International Prognostic Tool Predicts Outcomes in Angioimmunoblastic T-Cell Lymphoma

Researchers have created a novel prognostic score to predict outcomes in patients with angioimmunoblastic T-cell lymphoma (AITL), stratifying them into low-, intermediate-, and high-risk subgroups for overall survival (OS). The model, described in a recent Blood paper, is composed of the following factors: advanced age, elevated Eastern Cooperative Oncology Group (ECOG) performance status, increased C-reactive protein (CRP), and elevated ß2 microglobulin.

“Outcomes remain suboptimal for patients with AITL treated in the contemporary era, with particularly poor outcomes for high-risk patients,” said lead study author Ranjana Advani, MD, of Stanford University.

Dr. Advani and an international team of investigators analyzed data on 282 patients with AITL who were registered in the T-cell Project (TCP), a prospective registry of patients with peripheral T-cell lymphomas, between 2006 and 2018. All patients with mature T-cell or NK-cell lymphomas were included in the TCP registry at the time of diagnosis and prior to treatment initiation.

The primary endpoint of the study was five-year OS rate. Rates of progression-free survival (PFS) at five years represented the key secondary endpoint. Study authors compared outcomes between patients with early disease progression (defined as occurring within 24 months following diagnosis) and those without disease progression within 24 months.

In this analysis of global registry data, the geographic distribution of AITL was 7% in South America, 21% in North America, 21% in Europe, and 18% in Asia. Significantly fewer cases were reported in South America compared with other regions of the globe (p=0.002).

Overall, the median age at the time of diagnosis was 64 years (range = 22-80). Approximately 60% of the patients were male and 90% had advanced-stage disease. Up to 74% of patients had lymphadenopathy, while 31% had splenomegaly and 22% had hepatomegaly.

Most patients (81%) received anthracycline-containing chemotherapy, either with etoposide (16%) or without etoposide (65%). The remaining patients received treatment consisting of chemotherapy regimens without anthracyclines (11%) or supportive care (8%).

A total of 106 patients treated with curative intent achieved a complete response, while 37 patients achieved a partial response (overall response rate = 69%). The respective five-year OS and PFS estimates at the median follow-up of 58 months were 44% and 32%. The authors observed a trend toward a slightly higher five-year OS rate among patients who received chemotherapy with etoposide versus without etoposide (50% vs. 43%; p=0.769).

The 27 patients who underwent consolidative autologous hematopoietic cell transplantation (AHCT) had significantly greater five-year OS estimates compared with the 56 transplant-eligible patients who did not receive AHCT (89% vs. 52%, respectively; p=0.05). The five-year PFS estimate was also higher among patients who received transplant (79% vs. 31%; p=0.022).

In the overall population, a multivariate analysis revealed four factors associated with inferior PFS:

  • age ≥60 years (hazard ratio [HR] = 2.98; p=0.003)
  • ECOG performance status >2 (HR=4.3; p=0.001)
  • elevated CRP (HR=1.9; p=0.003)
  • elevated ß2 microglobulin (HR=3.21; p=0.002)

Using these four risk factors, the authors developed the AITL score, a new risk model for predicting survival outcomes in this patient population. The combined score stratified patients into low- (17%), intermediate- (23%), or high-risk (60%) profiles, corresponding with progressively lower likelihood of OS or PFS at five years.

For example, those classified as low risk had a five-year PFS estimate of 41%, while those classified as intermediate and high risk had five-year PFS estimates of 37% and 13% (p=0.003). The five-year OS estimates for low, intermediate, and high risk were 63%, 54%, and 21%, respectively (p=0.0003).

“Patients with a high-risk AITL score had particularly dismal outcomes,” the authors wrote. “These findings require validation in a prospective cohort of homogeneously treated patients.”

According to the investigators, the AITL score showed the best discriminant power with a lower Akaike’s information criteria (524.0) and higher concordance index (per Harrell C-statistic; 0.785) compared with the International Prognostic Index, Prognostic Index for T-cell lymphoma, and Prognostic Index for AITL (see TABLE).

The authors noted that these hypothesis-generating findings lay the framework for future studies, but “novel therapeutic approaches and better understanding of disease biology are required to improve outcomes.”

The authors report no relevant conflicts of interest.


Advani R, Skrypets T, Civallero M, et al. Outcomes and prognostic factors in angioimmunoblastic T-cell lymphoma: final report from the International T Cell Project [published online ahead of print, 2021 Mar 30]. Blood. 10.1182/blood.2020010387.