On August 19, 2026, Merck and Moderna announced that the first Phase 3 trial of an individualized cancer vaccine had met its endpoints in melanoma — and Moderna's share price rose 177% in one trading session, the biggest single-day percentage gain by an S&P 500 company this century [2][23]. This article explains what a personalized cancer vaccine is, why every previous attempt failed for 135 years, what the new trial did and did not show, and which cancers could be next.
Table of Contents
- What happened on August 19, 2026?
- What is a personalized cancer vaccine?
- Why did cancer vaccines fail for 135 years?
- How is a neoantigen vaccine made?
- What did the KEYNOTE-942 and INTerpath-001 trials show?
- Is it a cure? Can it prevent cancer?
- Could it work for other cancers?
- Cost and availability
- References
What happened on August 19, 2026?
The INTerpath-001 trial enrolled 1,137 patients whose Stage 2B–4 melanoma had been completely removed by surgery, leaving a high risk of recurrence. Participants were randomized two-to-one, double-blind, to receive either intismeran autogene — an mRNA vaccine custom-built against each patient's own tumor mutations — or a matching placebo. Both groups also received pembrolizumab (Keytruda), so any benefit came on top of standard treatment rather than instead of it [23].
The trial met its primary endpoint of recurrence-free survival and its key secondary endpoint of distant metastasis-free survival, with no new safety signals. The detailed effect sizes have not yet been released; they are being held for presentation at a medical congress [23].
The companies describe the result as the first positive Phase 3 trial for any individualized neoantigen therapy, and the first for any mRNA-based cancer treatment. Principal investigator Professor Georgina Long of Melanoma Institute Australia called it the first Phase 3 study to show that a therapy designed on "the unique mutational 'fingerprint' of a patient's own tumor" can reduce the risk of recurrence or death [23].
What is a personalized cancer vaccine?
A therapeutic cancer vaccine does not prevent infection the way a flu shot does. Its job is to hold up a molecular target and instruct the immune system's T cells to attack anything carrying that target — in this case, a patient's own tumor cells.
The "personalized" part comes from a 2015 insight by immunologists Ton Schumacher and Robert Schreiber. Tumor mutations create brand-new proteins, called neoantigens, that exist nowhere else in the body. Because the immune system's self-tolerance screening never encountered them, the T cells capable of attacking them still exist — they simply need a target list [15].
The catch: one patient's tumor mutations are not another's. Every target list is unique, which means an off-the-shelf product cannot work. Each dose has to be designed and manufactured for a single person.
Why did cancer vaccines fail for 135 years?
The idea is old. In the 1890s, New York surgeon William Coley observed tumors regressing after severe bacterial infections — including sarcoma patient Fred Stein, found alive and cancer-free seven years after being declared hopeless — and treated roughly a thousand patients with his heat-killed bacterial "toxins." A few responded remarkably; most did not, and nobody could predict which would be which [1][3][4][5]. In 1894 the Journal of the American Medical Association judged the toxins "an entire failure" as a cure, and immunotherapy spent decades on the margins [3].
Two observations kept the idea alive. In 1976, urologist Alvaro Morales instilled the BCG tuberculosis vaccine into the bladders of nine bladder-cancer patients; fifty years later it remains standard care for high-risk early bladder cancer [6]. And at the National Institutes of Health, Steven Rosenberg — who as a resident in 1968 had encountered a veteran whose metastatic stomach cancer had vanished without treatment — showed by 1988 that T cells extracted from a patient's melanoma, multiplied in the lab, and reinfused could shrink tumors [7][8].
Yet the vaccines built on that promise kept failing. A landmark 2004 analysis found an objective response rate of just 3.3% across 1,306 cancer-vaccine treatments [9]. The 2000s and 2010s added four large, expensive failures:
- MAGE-A3: 2,312 lung-cancer patients, disease-free survival identical to placebo (hazard ratio 1.02) [10]
- Canvaxin: halted Phase 3; final data showed median survival of 31.4 months with the vaccine versus 38.6 months with placebo [11]
- GVAX: Phase 3 in advanced prostate cancer halted early [12]
- PROSTVAC: a promising secondary survival signal in Phase 2 (its primary endpoint was missed), followed by a 1,297-man Phase 3 stopped for futility — survival identical across all arms [13][14]
The unifying explanation is central tolerance. Before birth, the immune system deletes T cells that react against the body's own proteins. Most failed vaccines pointed the immune system at proteins that tumors share with normal tissue — targets the relevant T cells had already been deleted for. Neoantigen vaccines were designed to solve exactly that flaw [15].
How is a neoantigen vaccine made?
Three steps, repeated from scratch for every single patient.
Step one: sequence the patient's tumor and healthy blood, and subtract. Any protein-coding mutation present in the tumor but not the blood joins the candidate list.
Step two: predict which mutant fragments the patient's immune system can actually display on the cell surface — an invisible target is a useless one. Neural-network models perform this prediction. The best-known public tool, NetMHCpan, has been in development since 2007 and was trained on more than 850,000 measured protein fragments by 2017 and over 13 million data points by 2020 [16][17]. Moderna's own selection algorithm is proprietary; what is public is its task — predict what is displayable and select up to 34 targets.
Step three: encode up to 34 selected neoantigens on a single strand of mRNA, wrap it in a lipid nanoparticle, and inject it [18].
The approach was first tested in humans in July 2017, in two small Nature studies published side by side. In one, 13 melanoma patients received personalized RNA vaccines; all eight who were disease-free at vaccination remained recurrence-free for up to 23 months [19]. In the other, a Boston team used protein fragments instead of RNA in six patients: four stayed recurrence-free at 25 months, and the two who progressed had complete tumor regression once an anti-PD-1 checkpoint drug was added [20].
What made scaling possible was the COVID-19 pandemic, which industrialized the exact mRNA manufacturing technology these vaccines needed [21].
What did the KEYNOTE-942 and INTerpath-001 trials show?
KEYNOTE-942 was the Phase 2b trial: 157 patients with resected melanoma, randomized to the personalized mRNA vaccine plus pembrolizumab or pembrolizumab alone. In December 2022 it met its pre-committed primary endpoint, with recurrence risk reduced by roughly 44% [22]. In the longer follow-up reported alongside the Phase 3 announcement, the reductions were 49% for recurrence or death and 59% for distant metastasis or death [23].
INTerpath-001 is the Phase 3 confirmation: 1,137 patients, double-blind, placebo-controlled. It met both its recurrence-free survival and distant metastasis-free survival endpoints; the exact hazard ratios are pending presentation at a medical congress [23]. Until those numbers are public, the honest summary is: the effect is real enough to clear a rigorous pre-specified statistical bar, and its precise size is not yet known.
Is it a cure? Can it prevent cancer?
Neither. Intismeran autogene is a treatment for people who already had melanoma removed — it lowers the risk of the cancer returning. It does not prevent melanoma, and "cure" is not a word the evidence supports.
A genuinely preventive cancer vaccine does already exist: the HPV vaccine, which blocks the virus behind most cervical cancer. In a Swedish registry study of 1.67 million women, invasive cervical cancer was diagnosed in 19 vaccinated versus 538 unvaccinated women, and those vaccinated before age 17 had an 88% lower risk [24]. In the United States, cervical-cancer deaths in women under 25 have fallen by more than 60% in a decade [25].
Could it work for other cancers?
Melanoma is among the most heavily mutated of all cancers — the most raw material for neoantigens — which makes it, in theory, one of the easier cancers for this approach. The open question is whether the result transfers to tumors with fewer mutations.
Merck and Moderna are running nine trials of this therapy: two in melanoma, four in lung cancer (three already at Phase 3), one in kidney, and two in bladder — one of which pairs the mRNA therapy with BCG, the same tuberculosis vaccine first instilled into bladders in 1976 [26]. The wider field is larger still: in May 2026, Nature Reviews Drug Discovery mapped the entire clinical pipeline of therapeutic cancer vaccines across dozens of tumor types [27].
That pipeline, more than the melanoma result alone, is why markets re-priced Moderna so dramatically: the bet is that a personalized platform proven in one cancer can be translated to many [2].
Cost and availability
The therapy is not yet FDA approved and is not available outside clinical trials. Pricing has not been announced, and manufacturing a bespoke product per patient is inherently expensive — though costs of sequencing and mRNA production have fallen steeply and continued innovation is expected now that the approach has a Phase 3 validation.
What to watch next: the full INTerpath-001 data at a medical congress, regulatory filings, and readouts from the lung, kidney, and bladder trials. If those repeat the melanoma pattern, August 19, 2026 will be remembered as the day a 135-year-old idea finally worked.
References
1. https://www.cancerresearch.org/blog/the-legacy-of-bessie-dashiell
3. https://pmc.ncbi.nlm.nih.gov/articles/PMC1888599/
4. https://www.thepharmacologist.org/william-coley
5. https://pmc.ncbi.nlm.nih.gov/articles/PMC7232517/
6. https://journals.asm.org/doi/10.1128/cmr.00194-23
7. https://pmc.ncbi.nlm.nih.gov/articles/PMC8210396/
8. https://pmc.ncbi.nlm.nih.gov/articles/PMC6237474/
9. https://pmc.ncbi.nlm.nih.gov/articles/PMC1435696/
10. https://pubmed.ncbi.nlm.nih.gov/27132212/
11. https://link.springer.com/article/10.1245/s10434-017-6072-3
12. https://www.fiercebiotech.com/biotech/cell-genesys-halts-vital-2-gvax-trial-advanced-prostate-cancer
13. https://pubmed.ncbi.nlm.nih.gov/20100959/
14. https://pmc.ncbi.nlm.nih.gov/articles/PMC6494360/
15. https://www.science.org/doi/10.1126/science.aaa4971
16. https://academic.oup.com/jimmunol/article/199/9/3360/7977122
17. https://academic.oup.com/nar/article/48/W1/W449/5837056
18. https://doi.org/10.1016/S0140-6736(23)02268-7
19. https://www.nature.com/articles/nature23003
20. https://www.nature.com/articles/nature22991
21. https://www.cnn.com/2021/11/12/health/covid-cancer-biontech-ugur-sahin/index.html
24. https://www.nejm.org/doi/full/10.1056/NEJMoa1917338
25. https://jamanetwork.com/journals/jama/fullarticle/2827212
