Medical Progress by 2026: Application of Gene Editing and Artificial Intelligence
Read more
TechCentral
techcentral.co.za

Medical Progress by 2026: Application of Gene Editing and Artificial Intelligence

At the annual meeting of the American Society of Clinical Oncology in May, results of a new drug for pancreatic cancer were presented, causing rare 42-second ovations. This success is part of a series of medical advancements this year, many of which are based on gene editing and artificial intelligence but remain in specialized press.

User X under the pseudonym Mgoes seeks to change this situation. On October 27, he published a post listing 10 such achievements, noting that the news 'is in press releases, and almost no one reads them.' Two days later, Elon Musk shared this post, stating that 'if you look closely enough, breakthroughs in everything are accelerating exponentially. An exponent always starts with something small.'

This account is pseudonymous and does not indicate scientific qualifications, but describes the owner as an investor in the venture fund 'SuperHuman Fund,' which focuses on 'human enhancement' and co-hosts the podcast Accelerate Bio, and has invested in some of the guests' companies.

Not all items on the list are breakthroughs, and only about half rely on gene editing or AI. However, collectively they demonstrate how these technologies are beginning to reach patients.

Gene editing alters the DNA instructions inside cells. The CRISPR-Cas9 technology, which earned Emma Sharpentie and Jennifer Doudna the Nobel Prize in Chemistry in 2020, became a medicine at the end of 2023 after regulators in the UK and US approved Casgevy for sickle cell anemia treatment. A newer version, known as base editing, works more like correcting a typo than cutting out an entire sentence.

Demos Hassabis and John Jumper from Google DeepMind received half of the 2024 Nobel Prize in Chemistry for AlphaFold, which predicts the structure of nearly all 200 million known proteins, helping scientists understand their functions. Artificial intelligence is now also used for selecting drug targets and developing molecules aimed at them.

This Year's Achievements

One result is a pill that helps extend the life of people with advanced pancreatic cancer. The drug daraxonrasib from Revolution Medicines approximately doubled the median survival time in a study involving 500 patients whose cancer had spread despite previous treatment: 13.2 months versus 6.7 months with chemotherapy. In August, it was approved by the U.S. Food and Drug Administration (FDA) under the name Rasonque. It blocks defective RAS proteins, which, according to the company, cause over 90% of pancreatic cancer cases.

Another achievement was a single gene-editing infusion preventing most episodes of painful swelling. In April, Intellia Therapeutics reported that a single administration of its CRISPR drug lonvo-z reduced episodes of hereditary angioedema by 87% compared to placebo. It edits a gene within the body to reduce the production of the protein causing the swelling, and 62% of patients had no episodes and did not require other therapy for six months. However, the procedure carries risks: US regulators suspended another internal treatment trial by Intellia after a patient developed severe liver damage and subsequently died.

A single treatment to lower cholesterol, potentially lifelong, was also developed. Researchers from Eli Lilly's subsidiary, Verve Therapeutics, reported that VERVE-102, which uses base editing to switch off the PCSK9 liver gene, reduced 'bad' cholesterol levels by 62% at the highest dose in an early study involving 35 people. The reduction was maintained for up to 18 months, although the longest observation period so far relates to patients on lower doses.

A new generation of weight loss injections has appeared. In May, Lilly announced that its weekly injection retatrutide helped people taking the maximum dose lose an average of 25% of their body weight over 80 weeks, including everyone who started treatment, although 11.3% on this dose discontinued due to side effects. This is a traditional drug mimicking three appetite and blood sugar hormones, unlike Ozempic, which mimics one.

AI helped find and design a drug for lung treatment. In September, Insilico Medicine began testing rentosertib in late-stage (Phase 3) trials in China for idiopathic pulmonary fibrosis—a disease that gradually causes lung scarring. AI analyzed biological data, identifying the TNIK enzyme as a target, and generative AI developed a drug to block it. The lead researcher of the trial, Professor Zuojun Xu, stated in an Insilico message that 'TNIK, the target identified by AI, has never been linked to fibrosis before.'

A vaccine was created to treat one patient's cancer. Computer programs identify up to 34 mutated targets unique to each patient's tumor, and the mRNA vaccine trains the immune system to search for them. In October, Merck and Moderna reported that their vaccine, when used with the immunotherapy drug Keytruda, reduced the risk of melanoma recurrence or death by 42% compared to Keytruda alone in a study of 1137 patients. It reduced the risk of cancer spread or death by 38%, less than 59% in an earlier, smaller study.

A drug to stimulate hair growth also emerged. In October, Veradermics announced that its pill met all goals in the second late-stage trial in men. This is minoxidil, a blood pressure control drug underlying Rogaine, repurposed into a slow-release tablet.

For some types of cancer, there is now the possibility of treatment without surgery. In October, the FDA granted preliminary, conditional approval for GSK's dostarlimab for rectal cancer, whose numerous genetic errors facilitate the immune system's detection of the tumor after the drug removes the brakes. Of the 154 patients in the main study, 71% received dostarlimab alone, requiring no surgery or other treatment, and showed no cancer progression.

Pig kidney as a bridge to transplantation. Tim Andrews lived a record 271 days without dialysis on a pig kidney containing 69 genetic edits until it was removed in October 2025 due to decreased function. In January, he received a human kidney, and The Lancet published this case, which his hospital describes as the first use of a pig kidney as a bridge to human transplantation.

DNA-exchanging cells. This is fundamental science, not medicine. A study published in the journal Cell in May showed the transfer of DNA fragments between cells through tiny tubes, carrying traits to recipient cells. However, this was laboratory work: researchers mixed two types of cells and introduced errors or damage to their DNA. It is unknown if this occurs in the body.

Results for pancreatic cancer, edema, cholesterol, and weight loss were published in the New England Journal of Medicine, while rectal cancer results were published in Annals of Oncology. Only the hair loss results are based exclusively on company data. Many trials are in early, small, or uncontrolled stages, and only two treatments have received approval.

The Start of Exponential Growth

An analysis conducted by Boston Consulting Group researchers in 2024 counted 75 AI-discovered molecules that entered clinical trials in humans between 2015 and 2023; this number grew by more than 60% annually. They passed initial safety tests in 80–90% of cases, which is 'significantly higher than historical industry averages.' However, in larger Phase 2 trials testing efficacy, only four out of ten were successful, which 'matches historical industry averages.'

No drug with both an AI-discovered target and an AI-designed molecule has yet been approved. Xu expects that rentosertib will require three to four years from the start of Phase 3 to receive approval 'under favorable conditions.'

Gene editing has progressed faster: from the first approved therapy Casgevy to the successful late-stage trial of internal editing lonvo-z in less than two and a half years. For now, the exponential growth Musk spoke of is evident in the number of new treatments entering trials, not yet in the number reaching patients.

Significance for South Africa

There are no records indicating that any of the 10 treatments are registered with the South African Health Products Regulatory Authority (Sahpra), which does not publish pending applications. Medical schemes, such as Momentum, exclude unregistered medicines that doctors can only obtain with special permission from Sahpra under Section 21 of the Medicines Act. Obesity is not a mandatory minimum benefit, so schemes are not obliged to cover weight loss injections. Casgevy, costing $2.2 million in the US, remains inaccessible in Africa.

Nevertheless, South Africans are participating in these processes. The University of Cape Town's Institute of Respiratory Medicine was a site for the lonvo-z trial, which was a 'groundbreaking first for South Africa and the African continent,' and centers in Johannesburg, Pretoria, and Cape Town participated in the melanoma vaccine trial. More than 30 South Africans underwent gene therapy for hemophilia in trials, and the H3D drug discovery center at UCT collaborates with AI developers to adapt malaria and tuberculosis drug dosages for African patients.

Similar stories

Expansion of Artificial Intelligence Use in Various Fields of Medicine Planned in Uzbekistan
Read more
podrobno.uz

Expansion of Artificial Intelligence Use in Various Fields of Medicine Planned in Uzbekistan

The application of artificial intelligence technologies will be increased in the healthcare sector of Uzbekistan. Starting next year, ten AI-based systems will be launched in cardiology, hepatology, ophthalmology, and hematology.

President Shavkat Mirziyoyev, speaking at a meeting with employees of the medical and pharmaceutical sectors, announced the introduction of new solutions. These systems will allow for remote monitoring of patient conditions, detection of diseases based on speech characteristics, and control over the safety of medicinal products. This expansion will increase the scope of AI application in healthcare, encouraging doctors to use modern technologies more actively.

Contests and Incentives in Digital Medicine

Competitions will be organized in each of the mentioned fields to determine the 'Best IT Operator' and the 'Best Digital Medicine Doctor'. The winners of these competitions will receive a reward of 100 million soms.

Previously, the head of state reported that starting next year, winners of regional stages of the competition among medical workers in Uzbekistan will be awarded electric vehicles. This year, the contests are being held for the first time, and the prize fund for the winners amounts to 30 million soms, but stronger motivational incentives are planned for the future.

Use of Artificial Intelligence for Content Review and Question Training for Enem 2026
Read more
g1.globo.com

Use of Artificial Intelligence for Content Review and Question Training for Enem 2026

Artificial Intelligence (AI) presents itself as a potential aid in preparation for the National High School Exam (Enem), offering students support in reviewing subjects, creating mock tests, and identifying weak points. However, specialists emphasize the need to check the information provided by these tools in reliable sources.

Vinícius Gonçalves, coordinator of the Computer Science and Systems Analysis and Development courses at Uninassau Teresina, suggests that an effective approach is to use past Enem exam papers to guide the answers generated by AI. He explained that the best tactic for students, especially when generating questions, is to input models of previous exams into the artificial intelligence, instructing it to base its new questions exclusively on this data.

This methodology allows participants to request exercises similar to those covered in previous years and review topics that are recurrent in the exam.

How to apply AI in content review

The technology can be used to create summaries, mind maps, and mock tests, in addition to clarifying conceptual doubts. Students also have the option to ask questions about specific subjects and monitor their own progress during study. According to Vinícius, the customization of learning depends on constant monitoring of performance. By recording their successes and difficulties in various tasks, the student can use this data to define the next focus areas for review.

The professor warns, however, that AI tools can produce false or invented data, a phenomenon called AI 'hallucination.' For this reason, he advises students to always validate answers and demand references to confirm the origin of the information. He advised: 'It's simple, always ask for the source. After you see an answer that doesn't cause you much discomfort or you have a study material and what the AI is saying does not match the study material, ask it to identify the source, so you can validate where this information is coming from.'

The central recommendation is to compare the obtained answers with official materials and reliable answer keys, since AI assists in explaining questions but should not be seen as an infallible source of information.

Precautions in the final phase before Enem

In the final stage of Enem preparation, the professor advises prioritizing exercise resolution over trying to absorb large volumes of new content all at once. During this period, AI can be useful for creating exercises based on old exams, helping candidates get used to the structure of the assessment.

Vinícius points out that regular practice helps the student identify testing patterns and develop efficient methods for solving questions. The tool also has the capacity to present new scenarios for topics frequently addressed in the test. However, the coordinator warns that the use of technology must not replace the student's critical thinking. If it is used only to deliver ready-made solutions, AI can foster dependence and compromise the development of autonomy.

He warned: 'You can turn it into a crutch instead of a tool, where the student becomes totally dependent on the answers from this artificial intelligence.'

Therefore, the guidance is for the student to try to solve the exercises before consulting the tool and subsequently check the answers in secure sources. In this way, AI acts as a learning support, without supplanting the effort and practice necessary for Enem.

Experts Discuss Using Artificial Intelligence to Address Gaps in India's Public Healthcare System
Read more
yourstory.com

Experts Discuss Using Artificial Intelligence to Address Gaps in India's Public Healthcare System

Startups are developing solutions to improve healthcare delivery in India, including tools for early disease detection and remote monitoring, which help frontline medical staff make more informed decisions. However, the main challenge lies in integrating these technologies into India's existing public health system and ensuring their accessibility.

At the TechSparks 2026 event, Dr. Gita Mandjunath, Founder and CEO of Niramai Health Analytix; Debraj Banerjee, Global Technology Investor and General Partner at Fundamentum Frontier Advisors; and Shekhar Sivasubramanian, Head of Wadhwani AI, discussed innovations in healthcare and the role of artificial intelligence, as well as the reasons why promising developments often fail to achieve widespread adoption.

According to the panel participants, startups face three interconnected challenges: the need for rigorous clinical validation, integration with existing public health infrastructure, and the requirement for patient capital to scale solutions across India's diverse healthcare ecosystem.

Shekhar Sivasubramanian, Head of Wadhwani AI, stated that 'there are no gaps in the system,' asserting that innovators should learn to build solutions considering real constraints rather than viewing them as obstacles. He emphasized that developers must account for realities such as inconsistent internet connectivity, incomplete data, and the necessity of operating in different languages and cultural contexts.

Gita Mandjunath noted that one of the primary barriers to technology adoption is the tendency to prioritize rapid product creation over its clinical validation. Although AI tools can be built in days using pre-existing datasets, a model that performs well in a lab does not guarantee reliable results when used in real medical practice.

She clarified: 'The question is not whether you create your model. The real problem is seeing if your algorithm works on real people.' Retrospective studies based on existing hospital data can be used for initial assessment, while prospective studies test the technology in a live clinical setting without allowing its predictions to influence treatment, and then compare the results with established clinical data.

Nevertheless, Mandjunath added that even these stages may be insufficient for public health implementation. A technology tested in a hospital must also be evaluated in the conditions where it is intended for use, including rural areas with limited diagnostic infrastructure. Before recommending a technology for wider application, the cost of implementation, deployment feasibility, and clinical effectiveness must be considered. She concluded that 'healthcare requires more patience,' and that founders need resilience and long-term commitment.

Beyond clinical validation, a significant challenge remains the incorporation of new technologies into existing government systems. Sivasubramanian from Wadhwani AI advised companies in public health to understand the systems used by both central and state authorities. His organization has teams working in health offices across 17 states, with groups of five to seven people in those offices.

This approach is necessary because technology implementation requires more than just developing a functional product. Teams must train users, understand local requirements, and adapt their solutions to existing workflows. He stressed that every application requires integration with primary healthcare infrastructure. Sivasubramanian noted that there are about 150,000 primary healthcare centers in India, and innovators must be prepared to operate at that scale. Without patience, persistence, and sufficient financial resources, a company can reach the halfway point, become discouraged, and abandon the project. He added that government officials implement these solutions, while innovators act as facilitators, and tech companies must work with them rather than assuming they can achieve change independently.

For medical technologies to reach underserved populations, their design must incorporate accessibility and acceptability. Gita Mandjunath cited Niramai's breast cancer screening technology as an example. The company uses thermal imaging and AI to assess breast cancer risk through a non-invasive screening process. The company's goal is to reduce screening costs to below 50 rupees at scale. This technology is also designed to eliminate barriers preventing women from undergoing screenings, including discomfort from procedures requiring physical contact or breast compression.

Mandjunath defined accessibility, acceptability, feasibility, and accuracy as key requirements for successful AI adoption in public health. Timeliness of results and clear accountability are also important. For investors, the task goes beyond determining technological functionality; they must assess whether the business can obtain regulatory approvals, meet financial requirements, and handle customer acquisition complexities.

Debraj Banerjee, General Partner at Fundamental Frontier Advisors, stated that high-tech companies require an ecosystem that supports the transition from product development to commercialization. He emphasized that 'there is no single formula for scaling medical innovations.' When evaluating AI-based products, he said that fundamental principles and their applicability to the product are considered. The product must be properly segmented, integrated into the relevant system, and localized. These factors help assess the probability of investment success.

Banerjee also noted that venture capital is only part of the entire ecosystem, and there needs to be development of a broader network of healthcare providers, regulators, investors, and commercial partners. The broader issue, in his view, is not simply creating better technology but building systems that allow it to reach users.

For Niramai, opportunities extend beyond India, as the company has screened around seven hundred thousand women and is working to expand access to its technology. Gita Mandjunath stated that the company's goal is to cover every woman, acknowledging that this will require sustained effort over many years. The company has also entered international markets. The same technology can be relevant in developed markets where healthcare systems seek ways to manage costs. Niramai has six centers in the US using its solution and has received FDA approval for co-use with existing screening methods. However, she added that expansion requires a different approach in each market, as regulatory requirements, reimbursement structures, and pricing models vary, and the solution cannot always be deployed identically everywhere. She concluded that Indian health innovators should not forgo international opportunities due to the need to create solutions for local needs.

Popular