Response to comments on “Impact of tumor size and nodal status on the recurrence of nonfunctional pancreatic neuroendocrine tumors ≤2 cm after curative resection: a multi‐institutional study of 392 cases”

We appreciate the comments from Tan and colleagues regarding our recent publication that examined the impact of tumor size and nodal status on postoperative recurrence of nonfunctional pancreatic neuroendocrine tumors (NF‐pNETs) ≤2 cm. In our study, only one in ten patients (39 of 392, 9.9%) underwent an enucleation, which was considerably lower than the 33.3% (70 of 210) of patients reported in the study by Sallinen et al. The reason for this difference was likely multifactorial and may have related to variations in tumor size, location, as well as surgeon preference. We agree that enucleation is an acceptable choice for select patients with pNETs such as small tumors located away from the main pancreatic duct, superior mesenteric vein or splenic vessels. Of note, patients undergoing enucleation are still at risk for complications and need to be monitored for pancreatic fistula, as well as associated pancreatic duct injuries. In fact, one meta‐analysis demonstrated that enucleation was associated with a higher incidence of pancreatic fistula versus standard pancreatectomy. In general, the performance of an atypical pancreatectomy, such as enucleation, is often associated with inadequate lymphadenectomy. For example, in our study, lymphadenectomy was commonly performed among the whole cohort (83.7%, 328 of 392), yet was only performed in 38.5% (15 of 39) of patients who underwent an enucleation. Importantly, 4 out of 15 (26.7%) patients who underwent tumor enucleation and nodal dissection had lymph node metastasis. In addition, nodal metastasis was strongly associated with tumor recurrence after resection (hazard ratio, 3.06; 95% confidence interval, 1.15‐8.17). In another study from our group, we recommended a minimum of eight lymph nodes to be evaluated for the optimal staging of patients with pNETs. When performing an enucleation, while lymph node sampling is feasible, attainment of this nodal threshold can be more challenging. In our study, the overall morbidity was 58.1% (227 of 392) with 87 (38.5%) patients experiencing a Clavien‐Dindo III‐IV complication. One patient died of cardiovascular disease within 90 days after surgery, and this patient was excluded from the study. In contrast, in the study by Wang et al, the authors included only patients undergoing pancreatic enucleation for pNETs or benign disease processes. As such, direct comparisons of the morbidity in the current study and the report by Wang et al may not be appropriate as the overwhelming majority of patients (90.1%, 353 of 392) in our study underwent more formal pancreatic resection, such as pancreatoduodenectomy, distal pancreatectomy, central pancreatectomy, as well as total pancreatectomy. In summary, we agree that a well‐balanced approach to operative risk, which takes into account the extent of surgery, should be considered for each individual patient with a small pNET. Data from our study did demonstrate, however, that a subset of patients had more aggressive tumor characteristics with NF‐pNETs ranging from 1.5 to 2 cm, such as a higher incidence of nodal metastasis, more advanced tumor grade and higher Ki‐67 levels. Taken together, the data suggest that surgical resection with lymphadenectomy should be strongly considered for patients with NF‐pNETs ≥1.5‐2.0 cm.