Surgeons at Hiroshima University Hospital counted the teeth in 339 patients before removing part or all of their pancreas, and the number they wrote down turned out to track how long those patients lived.
Patients who still had 21 or more natural teeth survived a median of 60.7 months after surgery for pancreatic ductal adenocarcinoma. Patients with 20 or fewer survived a median of 39.1 months. That is a gap of nearly two years in a disease whose expected course is about two years.
The work, led by first author Kaoru Seo with Kenichiro Uemura as principal investigator, was published in April in the Journal of Hepato-Biliary-Pancreatic Sciences and described in a later announcement from Hiroshima University. It covers consecutive patients who had radical pancreatectomy between 2014 and 2022, each of whom received a dental examination before the operation.
The association held after the researchers adjusted for tumor stage, lymph node metastasis, surgical margins, chemotherapy, and nutritional status. It is still an association. Nobody has shown that keeping teeth extends survival, and the researchers do not claim that it does.
The 21-Tooth Line Sits Almost Exactly Where Dentists Draw Theirs
The cutoff is the detail worth pausing on. The Hiroshima team split its patients at 21 teeth. American dental epidemiology already has a line in nearly the same place.
The Centers for Disease Control and Prevention defines loss of functional dentition as having fewer than 20 teeth remaining and severe tooth loss as having 8 or fewer. Adults have 32 permanent teeth, and third molars are commonly extracted, so a healthy older mouth typically holds 28.
That threshold captures a large slice of the population most likely to be diagnosed with this cancer. In CDC survey data covering 2017 through March 2020, about 1 in 10 U.S. adults aged 65 to 74 had lost all their teeth, rising to about 1 in 5 among those 75 and older.
Pancreatic cancer is most frequently diagnosed between ages 65 and 74, according to the National Cancer Institute. The two curves overlap almost perfectly.
Chewing Ability Failed to Explain the Survival Gap
The obvious explanation would be nutrition. Fewer teeth, worse chewing, worse eating, worse recovery from one of the most physiologically punishing operations in surgery.
The data did not cooperate. The researchers also assessed posterior occlusion, meaning whether the premolars and molars actually met when the patient bit down. That measure of chewing function was not independently associated with survival.
Uemura called that result surprising. "This suggests that tooth loss may represent more than oral function alone," he said, describing the count as a possible reflection of lifelong chronic inflammation, frailty, nutritional condition, lifestyle factors and systemic vulnerability accumulated over decades.
That reframes what is being measured. If chewing is not doing the work, then the teeth are functioning less as equipment and more as a record. They are a physical ledger of decades of gum disease, decay, smoking, dental access, and the money to pay for it.
What a Single Japanese Center Can and Cannot Show
The limitations are not incidental to the finding. All 339 patients were from a single hospital and were of Japanese ethnicity. The researchers say the results need to be tested in larger multicenter and international studies before anyone builds a risk score around them.
Confounding is the specific worry. Tooth count correlates with income, education, and smoking, and each of those independently predicts cancer outcomes. Adjusting for tumor stage and nutritional status does not remove socioeconomic position from the picture.
Median survival of 60.7 months after resection is also well above typical figures. That reflects a selected group: people healthy enough to undergo radical pancreatectomy at an academic center. Nothing here describes the majority of patients, most of whom are diagnosed too late for surgery.
Pancreatic cancer remains the third leading cause of cancer death in the United States. The American Cancer Society projects 67,530 new cases and 52,740 deaths for 2026.
Why Oral Health Keeps Turning Up in Cancer Data
This is not the first time the mouth has been a focus of cancer research. Uemura noted that periodontal pathogens, including Porphyromonas gingivalis, have been proposed to contribute to pancreatic carcinogenesis and tumor progression, though he was explicit that whether oral health itself influences postoperative survival remained unclear.
Prior work has linked poor dental health to prognosis in colorectal and esophageal cancer, and to complications after pancreatic surgery specifically. The Hiroshima team plans to directly investigate the biological connections among oral bacteria, chronic inflammation, and pancreatic cancer progression.
For readers, the practical takeaway is narrow. Nobody should treat a dental appointment as cancer prevention, and no medical guidance has changed. The paper describes a prognostic signal that might eventually help surgeons stratify risk before an operation, not an intervention.
The more useful reading may be the one the researchers landed on. "The mouth may reflect the body's long-term resilience," Uemura said. A surgeon who wants to know how much punishment a patient can absorb has, in most cases, already looked into it.
Key Questions Answered
What exactly did the researchers measure?
The number of natural teeth remaining at a preoperative dental exam, plus posterior occlusion, in 339 consecutive patients who had radical pancreatectomy at one Japanese hospital between 2014 and 2022.
Does this mean keeping your teeth improves cancer survival?
No. This is an observational association. The study cannot establish cause, and the researchers did not test whether dental treatment changes outcomes.
Why is the 21-tooth cutoff significant?
It sits close to the CDC's definition of functional dentition, which is fewer than 20 teeth remaining. That makes the threshold easy to compare against existing U.S. population data.
If chewing was not the reason, what is?
The researchers propose that tooth count reflects accumulated chronic inflammation, frailty, nutritional history, and lifestyle. That remains a hypothesis, not a demonstrated mechanism.
How generalizable are these numbers to U.S. patients?
Limited. One center, one ethnic group, and a surgically eligible population. The authors call for multicenter and international validation.
Should anyone change their care because of this?
No. Medical guidance has not changed. Routine dental care is worth having on its own merits, and questions about pancreatic cancer risk or treatment are best addressed with a qualified clinician.
