ABSTRACT
OBJECTIVES: The early-stage pancreatic cancer (e-PC; stage I/II) detection rate is quite low at approximately 25%. The aim of this study was to evaluate the feasibility of a social program (the Kishiwada Katsuragi project) wherein our hospital, which specializes in PC, and primary care medical offices (PMOs) used clinical findings to detect e-PC. METHODS: Patients with a score of ≥2 points on clinical findings were enrolled: symptoms of abdominal pain/back pain (1 point), new-onset diabetes (1 point), high amylase (AMY) and/or pancreaitc AMY (P-AMY) (1 point), high carbohydrate antigen 19-9 (1 point), and ultrasonography (US) findings including direct (e.g., a solid pancreatic tumor) and/or indirect findings (e.g., dilatation of a pancreatic diameter of ≥2.5 mm and/or cystic lesions) (2 points) were evaluated using the protocol for social programs. RESULTS: Between November 2014 and December 2016, 244 patients were enrolled by 41 PMOs as cooperative facilities, and 15 e-PC cases (53.6%) of the 28 PC patients were detected. The mean clinical finding score of the e-PC group (3.13 ± 1.9) was significantly higher than that of the overall non-PC group (2.1 ± 0.4) (p < 0.05). "High AMY/P-AMY" and "symptoms" were significantly more frequent in the e-PC group than in the non-PC group (p < 0.05). Although the sensitivity of direct findings by US was 40.0%, that of indirect-findings was 93.3% in the e-PC group. Nine and 6 of the 15 patients with e-PC were enrolled via general internal medicine offices (GIMs) and other PMOs without GIMs (general surgery, n = 3; urology, n = 2; otolaryngology, n = 1). CONCLUSION: This social program with collaborations between medical centers that specialize in PC and PMOs used clinical findings, suggesting that not only GIMs but also other PMOs and indirect findings by US may play an important role in improving the e-PC detection rate.
