EWALL prognostic index for the risk stratification of adult patients with ALL

Accurate risk stratification is important for optimal management of patients with acute lymphocytic leukemia (ALL).¹ Several risk factors are typically used to risk stratify patients, including age, white blood cell count, measurable residual disease status, and the presence or absence of high-risk genetic abnormalities; these factors are often applied independently, which may reduce the predictive capabilities.¹ The recently developed and validated UKALL integrated prognostic index uses continuous data and provides a flexible mechanism for the risk stratification of adolescents and young adults with ALL.¹

Recently, Enshaei et al.¹ developed and validated the European Working Group for Adult ALL prognostic index (EWALL-PI) using a modified version of the UKALL prognostic index to reflect the differences between pediatric and adult ALL. The findings were published in Blood Advances, which we summarize below.

Methods and patient population

• Data were collected from patients aged 15–67 years with newly diagnosed ALL from four clinical trials:
  • UKALL14 (NCT01085617)
  • NILG-ALL 10/07 (NCT00795756)
  • GINEMA-LAL1913 (NCT02067143)
  • PETHEMA-ALL-HR2011 (NCT01540812)
• The EWALL-PI method is shown in Figure 1.
• Study endpoints included overall survival (OS) and relapse rate.

Figure 1. Method for calculating the EWALL-PI score*

EOI, end of induction; EWALL-PI, European Working Group for Adult ALL prognostic index; GR-GEN, good risk genetic features; HR-GEN, high risk genetic features; MRD, measurable residual disease; WCC, white blood cell count.
*Adapted from Enshaei, et al.¹

Key findings

• EWALL-PI scores were calculated for a total of 778 patients (UKALL14, n = 253; NILG-ALL 10/07, n = 109; GINEMA-LAL1913, n = 108; PETHEMA-ALL-HR2011, n = 308).
• EWALL-PI scores were significantly higher for patients who experienced relapse and/or death vs those who did not.
• Univariate Cox regression analysis showed that each unit increase in EWALL-PI score was associated with an increased risk of relapse and increased risk of death by 24% and 32%, respectively, with no evidence of heterogeneity between the four trials.
• The EWALL-PI model outperformed all the original risk stratification models used in each trial for relapse risk and OS prediction.
• Threshold analysis and Youden’s index suggested that an EWALL-PI value of 2.50 was the optimal cutoff for identifying chemotherapy-treated patients with a low cumulative incidence of relapse and high OS at 3 years.
• The EWALL-P1 cutoff of 2.50 was validated across all four trials for the cumulative incidence of relapse and the UKALL14, NILG-ALL 10/07, and PETHEMA-ALL-HR2011 trials for OS (Figure 2).
• Subgroup analysis revealed that the prognostic effect of the EWALL-PI was robust across the major patient subgroups.

Figure 2. 3-year A CIR and B OS by EWALL-PI risk group across the UKALL14, NILG-ALL 10/07, GINEMA-LAL1913, and PETHEMA-ALL-HR2011 trials* 

CIR, cumulative incidence of relapse; EWALL-PI, European Working Group for Adult ALL prognostic index; OS, overall survival.
*Data from Enshaei, et al.¹