Proposal of a Novel Latin American International Prognostic Index (LATAM-IPI) for Diffuse Large B-Cell Lymphoma (DLBCL) By the Grupo Para El Estudio De Linfoproliferativos En Latino-America (GELL)

Introduction. There are different scoring systems to differentiate risk groups in patients with DLBCL treated with chemoimmunotherapy. Those systems have used the same 5 variables (age, performance status, LDH, stage, extranodal involvement) for 27 years. However, LATAM data have not been included in the development of previous scoring systems. It is important to mention that novel biological variables, such as albumin, beta-2-microglobulin (B2M) and platelet/lymphocyte ratio (PLR), have been reported and could improve discrimination (Villela et al. Blood 2019; 134Suppl_1: 1613). Therefore, we carried out a large, multinational study to develop and validate a LATAM-IPI score.

Methods. This is a retrospective cohort of 1030 patients with a diagnosis of DLBCL treated with standard chemoimmunotherapy with curative intent between 2010 and 2018. Data were obtained from 8 LATAM countries: Argentina, Colombia, Chile, Guatemala, Mexico, Paraguay, Peru, and Venezuela. The five classic IPI variables (age, ECOG, extranodal involvement, LDH, stage) were analyzed and albumin and PLR were added (Villela et al. Blood 2019; 134Suppl_1: 1613). B2M was not included because it was not requested regularly in all countries. Development of LATAM-IPI: The training set consisted of 85% of the sample, randomly selected, and the remaining 15% was reserved for internal validation. Using the training set, the univariate and multivariate association between clinical prognostic factors and OS was analyzed fitting Cox proportional-hazard models.

Outcomes. Clinical characteristics of the training (n=878) and internal validation (n=151) cohorts are shown in Table 1. There were no statistical differences in baseline characteristics between the cohorts. The median follow-up for the whole cohort was 36 months (IQR: 11-57). When exploring the classic IPI variables on the training set, all variables were associated with high risk of mortality [age 65-74, Hazard Ratio (HR) 1.24, 95% CI 0.96 to 1.58, p=0.08; age ≥75, HR 1.71, 95% CI 1.28 to 2.28, p=0.0003), ECOG (≥ 2, HR=2, 95% CI 1.61 to 2.53; p<0.0001), EN (≥2, HR=1.53, 95% CI 1.18 to 1.97; p=0.0012), stage (III/IV, HR=2.1, 95% CI 1.64 to 2.69; p<0.0001) and LDH (ratio 1.1-2.9, HR=1.55, 95% CI 1.22 to 1.97; p=0.0003; ratio ≥3, HR= 2.68, 95% CI 1.93 to 3.7, p<0.0001). Similarly, the biological variables Albumin (≤3.5 mg/dL, HR 2.37, 95% CI 1.9 to 2.95, p<0.0001) and PLR (≥273, HR= 1.52, 95% CI 1.23 to 1.87; p=0.0001) were associated with high risk of death. Next, these variables were evaluated by multivariate analysis. The independent variables were albumin (<3.5 mg/dL, HR 1.84, 95% CI 1.45 to 2.3, p<0.0001, 1 point), LDH (ratio 1.1 to 2.9, HR 1.30, 95% CI 1.02 to 1.67, p=0.03, 1 point; ratio ≥3, HR=1.84, 95% CI 1.31 to 2.5, p=0.0004, 2 points), advanced stage (HR 1.65, 95% CI 1.27 to 2.13, p=0.0001, 1 point), age (≥75, HR= 1.51, 95% CI 1.15 to 1.98, p=0.003, 1 point), and ECOG (≥2, HR 1.40, 95% CI 1.10 to 1.77, p=0.005). Now, for the development of LATAM-IPI, the groups were distributed as follows: 0 points, low; 1-3 points, intermediate; 4-6 points, high risk. There were no differences in the distribution of the risk groups between training and validation sets (Table 2). In the learning cohort, the 5-year OS rates for low, intermediate and high risk were 81%, 63% and 33%, respectively (p<0.0001). In the validation cohort, the 5-year OS rates for low, intermediate and high risk were 81%, 63% and 44%, respectively (p=0.02) (Figure 1).

Conclusions: Using multinational learning and validation cohorts including over 1,000 DLBCL patients treated with standard chemoimmunotherapy in LATAM, we developed a novel LATAM-IPI score using age ≥75 years, ECOG ≥2, advanced stage, LDH ratio (1.1-29 and ≥3) and albumin <3.5 mg/dl. Next steps are to disseminate our results with other involved researchers in LATAM to prospectively assess and reproduce our results. We expect this score will help to further define the prognosis of DLBCL patients in LATAM.