Pancreatoblastoma: A Descriptive and Comparative Analysis with Pancreatic Ductal Adenocarcinoma Using SEER Data.

BACKGROUND

Pancreatoblastoma (PB) is an exceptionally rare malignant epithelial neoplasm of the pancreas that recapitulates the developing pancreatic anlage. It is defined histologically by acinar-predominant differentiation with characteristic squamoid nests, and it may also show ductal and endocrine differentiation within the same tumor, features that distinguish it from acinar cell carcinoma and solid pseudopapillary neoplasm but that are readily overlooked. It occurs predominantly in young children, although adult-onset disease is well documented. We provide a population-based characterization of PB across the full age spectrum and benchmark it against pancreatic ductal adenocarcinoma (PDAC).

METHODS

Cases diagnosed between 2000 and 2021 were identified in the Surveillance, Epidemiology, and End Results (SEER) 17-registry database using site and histology codes. Cancer-specific survival (CSS) was estimated and compared, and Cox proportional hazards regression was used to explore associations with cancer-specific mortality.

RESULTS

Thirty-nine cases of PB were identified, compared with 155,924 cases of PDAC. The median age at diagnosis was 17 years (range, under 1 to 78 years); 12.8% (n = 5) were younger than 1 year. Males accounted for 69.2% (n = 27) of cases. The cohort was divided at the conventional pediatric-to-adult threshold of 18 years into a pediatric subgroup (age < 18 years; n = 20) and an adult subgroup (age ≥ 18 years; n = 19). CSS at 1 and 5 years was 95.0% and 83.5% in the pediatric subgroup, versus 67.7% and 24.6% in the adult subgroup (log-rank p < 0.001). Five-year CSS was 44.8% in males and 75.0% in females. In an exploratory multivariable model, older age was associated with higher cancer-specific mortality both as a dichotomous variable (adjusted hazard ratio [HR] for age ≥ 18 years 10.7, 95% confidence interval [CI] 2.4-48.2; p = 0.002) and, in a parallel model, as a continuous variable (adjusted HR 1.4 per 10-year increment, 95% CI 1.1-1.8; p = 0.002), indicating an age-mortality gradient. Male sex showed an association in the same direction that did not reach statistical significance (adjusted HR 3.3, 95% CI 0.96-11.6; p = 0.06). Relative to PDAC, PB was more frequently diagnosed in males and was associated with markedly superior survival (1- and 5-year CSS 81.8% and 54.9%, versus 28.8% and 4.0%).

CONCLUSIONS

Pancreatoblastoma is predominantly a malignancy of young males and carries a substantially more favorable prognosis than PDAC, but outcomes differ markedly across the age spectrum, with adult-onset disease showing considerably poorer survival. Translationally, these population-level estimates support age-stratified prognostic counseling, argue for the referral of adults to centers experienced in rare pancreatic tumors, and provide a rationale for prospective molecular profiling to determine whether adult and pediatric PBs are biologically distinct and whether Wnt/beta-catenin pathway activation is therapeutically actionable.