Surgery is the main stay of treatment for GEP NETs, but there is no consensus on optimal number of resected LNs. The effect of LN status and yield on relapse-free (RFS) and overall survival (OS) in patients (pts) with resected GEP NETs were evaluated.
Data on pts who underwent curative resection for GEP-NETs (Jan 02-Mar 17) were retrospectively analysed. Grade III tumours (Ki67>20%) were excluded. Kaplan-Meier and univariate/multivariable Cox-proportional hazard analyses were performed. Cut-point analysis was assessed to distinguish a binary categorisation of total LNs retrieved associated with RFS.
Of 217 pts, median (med) age was 59 yrs: 51% male. Primary tumour sites: small bowel (42%), pancreas (25%), appendix (18%), rectum (7%), colon (3%), gastric (2%), others (2%); grade 1 (G1): 77%, G2: 23%. LN cut-point value associated with RFS was 8; ≥8 LNs were retrieved in 106 pts, <8 in 45, and 0 or no record/documentation of LN retrieval in 66. Relapse was reported in 50 pts; 35 deaths. Med follow up times for all pts were 41 months (95% CI 36-51) and 71 months (95% CI 63–76) for RFS and OS respectively. On univariate analysis, there was no effect of LN ratio (number involved/number retrieved) on RFS: p = 0.1 or OS: p = 0.75. On univariate analysis, tumour necrosis (p = 0.021) and perineural infiltration (p = 0.016) were the only two variables significantly associated with OS; G (p = 1), TNM staging (p = 0.19) and surgical margin (p = 0.69) were not significantly associated with OS. Multivariable analysis for RFS included 4 variables of interest: perineural infiltration, LNs retrieved, positive LNs and localisation (Table).Table: 1325P
|Variable||Hazard Ratio||(95% CI)||p|
|Perineural infiltration||1.46||(0.74 - 2.69)||0.277|
|≥8 lymph nodes retrieved||2.70||(1.07 - 6.84)||0.036|
|Any lymph nodes positive||2.71||(0.88 - 8.30)||0.081|
|Pancreas (relative to 'other')||27.33||(2.54 - 294.08)||0.006|
|Small Bowel (relative to 'other')||32.44||(2.92 - 360.58)||0.005|
Removal of ≥ 8 LNs is associated with greater risk of relapse in G1 & G2 GEP NETs; localisation also has a significant association with RFS, necessitating stricter surveillance. Larger prospective studies are required to validate these findings.
Clinical trial identification
Legal entity responsible for the study
The Christie Hospital NHS Trust, Manchester, Uk.
Has not received any funding.
All authors have declared no conflicts of interest.