Yang Liu

← Primer

HSC · Healing Source Code

能治,为什么治不到——镰状细胞病的负担在非洲,治愈却在欧美 Curable, yet out of reach — sickle cell disease's burden is in Africa, its cure in Europe and America

2026-07-16 · 镰状细胞病, 可及性, Casgevy, 全球健康

2023 年,人类第一次拥有了一种能够治愈镰状细胞病的药。它的原理干净得近乎优雅:取出患者自己的造血干细胞,用 CRISPR 剪断一段调控序列,唤醒沉睡的胎儿血红蛋白,再把细胞送回骨髓。临床数据也漂亮——绝大多数接受治疗的人从此不再发作血管闭塞危象。而在 2021 年,全世界大约有五十万个婴儿带着这个病出生,其中的绝大多数,出生在没有一间授权治疗中心的国家——授权治疗中心不是监管概念,而是厂商自建并逐家认证的网络,而厂商至今只公开美国的名单。这两件事都是真的,而且它们之间的距离,并不是由药价决定的。

病人在哪里

镰状细胞病的地理分布不是偶然。携带一份 HbS 突变的人对疟疾更耐受,于是在疟疾长期流行的地方,这个突变被自然选择留了下来——代价是,当两份突变碰在一起,孩子就会得病。这条演化逻辑把疾病的重心牢牢压在了撒哈拉以南非洲,以及印度的部分地区。

2013 年,Piel 等人在《PLoS Medicine》上用各国 HbS 等位基因频率与人口出生数据建了一个模型,给出了至今仍被反复引用的一组数字:2010 年全球约有 305,800 名镰状细胞贫血新生儿,到 2050 年将升至约 404,200 名;其中 2010 年有 79% 出生在撒哈拉以南非洲,而随着人口结构变化,这个比例到 2050 年会升到 88%。仅尼日利亚、刚果民主共和国和印度三国,就占了 2010 年全球病例出生数的 57%。换句话说,这个病不但集中,而且正在朝着更集中的方向走。

十年后,GBD 2021 用另一套方法给出了更完整的账。2023 年,GBD 2021 Sickle Cell Disease Collaborators 在《The Lancet Haematology》报告:2021 年全球有约 515,000 名婴儿带着镰状细胞病出生,存活的患病人数从 2000 年的 546 万升至 774 万。而这项研究最刺目的一处,是它把两种死亡统计并排放在一起——按 GBD 的常规做法,每例死亡只归因于一个”根本死因”,这样算出的镰状细胞病死亡是 34,400 例;但如果计入所有”带着这个病死去”的人,总数是 376,000 例,几乎是前者的 11 倍。五岁以下儿童中有 81,100 例。以总死亡负担计,镰状细胞病在 GBD 评估的所有死因中排第 12 位;以单一根本死因计,它排第 40 位。

这个 11 倍的落差本身就是一份诊断书:它说明这些孩子并不是死于教科书上那个名字,而是死于肺炎、疟疾、严重贫血、感染——死于镰状细胞病让他们更容易死去的那些东西,然后被记在别的名下。数字之所以长期被低估,是因为统计系统看不见他们。

肯尼亚 Kilifi 的一项前瞻队列研究把这件事落到了具体的人身上。2019 年,Uyoga 等人在《The Lancet Global Health》报告,他们在 15,737 名婴儿中筛出 128 名镰状细胞病患儿,随访至五岁:患病儿童的死亡率是每千人年 58 例,而没有这个病的孩子是 2.4 例——校正后相差 23 倍。同一项研究里还有一个更平淡也更沉重的发现:那些被纳入专科门诊随访的患儿,死亡风险降到了约四分之一。救命的不是什么昂贵的东西,是有人在管。

治愈在哪里

最先松口的不是美国。2023 年 11 月 16 日,英国药品与医疗产品监管局(MHRA)率先批准了 Casgevy(exa-cel)——全球第一个获批的 CRISPR 基因编辑疗法,官方新闻稿的标题里直接用了”世界首例”。三周后的 12 月 8 日,美国 FDA 同日批准了两款针对镰状细胞病的基因治疗:Casgevy 与 Lyfgenia。FDA 自己的措辞是克制的:Casgevy 是”第一个获 FDA 批准的、使用 CRISPR/Cas9 的疗法”——限定在自己的辖区里。

它的临床证据经得起看。2024 年,Frangoul 等人在《New England Journal of Medicine》报告了 exa-cel 的 3 期结果:44 名 12 至 35 岁的重症患者接受治疗,中性粒细胞与血小板在每一位患者身上都成功植入;在随访足够长的 30 人中,29 人(97%)连续 12 个月以上不再发作重度血管闭塞危象,30 人全部不再因危象住院。这是一个非常干净的疗效信号。

但同一篇论文里,还写着这套疗法的另一半事实:在回输之前,患者要接受经药代动力学剂量调整的 busulfan 清髓预处理;安全性特征”总体上与清髓性 busulfan 预处理和自体造血干细胞移植一致”。也就是说,exa-cel 的疗效来自基因编辑,而它的风险与它的重量,主要来自那场移植。

这不是一个可以绕开的附加条件。它是这套疗法的物理形态。

一次”治愈”到底需要什么

把”治愈”拆开,你会看到一条很长的链条,每一环都是高收入国家医疗体系的产物。

首先要把造血干细胞从骨髓里请出来。在别的疾病里,这一步用 G-CSF 动员,便宜而成熟;但在镰状细胞病里不行。2009 年,Fitzhugh 等人在《Cytotherapy》梳理了当时全部 11 例接受 G-CSF 的镰状细胞病患者,其中 7 人发生了严重不良事件——血管闭塞发作、急性胸部综合征、多器官衰竭,以及死亡;各种预防措施都没有稳定的效果,文章的标题直接就是一个提议:是否该暂停使用。于是这条路被换成了 plerixafor。2018 年,Esrick 等人在《Blood Advances》证明,在换血之后单用 plerixafor 动员,可以安全地采到足够基因治疗使用的 CD34⁺ 细胞。请注意这句话里藏了什么:动员之前先要做换血,而换血本身需要血库、需要配型、需要一台单采机和会用它的人。

采集之后,细胞要被送进 GMP 工厂——一间对温度、洁净度、批次可比性、放行检验都有严苛要求的设施,在那里完成编辑、扩增、检定、冻存,再冷链送回。这中间是数周到数月,患者要在等待期里维持病情稳定,通常靠持续输血。

然后是最重的一环:清髓。busulfan 会把患者原有的骨髓清空,给编辑后的细胞腾出位置。这意味着数周的中性粒细胞缺乏期,意味着无菌病房、广谱抗生素、随时可得的血小板输注、处理感染与出血的团队。这一步的风险是真实的,它带来不育、器官毒性,以及远期第二肿瘤的忧虑——这些代价在临床试验里被明确记录,也是患者做决定时真正在权衡的东西。

回输之后还没结束。植入、免疫重建、长期随访——监管机构对基因治疗普遍要求长达十余年的随访,这需要一个能十年不丢失患者的医疗系统。

把这条链条并排看,一个事实就浮出来了:这不是”一支药”,而是”一整套医疗基础设施的一次调用”。而这套基础设施本身在全球的分布,早就有人量过。2015 年,Gratwohl 等人在《The Lancet Haematology》汇总了造血干细胞移植自 1957 年第一例以来的全球活动:到 2012 年前后累计约一百万例,由 75 个国家的 1,516 家移植中心报告完成——全世界 194 个 WHO 成员国里,只有 75 个上得了这张表。文章还给出了一条冷静到近乎残酷的观察:人均国民总收入低于 1,260 美元的国家,一例移植都没有做过。移植率与国家资源、团队数量、供者登记体系的完备程度直接相关。

这张图和镰状细胞病的分布图,几乎是彼此的负片。

降价能解决吗

Casgevy 与 Lyfgenia 的定价在百万美元量级,这个数字理所当然地成了公共讨论的焦点。但如果把价格设想成零,再回头看那条链条,就会发现故事并不会有太大改变。

免费的药,依然需要一间能做换血的血库、一台单采机、一间 GMP 工厂、一间无菌病房、一支能陪一个人扛过三周骨髓空窗期的团队,以及一套能追踪他十五年的随访系统。这些东西不在药的价签上,而在药的前提里。一个国家如果没有做过一例造血干细胞移植,它缺的不是那两百万美元,而是那两百万美元买不到的、需要十年才能长出来的东西。

这一点在 Casgevy 的授权治疗中心(ATC)网络上看得很清楚。ATC 不是监管概念,而是 Vertex 自建、逐家认证的网络:它的官方 ATC 定位器页面第一行就写着”仅限美国居民”——厂商向公众开放的名单里只有美国。2026 年 7 月 1 日,Vertex 宣布美国境内”已激活的 ATC 超过 75 家”。在美国之外,每在一个国家开出第一家 ATC,Vertex 都会专门发稿,比如 2024 年 1 月沙特获批时,新闻稿特意写明”沙特已有一家授权治疗中心激活”;而撒哈拉以南非洲,至今没有出现过这样一则公告。2025 年,Bukini 等人在《BMJ Global Health》里报告坦桑尼亚镰状细胞病中心的经验时,写下的是另一句话:在任何一个非洲国家,都没有正在进行的镰状细胞病基因治疗临床试验。连试验都还没有开始,遑论获批之后的治疗中心。

这也是为什么,在负担最重的地方,当下能救命的措施看起来一点也不前沿。Piel 等人在 2013 年的模型里估算,如果从 2015 年起普遍落实产前诊断、青霉素预防、疫苗接种这些基础手段,到 2050 年可以让约 530 万名镰状细胞贫血新生儿活下来;若再叠加大规模普筛,这个数字可达约 980 万,其中 85% 在撒哈拉以南非洲。而 2019 年,Tshilolo 等人在《New England Journal of Medicine》报告的 REACH 试验证明,羟基脲——一种上世纪的老药、口服、廉价——在撒哈拉以南非洲的儿童中是可行且安全的,并且降低了血管闭塞事件、输血需求,乃至疟疾与其他感染的发生率。

一边是每年五十万个新生患儿和一种一次性治愈手段,另一边是青霉素、疫苗和一片羟基脲。这两件事并不互相竞争,但它们的可及性差了几个数量级,而这差距的来源不在实验室,在物流、诊断、供应链和人。

真正的解法是把疗法本身变轻

如果障碍主要不是价格,而是这套疗法的形态,那么让它变得可及的路径,就只能是改变它的形态。这条路目前有两个方向,而且都不是设想,已经有实验证据在往前推。

第一个方向是拿掉化疗。清髓之所以必要,是因为要给新细胞腾出骨髓龛位;但腾位子未必非得用毒药。2019 年,Czechowicz 等人在《Nature Communications》报告,在小鼠中,用靶向 CD117(干细胞因子受体)的抗体药物偶联物可以选择性清除造血干细胞,在保留免疫系统的前提下完成有效移植。一旦预处理不再是清髓化疗,那间无菌病房、那三周的空窗期,以及背后一整套支持体系,才有可能被压缩。

第二个方向更彻底:不再把细胞取出来。2023 年,Breda 等人在《Science》报告了 CD117/LNP-mRNA——一种靶向造血干细胞表面 CD117 的脂质纳米颗粒,能直接在体内把 mRNA 送进造血干细胞;这项工作做在小鼠与体外的人源镰状细胞上。这篇论文在开头就写明了动机:现有方案副作用大、可及性有限。这个方向如果走通,动员、单采、GMP 制备、冻存冷链、回输——那条链条的绝大部分会一起消失,剩下的是一针。

需要说明的是,这两个方向目前都还停在临床前:证据来自小鼠模型和体外的人源细胞,而不是人。它们证明的是可行性,不是疗效——这和前面那些来自 3 期试验与前瞻队列的数字,不是同一个量级的证据,读的时候得分开来读。

这不是一条被产业忽略的路。早在 2019 年 10 月,NIH 与 Bill & Melinda Gates Foundation 就宣布共同投入至少 2 亿美元,目标写得很具体:在七到十年内,把安全、有效、持久的基因治疗推进到美国以及撒哈拉以南非洲相关国家的临床试验中。把”在非洲做试验”写进目标本身,是承认了一件事——如果一种疗法只能在有一百家移植中心的国家里做,那它从设计之初就已经把大多数患者排除在外了。

那两张地图

镰状细胞病是人类最早在分子层面被读懂的疾病之一,如今又成了第一个被 CRISPR 治愈的疾病。它在科学上被理解得如此透彻,以至于今天限制它的,几乎已经不再是生物学问题。

一张地图画着这个病在哪里——它压在撒哈拉以南非洲,而且比例还在上升。另一张地图画着造血干细胞移植能在哪里做——它止步于人均收入 1,260 美元这条线之外。当前这一代基因治疗,精确地生长在第二张地图上。

把这两张地图叠到一起的办法,不太可能是让第二张地图长到第一张那么大——那需要几代人。更现实的,是让疗法本身缩小到不再需要第二张地图:免清髓的预处理,体内递送的载荷,最终是一次注射。到那一天,“能治”和”治得到”之间的那段距离,才算真正被走完。

参考文献

  1. Piel FB, Hay SI, Gupta S, Weatherall DJ, Williams TN. Global burden of sickle cell anaemia in children under five, 2010-2050: modelling based on demographics, excess mortality, and interventions. PLoS Med. 2013;10(7):e1001484. DOI
  2. GBD 2021 Sickle Cell Disease Collaborators (Thomson AM, et al). Global, regional, and national prevalence and mortality burden of sickle cell disease, 2000-2021: a systematic analysis from the Global Burden of Disease Study 2021. Lancet Haematol. 2023;10(8):e585-e599. DOI
  3. Uyoga S, et al. The epidemiology of sickle cell disease in children recruited in infancy in Kilifi, Kenya: a prospective cohort study. Lancet Glob Health. 2019;7(10):e1458-e1466. DOI
  4. Frangoul H, et al. Exagamglogene Autotemcel for Severe Sickle Cell Disease. N Engl J Med. 2024;390(18):1649-1662. DOI
  5. Fitzhugh CD, Hsieh MM, Bolan CD, Saenz C, Tisdale JF. Granulocyte colony-stimulating factor (G-CSF) administration in individuals with sickle cell disease: time for a moratorium? Cytotherapy. 2009;11(4):464-471. DOI
  6. Esrick EB, et al. Successful hematopoietic stem cell mobilization and apheresis collection using plerixafor alone in sickle cell patients. Blood Adv. 2018;2(19):2505-2512. DOI
  7. Gratwohl A, et al. One million haemopoietic stem-cell transplants: a retrospective observational study. Lancet Haematol. 2015;2(3):e91-e100. DOI
  8. Tshilolo L, et al. Hydroxyurea for Children with Sickle Cell Anemia in Sub-Saharan Africa. N Engl J Med. 2019;380(2):121-131. DOI
  9. Czechowicz A, et al. Selective hematopoietic stem cell ablation using CD117-antibody-drug-conjugates enables safe and effective transplantation with immunity preservation. Nat Commun. 2019;10(1):617. DOI
  10. Breda L, et al. In vivo hematopoietic stem cell modification by mRNA delivery. Science. 2023;381(6656):436-443. DOI
  11. U.S. Food and Drug Administration. FDA Approves First Gene Therapies to Treat Patients with Sickle Cell Disease. Press announcement, December 8, 2023. 链接
  12. National Institutes of Health. NIH launches new collaboration to develop gene-based cures for sickle cell disease and HIV on global scale. News release, October 23, 2019. 链接
  13. Medicines and Healthcare products Regulatory Agency (MHRA). MHRA authorises world-first gene therapy that aims to cure sickle-cell disease and transfusion-dependent β-thalassemia. Press release, 16 November 2023. 链接
  14. Vertex Pharmaceuticals. Find a CASGEVY® Authorized Treatment Center. (页面首行标注 “For US residents only.”;厂商仅公开美国 ATC 名单) 链接
  15. Vertex Pharmaceuticals. Vertex Announces US FDA Approval for Expanded Use of CASGEVY® for the Treatment of People Ages 2 Years and Older With Sickle Cell Disease or Transfusion-Dependent Beta Thalassemia. Press release, July 1, 2026.(原文:“Today, there are more than 75 activated ATCs in the U.S.”) 链接
  16. Vertex Pharmaceuticals. Vertex Announces Approval of First CRISPR/Cas9 Gene-Edited Therapy, CASGEVY™, for the Treatment of Sickle Cell Disease (SCD) and Transfusion-Dependent Beta Thalassemia (TDT) in Kingdom of Saudi Arabia. Press release, January 9, 2024.(原文:“One Authorized Treatment Center (ATC) in Saudi Arabia is already activated.”) 链接
  17. Bukini D, et al. Strengthening advanced therapy for sickle cell disease in Africa: experience from sickle cell disease centre in Dar es Salaam, Tanzania. BMJ Glob Health. 2025;10(1):e017878.(原文:“There are no ongoing clinical trials related to gene therapy for SCD, in any African country.”) DOI

In 2023, humanity acquired for the first time a drug capable of curing sickle cell disease. Its principle is clean to the point of elegance: take out the patient’s own hematopoietic stem cells, use CRISPR to cut a regulatory sequence, wake the sleeping fetal hemoglobin, and put the cells back into the marrow. The clinical data are handsome too—the overwhelming majority of those treated never suffer a vaso-occlusive crisis again. And in 2021, roughly five hundred thousand babies worldwide were born carrying this disease, the vast majority of them in countries without a single authorized treatment center—an authorized treatment center being not a regulatory concept but a network the manufacturer builds itself and certifies one center at a time, and to this day the manufacturer publishes only the American list. Both of these things are true, and the distance between them is not set by the price of the drug.

Where the patients are

The geographic distribution of sickle cell disease is no accident. People carrying a single copy of the HbS mutation are more resistant to malaria, and so in places where malaria has long been endemic, natural selection kept the mutation—at the cost that when two copies meet, the child falls ill. This evolutionary logic pressed the center of gravity of the disease firmly onto sub-Saharan Africa, and parts of India.

In 2013, Piel and colleagues published in PLoS Medicine a model built from national HbS allele frequencies and population birth data, yielding a set of numbers still cited again and again: in 2010 there were about 305,800 newborns with sickle cell anaemia worldwide, rising to about 404,200 by 2050; of these, 79% were born in sub-Saharan Africa in 2010, and with shifting demographics that share will rise to 88% by 2050. Nigeria, the Democratic Republic of the Congo, and India alone accounted for 57% of the world’s affected births in 2010. In other words, the disease is not only concentrated—it is moving toward being more concentrated still.

A decade later, GBD 2021 produced a fuller accounting by another method. In 2023, the GBD 2021 Sickle Cell Disease Collaborators reported in The Lancet Haematology: in 2021, about 515,000 babies worldwide were born with sickle cell disease, and the number of people living with it rose from 5.46 million in 2000 to 7.74 million. The most piercing part of that study is that it set two mortality counts side by side—under GBD’s usual practice, each death is attributed to a single “underlying cause,” and counted that way sickle cell disease deaths came to 34,400; but if you count everyone who died with the disease, the total is 376,000, nearly eleven times the former. Among children under five there were 81,100. By total mortality burden, sickle cell disease ranked 12th among all causes of death GBD assessed; by single underlying cause, it ranked 40th.

That elevenfold gap is itself a diagnosis: it shows these children are not dying under the name in the textbook, but of pneumonia, malaria, severe anaemia, infection—of the things sickle cell disease makes it easier for them to die of, and then being recorded under another name. The numbers have long been underestimated because the statistical systems cannot see them.

A prospective cohort study in Kilifi, Kenya, brought this down to specific people. In 2019, Uyoga and colleagues reported in The Lancet Global Health that among 15,737 infants they identified 128 children with sickle cell disease and followed them to age five: mortality among affected children was 58 per 1,000 person-years, against 2.4 among children without the disease—a 23-fold difference after adjustment. The same study held a plainer and heavier finding: for those children enrolled in specialist clinic follow-up, the risk of death dropped to about a quarter. What saved lives was nothing expensive. It was somebody keeping track.

Where the cure is

The first to give way was not the United States. On 16 November 2023, the UK’s Medicines and Healthcare products Regulatory Agency (MHRA) was the first to authorize Casgevy (exa-cel)—the world’s first approved CRISPR gene-editing therapy; the headline of the official press release used the phrase “world-first” outright. Three weeks later, on 8 December, the US FDA approved on the same day two gene therapies for sickle cell disease: Casgevy and Lyfgenia. The FDA’s own wording was restrained: Casgevy is “the first FDA-approved therapy utilizing CRISPR/Cas9”—bounded within its own jurisdiction.

Its clinical evidence holds up under inspection. In 2024, Frangoul and colleagues reported the phase 3 results of exa-cel in the New England Journal of Medicine: 44 severely affected patients aged 12 to 35 were treated, and neutrophil and platelet engraftment succeeded in every single one; among the 30 with sufficient follow-up, 29 (97%) went 12 consecutive months or more without a severe vaso-occlusive crisis, and all 30 were free of hospitalization for crises. That is a very clean efficacy signal.

But the same paper also states the other half of the facts about this therapy: before reinfusion, patients undergo myeloablative conditioning with pharmacokinetically dose-adjusted busulfan; the safety profile was “generally consistent with myeloablative busulfan conditioning and autologous hematopoietic stem-cell transplantation.” Which is to say: exa-cel’s efficacy comes from the gene editing, while its risk and its weight come mainly from that transplant.

This is not an add-on condition that can be sidestepped. It is the physical form of the therapy.

What one “cure” actually requires

Take “cure” apart and you see a very long chain, every link of which is a product of a high-income country’s health system.

First the hematopoietic stem cells must be coaxed out of the marrow. In other diseases this step uses G-CSF mobilization—cheap and well established; but not in sickle cell disease. In 2009, Fitzhugh and colleagues reviewed in Cytotherapy all 11 sickle cell disease patients who had then received G-CSF, of whom 7 suffered serious adverse events—vaso-occlusive episodes, acute chest syndrome, multi-organ failure, and death; no preventive measure worked reliably, and the paper’s title was itself a proposal: time for a moratorium? So this route was replaced by plerixafor. In 2018, Esrick and colleagues demonstrated in Blood Advances that mobilization with plerixafor alone, following exchange transfusion, can safely collect enough CD34⁺ cells for gene therapy. Note what is buried in that sentence: before mobilization comes exchange transfusion, and exchange transfusion itself requires a blood bank, requires matching, requires an apheresis machine and someone who knows how to run it.

After collection, the cells must be shipped to a GMP facility—a plant with exacting requirements for temperature, cleanliness, batch comparability, and release testing—where they are edited, expanded, tested, cryopreserved, and shipped back on a cold chain. That takes weeks to months, during which the patient must be kept clinically stable, usually with continued transfusion.

Then comes the heaviest link: myeloablation. Busulfan empties out the patient’s existing marrow to make room for the edited cells. That means weeks of neutropenia; it means a sterile ward, broad-spectrum antibiotics, platelet transfusions available at any hour, a team to handle infection and bleeding. The risk of this step is real: it brings infertility, organ toxicity, and the long-term worry of secondary malignancy—costs explicitly recorded in the clinical trials, and the things patients are genuinely weighing when they decide.

Nor does it end at reinfusion. Engraftment, immune reconstitution, long-term follow-up—regulators generally require more than a decade of follow-up for gene therapies, and that takes a health system that will not lose the patient in ten years.

Lay this chain out side by side and a fact surfaces: this is not “a drug,” it is “a single invocation of an entire medical infrastructure.” And the global distribution of that infrastructure has long since been measured. In 2015, Gratwohl and colleagues pooled in The Lancet Haematology the worldwide activity of hematopoietic stem cell transplantation since the first case in 1957: about one million cumulative cases by around 2012, reported by 1,516 transplant centers in 75 countries—of the world’s 194 WHO member states, only 75 made it onto that table. The paper also offers an observation so level-headed it is nearly cruel: countries with a gross national income per capita below US$1,260 had never performed a single transplant. Transplant rates correlate directly with national resources, team numbers, and how complete the donor registry system is.

That map and the map of sickle cell disease are almost each other’s negatives.

Can a price cut solve it?

Casgevy and Lyfgenia are priced on the order of a million dollars, and that figure has, naturally enough, become the focus of public discussion. But set the price to zero and look back at that chain, and you find the story does not change much.

A free drug still needs a blood bank that can do exchange transfusion, an apheresis machine, a GMP facility, a sterile ward, a team that can carry a person through three weeks with an empty marrow, and a follow-up system that can track him for fifteen years. None of this is on the drug’s price tag; it is in the drug’s preconditions. If a country has never performed a single hematopoietic stem cell transplant, what it lacks is not the two million dollars—it is the thing two million dollars cannot buy, the thing that takes ten years to grow.

This is plain to see in Casgevy’s network of authorized treatment centers (ATCs). An ATC is not a regulatory concept but a network Vertex builds itself and certifies one center at a time: the first line of its official ATC locator page reads “For US residents only”—the only list the manufacturer opens to the public is the American one. On 1 July 2026, Vertex announced that within the United States there were “more than 75 activated ATCs.” Outside the United States, every time a first ATC opens in a country, Vertex issues a dedicated release—when Saudi Arabia’s approval came in January 2024, for instance, the press release noted specifically that “one Authorized Treatment Center in Saudi Arabia is already activated”—and for sub-Saharan Africa, no such announcement has ever appeared. In 2025, when Bukini and colleagues reported in BMJ Global Health the experience of a sickle cell disease center in Tanzania, what they wrote was a different sentence: there are no ongoing clinical trials of gene therapy for sickle cell disease in any African country. The trials have not even begun, let alone the treatment centers that would follow approval.

This is also why, in the places where the burden is heaviest, the measures that can save lives right now look not at all cutting-edge. In their 2013 model, Piel and colleagues estimated that if prenatal diagnosis, penicillin prophylaxis, and vaccination—these basic tools—were universally implemented from 2015, about 5.3 million newborns with sickle cell anaemia could be kept alive by 2050; layer large-scale universal screening on top and the figure reaches about 9.8 million, 85% of them in sub-Saharan Africa. And in 2019, the REACH trial reported by Tshilolo and colleagues in the New England Journal of Medicine demonstrated that hydroxyurea—an old drug from the last century, oral, cheap—is feasible and safe in children in sub-Saharan Africa, and reduced vaso-occlusive events, transfusion needs, and even the incidence of malaria and other infections.

On one side, half a million new patients born each year and a one-shot cure; on the other, penicillin, vaccines, and a hydroxyurea tablet. The two do not compete with each other, but their accessibility differs by orders of magnitude, and the source of that gap is not in the laboratory. It is in logistics, diagnostics, supply chains, and people.

The real fix is to make the therapy itself lighter

If the obstacle is mainly not the price but the form of this therapy, then the only path to making it accessible is to change its form. There are currently two directions here, and neither is a thought experiment—experimental evidence is already pushing both forward.

The first direction is to take away the chemotherapy. Myeloablation is necessary because room must be made in the marrow niche for the new cells; but making room need not require poison. In 2019, Czechowicz and colleagues reported in Nature Communications that in mice, an antibody-drug conjugate targeting CD117 (the stem cell factor receptor) can selectively deplete hematopoietic stem cells, achieving effective transplantation while preserving immunity. Once conditioning is no longer myeloablative chemotherapy, that sterile ward, those three weeks of aplasia, and the whole support system behind them can begin to be compressed.

The second direction is more thoroughgoing: stop taking the cells out at all. In 2023, Breda and colleagues reported in Science on CD117/LNP-mRNA—a lipid nanoparticle targeting CD117 on the surface of hematopoietic stem cells, able to deliver mRNA directly into hematopoietic stem cells in the body; this work was done in mice and in human sickle cells in vitro. The paper states its motivation right at the opening: existing approaches have substantial side effects and limited accessibility. If this direction works out, mobilization, apheresis, GMP manufacturing, cryopreservation and cold chain, reinfusion—the great majority of that chain disappears together, and what remains is one injection.

It must be said that both directions still stop at the preclinical stage: the evidence comes from mouse models and human cells in vitro, not from people. What they demonstrate is feasibility, not efficacy—this is not the same order of evidence as the numbers above from phase 3 trials and prospective cohorts, and the two must be read apart.

This is not a road the industry has ignored. As early as October 2019, the NIH and the Bill & Melinda Gates Foundation announced a joint commitment of at least $200 million, with the goal written out concretely: to advance safe, effective, durable gene therapies into clinical trials within seven to ten years, in the United States and in relevant countries in sub-Saharan Africa. Writing “trials in Africa” into the goal is itself an admission of something—that if a therapy can only be done in countries with a hundred transplant centers, then from the moment it was designed it had already excluded most of its patients.

Those two maps

Sickle cell disease is one of the first diseases humanity ever read at the molecular level, and it has now become the first disease cured by CRISPR. It is understood so thoroughly in scientific terms that what limits it today is almost no longer a question of biology.

One map shows where the disease is—pressed onto sub-Saharan Africa, with the share still rising. The other map shows where hematopoietic stem cell transplantation can be done—stopping short at that line of US$1,260 per capita income. This generation of gene therapy grows precisely on the second map.

The way to lay these two maps over each other is unlikely to be growing the second map until it is as large as the first—that would take generations. The more realistic route is to shrink the therapy itself until it no longer needs the second map: conditioning without myeloablation, a payload delivered in vivo, and finally a single injection. Only on that day will the distance between “curable” and “reached” have truly been walked.

References

  1. Piel FB, Hay SI, Gupta S, Weatherall DJ, Williams TN. Global burden of sickle cell anaemia in children under five, 2010-2050: modelling based on demographics, excess mortality, and interventions. PLoS Med. 2013;10(7):e1001484. DOI
  2. GBD 2021 Sickle Cell Disease Collaborators (Thomson AM, et al). Global, regional, and national prevalence and mortality burden of sickle cell disease, 2000-2021: a systematic analysis from the Global Burden of Disease Study 2021. Lancet Haematol. 2023;10(8):e585-e599. DOI
  3. Uyoga S, et al. The epidemiology of sickle cell disease in children recruited in infancy in Kilifi, Kenya: a prospective cohort study. Lancet Glob Health. 2019;7(10):e1458-e1466. DOI
  4. Frangoul H, et al. Exagamglogene Autotemcel for Severe Sickle Cell Disease. N Engl J Med. 2024;390(18):1649-1662. DOI
  5. Fitzhugh CD, Hsieh MM, Bolan CD, Saenz C, Tisdale JF. Granulocyte colony-stimulating factor (G-CSF) administration in individuals with sickle cell disease: time for a moratorium? Cytotherapy. 2009;11(4):464-471. DOI
  6. Esrick EB, et al. Successful hematopoietic stem cell mobilization and apheresis collection using plerixafor alone in sickle cell patients. Blood Adv. 2018;2(19):2505-2512. DOI
  7. Gratwohl A, et al. One million haemopoietic stem-cell transplants: a retrospective observational study. Lancet Haematol. 2015;2(3):e91-e100. DOI
  8. Tshilolo L, et al. Hydroxyurea for Children with Sickle Cell Anemia in Sub-Saharan Africa. N Engl J Med. 2019;380(2):121-131. DOI
  9. Czechowicz A, et al. Selective hematopoietic stem cell ablation using CD117-antibody-drug-conjugates enables safe and effective transplantation with immunity preservation. Nat Commun. 2019;10(1):617. DOI
  10. Breda L, et al. In vivo hematopoietic stem cell modification by mRNA delivery. Science. 2023;381(6656):436-443. DOI
  11. U.S. Food and Drug Administration. FDA Approves First Gene Therapies to Treat Patients with Sickle Cell Disease. Press announcement, December 8, 2023. Link
  12. National Institutes of Health. NIH launches new collaboration to develop gene-based cures for sickle cell disease and HIV on global scale. News release, October 23, 2019. Link
  13. Medicines and Healthcare products Regulatory Agency (MHRA). MHRA authorises world-first gene therapy that aims to cure sickle-cell disease and transfusion-dependent β-thalassemia. Press release, 16 November 2023. Link
  14. Vertex Pharmaceuticals. Find a CASGEVY® Authorized Treatment Center. (The first line of the page states “For US residents only.”; the manufacturer publishes only the US ATC list.) Link
  15. Vertex Pharmaceuticals. Vertex Announces US FDA Approval for Expanded Use of CASGEVY® for the Treatment of People Ages 2 Years and Older With Sickle Cell Disease or Transfusion-Dependent Beta Thalassemia. Press release, July 1, 2026. (Original: “Today, there are more than 75 activated ATCs in the U.S.”) Link
  16. Vertex Pharmaceuticals. Vertex Announces Approval of First CRISPR/Cas9 Gene-Edited Therapy, CASGEVY™, for the Treatment of Sickle Cell Disease (SCD) and Transfusion-Dependent Beta Thalassemia (TDT) in Kingdom of Saudi Arabia. Press release, January 9, 2024. (Original: “One Authorized Treatment Center (ATC) in Saudi Arabia is already activated.”) Link
  17. Bukini D, et al. Strengthening advanced therapy for sickle cell disease in Africa: experience from sickle cell disease centre in Dar es Salaam, Tanzania. BMJ Glob Health. 2025;10(1):e017878. (Original: “There are no ongoing clinical trials related to gene therapy for SCD, in any African country.”) DOI