The study involves the enrollment of 34 patients diagnosed with advanced thymic, pulmonary and duodeno-pancreatic well-differentiated high grade neuroendocrine tumors (Ki-67 > 20%). The objective of this retrospective single-centre translational study will be to explore whether patients differ clinically in terms of diagnosis and treatment management. Currently, well differentiated high grade pulmonary NETs are managed using extrapolated algorithms from duodeno-pancreatic NETs, underlining a significant unmet clinical need. This is likely due to the rarity, uncertain pathological and molecular classification, and heterogeneous clinical course of well differentiated high grade pulmonary NETs.

In this study a retrospective data-base of pulmonary, thymic and duodeno-pancreatic NETs with Ki-67 > 20% will be created in order to analyze diagnostic and therapeutic pathways, clinical outcomes, imaging, disease evolution and molecular profiling. This study will adopt a hypothesis-generating approach to explore whether patients in these distinct groups differ clinically in terms of diagnosis and treatment management.

The LINEAR study aims to address unmet medical clinical needs in LNETs. This project specifically focuses on lung and thymic advanced NETs with Ki-67 > 20%, a rare subtype of lung cancer subtypes characterized by heterogeneous biological behaviour and variable clinical course. This contrasts with duodeno-pancreatic NETs, for which a higher level of evidence currently guides treatment sequencing. The molecular landscape and optimal therapeutic strategies for thymic and LNETs remain under investigation and are currently based on pathological features and metabolic imaging findings. Some LNETS present a carcinoids morphology but exhibit elevated Ki67 indices (often exceeding 20-30%), and these and appear to share similar behaviour and clinical characteristics with well differentiated high grade duodeno-pancreatic NETs (ki-67>20%). Such clinical cases fall into a "grey zone" where treatment prioritization is challenging due to limited data and lack of clear guidelines.

The primary endpoint of this study will be to compare therapeutic algorithm applied to well-differentiated duodeno-pancreatic, thymic and lung NETs with high proliferative indices (Ki-67 > 20%) based on the hypothesis that patients belonging to these distinct groups do not differ significantly from a clinical point of view in terms of diagnosis and treatment management.

Secondarily we will investigate the correlation of genomic alterations with patients' outcome and treatment activity and efficacy.

• To compare overall survival (OS) in the two groups.
• To compare first line progression-free survival (PFS) and time to progression (TTP) in the histological groups.
• To assess treatment response across lines of therapy, including response rate (RR), disease control rate (DCR), and treatment duration in both cohorts.
• To correlate specific genetic alterations with clinical outcomes (OS, PFS).
• To explore potential actionable molecular targets and their distribution across the two tumor origins.

Inclusion Criteria:

• Histologically confirmed diagnosis of well-differentiated high grade neuroendocrine tumor (Ki-67 > 20% according to WHO 2022) performed or reviewed by a NEN-dedicated pathologist.
• Primary tumor site:thyme, lung and duodenum-pancreas NETs.
• Advanced stage of tumor disease and Any number of lines of therapy
• Sufficient available clinical data on diagnosis, treatments, outcomes.

Exclusion Criteria:

• Poorly differentiated neuroendocrine carcinomas (NECs), GEP NET G1/G2, pulmonary carcinoid with Ki-67 < 20%.
• Diagnosis of mixed neuroendocrine non-neuroendocrine neoplasms (MiNENs)
• Inadequate or unavailable tumor tissue for molecular analysis.
• Incomplete clinical records or follow-up.
• Other primary sites, except lung or pancreas.

Histologically confirmed diagnosis of thymic, pulmonaRy and pancreatic well differentiated high grade neuroendocrine tumors(Ki-67 > 20% according to WHO 2022)