Development Of A Nomogram For Predicting The Biological Malignant Behavior Of Solid Pseudopapillary Neoplasms In The Pancreas

Objective:Solid pseudopapillary neoplasms of the pancreas are indolent rare tumors with a malignant potential.The risk factors associated with the biological malignant behavior of solid pseudopapillary neoplasms of the pancreas are still unclear,meanwhile predicting the malignant behavior of solid pseudopapillary neoplasms remains an important goal.Based on this study,clinicians were guided to develop individualized treatment and monitoring plans to more competently manage patients by preoperatively predicting which kinds of patients with solid pseudopapillary neoplasms of the pancreas have or may have biological malignant behavior such as metastasis or recurrence in the future.Methods:A retrospective analysis on patients with solid pseudopapillary neoplasms in the pancreas who underwent surgical treatment in the First Hospital of Jilin University from January 2010 to June 2022 was conducted.All of them have received a follow-up of 88 months on average(range: 6-152 months).According to WHO guidelines,biological malignant behavior was defined as solid pseudopapillary neoplasms of the pancreas locally invading adjacent structures,developing recurrence or had systemic metastasis,either at the initial diagnosis or later.Based on this guideline,the tumors were divided into the benign and the malignant group.The clinical baseline data,pathology,imaging and laboratory indicators of patients were analyzed through univariate and multivariate logistic regression to identify the independent risk factors associated with the malignant behavior of solid pseudopapillary neoplasms in their pancreas,and a predictive model was established in the form of a nomogram.The discriminative and calibration ability of the model was verified by 1000 self-sampling replicates.Results:Of the 208 patients pathologically diagnosed with solid pseudopapillary neoplasms in the pancreas,there were 53 cases(25.5%)in the malignant behavior group and 155(74.5%)in the benign behavior group.The presence of clinical symptoms among patients was statistically significant when distinguishing the benign behavior group from the malignant behavior group.A univariate logistic regression analysis showed that among the inflammatory markers,differences in LMR,PNI and SIRI were statistically significant between the benign and the malignant behavior group,whereas SII,PLR and NLR were not statistically significant.Among the hematological indicators,elevated tumor markers and the presence of concomitant viral hepatitis did not differ statistically between the benign and the malignant behavior group.In preoperative imaging,large tumor sizes,incomplete capsules and indistinct margins were significantly different between the benign and the malignant behavior group.In contrast,distal pancreatic atrophy and dilated pancreaticobiliary ducts were not statistically different between the above groups.In multivariate logistic regression analysis,clinical symptoms(P < 0.001),unclear tumor margins(P = 0.001),incomplete tumor capsules(P = 0.005),maximum tumor diameters ≥ 7.2 cm(P = 0.003)and prognostic nutritional index values < 47.45(P = 0.007)were independent risk factors of solid pseudopapillary neoplasms in the pancreas with malignant behavior.A nomogram model was successfully established to predict the malignant behavior of solid pseudopapillary neoplasms in the pancreas.The area under the receiver operating characteristic curve was 0.856.The calibration prediction curve was in good agreement with the standard curve.Conclusion:Solid pseudopapillary neoplasms in the pancreas without biological malignant behavior have a low recurrence rate and an excellent long-term survival rate after surgical resection.The nomogram model based on clinical symptoms,inflammatory markers and imaging features had a high application value for the preoperative prediction of the malignant behavior of solid pseudopapillary neoplasms in the pancreas.The above information can help surgeons develop individualized treatment modalities,management and monitoring plans to manage patients more precisely.