2026 has been one of the most remarkable years in cancer treatment in a long time. A personalized mRNA vaccine beat the standard of care in a large melanoma trial. A daily pill roughly doubled survival in advanced pancreatic cancer. And blood tests are helping doctors decide who needs chemotherapy and who can safely skip it. Here’s what the research actually shows, including the setbacks.
Prefer to listen? This article goes with Episode 3 of The Larson Health Podcast on YouTube.
Quick takeaways
-
Personalized mRNA vaccines work, at least in melanoma. Adding a custom vaccine to Keytruda cut the risk of melanoma coming back or death by 42% in a phase 3 trial of 1,137 patients.
-
Breast cancer vaccines look promising too. In 14 women with triple-negative breast cancer, a personalized mRNA vaccine triggered immune responses in all of them, and 11 stayed relapse-free for up to six years.
-
A pancreatic cancer vaccine shows durable results. In a small early trial, 7 of 8 patients whose immune systems responded were still alive four to six years after surgery.
-
Not every vaccine trial succeeds. The same pancreatic vaccine failed when given alone in colorectal cancer, and that trial was stopped.
-
The “undruggable” RAS gene has been hit. A pill called daraxonrasib roughly doubled median survival in previously treated metastatic pancreatic cancer, from 6.7 to 13.2 months.
-
“Guided missile” drugs are moving earlier. The antibody-drug conjugate Enhertu kept HER2-positive metastatic breast cancer in check for a median of more than three years.
-
Blood tests can spare people from chemo. In colon cancer, a tumor DNA blood test nearly halved chemotherapy use without hurting survival.
-
Some rectal cancers can vanish with immunotherapy alone. In one trial, dostarlimab cleared the tumor in all 49 patients with mismatch repair deficient rectal cancer, with no chemo, radiation or surgery.
-
None of this replaces your care team. These are questions to bring to your oncologist, not decisions to make alone.
Most vaccines prevent disease. Therapeutic cancer vaccines are different: they’re given to people who already have cancer, usually after surgery, to help the immune system find and destroy any cancer cells left behind.
Why does that matter? Even when surgery removes a tumor completely, microscopic cancer cells may already have spread. Doctors call this minimal residual disease, and it’s the reason cancer can come back months or years later.
Here’s how a personalized mRNA cancer vaccine is made:
-
The tumor is sequenced. After surgery, the lab reads the tumor’s DNA and compares it with the patient’s healthy DNA.
-
Targets are chosen. Cancer cells make abnormal proteins called neoantigens that healthy cells don’t. Computer algorithms pick the ones most likely to trigger a strong immune response. One pancreatic vaccine encodes up to 20 of them.
-
The vaccine is built. Those targets are written into messenger RNA, the same basic technology used in COVID-19 vaccines. Production takes about six weeks.
-
The immune system learns. After the shot, the body’s cells make small pieces of the neoantigens, and the immune system learns to hunt any cancer cell carrying them.
These vaccines are usually paired with a checkpoint inhibitor like Keytruda (pembrolizumab). Cancer cells use an “off switch” to stop immune cells from attacking. Checkpoint inhibitors block that switch. A simple way to picture it: the vaccine shows the immune system what to attack, and the checkpoint inhibitor takes the brakes off so it can.
The melanoma breakthrough: INTerpath-001
On August 19, 2026, Merck and Moderna announced that their personalized mRNA vaccine, intismeran autogene, succeeded in a phase 3 trial. It was the first positive phase 3 result ever for an mRNA cancer treatment and for any personalized neoantigen therapy.
Who was in the trial
-
1,137 people with high-risk melanoma (stage IIB to IV) whose tumors had been completely removed
-
Randomly assigned 2 to 1 to get the vaccine plus Keytruda, or a placebo plus Keytruda (today’s standard treatment)
-
Up to nine vaccine doses, every three weeks
What it found (full results presented at the ESMO conference in October 2026)
-
42% lower risk of the cancer coming back or death (hazard ratio 0.58)
-
82% vs 70% of patients were recurrence-free at two years
-
88.5% vs 80.2% were free of distant spread at two years
-
The benefit appeared after the first six months, once the vaccine course was done. From then on, the risk was 58% lower. That fits how the immune system works: it takes time to learn new targets.
This builds on an earlier phase 2 trial, KEYNOTE-942, where the benefit held up over five years of follow-up.
What’s still unknown: Overall survival is still being measured, and the vaccine isn’t approved yet. One expert told ABC News he expects approval, likely by early 2027. Cost and the logistics of making a custom vaccine for every patient are real open questions. A similar phase 3 trial is now underway in lung cancer.
A breast cancer vaccine: early but encouraging
Triple-negative breast cancer makes up about 15% of breast cancers. It lacks the three receptors many breast cancer drugs target, tends to affect younger women, and often comes back early.
In February 2026, BioNTech and researchers in Mainz, Germany, published a first-in-human study of a personalized mRNA vaccine for it in the journal Nature.
-
14 women with early-stage triple-negative breast cancer got the vaccine after surgery and chemotherapy
-
All 14 developed immune cells that recognized their tumor targets, and most responded to several targets at once
-
11 of 14 stayed relapse-free for up to six years
-
One woman with a weaker vaccine response went into complete remission after a checkpoint inhibitor, another sign that the combination matters
The caveat: 14 patients with no comparison group is a small, exploratory study. But it suggests the approach may work well beyond melanoma.
Pancreatic cancer: a vaccine that keeps working
Pancreatic cancer has a five-year survival rate of about 13%, and it usually returns even after surgery. A team at Memorial Sloan Kettering, led by Dr. Vinod Balachandran, has been testing a personalized mRNA vaccine called autogene cevumeran (from BioNTech and Genentech).
The trial
-
16 patients had their tumors surgically removed
-
They then got the vaccine, a checkpoint inhibitor (atezolizumab) and standard chemotherapy
-
Half of them, 8 of 16, had a strong immune response to the vaccine
Six-year results (AACR meeting, April 2026)
-
7 of 8 responders (about 88%) were alive four to six years after surgery
-
Only 2 of 8 non-responders were alive, with a median survival of 3.4 years
-
Responders still had active cancer-fighting T cells up to six years later, acting like long-lived immune memory
The important caveat: 16 patients is very small, and the study wasn’t randomized. People whose immune systems responded might have done better anyway. A large randomized trial, IMCODE003, is now testing the vaccine with atezolizumab and chemotherapy against chemotherapy alone.
The setback: when the same vaccine failed
In August 2026, BioNTech and Genentech stopped a phase 2 trial of the same vaccine in colorectal cancer.
-
It tested the vaccine alone, without a checkpoint inhibitor, in people whose colorectal cancer was removed but who still had tumor DNA in their blood
-
An independent safety board found the vaccine was unlikely to show a benefit
-
Genentech confirmed more deaths in the vaccine group than in the comparison group
-
No new safety signals were found for the vaccine itself
Why the difference? Researchers point to two factors. Colorectal cancers like these are often “immunologically cold”, meaning they hide well from the immune system. And the successful trials paired the vaccine with a checkpoint inhibitor, while this one used the vaccine alone.
The lesson: a promising early result is a reason to run a bigger trial, not proof that something works everywhere.
Targeting the “undruggable”: daraxonrasib
The RAS genes, including the famous KRAS, act like a switch that tells cells to grow. When they mutate, the switch can get stuck “on.” RAS mutations drive a huge share of pancreatic cancers. For decades, scientists considered RAS proteins nearly impossible to block with drugs.
Daraxonrasib, from Revolution Medicines, is a once-daily pill designed to block RAS while it’s switched on, across many different RAS variants.
The RASolute 302 trial (presented at ASCO in May 2026 and published in the New England Journal of Medicine)
-
500 people with metastatic pancreatic cancer that had progressed after earlier treatment
-
Randomly assigned to daraxonrasib or standard chemotherapy
What it found
-
60% lower risk of death
-
Median survival of 13.2 months vs 6.7 months, roughly double
-
About 32% vs 11% saw their tumors shrink
-
Fewer serious side effects: about 11% vs 19%. The most common severe side effects of the pill were rash and mouth sores.
-
Patients’ pain, overall health and quality of life held up longer
Perspective: This isn’t a cure, and 13 months is still too short. But in a disease where progress has been slow, roughly doubling survival is a milestone. The FDA has authorized an expanded access program, and three more phase 3 trials are testing the drug earlier in pancreatic cancer and in lung cancer.
“Guided missiles”: antibody-drug conjugates
Traditional chemotherapy travels through the whole body, which is why it causes hair loss, nausea and low blood counts. An antibody-drug conjugate (ADC) links a chemotherapy drug to an antibody that latches onto a specific marker on cancer cells. The antibody acts like a homing device, delivering the chemo straight to the tumor.
The best-known ADC is trastuzumab deruxtecan (Enhertu), which targets a marker called HER2. It’s already approved in more than 75 countries for certain breast cancers.
The DESTINY-Breast09 trial (2025)
-
About 1,160 patients with HER2-positive breast cancer that had spread, getting their first treatment
-
Enhertu plus the antibody pertuzumab, compared with the long-standing standard of chemo plus two antibodies
What it found
-
44% lower risk of the cancer progressing or death
-
Median of about 40.7 months vs 26.9 months before the cancer worsened, more than three years with Enhertu
-
More side effects: 5 treatment-related deaths vs 1 with standard treatment. Doctors watch closely for a known risk of serious lung inflammation.
-
It’s not yet known whether using it first helps people live longer overall
These drugs are also moving earlier. A second trial, DESTINY-Breast05, found that Enhertu given after surgery helped more women with high-risk early HER2-positive breast cancer stay free of invasive cancer than the current standard. Dozens of other ADCs are in development.
A blood test that decides who needs chemo
After colon cancer surgery, doctors often recommend chemotherapy to kill hidden cancer cells. But many patients, especially with stage II colon cancer, are already cured by surgery alone. They go through months of side effects for no benefit, and until recently there was no reliable way to tell who needed chemo.
Circulating tumor DNA (ctDNA) testing changes that. After surgery, a blood test looks for DNA fragments carrying that patient’s specific tumor mutations. If they’re still there, cancer cells likely remain. If not, surgery likely got it all.
The DYNAMIC trial (New England Journal of Medicine, 2022; five-year results in Nature Medicine, 2025)
-
455 people with stage II colon cancer after surgery
-
Some had chemo decisions made the usual way; others got chemo only if their blood test showed ctDNA
What it found
-
Chemotherapy use dropped from about 28% to about 15%
-
Two-year recurrence-free rates were about the same: 93.5% vs 92.4%
-
Five-year overall survival was about 94% in both groups
-
Without treatment, ctDNA-positive patients had an estimated recurrence risk above 80%, so the test found the people who truly needed chemo
Smarter care sometimes means less care. A follow-up trial, DYNAMIC-III, has tested the approach in stage III colon cancer, and ctDNA tests are being studied across many cancers.
The rectal cancer trial where every tumor disappeared
Some tumors are mismatch repair deficient (dMMR), meaning the cell’s DNA “spell-checker” is broken. This is common in people with Lynch syndrome, an inherited condition. These tumors carry so many mutations that they’re unusually visible to the immune system, which makes them very sensitive to checkpoint inhibitors. About 5–10% of locally advanced rectal cancers are dMMR.
Standard treatment for locally advanced rectal cancer usually means chemotherapy, radiation and surgery. That can cause lasting bowel, bladder and sexual problems, and sometimes a permanent colostomy bag.
What Memorial Sloan Kettering tried: Dr. Andrea Cercek’s team gave patients the checkpoint inhibitor dostarlimab every three weeks for six months, before anything else.
What happened
-
In the first 14 patients (New England Journal of Medicine, 2022), every tumor disappeared, with no chemo, radiation or surgery
-
By 2025, that 100% complete response held across 49 patients, and about 92% were still cancer-free two years later
-
About 65% of patients with other dMMR cancers also had complete responses, and the vast majority of responders avoided surgery
-
Tumor DNA in the blood typically disappeared in about six weeks, months before scans confirmed it
-
In a larger global trial, AZUR-1 (October 2026), 76% had no remaining cancer after dostarlimab alone. That’s lower than the original results, which is common as treatments reach larger groups, but still remarkable.
The FDA gave dostarlimab Breakthrough Therapy designation for this use in December 2024. It’s a powerful example of why tumor genetic testing matters.
What’s coming next
-
CAR-T for solid tumors. CAR-T engineers a patient’s own immune cells to attack cancer. It already works well in some leukemias and lymphomas. In a small New England Journal of Medicine study, three patients with recurrent glioblastoma saw their brain tumors shrink rapidly, though the tumors eventually returned. Newer “armored” CAR-T cells cleared glioblastoma in many mice in 2026 research, but mouse results often don’t hold up in people.
-
Metabolic health during treatment. Health-records studies found colorectal cancer patients on GLP-1 drugs had better survival. Randomized trials are needed to know whether adding a GLP-1 to cancer treatment truly helps.
-
Artificial intelligence. AI already helps read blood test patterns, choose vaccine targets, and analyze scans and pathology slides.
Questions to ask your oncologist
If you or someone you love is going through cancer treatment, these are worth bringing up:
-
Has my tumor been genetically tested, and does it have mutations a targeted drug could address?
-
Would a ctDNA blood test help guide my treatment after surgery?
-
Are there clinical trials I might qualify for, including personalized vaccine trials? (You can also search clinicaltrials.gov with your care team.)
-
What are the side effects and quality-of-life tradeoffs of each option?
Frequently asked questions
Is there a cancer vaccine I can get now? Preventive vaccines for HPV and hepatitis B are widely available and prevent the cancers those viruses cause. Personalized mRNA vaccines that treat cancer are still investigational. The melanoma vaccine may be approved as soon as early 2027, according to one expert, but it isn’t available outside clinical trials yet.
How is an mRNA cancer vaccine different from the COVID vaccine? Both use messenger RNA to teach the immune system a target. The COVID vaccine teaches everyone the same virus protein. A personalized cancer vaccine is custom-made from each patient’s own tumor mutations, so no two are the same.
Do personalized cancer vaccines work for every cancer? Not so far. They showed strong results in melanoma and promising early results in pancreatic cancer, both when paired with a checkpoint inhibitor. A trial giving the vaccine alone in colorectal cancer was stopped after more deaths in the vaccine group.
What is daraxonrasib, and who can get it? It’s an experimental daily pill that blocks RAS, a gene that drives most pancreatic cancers. In a phase 3 trial it roughly doubled median survival in previously treated metastatic pancreatic cancer. It isn’t fully approved yet, but the FDA has authorized an expanded access program for some eligible patients. Ask an oncologist whether you qualify.
What is a ctDNA test? A blood test that looks for tiny fragments of DNA from a patient’s tumor. After surgery, it can show whether cancer cells are likely still in the body. In stage II colon cancer, using it to guide chemotherapy nearly halved chemo use without hurting survival.
What is an antibody-drug conjugate? A “guided missile” drug that attaches chemotherapy to an antibody that seeks out a marker on cancer cells, so more of the drug reaches the tumor. Enhertu, which targets HER2, is the best-known example. These drugs can still cause serious side effects, including lung inflammation.
Can immunotherapy replace surgery for rectal cancer? For some patients, possibly. In people with mismatch repair deficient rectal cancer, the drug dostarlimab cleared the tumor in every patient in one trial, letting them skip chemo, radiation and surgery. This applies only to that specific tumor type, so ask whether your tumor has been tested for mismatch repair status.
Can GLP-1 drugs help during cancer treatment? It’s an open question. Some health-records studies found better survival in colorectal cancer patients taking GLP-1s, but those studies can’t prove cause and effect. Randomized trials are needed, and GLP-1s aren’t approved to treat cancer.
This post is for education, not medical advice. Talk with your care team before making treatment decisions.
Listen to the full series
Want the full story? Episode 3 of The Larson Health Podcast walks through every study in detail, and it wraps up our three-part series on the evolving science of cancer: early detection, prevention and treatment. Watch and subscribe on YouTube at @Thecoachkimlarson.
Brought to you by Larson Nutrition, personalized metabolic health coaching that starts with a free health assessment, and Larson Botanicals, small-batch organic skincare and soy-coconut candles made for sensitive skin. Their gift packages and candles make great holiday gifts.