Incidence, Survival Outcome, and Prognostic Nomogram of Patients with Angioimmunoblastic T-cell Lymphoma: a Population-based Analysis.

ABSTRACT

OBJECTIVE: Due to the rarity of angioimmunoblastic T-cell lymphoma (AITL), very limited data concerning its incidence patterns and prognostic factors are available. This study aimed to explore the incidence, characteristics, survival outcomes, and prognostic factors of AITL. METHODS: Age-adjusted incidence and temporal trends were calculated based on 1247 AITL patients from the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER)-13 database. A total of 1525 AITL patients from the SEER-18 database and 43 patients from our single center were included for survival analysis and nomogram construction. RESULTS: The age-adjusted incidence for overall cohort was 0.123 [95% confidence interval (CI), 0.117-0.131) per 100 000 during 1992-2017. The overall incidence increased steeply at the rate of 15.3% (95%CI 11.0%-19.8%, P<0.001) per year during 1992-2004, but remained stable during 2004-2017 (P=0.200). Similar incidence trends were observed in age, sex, and stage subgroups. The final nomograms consisted of four variables age at diagnosis, sex, Ann Arbor stage, and primary site. The concordance index (C-index) of the nomogram for 5-year overall survival prediction was 0.717, 0.690 and 0.820 in the training cohort, validation cohort-1 and cohort-2, respectively. Regarding the disease-specific survival (DSS), the nomogram also demonstrated a good discrimination level, with the C-index for predicting the probability of DSS at 5 years of 0.716, 0.682 and 0.813 for the three cohorts, respectively. The calibration displayed good concordance between the nomogram-predicted and actual observed outcomes. CONCLUSION: The age-adjusted incidence for AITL was low during 1992-2017. The incidence continuously increased during 1992-2004, but remained stable during 2004-2017. The nomograms as proposed may provide a favorable and accurate prognostic survival prediction in AITL.