Last updated 24 September 2026. Reading time about 15 minutes.
Every January, the editors at MIT Technology Review spend months arguing about which technologies deserve a place on their annual list. The 2026 edition is the 25th, and it runs from batteries to nuclear reactors to AI data centres. Biology gets a generous share of it. I wanted to look at that biological side properly, and to look beyond a single country while doing it.
This guide covers ten biology-related technologies that, as of late September 2026, are producing real results, real regulation, or real arguments. Some are already treating patients. Some are a handful of people into their first trials. One has never been released outside a sealed laboratory. For each, I have tried to say what the evidence shows, what it does not, and which region is driving the story.
“Chance favours only the prepared mind.”
Louis Pasteur, lecture at the University of Lille, 1854
In this article
The ten at a glance
Before the detail, here is a quick map of where each technology stands today. The placement is my own editorial judgement, based on the sources listed at the end, and it measures how far each one has travelled rather than how important it is.
| Technology | Lab proof of concept | Early human trials | Late-stage or pivotal trials | Regulation adopted | Commercially available |
|---|---|---|---|---|---|
| 1. Personalised gene editing Single-patient use; FDA framework in draft | |||||
| 2. AI-designed genomes Bacteriophages only | |||||
| 3. AI-discovered drugs Phase III started July 2026 | |||||
| 4. mRNA cancer vaccines Phase 3 topline positive; not yet approved | |||||
| 5. In vivo CAR-T Five-patient lupus report | |||||
| 6. Pig organ transplants Registered trials under way | |||||
| 7. Organoids and chips FDA roadmap and 2026 rulemaking | |||||
| 8. Gene drive mosquitoes No wild release to date | |||||
| 9. Genome-edited crops EU law adopted; applies 2028 | |||||
| 10. Embryo screening Sold in the US; professional bodies object |
Figure 1. Development stage of each technology, September 2026 (editorial classification).
1. Personalised gene editing
Regional focus: United States and India
In February 2025, an infant in Philadelphia known publicly as KJ Muldoon received a gene-editing treatment built for him and, as far as anyone knows, nobody else. He had a severe form of CPS1 deficiency, a rare liver disorder in which the body cannot clear the ammonia produced when protein is digested. Teams at the Children’s Hospital of Philadelphia and Penn Medicine designed the therapy, and the New England Journal of Medicine reported the case that May. By early March 2026, he was walking.
What matters for everyone else is the regulatory follow-through. In November 2025, FDA leaders Marty Makary and Vinay Prasad described a “plausible mechanism pathway” in the NEJM, and on 23 February 2026 the agency published draft guidance. For now it covers genome editing and RNA-based therapies that target the root cause of a rare disease. CHOP and Penn researchers have already used the framework to plan a customisable prime-editing platform for infantile-onset urea cycle disorders, reported in the American Journal of Human Genetics in March 2026.
The caution is fair. Kiran Musunuru, who helped care for KJ, described the FDA’s first announcement as “light on specifics”, and STAT reported ethicists warning that the pathway could become a Pandora’s box. Cost is the other worry.
“My biggest ethical concern is actually access and inequality.”
Jennifer Doudna, Nobel laureate in Chemistry, speaking to Forbes, February 2026
India is trying to answer that concern directly. BIRSA 101, a CRISPR therapy for sickle cell disease developed at CSIR-IGIB and licensed to the Serum Institute of India, was launched in November 2025 as part of the National Sickle Cell Anaemia Elimination Mission, which aims to end the disease by 2047. The science minister described the approach as “precise genetic surgery”. Overseas CRISPR treatments for the disease reportedly cost INR 20 to 25 crore per patient. The indigenous platform is expected to cost around INR 50 lakh, with clinical trials due to begin in 2026 at centres including AIIMS Delhi.
Figure 2. Sickle cell gene therapy: reported versus expected cost
Data: Observer Research Foundation (November 2025). Prices are reported estimates and an expected target, not final market prices. 1 crore = 100 lakh.
2. AI-designed genomes
Regional focus: United States
Richard Feynman left a line on his Caltech blackboard when he died in 1988: “What I cannot create, I do not understand.” Synthetic biologists quote it constantly. In August 2026 it acquired a strange new twist.
On 6 August, Science published work from Brian Hie’s group at Stanford and the Arc Institute in which two genome language models, Evo 1 and Evo 2, wrote complete bacteriophage genomes from scratch. Phages infect bacteria (here, E. coli), not people. Of roughly 300 AI-designed genomes that were synthesised and tested, 16 produced working viruses. Some killed E. coli strains that had resisted natural phages, and one used a DNA-packaging protein unlike those of its natural relatives.
Figure 3. Evo-designed phage genomes: how many worked
Data: Science (6 August 2026) as reported by eWeek and C&EN. Test count is approximate.
A hit rate near five percent sounds low until you remember that a random string of DNA would produce nothing at all. Sceptics still have a point. A phage is about as simple as a genome gets, no phage therapy has yet been approved by a major regulator, and biosecurity researchers quoted by BioTechniques want firmer guardrails around this kind of design. Hie’s own argument, as reported by Stanford, is that the same tools strengthen defences against both natural pandemics and engineered threats.
3. AI-discovered drugs
Regional focus: China
For years, “AI drug discovery” mostly meant a slide in an investor deck. Rentosertib, an oral TNIK inhibitor for idiopathic pulmonary fibrosis (IPF), gives the field a track record. Insilico Medicine says the compound, produced on its Pharma.AI platform, moved from target discovery through Phase 0 and Phase 1 trials in under 30 months. In a Phase 2a trial of 71 patients at 22 sites in China, published in Nature Medicine in 2025, people on 60 mg daily gained an average of 98.4 mL of forced vital capacity over 12 weeks, while the placebo group lost 20.3 mL. On 7 July 2026 the company announced the start of Phase III.
Keep the caveats on the same page as the excitement. The Phase 2a trial lasted 12 weeks and its primary goal was safety. A field survey compiled by IntuitionLabs found a median of 6.5 years from company founding to a first Phase 1 trial across 63 AI-native firms, and plenty of programmes have been dropped along the way. AI has sped up parts of the pipeline. It has not abolished failure.
4. Personalised mRNA cancer vaccines
Regional focus: Australia and the United States
Personalised cancer vaccines spent two decades as a promising idea that rarely survived a large trial. On 19 August 2026, Merck and Moderna announced that the Phase 3 INTerpath-001 trial had met its primary goal. It enrolled 1,137 people whose stage IIB to IV melanoma had been completely removed by surgery. Adding the individualised mRNA vaccine intismeran autogene to pembrolizumab (Keytruda) lengthened recurrence-free survival compared with pembrolizumab alone, and it improved distant metastasis-free survival too. Reports describe it as the first positive Phase 3 result for an mRNA cancer therapy and for an individualised neoantigen therapy.
The mechanism is worth a moment. Each dose is made for one person. The tumour is sequenced, and the manufacturer builds an mRNA that encodes up to 34 of its unique mutations. The immune system learns to recognise cells carrying those markers, while pembrolizumab keeps T cells switched on.
“Today’s results represent a landmark moment for adjuvant melanoma treatment.”
Georgina Long, principal investigator, Melanoma Institute Australia and University of Sydney
Figure 4. Earlier Phase 2b result that the Phase 3 trial set out to confirm
Data: five-year KEYNOTE-942 follow-up presented at ASCO 2026, as reported by Dermatology Times and C&EN. The Phase 3 effect size has not been published.
What is missing is the size of the Phase 3 benefit. The companies have released only topline statements and plan to present full data at a medical meeting, so approval, large-scale manufacturing and cost are all still open questions.
From my desk
[A second personal note fits well here. If someone you know has faced cancer, or if you have ever waited for a test result, say so plainly and briefly. One honest, specific detail beats a paragraph of general feeling. Example to rewrite: “I have never seen a trial headline land the way this one did in my family group chat. Everyone had a story about someone who had been through melanoma.”]
5. In vivo CAR-T cell therapy
Regional focus: China
Conventional CAR-T therapy means removing a patient’s T cells, engineering them in a specialised facility, and infusing them back after chemotherapy. It works, but it is slow and expensive. The in vivo approach skips the factory: lipid nanoparticles carry the instructions as mRNA and reprogram T cells inside the body.
In September 2025, the New England Journal of Medicine published the first human data for this idea in autoimmune disease. Researchers in Hefei, working with the Shenzhen company MagicRNA, treated five people with long-standing, treatment-resistant lupus. The product, HN2301, was given intravenously without prior lymphodepletion. The company reported that up to 60 percent of circulating CD8 T cells were reprogrammed, circulating B cells were fully depleted, and disease-activity scores fell by as much as 20 points within three months, with no neurotoxicity or other severe adverse events.
Five patients and three months of follow-up is a signal, not a verdict. RheumNow described the clinical results as more mixed than the headline, and a 2026 review notes that durability and best dosing remain to be established. A trial in refractory Graves’ disease (NCT07333677) is recruiting at Shanghai Zhongshan Hospital, so more evidence should arrive soon.
6. Gene-edited pig organs
Regional focus: United States
About 90,000 people in the United States are waiting for a kidney, and roughly 11 on the list die each day, according to United Network for Organ Sharing figures cited in 2025. That gap explains why gene-edited pig kidneys moved from dramatic one-off operations to registered clinical trials.
Two designs are in play. United Therapeutics’ UKidney carries 10 edits: six human genes added and four pig genes removed. The eGenesis kidney, EGEN-2784, carries many more, including inactivation of the pig’s embedded retroviruses. FDA cleared the first application in February 2025 and the second in September 2025. The first transplant in United Therapeutics’ EXPAND trial took place at NYU Langone on 3 November 2025, and the study enrols dialysis patients aged 55 to 70 who are unlikely to receive a human kidney soon.
Figure 5. Days a pig kidney functioned in living recipients
Data: CNN (September 2025) and Artificial Organs review (2026). Both kidneys were later removed; these are two cases, not a trial average.
“We are entering a transformative era in organ transplantation.”
Mike Curtis, chief executive, eGenesis
The unknowns are large: long-term rejection, the risk of animal-borne infection, cost, and how many organs herds can supply. Curtis’s optimism is understandable, but the honest reading of the data is that we have proof of function for months, not yet for decades.
7. Organoids and organ-on-a-chip
Regional focus: United States, with global uptake
The assumption that animal studies must come before human ones is being rewritten. On 10 April 2025, FDA announced a plan to reduce, refine or replace animal testing for monoclonal antibodies and other drugs, using what it calls New Approach Methodologies (NAMs): lab-grown organoids, organ-on-a-chip devices and AI-based toxicity models. NIH followed on 29 April 2025 by opening its Office of Research Innovation, Validation, and Application (ORIVA).
The pace has continued into this year. FDA published a one-year progress report on 20 April 2026, and its NAMs page now lists 2026 draft guidance on how to validate and use these methods, along with streamlined nonclinical safety studies for cancer biologics. This month the agency issued a direct final rule updating its terminology so that regulations recognise cell-based assays, organs-on-chips and computer models alongside animal testing. FDA stresses that the rule does not require any particular method or change its evidentiary standards.
The motivation is simple. Roughly nine in ten drugs that pass preclinical evaluation fail in human trials, a figure repeated across recent reviews, partly because animal biology predicts human responses poorly. Organ chips add fluid flow, mechanical forces and immune cells that static dishes lack. It is early, though. Validation is the bottleneck, and some developers argue that whole-organ testing still deserves a place in the mix.
8. Gene drive mosquitoes
Regional focus: Africa (Tanzania, Burkina Faso, Uganda)
“In Nature nothing exists alone.”
Rachel Carson, Silent Spring, 1962
Malaria killed roughly 600,000 people in Africa in 2024, according to a 2026 paper on post-release safety monitoring, and progress toward elimination has stalled. Gene drives push a chosen trait through a mosquito population faster than ordinary inheritance allows. They are among the boldest tools proposed, and the only technology on this list that has never left the laboratory.
The Tanzania-led Transmission Zero project reported in December 2025 a gene-drive strain that stops malaria parasites developing inside the mosquito, rather than killing the insect. It borrows two small proteins, one first found in the African clawed frog and one in honeybees. The work is done in a purpose-built Tanzanian facility, with an island trial in Lake Victoria planned for a later phase. Tanzanian scientists Alphaxard Manjurano and Dickson Lwetoijera wrote that gene-drive mosquitoes offer “unprecedented hope”.
Then came Burkina Faso. On 11 August 2025, the Target Malaria team released about 16,000 male-biased mosquitoes, which were not gene drives, after approvals from national biosafety, environmental and ethics bodies. A week later, authorities ordered all activity suspended, and on 22 August the ministry ended the project, citing biosafety risks, limited impact and scientific sovereignty. Nature Africa reported that laboratories were sealed and samples destroyed. Target Malaria says no gene-drive mosquito has ever been released into the wild.
Whatever your view of the technology, the episode shows that paper approvals are not the same as public consent. Any future field trial in Africa will be judged as much by trust as by data.
9. Genome-edited crops
Regional focus: Europe
Europe spent two decades arguing about gene-edited plants. A 2018 ruling by the EU Court of Justice put them under the strict GMO rules, and the Commission proposed easing that in 2023. On 17 June 2026, Parliament and Council adopted Regulation (EU) 2026/1388. It was published on 26 June, is in force from mid-July 2026, and applies from 17 July 2028 after a two-year implementation period for secondary rules.
The regulation sets up two tiers. Category 1 plants, whose changes could occur naturally or through conventional breeding, are treated like conventional varieties: they are not classed as GMOs and carry no GMO label, although seed and other reproductive material is labelled. Category 2 plants stay under the GMO framework with authorisation, traceability and labelling. All NGT plants remain banned from organic production, and the text adds transparency measures on patents. The US Department of Agriculture’s Foreign Agricultural Service says it brings the EU closer to other major agricultural markets.
Separately, the Commission’s Biotech Act proposal of 16 December 2025, aimed mainly at health biotechnology, is still moving through Parliament and Council. Law firm Morgan Lewis expected adoption after the end of 2026.
10. Polygenic embryo screening
Regional focus: United States, with global debate
Polygenic embryo screening sequences embryos created through IVF and ranks them by statistical risk for conditions such as heart disease and diabetes. MIT Technology Review’s 2026 issue puts “designer babies” among its headline technologies, and US companies including Orchid, Nucleus Genomics and Herasight sell versions of the service. According to its own materials, Herasight also offers scores for traits outside disease, including cognitive ability.
The evidence has not caught up with the marketing. In December 2025, the ethics and practice committees of the American Society for Reproductive Medicine concluded that the technology is not ready for clinical use and should not currently be offered as a reproductive service. A peer-reviewed interview study of US clinicians and patients found that no regulation dictates what can be screened in embryos or for what purpose, though laboratories are regulated for quality. Clinicians largely wanted some rules; patients mostly opposed restrictions. A 2026 Frontiers analysis comparing ten countries expects use to grow regardless.
I do not think this one has a clean answer. Screening out a devastating single-gene disease is one thing. Ranking embryos by a probabilistic score for a complex trait is another, and the science is weaker exactly where the marketing is loudest.
Regional snapshot: five continents
The same technologies look different depending on where you stand. Here is what stood out on each continent, with the policy anchor in the last column.
| Region | What stood out | Policy anchor |
|---|---|---|
| Asia | China produced early first-in-human data for in vivo CAR-T in lupus and hosted the Phase 2a trial of an AI-discovered fibrosis drug, now in Phase III. India launched BIRSA 101 for sickle cell disease. | India’s BioE3 policy (August 2024) and the National Sickle Cell Anaemia Elimination Mission (to 2047) |
| North America | Personalised CRISPR, pig kidney trials, AI-designed phages and embryo screening all originate here. | FDA plausible mechanism draft guidance (February 2026); FDA and NIH animal-testing reforms |
| Australia | Georgina Long of Melanoma Institute Australia is principal investigator of the INTerpath-001 melanoma vaccine trial. | Gene Technology Act 2000 and the Office of the Gene Technology Regulator; the TGA authorises clinical trials |
| Europe | The EU adopted its gene-edited plant law and is negotiating a Biotech Act. | Regulation (EU) 2026/1388; Biotech Act proposal COM(2025) 1022 |
| Africa | Tanzania leads gene-drive design work while Burkina Faso ended a field programme. | Burkina Faso ministerial decision of 22 August 2025; national biosafety agencies across the continent |
Figure 6. Regional summary. Australia’s entry reflects the regulatory system and a leading trial role, not a claim that most of the science originates there.
Policy and law timeline, 2025 to 2026
Figure 7. Key regulatory and research milestones, April 2025 to September 2026.
What to watch over the next twelve months
Five threads are worth following. First, whether FDA finalises the plausible mechanism guidance and how it proceeds after the leadership departures reported at the agency in mid-2026. Second, when Merck and Moderna present full INTerpath-001 data and file with regulators. Third, the pace of the EU’s secondary legislation for gene-edited plants and the fate of the Biotech Act. Fourth, whether Insilico’s Phase III and the EXPAND transplant trial deliver on early signals. Fifth, whether any African country accepts a gene-drive field trial after the Burkina Faso decision.
Frequently asked questions
Which of these technologies can patients use today?
Among the ten, only embryo screening is sold as a commercial service, and professional bodies dispute its readiness. Personalised gene editing has been used in a single patient, and the rest are in trials or in regulation. The one approved CRISPR medicine, Casgevy for sickle cell disease and beta thalassaemia, sits alongside this list rather than on it.
Are gene-edited crops the same as GMOs in the EU?
Not under the new rules. Category 1 plants are treated like conventional varieties, while Category 2 plants remain under the GMO framework. The regulation applies from 17 July 2028.
Does the FDA now ban animal testing?
No. FDA is reducing reliance on it and recognising alternative methods. Its September 2026 rule updates terminology and does not require a particular method.
Where can I check the latest status?
Start with ClinicalTrials.gov for trial status, the FDA New Approach Methodologies page, the European Commission’s biotechnology pages, and Australia’s OGTR site.
Closing thoughts
Reading across these ten, one pattern keeps returning. The science is moving faster than the systems that pay for it, regulate it and earn public trust for it. KJ’s treatment took six months to build, and the rules are still catching up. A melanoma vaccine has a positive Phase 3 result, and nobody yet knows what it will cost. An African government shut down a mosquito programme that had every paper approval.
That is not a reason to slow down. It is a reason to read carefully, ask who benefits, and notice which claims rest on peer-reviewed evidence and which rest on a press release.
Disclaimer. This article is general information, not medical, legal or investment advice. Facts were checked against the sources below as of 24 September 2026, and several of these stories are changing quickly. Where a figure comes from a company statement, the text says so.
Sources and further reading
- MIT Technology Review, 10 Breakthrough Technologies 2026
- STAT News, FDA’s plausible mechanism pathway raises concerns (26 Jan 2026)
- Fierce Biotech, FDA illuminates approval pathway for bespoke gene editing therapies (24 Feb 2026)
- Children’s Hospital of Philadelphia, prime editing platform and FDA framework (31 Mar 2026)
- Innovative Genomics Institute, CRISPR Clinical Trials: A 2026 Update
- Forbes, Jennifer Doudna’s plan to bring gene editing to the masses (Feb 2026)
- Press Information Bureau (India), launch of BIRSA 101 (Nov 2025)
- Observer Research Foundation, Birsa-101 and affordable CRISPR therapies
- Stanford Report, AI designs a novel E. coli killer (6 Aug 2026)
- C&EN, AI program designs new bacteriophages (Aug 2026)
- eWeek, Scientists use AI to design 16 functional viruses (Aug 2026)
- BioTechniques, reactions to agentic bacteriophage design
- Insilico Medicine, Phase III initiation for rentosertib (7 Jul 2026)
- IntuitionLabs, AI-discovered drugs in clinical trials 2026
- AJMC, Moderna and Merck mRNA cancer vaccine succeeds in late-stage trial (19 Aug 2026)
- STAT News, mRNA cancer vaccine melanoma trial (19 Aug 2026)
- C&EN, an mRNA cancer vaccine succeeds in late-stage trial
- Melanoma Research Alliance, new milestone in melanoma
- CURE, first personalised cancer vaccine to pass a Phase 3 trial
- New England Journal of Medicine, In vivo CD19 CAR T-cell therapy for refractory SLE (2025)
- RheumNow, CAR-T product shows early promise in lupus
- ClinicalTrials.gov, NCT07333677 in vivo CAR-T for Graves’ disease
- NYU Langone Health, first pig kidney clinical trial transplant
- American Kidney Fund, FDA clears first pig kidney trials
- CNN, pig kidney transplants take a step forward (Sep 2025)
- Artificial Organs, Kidney xenotransplantation enters the clinical trial era (2026)
- Healthcare Brew, pig kidney pioneers talk next steps (Mar 2025)
- FDA, New Approach Methodologies
- FDA, plan to phase out animal testing requirement (10 Apr 2025)
- Life Science Daily, CRISPR gene editing in 2026
- FDA roadmap one-year progress commentary (PubMed)
- Science, Burkina Faso halts gene drive project
- Nature Africa, mosquito gene drive cancellation disrupts malaria research
- Ecofin Agency, Burkina Faso halts a malaria programme
- Target Malaria, activities suspended in Burkina Faso
- Target Malaria, common questions about gene drive
- Post-release safety monitoring of gene drive mosquitoes in sub-Saharan Africa (2026)
- Singularity Hub, new gene drive stops malaria without killing mosquitoes
- GMP+ International, new EU regulation on new genomic techniques
- British Agriculture Bureau, EU approves new genomic techniques for plants
- USDA Foreign Agricultural Service, EU Parliament passes NGT regulation
- Morgan Lewis, European Commission proposes Biotech Act
- European Commission, European Biotech Act
- American Society for Reproductive Medicine, report on polygenic embryo screening (Dec 2025)
- NPR, labs say they can screen embryos for specific traits (May 2026)
- Fertility and Sterility, governance of polygenic embryo screening
- Frontiers in Reproductive Health, precautions for polygenic embryo selection (2026)
- Office of the Gene Technology Regulator (Australia)
- India BioE3 policy and biomanufacturing centres (PMC)
All figures in this article are original visualisations built from the cited data. Quotations are short and credited to their speakers.
0 Comments