RESUMEN
OBJECTIVE: Our objective was to develop an approach for selecting combinatorial markers of pathology from diverse clinical data types. We demonstrate this approach on the problem of pancreatic cyst classification. MATERIALS AND METHODS: We analyzed 1026 patients with surgically resected pancreatic cysts, comprising 584 intraductal papillary mucinous neoplasms, 332 serous cystadenomas, 78 mucinous cystic neoplasms, and 42 solid-pseudopapillary neoplasms. To derive optimal markers for cyst classification from the preoperative clinical and radiological data, we developed a statistical approach for combining any number of categorical, dichotomous, or continuous-valued clinical parameters into individual predictors of pathology. The approach is unbiased and statistically rigorous. Millions of feature combinations were tested using 10-fold cross-validation, and the most informative features were validated in an independent cohort of 130 patients with surgically resected pancreatic cysts. RESULTS: We identified combinatorial clinical markers that classified serous cystadenomas with 95% sensitivity and 83% specificity; solid-pseudopapillary neoplasms with 89% sensitivity and 86% specificity; mucinous cystic neoplasms with 91% sensitivity and 83% specificity; and intraductal papillary mucinous neoplasms with 94% sensitivity and 90% specificity. No individual features were as accurate as the combination markers. We further validated these combinatorial markers on an independent cohort of 130 pancreatic cysts, and achieved high and well-balanced accuracies. Overall sensitivity and specificity for identifying patients requiring surgical resection was 84% and 81%, respectively. CONCLUSIONS: Our approach identified combinatorial markers for pancreatic cyst classification that had improved performance relative to the individual features they comprise. In principle, this approach can be applied to any clinical dataset comprising dichotomous, categorical, and continuous-valued parameters.
Asunto(s)
Biomarcadores de Tumor/análisis , Quiste Pancreático/diagnóstico , Neoplasias Pancreáticas/diagnóstico , Adulto , Anciano , Anciano de 80 o más Años , Biomarcadores , Cistoadenoma/diagnóstico , Femenino , Humanos , Persona de Mediana Edad , Quiste Pancreático/cirugía , Estudios Retrospectivos , Sensibilidad y EspecificidadRESUMEN
OBJECTIVE: To investigate whether gastrin has regenerative effect on the pancreas and in particular whether it prevents the atrophy of the distal pancreas after resection of pancreas in humans. SUMMARY BACKGROUND DATA: Although pancreatic regeneration after resection is well documented in animals, atrophy rather than regeneration of the distal remnant pancreas commonly occurs following pancreatoduodenectomy in humans. Of the many factors involving pancreatic regeneration, gastrin has been shown to have trophic effect on the pancreas in an animal model. METHODS: Between March 1999 and May 2000, a randomized prospective study was performed in 56 patients who underwent pylorus-preserving pancreatoduodenectomy for periampullary neoplasms. Patients were allocated to either a lansoprazole group or a control group. The lansoprazole members were given oral lansoprazole (30 mg/d) over 12 weeks postoperatively to induce hypergastrinemia. During the study period, 19 patients were excluded for different reasons; in the end a total of 37 patients (lansoprazole, n = 18; control, n = 19) were eligible for study. The volume of the distal pancreas as determined using thin-sectioned spiral CT data, nutritional status, and endocrine (insulin level, glucose tolerance test) and exocrine function (stool elastase) of the pancreas and serum gastrin levels were measured before surgery and 3 months after surgery. The two groups were clinically comparable. RESULTS: Serum gastrin level was elevated in the lansoprazole group. In this group, the mean volume of the distal pancreas was reduced by 10% after pylorus-preserving pancreatoduodenectomy, whereas severe pancreatic atrophy occurred in the control group. Postoperative insulin and stool elastase levels were higher in the lansoprazole group than in the control group. CONCLUSIONS: This study is the first prospective randomized trial of induced hypergastrinemia on the regeneration of the pancreas in humans. It may be possible to use induced hypergastrinemia in the treatment or prevention of pancreatic insufficiency following resection or injury.