Cancer is a large group of diseases that can start in almost any organ or tissue of the body when abnormal cells grow uncontrollably, and go beyond their usual boundaries to invade adjoining parts of the body. According to the World Health Organization (WHO), it is the second most common cause of death globally, accounting for millions of deaths every year. Lung, prostate, colorectal, stomach and liver cancer are the most common types of cancer in men, while breast, colorectal, lung, cervical and thyroid cancer are the most common among women. However, these are not necessarily the deadliest forms of cancer.

What makes cancer the deadliest depends upon how many people have it and what percentage of those people actually survive. Cancer researchers determine this on the basis of five-year relative survival. This is the percentage of people who are expected to survive the effects of a given cancer, excluding their risk of other possible causes of death, for five years past a diagnosis. It is also important to note that what makes cancer really deadly is that practically no cure for it. A cure for cancer would imply that there are no cancerous cells remaining in the body.

Here are the 5 deadliest cancers in the U.S., according to SEER five-year relative survival data for cases diagnosed between 2014 and 2020.

1. Pancreatic cancer occurs when cells in your pancreas, a gland in your abdomen that aids digestion, mutate and multiply out of control, forming a tumour. Major risk factors include smoking, obesity, diabetes, chronic pancreatitis, certain genetic mutations and environmental chemical exposure.

2. Esophageal cancer develops in the oesophagus, which is the tube that connects your throat to your stomach.

3. Liver cancer and intrahepatic bile duct cancer originate in the liver or bile ducts, often linked to hepatitis infections, heavy alcohol use, obesity, and aflatoxin exposure.

4. Lung and bronchus cancer primarily caused by smoking, secondhand smoke, and environmental pollutants, affects the lungs and airways, making it the leading cause of cancer death in the US.

5. Acute myeloid leukaemia (AML) is an aggressive blood and bone marrow cancer that progresses rapidly, often linked to genetic mutations, radiation exposure, and certain chemicals.

ALSO READ: Why Are Lifestyle Factors Making Millennials Vulnerable To Cancer?

Stopping GLP-1 drugs such as Ozempic, Wegovy, Mounjaro and Zepbound may cause some of their cardiovascular benefits to fade, a large new study suggests.

Researchers from Washington University in St. Louis found that people who stopped GLP-1 treatment had a progressively higher risk of major cardiovascular events, including heart attack, stroke and death. After two years off treatment, their risk was 22% higher than among those who continued taking the drugs.

The findings, published in BMJ Medicine, are significant as GLP-1 drugs are widely used to treat type 2 diabetes and obesity and have also been shown to provide cardiovascular benefits.

What Did The Study Find?

Researchers analyzed health data from 333,687 US veterans with type 2 diabetes over up to three years. Of these, 132,551 had been prescribed GLP-1 drugs, while 201,136 had been prescribed sulfonylureas, another class of diabetes medicines.

Treatment status was reassessed every six months. About 26% of GLP-1 users stopped treatment completely, while another 23% had a treatment gap of at least six months before restarting.

The longer people remained off treatment, the greater their cardiovascular risk.

After one year without restarting, the risk of major cardiovascular events was 14% higher than among continuous users. After two years, it was 22% higher.

"There is enormous exuberance about starting GLP-1 drugs, but not nearly enough attention to what happens when people stop," said senior author Ziyad Al-Aly, a clinical epidemiologist at Washington University.

Also read: Are GLP-1 Drugs Safe for Children? Study Finds Nutritional Deficiency in Nearly 17% Within a Year

Continuous Treatment Showed Greater Benefit

People who remained on GLP-1 drugs throughout the three-year study had an 18% lower risk of major cardiovascular events than those taking sulfonylureas.

That translated to about four fewer major cardiovascular events per 100 people over three years.

Those who remained on treatment for two or 2.5 years before stopping also had lower cardiovascular risk, by 7% and 15%, respectively.

However, people who stopped before 18 months did not have a significant reduction in cardiovascular risk compared with those taking sulfonylureas by the end of the study.

Treatment Gaps May Weaken Heart Protection

Interrupting GLP-1 treatment and later restarting it was associated with less cardiovascular protection. While continuous users had an 18% reduction in major cardiovascular events, those who stopped and later restarted had an average 12% reduction.

Even a six-month treatment gap was associated with a 4% to 8% increase in cardiovascular risk compared with continuous use.

Al-Aly said many patients stop GLP-1 drugs because of "cost, side effects or shortages."

"When they stop, it's not just weight that comes back; they experience a resurgence in inflammation, blood pressure, and cholesterol. Weight regain is visible; the metabolic reversal is not," he said.

"Our data suggest this metabolic whiplash is detrimental to heart health," Al-Aly added.

Read More: EXCLUSIVE: GLP-1 Drugs Are The ‘New Statins’, Says University Hospital Birmingham Professor

Can Restarting GLP-1 Drugs Restore The Benefits?

Restarting treatment appeared to restore some cardiovascular protection, but not all of it.

The researchers said the findings suggest that cardiovascular benefits gained during GLP-1 treatment may diminish after treatment is interrupted.

"Clinicians should treat adherence to GLP-1 treatment as an important outcome in its own right -- not an afterthought," Al-Aly said.

He added that healthcare systems should help patients manage side effects, address cost barriers and understand that GLP-1 drugs are used to manage chronic conditions.

The human brain has always been considered to be a single organ. But according to Stanford Medicine researchers, the brain's growth history indicates that it originates from two different populations of early cells that follow separate developmental paths.

The findings, published in Nature Neuroscience, suggest that the front and back parts of the brain develop from different types of progenitor cells, questioning the long-believed idea that the entire brain originates from one common cell.

An important point to note is that this does not mean that humans have two completely separate brains. The two developmental systems ultimately fuse together and function as one brain.

Mechanism Of Brain Forming From Two Cell Groups

During early development, cells called progenitor cells enable the production of specialised brain cells, including neurons. Researchers found that the cells that form the forebrain and midbrain follow a different developmental pathway from those that form the hindbrain, which includes the brainstem.

The forebrain is involved in neurological functions like language, consciousness and higher-level thinking. The hindbrain contains structures that control essential functions including breathing, heartbeat, sleep and swallowing.

Kyle Loh, associate professor of developmental biology at Stanford Medicine and senior author, said, “We’ve shown for the first time that the front of the brain arises from a totally different progenitor cell than the back of the brain.”

He added, “Our discovery means that we can now grow neurons from the back of the brain, the hindbrain, in a petri dish and study their functions.”

Also read: Is Controlling BP Enough After a Brain Bleed? Doctors Explain How to Prevent Another Stroke

A Significant Discovery

Scientists have been able to grow several types of human brain cells in the laboratory, but creating hindbrain neurons has been challenging.

The new findings give some clarity about why some previous approaches were unsuccessful. Researchers were attempting to transform forebrain or midbrain progenitor cells into hindbrain cells, but the Stanford team found that these cells are fundamentally different from the beginning of development.

Jokhai, one of the researchers involved in the study, explained, “Previous attempts to make hindbrain neurons likely tried to coax forebrain and midbrain progenitors into hindbrain cells, which our study shows is not possible.”

He added, “In stem cell biology, people are always fixated with creating the end cell type, like the neuron. But it’s important to begin at the earliest stages of embryonic development.”

Also read: Brain Health After 40: How Diabetes, Menopause And Early Detection Affect Dementia Risk | World Alzheimer’s Day

Study's Implications On Brain Disease Research

The discovery could have various practical implications for neurological research. The hindbrain and brainstem contain neurons involved in breathing, swallowing and movement, meaning researchers need reliable human cells to study diseases that damage these systems.

Stanford scientists say the new findings allowed them to understand and generate functional human hindbrain neurons in the laboratory. This could provide new models for studying conditions like amyotrophic lateral sclerosis (ALS) and spinal muscular atrophy (SMA).

Instead of studying these diseases only in animals or using cells that do not accurately represent the affected brain region, researchers may now be able to grow relevant human neurons and investigate how they develop, malfunction and respond to different treatments.

The finding could therefore change not just how scientists understand the human brain's anatomy, but how they understand its origins and how they recreate different brain regions in the laboratory.

As artificial intelligence (AI) rapidly reshapes health research and clinical care, Anthropic and medical knowledge platform OpenEvidence have announced a collaboration to provide AI-powered clinical decision support to healthcare providers in dozens of low- and middle-income countries.

OpenEvidence answers doctors’ clinical questions using peer-reviewed research and treatment guidelines. The platform, already free to clinicians in the US and Europe, will now be extended to providers in countries where access to medical literature, specialist expertise and continuing education may be limited.

The initiative is being rolled out in about 100 countries, including Uganda, Angola, Sudan, Haiti and Mongolia, according to a list provided by OpenEvidence.

“Access to medical knowledge shouldn’t depend on geography,” OpenEvidence founder Daniel Nadler said.

How Can Clinical AI Help Doctors?

Also read: Anthropic Says Claude Blocked Bird Flu, Chikungunya Research: How AI Could Enable Biological Misuse

Rajeev Jayadevan, former president of IMA Cochin and convener of the Research Cell, Kerala, told HealthandMe that doctors in Western countries already have access to OpenEvidence, which has also been incorporated into electronic medical records at major US hospitals.

He said the platform could help doctors quickly access medical evidence at the point of care.

“The advantage is that the doctor will have instant access to relevant research papers published in multiple journals, all at once, while asking a specific question about a patient while at the bedside,” Dr Rajeev said.

He called it “a major time-saving step for the busy clinician, who may not have the time to go to the library, sit down with medical journals or textbooks, or even do internet-based searches to look for relevant papers”.

Dr Rajeev also said wider access to medical information could help reduce inequalities in healthcare.

“Medical information that directly helps patients must be available to the doctor free of charge,” he said.

However, AI-based clinical tools also have limitations, particularly when used across different healthcare systems.

“Potential disadvantages include the need for doctors to be trained in digital health literacy and the fact that the answer provided by AI algorithms heavily depends on what the doctor asked for, and in what format,” Dr Rajeev said.

“Some of these issues can be ironed out by using templates, so that important information or keywords are not missed by the doctor who enters the query. Other risks include relying on data from other countries and treatment options that are unavailable or expensive for the local community,” he added.

What Are the Ethical Risks of Clinical AI?

Read More: Google DeepMind’s AlphaGenome Atlas Maps 9 Billion Mutations in Human Genome; To Reveal Genetics, Rare Disease Clues

The expansion of AI in healthcare comes as the WHO has separately called for stronger ethics oversight of AI-related health research.

In a report published this week, WHO warned that rapid advances in AI are creating challenges around privacy, bias, fairness, transparency and accountability. It said existing research ethics systems may need additional expertise to assess risks associated with AI and large datasets.

Dr Rajeev said AI-driven research, particularly when it involves large datasets crossing national borders, may require ethics committees to include experts in computer science, data ethics and bioinformatics.

“Ethics Committees by themselves will need to be upgraded to incorporate computer scientists, data ethicists, and bioinformaticians.”

WHO also said oversight should extend beyond ethics committees to researchers, funders, scientific journals, data governance bodies, professional societies and regulators.

Why This Matters

For low- and middle-income countries, clinical AI could help bridge gaps in access to medical literature and specialist knowledge. At the same time, WHO has highlighted the need for local capacity and safeguards to ensure that AI systems developed largely in higher-income settings are appropriate for different healthcare systems and do not introduce new inequities.

Dr. Rajendra Pratap Gupta, chairman, Health Parliament and co-chair, Global Policy Network on AI, UN Internet Governance Forum told HeathandMe that accountability is key when it comes to AI in healthcare.

“Anthropic’s partnership with OpenEvidence shows that clinical AI is moving rapidly from experimentation to global deployment. WHO’s warning reminds us that innovation cannot outrun ethics. In my view, accountability must be the central pillar of AI governance, because when technology influences clinical decisions, responsibility for evidence, safety and patient outcomes must never become blurred,” added the expert.