National population-based analysis of demographics and outcomes of patients diagnosed with extranodal NK/T-cell nasal type lymphoma in the US: Does ethnicity play a role?

Abstract

e18772 Background: Extranodal Natural Killer/T-cell Lymphoma Nasal Type (ENKTL-NT) in an uncommon malignancy with rising incidence in the United States (US) despite previously being described only in Asia and Latin America (J Oral Maxillofac Pathol PMC3125667 Crit Rev Oncol Hematol PMID33493634). Although primarily occupying the nasal cavity, other locations may be affected, which makes its presentation ambiguous. Diagnosis is important as omitting radiation has been associated with detrimental survival (J Oncol Pract PMID31600462). This is the first national population-based analysis evaluating ethnic differences for Hispanics (HI) diagnosed with ENKTL-NT. Methods: Data were analyzed on ENKTL-NT patients in the US reported to the Surveillance, Epidemiology, and End Results (SEER) database between 2000-2018. Demographic, clinical and survival patterns were analyzed. Kaplan-Meier and Cox regression analyses were used when appropriate. Multivariate analysis and propensity score matching were performed with adjustment for age, stage and B-symptoms. Results: 1,205 patients (HI=409, Non HI (NH)=796) were diagnosed with ENKTL-NT. Male sex predominated in both. HI were diagnosed at younger median age, 43 years (y) vs 59y in NH [p <0.001]. Most of HI and NH were whites 98% vs 58%; however, for NH other races were noted. Most of the patients were diagnosed between 2010-2014. For both, the presence of B symptoms was predominantly unknown. At diagnosis stage I was mainly for HI while unknown stage was most likely noted for NH [p = 0.013]. HI were less likely than NH to receive radiation as part of their treatment, 47% vs 57% respectively [p = 0.003]. Median survival for HI was 1.7y vs 1.2y for NH; survival probability at 2, 5 and 10y for HI corresponded to 0.467, 0.413 and 0.355, while for NH it was 0.444, 0.352 and 0.292, respectively. OS probability at 10y had no statistically significant difference [p = 0.06]. On multivariate analysis, those who were older than 80y and between 60-80y, had worse OS compared to those younger than 60y (HR 2.3 - 95% CI: 1.7 – 3 and 1.2 - 95% CI: 1 – 1.5). Regarding stage, those who were at stage III/IV had worse OS than those at early stage I/II (HR 2.4 -95% CI: 1.7 - 3.2 and HR 3.4 -95% CI: 2.7 - 4.30). Conclusions: This national population-based study identified statistical differences in HI diagnosed with ENKTL-NT. Interestingly, HI were less likely to receive radiation therapy which is the mainstay treatment and key determinant for survival. The fact that there was no statistical difference in survival for HI and NH may underrepresent the improved survival from HI as they are not optimally treated and still living as longer as the NH. Additional research with biological variables should be carried out to better understand this oncological outcome and potentially offer improved targeted treatments for this ethnic minority.