Validating the Halp Score (Hemoglobin,Albumin, Lymphocytes, and Platelets) and the Neutrophil/Lymphocyte Ratio (NLR) As Prognostic Factors for Overall Survival in Patients with Diffuse Large B-Cell Lymphoma. Retrospective Analysis By the Grupo De Estudio De Latino America De Linfoproliferativos (GELL)

Introduction. Diffuse large B-cell lymphoma (DLBCL) is the most common type of malignant lymphoid neoplasm. The international Prognostic Index (IPI) and its variants are the main prognostic tools used in DLBCL with value in the rituximab era. Certain molecular biomarkers and genetic signatures in DLBCL have been identified, but the cost and unavailability in Latin America (LATAM) are an issue. Therefore, using accesible tools in LATAM is a potencially unmet need. Vlatka (2021), reported a new score (HALP: hemoglobin, albumin, lymphocytes, and platelets) for overall survival (OS) in DLBCL. These parameters are easy to access in LATAM centers. Thus, we aim to validate this novel OS score in the GELL database, and expanding database from LATAM countries.

Methods. This was a retrospective analyses of patients with de novo DLBCLtreated with curative intent between 2010 and 2018. The classic 5 international prognostic index variables [age, ECOG, Extranodal (EN) involvement, Lactate dehydrogenase (LDH), and advanced stage] and low serum albumin defined as ≤3.5mg/dL(divided in low, 3.4-2.5 mg/dL, and very low, ≤2.4 mg/dL), as previously described by our group (Villela,2018&2019). We also calculated the NLR considering an adverse prognostic factor >4, as previously published by us (Beltran&Villela,2020). The HALP score was calculated using the following formula: hemoglobin(g/L) x albumin (g/L) x absolute lymphocyte count (k/μL) divided by platelets (k/μL). The ROC method was used to calculate the HALP cut-off. Demographic characteristics are reported using descriptive statistics. Cox proportional-hazard regression model was used to evaluate parameters associated with OS, and survival curves were estimated with the Kaplan-Meier (KM) method.

Outcomes. 1407 patients were included, who were treated with standard RCHOP (n=1112,79%), RminiCHOP (n=93,7%), REPOCH (n=111,8%), and CHOP (n=91,6%). The median follow-up was 36 months (IQR: 7 to 56). NLR >4 was observed in 18.5%. The median HALP score was 24 (IQR, 12 to 40), and the cut-off of ≤13 (AUC 0.58;95%CI 0.56 to 0.61; p<0.0001) was considered an adverse prognostic factor, which was observed in 451 patients (32.7%). Female sex, ECOG >1, EN >1. high LDH, advance stage, low and very low albumin, and NLR>4 were associated with HALP≤13, but age was not (Table 1). Patients with HALP ≤13 had a lower 3-years OS rate than HALP>13 (48% vs. 66%, respectively; p<0.001). Table 2 shows the univariate & multivariate analysis of the variables with independent influence on OS, including HALP & NLR. In multivariate analysis, HALP and EN involvement were left out of the model ( Harrell´s C-Index, 0.73;95%CI 0.69 to 0.76).

Conclusion: NLR>4, but not HALP ≤13, could prognosticate inferior OS in LATAM patients with DLBCL treated with curative intent. The adverse prognostic value of NLR>4 should be validated prospectively in other cohort of DLBCL patients.