麻豆传媒

SEATTLE -- Patients with small (<1 cm) pancreatic neuroendocrine tumors (pNETs) had a 5-year survival of 100% with surveillance, data from a large retrospective study showed.

Surgery for small pNETs was associated with worse survival, though the difference from surveillance did not differ significantly. Patients with intermediate- and larger-size tumors fared significantly better with surgery, as reported here at the North American Neuroendocrine Tumor Society annual symposium.

"To our knowledge, this retrospective analysis is the largest on well-differentiated pancreatic neuroendocrine tumors," said Hussein Assi, MD, of the University of Oklahoma Stephenson Cancer Center in Oklahoma City. "Active surveillance is potentially a safe approach for pancreatic neuroendocrine tumors less than 1 cm. Larger tumors likely need active intervention."

"Prospective, randomized clinical trials are needed in order to conquer the current challenges of patients with pancreatic neuroendocrine tumors," he added.

Surgery for pNETs achieves good disease control in a majority of patients but is associated with significant morbidity and a significant risk of complications, which reached half of patients in one series. Complications can include pancreatic fistula, diabetes, and severe hemorrhage, Assi noted. Additionally, the surgery is associated with "non-negligible" mortality risk.

Studies to identify patient and disease characteristics that influence survival yielded inconsistent results. Although most patients ultimately have surgery, optimal treatment remains controversial, Assi continued. The National Comprehensive Cancer Network recommends consideration of nonsurgical options for pNETs <1 cm, whereas the European Neuroendocrine Tumor Society established a cutoff of <2 cm.

Assi and colleagues investigated the relationship between surgical resection and overall survival (OS) in patients with pNETs, according to tumor size. Specifically, they sought to determine whether surgery improved survival in patients with small, well-differentiated pNETs. They also wanted to see whether they could identify prognostic factors associated with OS in patients with pNETs.

Investigators searched the National Cancer Database to identify patients with newly diagnosed pNETs during 2004-2015. After exclusion of patients with grade 3 tumors, metastatic disease, and missing data, the analysis comprised 2,004 patients with well-differentiated pNETs.

Grouping by tumor size showed 220 patients had pNETs <1 cm, 794 had tumors 1-2 cm, and 990 had tumors >2 cm. Overall, 1,781 patients had surgery and 223 had nonsurgical treatment, including 31 patients with tumors <1 cm, 93 with tumors 1-2 cm, and 99 with tumors >2 cm.

Patients who did not have surgery were significantly younger (median age 68 vs 60 for the surgery group, P<0.0001). The cohort had an even distribution of men and women and did not differ significantly by race or comorbidity score.

Considering all tumors irrespective of size, nonsurgical management was associated with location in the pancreas head or body whereas surgery more often involved tumors in the pancreas tail or other primary sites (P<0.0001), stage 1 disease (P<0.0001), unknown regional lymph node status (P<0.0001), and treatment with chemotherapy (P<0.0001) or radiation therapy (P=0.0018). Tumor size did not influence the decision to have surgical or nonsurgical treatment (P=0.1701).

An unadjusted analysis of 5-year OS with no surgery versus surgery by tumor size yielded the following values:

• <1 cm: 100% versus 91.7% (P=0.1358)
• 1-2 cm: 86.5% versus 95.2% (P=0.0011)
• >2 cm: 80.2% versus 93.1% (P<0.001)

After adjustment, the hazard ratio (HR) for OS for no surgery versus surgery was 1.92 (95% CI 0.17-21.53, P=5.986) for tumors <1 cm, 0.4 (95% CI 0.18-0.95, P=0.0364) for tumors 1-2 cm, and 0.34 (95% CI 0.16-0.72, P=0.0050) for tumors >2 cm.

"Patients with tumors less than 1 cm were almost twice as likely to be alive at 5 years with surveillance than with surgery, even though the difference was not statistically significant," Assi told 麻豆传媒. "These are the best data we have on the issue, and they suggest that surveillance is at least as good as surgery for patients with small pancreatic neuroendocrine tumors. Clearly, surgery led to better survival for patients with larger tumors. The only way to get a more definitive answer would be to conduct a randomized, controlled clinical trial."

Comment