Adjuvant chemotherapy in nonmetastatic goblet cell carcinomas: A population-based analysis.

Abstract

e16203 Background: Goblet cell carcinoma (GCC) is a rare mixed endocrine-neuroendocrine tumor arising almost exclusively in the appendix. The optimal management of these patients is still unclear, given GCC rarity and the difficulty in proper pathology diagnosis. We sought to explore the efficacy of adjuvant chemotherapy (ACT) in GCC extracted from the Surveillance, Epidemiology and End Result (SEER) US registry. Methods: Patients with pathology diagnosis of GCC were identified in the SEER registry by the 8243 ICD-09 code. Data about sex, age, tumor stage at diagnosis, number of analyzed and positive lymph-nodes, chemotherapy and survival were collected. Lymph node ratio (LNR) was calculated as the ratio between the number of metastatic lymph-nodes and removed lymph nodes. The best cutoff to predict survival state at 5-year from diagnosis was calculated. The primary endpoint was overall survival (OS). Results: Overall, 1055 GCC patients (51.7% male, median age 57 years) were identified. The median tumor diameter was 20 mm. According to the American Joint Committee on Cancer staging manual 7th edition, 128 patients (12.1%) had nodal involvement (N+): 95 were N1 and 33 were N2, while 66 (6.3%) had distant metastasis (M+). Prognostic LNR cutoff was 0.16. Using this cutoff, LNR was ≤0.16 in 674 patients (63.9%), and > 0.16 in 125 patients (11.8%). The median OS was 232 months (95% confidence interval [95%CI]: 153.4-310.5). Overall, 5-year survival rate (OS-5) was 73.4% (N = 453). At univariate analysis age, tumor diameter, M+, N+, number of lymph nodes removed, number of metastatic lymph nodes and LNR were significantly associated with the risk of death. At multivariate analysis, age, M+, N+, number of removed lymph nodes, and number of metastatic lymph nodes retained their association. After excluding M+ and N+ patients, 897 localized GCC patients (52.8% male) were analyzed. Fifty-five patients (6.1%) received ACT and OS-5 was 83.8% (N = 425). CT was administered more often in tumors with higher histological grade, higher T stage and greater tumor diameter. At the multivariate analysis, only age and number of removed lymph nodes were independently associated with the risk of death. Notably, ACT was not associated with increased survival. Ninety-two patients (57.6% male) had nodal involvement without distant metastases: 73 were N1 and 19 were N2. In 56 patients (60.9%) LNR was ≤0.16, while it was > 0.16 in 35 (38.0%). Thirty-five patients (38%) received ACT, without significant imbalances. OS-5 was 45.2% (N = 28). At univariate analysis, age, N2, number of metastatic lymph nodes and LNR were significantly associated with the risk of death. At multivariate analysis, only the number of metastatic lymph nodes retained its association. Of note, ACT was not associated with increased survival. Conclusions: In GCC, ACT was not associated with increased survival in our population-based analysis, irrespective of nodal involvement.