CRS Research Digest • Issue 01
June 9, 2026
NCCN Guidelines Expand to 60+ Tumor Types
Innovation
The National Comprehensive Cancer Network (NCCN) has grown its Clinical Practice Guidelines from just 8 tumor types in 1996 to more than 60 tumor types, subtypes, and clinical topics today. At their 2026 Annual Conference, oncologists presented key updates and new recommendations across a range of cancers. This matters because these guidelines shape how doctors treat patients worldwide, and they grow more precise every year.
How it was studied
This wasn't a single study, but rather a collection of expert consensus updates presented at the NCCN 2026 Annual Conference. Oncologists reviewed the latest research and clinical evidence for each cancer type and updated the treatment recommendations accordingly.
Read next
Visit the full NCCN Guidelines at nccn.org for your cancer type of interest. The updates cover everything from early detection to advanced disease management.
AI and data lesson
No AI was used to generate the guidelines themselves, but AI is increasingly being integrated into the diagnostic and treatment-planning tools that oncologists use alongside these guidelines.
Why this matters for students
These guidelines are the backbone of modern cancer care. As you read research papers or hear about new treatments, they're being evaluated against these evidence-based standards. Understanding that guidelines evolve shows how cancer care gets better over time.
AI and Chest CT Scans Could Transform Lung Cancer Screening
Innovation
Researchers propose that low-dose CT scans obtained during routine lung cancer screening should become a public research resource. Why? Because AI trained on millions of these scans could do more than just find lung nodules. It could also flag emphysema, coronary artery disease, atrial fibrillation, and other conditions hiding in the same image. One scan, multiple early warnings.
How it was studied
This is a perspective paper published in Frontiers in Oncology. The authors aren't reporting new data from a trial, but rather proposing a strategy to unlock data we already have. They argue that sharing these CT scans (with patient privacy protected) would accelerate AI development.
Read next
Look up the paper in Frontiers in Oncology, "Beyond Nodule Detection: AI, Chest CT, and a Vision for Population Health."
AI and data lesson
This is a perfect example of how AI needs data to learn. The models are trained on thousands of real chest CT scans. The databases here are hospital CT imaging archives and lung cancer screening programs. The AI approach is supervised learning: show the model a scan, tell it what diseases are present, and it learns to recognize the patterns.
Why this matters for students
This shows how one dataset can power multiple discoveries. It's also a real-world example of the privacy vs. progress tradeoff in medical AI, and it shows why researchers push for secure data-sharing frameworks.
GLP-1 Drugs May Lower Obesity-Related Cancer Risk
Innovation
New research suggests that glucagon-like peptide-1 receptor agonists (GLP-1 RAs), drugs first developed for diabetes but now famous for weight loss, may lower the risk of developing obesity-related cancers (like colon, breast, and endometrial cancer) in obese people without diabetes, at least in the short term.
How it was studied
This is observational research showing an association between GLP-1 RA use and lower cancer risk. The researchers looked at health records of obese, nondiabetic patients who started GLP-1 RAs and compared their cancer rates to similar patients who didn't use the drug.
Read next
Search for recent papers on GLP-1 RAs and cancer risk. The field is moving quickly, so there are several 2024-2026 studies on this topic.
AI and data lesson
No AI was explicitly used here; this is epidemiological analysis. But large-scale health databases (EHR systems, insurance claims databases) are increasingly analyzed with machine learning to spot patterns and control for confounding factors (things that might explain the results besides the drug itself).
Why this matters for students
This is how we discover new uses for existing drugs. It's also a reminder that weight loss itself is preventive, and that sometimes a drug's side effect (here, dramatic weight loss) matters as much as its original purpose.
Alcohol and Health: No Safe Level
Innovation
A large study published in the Journal of Studies on Alcohol and Drugs found that there is no safe level of alcohol consumption when it comes to cancer risk. Even low levels of drinking are associated with elevated health risks. Higher consumption increases the risks of cancer, cardiovascular disease, and death/disability progressively.
How it was studied
This was a comprehensive analysis looking at health outcomes across different levels of alcohol consumption. The researchers reviewed existing data on alcohol's effects across multiple diseases and used statistical modeling to estimate risk.
Read next
Find the paper by George et al in the Journal of Studies on Alcohol and Drugs. The findings challenge older research that suggested small amounts of alcohol were protective.
AI and data lesson
This kind of meta-analysis often uses AI-assisted text mining to identify and extract data from thousands of studies. Machine learning can also help identify confounding variables (things other than alcohol that might explain the results).
Why this matters for students
This is an example of how science evolves. Earlier research suggested light drinking was harmless, but newer, larger analyses contradict that. It's a reminder to always check the most recent evidence.
Digital Tool Boosts Patient Engagement in Clinical Trials
Innovation
The Alliance for Clinical Trials in Oncology developed a new online tool called the Participant Engagement Portal (PEP). In a pilot test, 84% of participants reported a positive experience using it. The tool helps patients understand trials, track their own health data, and stay connected to their research team.
How it was studied
This was a pilot project testing the new portal with a group of clinical trial participants. The researchers measured user satisfaction and engagement through surveys and usage data.
Read next
Look for publications from the Alliance for Clinical Trials in Oncology about PEP and patient engagement in trials.
AI and data lesson
No AI was mentioned directly in this item, but modern patient portals often use AI for personalized messaging, appointment reminders, and flagging health data that might need attention.
Why this matters for students
If you ever participate in a clinical trial, tools like this make the experience better and help researchers collect better data. It's also a reminder that technology can make medicine more patient-centered.