NANETS2023 Clinical – Chemo/SSA/Biologics (17 abstracts)
1Helen Diller Family Comprehensive Cancer Center, University of California San Francisco, San Francisco, CA; 2Department of Epidemiology and Biostatistics, University of California San Francisco, San Francisco, CA; 3Department of Radiology and Biomedical Imaging, University of California San Francisco, San Francisco, CA; 4Department of Medicine, Division of Hematology/Oncology, University of California San Francisco, San Francisco, CA; 5Department of Surgery, University of California San Francisco, San Francisco, CA; 6Department of Pathology, University of California San Francisco, San Francisco, CA
Background: G3NETs demonstrate variable clinical behavior (Ki-67>20-100%). Treatment extrapolated from grade 1/2 (G1/2) NETs and neuroendocrine carcinomas depending on clinicopathologic features. Therapy for treatment-emergent G3NETs remains especially ill-defined.
Methods: Retrospective chart review of patients with de novo G3NETs or progressing from low-to-high grade (L-H) (G1/2 at diagnosis and G3 >=3 months later). Treatment patterns compared by Pearson chi-squared tests among patients with complete treatment information. Overall survival (OS) from stage IV G3NETs diagnosis estimated by Kaplan-Meier.
Results: n=50 patients with G3NETs: 18% nonwhite, 46% female, median age at G3 62 (29-81). Primary site: pancreas 60%, other gastrointestinal 22%, unknown 18%. Median follow-up from stage IV G3 48 months (mo). De novo G3NETs in 38% (19/50) with median Ki-67 30% (range 21-70%), Ki-67<=55% in 79%. Among these, 89% (17/19) stage IV and 11% (2/19) locoregional at diagnosis (treated surgically before developing metastases). Median OS was 35 mo (95% CI 20-54). L-H G3NETs in 62% (31/50): median Ki-67 at G3 diagnosis 38% (21-75%), Ki-67<=55% in 87%. Among these, 77% (24/31) stage IV at G1/G2 diagnosis. Median time to L-H 58 mo (range 3-391 mo); 97% (30/31) stage IV. Therapy prior to G3 diagnosis: SSA 81%, surgery 48%, liver-directed therapy (LDT) 41%, cytotoxic chemotherapy (CC) 39%, radiation 32%, peptide receptor radionuclide therapy (PRRT) 29%, and targeted agents (e.g. everolimus, sunitinib) 26%. Median OS 16 mo (95% CI 9-22). Treatment for advanced G3 listed in Table 1.Table 1. First-line (1L) and second-line (2L) treatment for metastatic G3NETs (
De novo G3NETs (n=19) | L-H G3NETs (n=25) | |||
1L,N(%) | 2L,N(%) | 1L,N(%) | 2L,N(%) | |
SSA^ | 9(47) | 0 | 1(4) | 0 |
Cisplatin/carboplatin-based | 8(42) | 3(16) | 4(16) | 2(8) |
Other CC*/Immunotherapy | 0 | 7(37) | 4(16) | 5(20) |
Everolimus | 0 | 0 | 0 | 2(8) |
PRRT | 0 | 1(5) | 2(8) | 1(4) |
Radiation | 0 | 1(5) | 2(8%) | 3(12) |
LDT | 0 | 3(16) | 6(24) | 4(16) |
Surgery (primary or metastasis) | 2(11) | 2(11) | 6(24) | 2(8) |
*capecitabine/temozolomide, FOLFOX^Pearson chi-squared test indicated significant differences in 1L (p<0.001) |
Conclusion: Higher prevalence of L-H compared to de novo G3NETs was observed at our institution. While there is extensive overlap, de novo G3NETs more commonly treated with SSA. Further research is needed to compare the outcomes and molecular underpinnings of apparent de novo G3NETs vs L-H, and the impact of primary tumor site, Ki67, and prior therapy. Understanding these differences will help elucidate whether these subtypes are truly distinct entities (with one being treatment-emergent) or simply reflect the natural history of the disease and intrapatient heterogeneity.
Abstract ID 23806