrolled patients ages 12 to 50 with severe sickle cell disease, defined as at least two severe vaso-occlusive events per year over the prior two years [s1]. Each patient received myeloablative conditioning with busulfan — chemotherapy that clears existing bone marrow — followed by a single infusion of their own gene-edited stem cells [s1]. As of the October 29, 2024 data cutoff, 28 patients had been treated, with a median follow-up of just 9.5 months, ranging from 0.7 to 25.2 months across the group [s1].

The results

Among the 27 patients who achieved both neutrophil and platelet engraftment by the cutoff date, engraftment took a median of 23 and 25 days respectively [s1]. At month six, among the 18 patients with at least six months of data available, mean total hemoglobin had risen from 9.8 g/dL at baseline to 13.8 g/dL, and mean fetal hemoglobin — nearly absent at baseline at 2.5% — had climbed to 48.1% of total hemoglobin, with both measures sustained afterward [s1]. Only one of the 28 patients had a severe vaso-occlusive event after infusion; the rest had none [s1]. Reported adverse events were consistent with what is expected from busulfan-based conditioning and stem cell transplantation more broadly, rather than reflecting a distinct safety signal from the gene-editing step itself [s1].

The caveat the paper states directly

The study was stopped before its planned enrollment or follow-up duration were reached — not because of a safety problem, but, as the authors describe it, "on the basis of the sponsor's reassessment of clinical development priorities" [s1]. The paper's authors are explicit that "results of an analysis that was not prespecified are reported" [s1] — meaning the specific way the 28 patients' outcomes were analyzed and presented was decided after the fact, rather than locked in before the data were collected, as is standard practice for a definitive clinical trial analysis.

That distinction matters for how much weight the results should carry. A prespecified analysis in a trial that reaches its planned size and follow-up duration is designed to protect against the kind of selective emphasis that can make small, early datasets look better than they will hold up as more data accumulate. A non-prespecified analysis of an early-terminated trial does not automatically mean the results are wrong — but it does mean they haven't been tested against the safeguards that make clinical trial results reliable evidence on their own. With a median follow-up under ten months and only 18 of 28 patients contributing six-month data, this remains an early, small, uncontrolled dataset — there was no placebo or active comparator arm, consistent with the ethical constraints of testing an irreversible one-time gene therapy, but a limitation nonetheless for judging effect size.

Why the therapy was reportedly still being developed despite the trial's end

The paper does not explain why the sponsor's development priorities shifted, and this article is not in a position to speculate on that decision. What is verifiable from the published data alone is that the results available — strong hemoglobin normalization, a near-elimination of severe pain crises in 27 of 28 patients, and a safety profile consistent with the known risks of myeloablative conditioning — support, in the authors' own words, "further investigation of this gene-editing approach in the treatment of severe sickle cell disease" [s1], not a conclusion that the question is already settled.

This article is informational and is not medical advice.

Sources

• [s1] Hanna R, Frangoul H, Pineiro L, et al. (RUBY Investigators). "CRISPR-Cas12a Gene Editing of HBG1 and HBG2 Promoters to Treat Sickle Cell Disease." New England Journal of Medicine, 1 April
  1. https://doi.org/10.1056/NEJMoa2415550