Incidence and Predictors of Second Primary Cancers in Patients With Neuroendocrine Tumors

Neuroendocrine tumors (NETs) are rare heterogenous tumors arising from the gastrointestinal tract and bronchopulmonary system.^1-3 Overall NET incidence has been steadily rising in North America.^2,3 Although survival varies based on clinicopathologic factors, prognosis is favorable for most with a median survival of 9 years.² As a personal history of cancer is associated with a higher risk of second primary cancers (SPCs) for multiple malignant neoplasms, patients with NETs may be at risk.^4,5 We evaluated the risk and predictors of SPCs after a NET diagnosis.

We performed a retrospective cohort study of individuals diagnosed with NETs from 2000 to 2016, using the Surveillance, Epidemiology, and End Results (SEER) registry. Patients aged 20 years or older with gastrointestinal, pancreatic, or lung primary invasive NETs from the International Classification of Diseases for Oncology, third edition (ICD O-3) site and histology codes were included.² Patients with the index tumor identified from autopsy/death certificate and SPCs diagnosed within 2 months of the primary were excluded. Standardized incidence ratios (SIRs) were calculated from the number of observed SPCs divided by the expected number. The expected number was estimated from incidence rates stratified by age, sex, race, and diagnosis year of the population. Factors associated with SPCs were assessed using Fine and Gray models accounting for the competing risk of death. Analyses were performed using SEER Stat (version 8.3.8) and SAS statistical software (version 9.4, STAT Corp). Methods are further described in the Supplement.

Among 58 596 patients with NETs, 4612 (7.9%) had SPCs. Neuroendocrine tumor histology was observed in 16.2% of the SPCs. Median (IQR) follow-up was 89 (49-133) and 44 (15-93) months for patients with and without SPCs, respectively. Median (IQR) time to SPC was 40 (17-77) months. The 5-year cumulative incidence of SPC was 5.4% (95% CI, 5.2%-5.6%) for all, 5.9% (95% CI, 5.6%-6.2%) for gastrointestinal, 3.8% (95% CI, 3.3%-4.4%) for pancreas, and 4.8% (95% CI, 4.4%-5.2%) for lung NETs.

Patients with NETs tumors had increased risk of SPCs (Table 1) with site-specific differences. Patients with gastric NETs had increased risk of esophageal (SIR, 2.81; 95% CI, 1.13-5.80), small intestine (SIR, 9.85; 95% CI, 5.25-16.85), pancreatic (SIR, 2.59; 95% CI, 1.60-3.95), and liver SPCs (SIR, 2.76; 95% CI 1.38-4.93). Patients with appendiceal NETs had increased risk of small intestine (SIR, 20.74; 95% CI, 11.04-35.46), colorectal (SIR 3.04, 95% CI 2.09-4.27), and lung SPCs (SIR 1.62, 95%CI 1.05-2.40). Patients with colon NETs had increased risk of small intestine SPCs (SIR, 7.76; 95% CI, 3.72-14.28). Patients with rectal NETs had increased risk of pancreatic (SIR, 1.45; 95% CI, 1.02-1.99), lung (SIR, 1.20; 95% CI, 1.02-1.41), and prostate SPCs (SIR, 1.43; 95% CI, 1.25-1.63). Patients with small intestine NETs had increased risk of liver (SIR, 1.59; 95% CI, 1.02-2.37) and prostate (SIR, 1.35; 95% CI, 1.17-1.55), but not colorectal SPCs (SIR, 0.80; 95% CI, 0.62-1.01). Patients with pancreatic NETs had increased risk of gastric (SIR, 2.49; 95% CI, 1.24-4.46), and small intestine SPCs (SIR, 8.79; 95% CI, 4.54-15.36).

Discrete age cohorts and black race and ethnicity were associated with SPCs (Table 2). Female sex, higher NET grade, and stage were inversely associated with SPCs. Stratified analysis by NET site showed similar results for gastrointestinal and lung NETs. Among pancreatic NETs, only distant disease was inversely associated with SPCs (SHR, 0.41; 95% CI, 0.30-0.55).

In this population-based analysis, there was an increased risk of SPCs after NET diagnosis. At 5 years, for every 100 patients with NET, 5 were diagnosed with an SPC, representing 35% more cases than expected in the general population. The patterns observed were distinct from those expected from known NET-related genetic syndromes.⁶ Limitations include the histologic inclusion of NET SPCs, and potential misclassification of recurrence as SPC, and the shorter follow-up among patients without SPCs potentially underestimating SPC risk. These data highlight the need to consider detection strategies for SPCs to NET surveillance tailored to NET-specific and patient factors. Further investigations are warranted regarding specific surveillance regimens and prognostic implications.

Accepted for Publication: July 21, 2021.

Published Online: September 30, 2021. doi:10.1001/jamaoncol.2021.4531

Corresponding Author: Julie Hallet, MD, Odette Cancer Centre, Sunnybrook Health Sciences Centre, 2075 Bayview Ave, Ste T2-063, Toronto, ON M4N 3M5, Canada (julie.hallet@sunnybrook.ca).

Author Contributions: Dr Bateni had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Bateni, Law, Singh, Assal, Hallet.

Acquisition, analysis, or interpretation of data: Bateni, Coburn, Singh, Myrehaug, Assal, Hallet.

Drafting of the manuscript: Bateni, Singh, Hallet.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Bateni, Coburn, Singh, Hallet.

Obtained funding: Hallet.

Administrative, technical, or material support: Bateni, Law, Singh, Assal, Hallet.

Supervision: Coburn, Law, Singh, Hallet.

Conflict of Interest Disclosures: Dr Coburn reported other from AstraZeneca Canada Honorarium outside the submitted work. Dr Law reported other from Ipsen Advisory Board Member outside the submitted work. Dr Singh reported personal fees from IPSEN and grants from NOVARTIS/AAA outside the submitted work. Dr Myrehaug reported personal fees from AAA/Novartis and grants from Ipsen during the conduct of the study. Dr Hallet reported personal fees from Ipsen Biopharamceuticals Canada speaker honoraria and personal fees from Advanced Accelerator Applications speaker honoraria during the conduct of the study. No other disclosures were reported.