麻豆传媒

SEATTLE -- Surgery for pancreatic neuroendocrine tumors (pNETs) significantly improved survival for all patients except those with the earliest-stage disease, a retrospective review of 4,700 patients showed.

Among patients with stages II-IV pNETs, surgery reduced the mortality hazard by 70%-80% as compared with patients who did not have surgery. Only patients with stage I disease did not live longer with surgery.

"After adjustment for other variables, receipt of surgery for stage IV patients showed a 71% decreased hazard of death," reported Benjamin D. Powers, MD, of Moffitt Cancer Center in Tampa, Florida, and colleagues at the North American Neuroendocrine Tumor Society annual symposium. "This observation suggests that local control of the primary pNET through surgical resection drives outcomes, even in patients with advanced stage, and should be considered in the multidisciplinary management of pNETs."

Management of patients with pNETs by clinical grade and stage has a controversial history. The latest (8th edition) version of the American Joint Committee on Cancer (AJCC) staging manual attempted to address the controversy with a new pNET staging system that excluded poorly differentiated tumors (neuroendocrine carcinoma).

Powers' group retrospectively applied the new AJCC criteria to a large group of patients with surgically treated pNETs representing all stages of disease. Their primary objective was to determine whether the impact of surgery on survival of patients with stages I-IV moderately or well-differentiated tumors.

Data for the study came from the National Cancer Database and included 4,739 adults treated for pNETs from 2004-2013. Investigators excluded patients with poorly or undifferentiated tumors, patients with multiple cancers, and patients with incomplete staging data.

The results showed that 3,866 (81.6%) patients underwent surgical resection. Patients who did not have surgery served as the reference. The analysis included 1,202 patients with stage I disease, 1,853 with stage II/III, and 1,084 with stage IV pNETs.

"Drivers of surgical utilization varied significantly by stage," the authors reported. "For clinical stage I patients, surgery was more likely for patients with body or tail tumors relative to pancreatic head tumors and less likely for patients 70 or older. For stage II/III patients, age 70 or older, black race, and community facility type were associated with decreased odds of surgery. Patients with body or tail tumors were more likely to receive surgery."

Complete survival data were available for 3,551 patients, and 705 patients died. Surgery offered no survival benefit to patients with stage I tumors. In contrast, surgery reduced the hazard for death by 77% among patients with stage II/III disease (95% CI 0o.17-0.31) and by 71% for patients with stage IV disease (95% CI 0.23-0.38).

Patients with primary tumor location other than the pancreatic head had numerically (nonsignificant) lower hazards. Patients with Charlson-Deyo comorbidity scores of 1+ had significantly higher odds of worse survival across all four disease stages, as did patients treated at community versus academic centers.

Comment