Medical Students’ Perceptions and Attitudes Toward English as a Medium of Instruction at the Faculty of Medicine and Pharmacy of Rabat: Cross-Sectional Study

Background: The global adoption of English-medium instruction (EMI) in higher education has introduced complex implementation challenges, the severity of which often depends on the resources available within specific educational contexts. Evidence remains limited in public medical schools in under-resourced, non-Anglophone countries, which serve a socioeconomically and educationally diverse student population. In such settings, the direct transfer of existing EMI integration models presents significant practical challenges. Objective: This exploratory, observational, monocentric study aimed to investigate medical students’ perceptions and attitudes toward EMI implementation at the Faculty of Medicine and Pharmacy of Rabat (FMPR) in Morocco and to evaluate their readiness for EMI adoption. Methods: A cross-sectional survey was administered to 102 second-cycle medical students at FMPR. The 23-item questionnaire included Likert-type scales and multiple-choice questions across 4 domains: demographic data, self-reported English language proficiency, language perceptions and attitudes, and EMI needs. Bivariate, univariate, and multivariate logistic regression analyses were conducted to determine the effects of the explanatory variables on EMI choice. Results: The sample comprised 65 female and 37 male participants. Most reported using English speaking and listening skills “often” on a daily basis, while writing skills were reported as being used “rarely.” Participants rated their general English proficiency and their academic listening and reading skills as “good.” Overall, 82.4% (84/102) of students were strongly in favor of implementing EMI at FMPR. Bivariate analysis showed significant positive associations between EMI choice and age (=.04), course year (<.001), interest in learning English (=.006), and the belief that English should be the language of higher education (=.02). Preference for EMI increased alongside course year, peaking at 96.4% (27/28) among students in the third and fourth years. Multivariate logistic regression confirmed that being in the fourth vs third course year (adjusted odds ratio 22.12, 95% CI 2.56‐190.84; =.005) and having an interest in learning English (adjusted odds ratio 5.12, 95% CI 1.05‐25.03; =.04) were significant positive predictors of EMI choice. Conclusions: Medical students at FMPR exhibit highly positive perceptions and attitudes toward the potential implementation of EMI, despite variations in their self-reported English readiness. These findings provide actionable insights for the successful integration of EMI that extend beyond the Moroccan context, offering a valuable framework for medical institutions in other under-resourced, low- and middle-income, non-Anglophone settings across the Global South.

Gene Regulation Map Uncovers Heart Failure Targets

Scientists have created the most detailed map to date of single-cell gene regulation in heart failure. Published today in Science, the study integrates multiple layers of genomics data to identify new therapeutic targets for the world’s leading cause of death.

“Our goal is to use this atlas to discover targets that we can act on therapeutically,” said Neil C. Chi, MD, PhD, professor of medicine at UC San Diego School of Medicine. “Right now, one of the biggest limitations in cardiology is not the lack of tools, but the lack of targets. This kind of data changes that.”

While previous genetic studies have uncovered many genetic changes linked to heart failure, more than 85% of them were found in noncoding DNA regions, making it difficult to understand how they contribute to the condition and develop targeted therapeutics. Because of this, treatment options for heart failure remain limited.

“What makes this study unique is the ability to integrate multiple layers of genome regulation in the same cells,” said Bing Ren, PhD, professor emeritus of cellular and molecular medicine at UC San Diego School of Medicine and scientific director and CEO of the New York Genome Center. “These technologies allow us to look beyond which genes are active to understand how the genome is organized and controlled, revealing regulatory elements and interactions that were previously inaccessible.”

The researchers analyzed more than 750,000 individual heart cells from 36 participants with and without heart failure. Results revealed that heart failure is associated with major shifts in cell composition, with a reduction in the number of cardiomyocytes and an increase of fibroblasts and immune cells. In failing hearts, over 10,000 genes showed altered expression patterns and more than 50,000 DNA regions showed changes in chromatin states that altered the ability of proteins to interact with them. 

Compared to other cell types, fibroblasts and cardiomyocytes showed the most extensive remodeling of their gene regulation networks. These cells also showed several distinct cell states as they progressed from healthy to diseased—in the case of fibroblasts, transforming into activated fibroblasts and myofibroblasts that contribute to scarring and fibrosis. 

“These intermediate states are where the disease is actively unfolding,” said Chi. “If we can understand and target those transitions, we may be able to intervene earlier and more effectively.”

Integrating the single-cell atlas with genome-wide association data, the team showed that genetic changes are concentrated in specific regulatory regions active in certain cell types, particularly in cardiomyocytes. These findings suggest that targeting cardiomyocyte genes may be more effective than targeting genes found on other cell types when treating heart disease. 

“This is a higher-order view of disease biology,” said Chi. “Instead of just asking which genes are turned on or off, we’re now understanding how their regulation is controlled across the genome—and that’s where most disease risk actually resides. By connecting genetic risk, gene regulation and cell-specific disease processes, this study provides a blueprint for precision therapies in heart failure. It opens the door to targeting the right mechanisms in the right cells at the right time.”

The post Gene Regulation Map Uncovers Heart Failure Targets appeared first on Inside Precision Medicine.

Jagged Little Pill: With Merck’s Oral PCSK9 Inhibitor, Does LDL-C Care Change?

The story of PCSK9 is an archetypal chapter from the precision medicine book of successes. It opens with the discovery of rare genetic variants that naturally lower PCSK9 activity, showing how well lowering low-density lipoprotein cholesterol (LDL-C) protects against cardiovascular disease. That discovery leads to a renaissance in lipid management, with monoclonal antibodies, RNA-based therapies, and experimental gene-editing methods to permanently silence the gene. The latest advance in delivering PCSK9-inhibiting therapies is perhaps also the simplest: a pill. 

Last week, the FDA approved Merck’s once-daily oral PCSK9 inhibitor LIPFENDRA (enlicitide), making it the first pill to target one of cardiovascular medicine’s most validated therapeutic pathways. In two pivotal Phase III studies, enlicitide reduced placebo-adjusted LDL-C by as much as 60% in adults with hypercholesterolemia, including those with heterozygous familial hypercholesterolemia (HeFH), when added to diet and exercise.  

Seth S. Martin, MD, professor and preventive cardiologist at Johns Hopkins, believes cheaper, oral PCSK9 inhibition—a 30-day supply of LIPFENDRA costs $315 compared to injectable PCSK9 inhibitors that cost $500-$600 per month—could help expand the use of one of cardiology’s most effective therapeutic strategies.  

“I see oral PCSK9 inhibition as a meaningful advance that can help build on momentum in this regard by expanding the options available to patients and clinicians,” Martin told Inside Precision Medicine. “Having a once-daily oral medication as an option can significantly broaden the use of PCSK9-targeted therapy. The more options we have, the better positioned we are to collaborate with our patients in finding treatment approaches that work and close gaps in care.” 

The approval could solve one of preventive cardiology’s biggest problems: implementing highly effective LDL-lowering therapies. “Atherosclerotic cardiovascular disease (ASCVD) remains the leading cause of death, and elevated LDL-C is one of its most important modifiable drivers,” Martin said. “If we’re going to knock ASCVD off the top of that list, we’ll need to scale up the use of effective LDL-C-lowering therapies to consistently hit LDL-C goals that we set in our 2026 ACC/AHA/Multisociety Dyslipidemia Guideline.” 

Molecular innovation meets medicinal chemistry 

Enlicitide is a macrocyclic peptide that binds PCSK9 with high affinity while remaining orally bioavailable—a feat many researchers once considered unattainable.   

“For years, many medicinal chemists believed it simply wasn’t possible to develop an oral molecule that could inhibit the PCSK9–LDL receptor interaction the way monoclonal antibodies do,” Christie Mitchell Ballantyne, MD, chief of cardiology at Baylor College of Medicine and an investigator in the enlicitide clinical development program, told Inside Precision Medicine. “This isn’t a conventional small molecule. It’s a macrocyclic peptide, and developing it required extraordinary chemistry.”  

The challenge went beyond finding a PCSK9-binding molecule. Ballantyne said it took multiple rounds of innovation to optimize the lead compound for oral absorption, metabolic stability, and scalable manufacturing after screening massive macrocyclic libraries. With these manufacturing advances, which may be as important as molecular design, Merck was able to introduce enlicitide at a lower price than injectable PCSK9 inhibitors and even some oral LDL-lowering therapies by simplifying chemical synthesis, potentially overcoming one of the main barriers to the drug class’s use.  

Two Phase III CORALreef studies supported FDA approval. In the 2,904-patient CORALreef Lipids trial, adults with hypercholesterolemia at high cardiovascular risk had a placebo-adjusted 56% LDL reduction at 24 weeks, increasing to 60% with revised post hoc data handling that excluded biologically implausible baseline LDL-C values. Enlicitide also reduced non-high-density lipoprotein (HDL) cholesterol by 54% and apolipoprotein B by 50%. The 303-patient CORALreef HeFH study produced similarly robust findings, with placebo-adjusted LDL reductions of 59% alongside significant decreases in non-HDL cholesterol and ApoB. 

Overall, Ballantyne said, the Phase III program largely confirmed what decades of PCSK9 biology had led researchers to expect, lining up remarkably well with what has been seen from monoclonal antibodies. 

Karol Watson, MD, PhD, professor of medicine/cardiology at the David Geffen School of Medicine at UCLA and co-director of the UCLA Program in Preventive Cardiology, expects the trial to validate what decades of LDL biology have consistently suggested, while emphasizing that definitive evidence still matters. “Based on everything we’ve learned over the last several decades, I think it’s very reasonable to expect that those LDL reductions will translate into fewer cardiovascular events,” Watson told Inside Precision Medicine. “But we’ll let the outcomes trial give us the definitive answer.” 

For example, although LDL-C is one of medicine’s most extensively validated surrogate biomarkers, regulators appropriately distinguish between lowering cholesterol and demonstrating reductions in heart attacks, strokes, and cardiovascular death. 

 Merck’s ongoing CORALreef Outcomes trial aims to answer that question directly. The study has enrolled more than 14,500 high-risk patients and is evaluating whether enlicitide reduces major cardiovascular events in addition to lowering LDL-C. Ballantyne noted that investigators continue collecting long-term safety and efficacy data through both the outcomes trial and extension studies. 

Making prevention practical 

PCSK9 inhibitors have reduced LDL-C by roughly 60% and lowered cardiovascular events for nearly a decade. Yet despite strong clinical evidence, they remain substantially underused in routine practice. The reasons have been familiar: high initial prices, burdensome insurance requirements, physician inertia, and patient reluctance to begin lifelong injectable therapy for a condition that causes no immediate symptoms.  

“The challenge has never really been whether PCSK9 inhibition works,” Ballantyne said. “These have been excellent drugs. But they’ve had an access problem.” 

For Ballantyne, the promise of enlicitide came through while taking a look at the CORALreef safety data, where findings closely mirrored the placebo. Adverse event-related discontinuation rates were similar across treatment groups, while diarrhea and dizziness occurred infrequently. 

“When I review a placebo-controlled trial, one of the first things I examine is whether patients discontinue therapy more often than placebo,” Ballantyne said. “If discontinuation rates are similar, that’s generally very good news because it tells you patients can tolerate the medication.”  

Watson believes oral therapy addresses one of those barriers by giving physicians another option for patients who hesitate when injections enter the conversation. “Having an oral PCSK9 inhibitor is a really important step forward because it gives us another option for patients who just don’t want an injectable medication,” she said. “Many patients do perfectly well with injections, but for others it’s a real barrier. Sometimes they put off starting treatment, and sometimes they never start it at all. If we can offer a pill instead, I think we’ll be able to get more patients on effective therapy sooner.” 

That flexibility may be especially valuable for patients with established cardiovascular disease or HeFH who remain above guideline-recommended LDL targets despite maximally tolerated statins and ezetimibe. “We now have another way to lower LDL cholesterol without asking patients to make the jump to an injectable if they’re reluctant,” Watson said. 

Still, Watson cautions against assuming an oral formulation will automatically solve the adherence problem. “An oral medication has the potential to improve adherence for some patients because many people would rather take a pill than give themselves an injection,” she said. “But we have to remember another group of patients hates taking daily medications, so they will prefer an every 2-week injectable.” 

 Keith C. Ferdinand, MD, professor of medicine at the Tulane University School of Medicine and the Tulane Heart & Vascular Institute, shares that assessment. “Undertreatment is widespread and common,” Ferdinand told Inside Precision Medicine. “PCSK9 inhibitor uptake has been poor, but probably due to early access barriers and costs. But generic high-intensity statins, costing much less, are underutilized, and most ASCVD patients do not obtain a [cholesterol target of <70 mg/dL]. If an oral PCSK9 inhibitor has cost barriers, uptake will be poor and may not increase goal attainment.” 

For Ferdinand, introducing another effective drug is only part of the solution. Measuring whether clinicians actually achieve LDL targets may be equally important. “Oral PCSK9 inhibitors may be costly and underutilized,” he said. “A performance measure of goal attainment would have to overcome therapeutic inertia. Clinicians need a report card on their care and outcomes and respond to high LDL-C. Widespread coverage with easy prior authorization is key.” 

A new chapter for PCSK9—and precision medicine  

The arrival of the first oral PCSK9 inhibitor also comes as cardiovascular medicine enters another transformative era. Several investigational gene-editing therapies are designed to permanently disable PCSK9 after a single treatment, raising the possibility that lifelong cholesterol management could eventually become a one-time intervention. 

At first glance, the technologies appear to compete. Ballantyne sees them as addressing different clinical needs. “The numbers you’re seeing are on top of statin therapy,” he said. “Patients didn’t stop their statins—they received gene editing in addition to their daily medication.” That context changes the comparison. “If you’re still taking a pill every day,” Ballantyne said, “and you could instead take a pill that combines a statin with an oral PCSK9 inhibitor, that’s another way of thinking about the problem.” 

Rather than viewing oral therapy and gene editing as mutually exclusive, he expects cardiovascular prevention to evolve into a spectrum of treatment options. Most patients may continue using inexpensive oral medications, while one-time genetic interventions could eventually serve carefully selected populations for whom lifelong adherence is particularly challenging or extraordinarily high LDL levels warrant more aggressive intervention. Ballantyne believes the field will ultimately determine which patients benefit most from each approach based on durability, safety, cost, practicality, and patient preference rather than assuming a single technology will replace all others. 

After all, PCSK9 remains the target that human genetics first illuminated more than two decades ago. What is changing is how clinicians and patients may choose to intervene. “The more options we have, the better positioned we are to collaborate with our patients in finding treatment approaches that work and close gaps in care,” Ballantyne said. “If oral PCSK9 inhibitors help more patients initiate treatment, stay on therapy, and achieve guideline-recommended LDL-C levels, they can meaningfully improve cardiovascular disease prevention over the next several years and beyond.” 

Whether enlicitide ultimately becomes transformative will depend on factors extending well beyond medicinal chemistry. “This has the potential to be more than just another cholesterol-lowering drug,” Watson said. “Whether it becomes truly practice-changing will depend on several things: cost, access, and whether clinicians are comfortable incorporating it into routine care.” 

Watson expects the cardiovascular outcomes data to determine just how broadly clinicians embrace the therapy. “Given what we know about LDL-C as a causal driver of atherosclerotic cardiovascular disease and given the amount of LDL lowering this drug produces, I’d be surprised if we didn’t see cardiovascular benefit,” she said. “If those outcome data are positive, I think this could become another really valuable tool in our toolbox. I don’t see it replacing statins or ezetimibe, but I do see it helping many more high-risk patients actually reach their LDL goals.” 

For Ballantyne, however, the approval represents something even larger than another effective lipid-lowering therapy. “It’s a remarkable science story,” he said. “You start with human genetics, then monoclonal antibodies, then RNA therapies, now macrocyclic peptides. It’s another example of how understanding biology can lead to entirely new therapeutic approaches.” 

For patients who continue to struggle to reach recommended LDL targets despite existing therapies, that story has now produced something both novel and reassuringly familiar: a once-daily pill. 

The post Jagged Little Pill: With Merck’s Oral PCSK9 Inhibitor, Does LDL-C Care Change? appeared first on Inside Precision Medicine.

STAT+: Popular insulin product enters shortage as other pens leave market

People with diabetes who use a popular long-acting insulin pen made by Sanofi have had to become pharmacy scouts in recent weeks as they encounter supply issues.

The Lantus Solostar pen containing 3 milliliters of insulin glargine has been running low for several weeks because of increased demand “driven by broader market dynamics,” a Sanofi spokesperson told STAT. The issue is not technically considered a shortage by the Food and Drug Administration, but patients are nonetheless having trouble finding the medication.

“Our priority is ensuring patients have access to their medications,” the spokesperson said. “This is a temporary situation that we anticipate will improve over the coming weeks.”

Continue to STAT+ to read the full story…

“Physical Twins” of Human Arteries Predict Individual Stroke Risk 

Scientists in Australia have developed an artery-on-a-chip platform that can replicate a patient’s exact vascular structure to better assess their risk of ischemic stroke. Results published today in Cell Biomaterials show that differences in artery shape explain significant differences in stroke risk seen between patients with similar levels of artery narrowing. 

“Our findings suggested that three-dimensional vascular shape and local flow disturbances matter far more than simple narrowing,” said Yunduo Charles Zhao, graduate student at the University of Sydney and the Heart Research Institute in Newtown, Australia. “We hope these tools will allow us to study drugs aimed at reducing the risk of stroke and to eventually provide personalized treatments for each patient based on their anatomy.” 

Despite significant advances in imaging and management, stroke risk stratification is still very imprecise. Nearly 20% of patients receiving what is considered an optimal antiplatelet therapy continue to experience recurrent strokes, reflecting an incomplete understanding on how complex mechanical and biological factors determine whether clots grow, stabilize, or cause a stroke. 

“This idea was born out of a critical clinical gap,” said Lining Arnold Ju, PhD, associate professor at the University of Sydney in Australia and senior author of the study. “We know that even patients at ‘low risk’ can suffer from severe or fatal strokes. We wanted to find a better way to predict this risk.” 

Ju’s team used high-resolution 3D printing to create a “physical twin” that replicated the three-dimensional structure of the carotid artery from six patients who had previously experienced stroke of the carotid artery. To capture the full complexity of each patient’s unique physiology, the model also included the thrombogenic matrix and endothelium tissues, and simulated blood flow through the chip. 

“Our work recreates precise, patient-specific carotid artery geometries,” said Ju. “The physical twin also uses cells that more closely mimic the dynamics of blood flow in these structures.”

Using computer simulations, the researchers modeled blood flow through each carotid artery, uncovering substantial differences in local blood flow despite similar degrees of narrowing. They then used a laser to create an injury in the physical twins and study how blood clots formed, revealing that subtle differences in an artery’s shape could lead to strikingly different clotting responses. 

The researchers are currently recruiting patients with a stroke history for a clinical trial designed to evaluate the potential of this technology to improve diagnosis and treatment in underserved stroke patients. Down the line, the artery-on-a-chip physical twins could find applications in other cardiovascular conditions, such as peripheral artery disease, deep-vein thrombosis, and aneurysms. 

 

The post “Physical Twins” of Human Arteries Predict Individual Stroke Risk  appeared first on Inside Precision Medicine.

How AI helps scientists design the next generation of medicines

Designing and developing a new medicine is an expensive, failure-prone scientific challenge. A new drug can take many years to develop, at the cost of a significant investment. And even then, most possible candidates never reach the patient. For biologic medicines, therapies made from engineered proteins rather than synthetic chemistry (which are often used to treat conditions across most major acute and chronic diseases), the complexity is even greater.

Scientists explore vast quantities of possible molecules, looking for the rare few that will bind to the right target, remain stable in the human body, and be manufacturable at scale. Today, AI is speeding up these processes and has quickly become a core part of the infrastructure in pharmaceutical R&D.

AI-assisted design is a growing part of how biologic drug candidates are developed, and companies like AstraZeneca are actively building its engineering teams to push this further. “Everything we do, whether it’s design, make, test, or analyze, is now computationally enhanced,” says Puja Sapra, senior vice president and head of R&D biologics engineering and oncology targeted discovery at AstraZeneca. “The cycle times are getting shorter while productivity and innovation increase.”

Sapra explains that AstraZeneca’s approach follows a build-measure-learn loop. AI generates or prioritizes candidate molecules computationally, predicting which designs are most likely to succeed. Scientists then focus lab resources only on the top-ranked candidates. This leads to a tighter feedback cycle with fewer dead ends, faster iteration, and the ability to go after disease targets that were previously considered untreatable by medicine. Because the number of possible molecular combinations far exceeds what any human team can systematically explore, using AI to narrow and refine the options for testing has become a major focus in biologics drug design.

Navigating complex drug design problems

Beyond accelerating timelines, AI is also being applied to the discovery of entirely new classes of medicines. Traditional biologics typically target one disease pathway. The next generation of drugs can hit multiple targets simultaneously or precisely deliver therapeutic payloads to specific cells. Achieving this requires optimization across many variables at once. Looking ahead AI-driven models could help design these increasingly complex, multi-specific biologics, explains Puja Sapra. “For example,” she continues, “such models could help identify which two or three targets to prioritize based on the underlying biology, then optimize across multiple parameters to balance a molecule’s potency, stability, manufacturability, and safety.” “Drugging the undruggable is becoming a reality,” Sapra says. “These technologies will eventually enable us to develop medicines against targets once thought impossible to reach. The potential for benefit to patients is remarkable.”

The data moat

McKinsey estimates that generative AI, combined with other computational tools, could cut drug discovery timelines by as much as 50%. But every AI model is only as good as its training data. In drug discovery, that means ample quantities of high-quality biological data. Experiments can provide a rich source of such data. Whether they succeed or fail, each experiment generates a signal about what does and does not work.

“Data is our differentiator,” says Sapra, explaining how the company’s datasets are proprietary and multimodal and include molecular structures, binding measurements, safety profiles, and manufacturing outcomes. “We’ve built an intentionally diverse portfolio across multiple disease areas and drug types. All of that data empowers us to fine-tune frontier AI models with richer, more representative training sets.” She continues, “Further, we have invested in deep screening technologies to generate additional datasets required in volume to constantly refine and validate our models.”

Building an autonomous discovery engine

To bring all of that data together in one place, AstraZeneca is building what it calls a “lab of the future” facility in Kendall Square, Cambridge, Massachusetts where AI and robotic automation will be able to form a continuous, closed-loop discovery system. “Where a self-driving car uses sensors and models to navigate its environment, this system uses AI to make predictions, robotic systems to execute experiments, and instruments to generate data,” explains Sapra. That data feeds directly back into the models, accelerating each subsequent cycle.

“Throughout, scientists will remain central to the process, providing the oversight, judgement, and strategic direction that ensure outputs are explainable, tolerable, and directed toward potential patient benefit,” she adds.

Eventually, automated high-throughput systems will be able to make and evaluate thousands of molecular interactions on a weekly basis. “This will generate AI-ready data at a scale that traditional workflows cannot match,” Sapra says. “Robotic sample handling, automated quality checks, and integrated data pipelines also have the potential to help accelerate early drug development timelines significantly.”

The next frontier: Generating medicines from scratch

Ultimately, Sapra says, the end-state vision for AI in biologic drug discovery is what the field calls “de novo” design. For this, the goal is for AI to generate entirely new protein sequences that precisely fit the desired drug properties. This includes designing the structure, predicting safety, how it will behave in the body and how to make it manufacturable.

“The field is making great progress toward a completely AI-generated biologic, designed from scratch all the way to a clinical candidate,” Sapra says. “As we continue to leverage frontier models and fine-tune them with the right datasets, we bring ourselves closer to this reality. I believe it will come. It’s a matter of time.”

Several key elements are needed to reach this point, however. First is richer and more standardized training data across the industry. Second, robust evaluation benchmarks for AI-generated candidates. And third, teams that know how to work at the intersection of machine learning and biology. Of all the prerequisites, however, safety prediction may be the most consequential, and perhaps the least discussed, Sapra says.

“One of the hardest problems in de novo design is predicting whether a computationally generated molecule will be safe in the human body,” Sapra explains. AstraZeneca is tackling this with what amounts to virtual clinical trials. These are advanced cell systems and micro-scale organ models that function as physical testbeds, paired with AI that learns from their outputs.

 “These systems have the potential to generate enhanced biological signals without traditional testing bottlenecks, and they’re a critical missing piece in closing the loop between AI-generated designs and clinical-ready candidates,” Sapra adds.

A shift currently underway is the move toward agentic AI systems that can simultaneously generate molecule candidates and predict how efficacious and safe they are likely to be. These autonomous workflows can connect disease-level insights directly to molecule design, bridging what were previously separate data silos. “The complexity of the biology goes hand-in-hand with the design of the molecule,” summarizes Sapra.

Human talent unlocks AI potential

The transformation underway in biologics is not just about technology. “With more autonomous systems, human oversight remains at the heart of this approach—ensuring explainable and ethical AI for the benefit of patients,” says Sapra.

For scientists, working with AI is a collaborative process. “Scientists will work hand-in-hand with these model systems,” she says. “There will be a world where models will design molecules, then scientists will work with the systems to test those molecules and put all that data together.” Through this process of human checks, balances, and judgement calls, the models will evolve and constantly improve, ultimately with potential to benefit patients.

For engineers, designing and building effective systems ready for human-AI collaboration will mean ensuring high levels of model transparency and explainability. According to Sapra, AstraZeneca’s engineering teams include data scientists, automation specialists, and AI engineers, who are developing systems that act as “thinking partners” rather than black boxes. “Engineers are designing systems that generate, validate, and learn at speed. And the problems are genuinely hard: Multimodal data fusion, closed-loop optimization, uncertainty quantification, and interpretability at the point of clinical decision-making,” she adds.

In taking on such technically demanding challenges, engineers and scientists have the opportunity to contribute to the research and development of potentially life-changing treatments for many diseases, says Sapra. “The biologic medicines we can develop today, and those we’ll design tomorrow, depend on combining world-class AI and engineering talent with deep scientific expertise.”

This article has been initiated and funded by AstraZeneca.  Z4-85058, July 2026.

This content was produced by Insights, the custom content arm of MIT Technology Review. It was not written by MIT Technology Review’s editorial staff. It was researched, designed, and written by human writers, editors, analysts, and illustrators. This includes the writing of surveys and collection of data for surveys. AI tools that may have been used were limited to secondary production processes that passed thorough human review.

STAT+: What is a ‘world model’? Nabla’s Alex LeBrun explains

You’re reading the web edition of STAT’s AI Prognosis newsletter, our subscriber-exclusive guide to artificial intelligence in health care and medicine. Sign up to get it delivered in your inbox every Wednesday. 

Every time I interview someone from an AI scribe company, I ask if it’s OK to record the interview for my notes and make a joke about the ironic situation.

Turns out, not everyone is asking, and people have started to record their daily conversations, such as first dates, without asking. “This Conversation is Being Recorded. They All Are” is the Wall Street Journal’s headline. What a hellscape.

Continue to STAT+ to read the full story…