Anaplastic Large Cell Lymphomas (ALCL) are rare T-cell neoplasms grouped according to whether they express the fusion protein anaplastic lymphoma kinase (ALK+) or not (ALK-). ALK+ ALCL has consistently been found to have a favorable outcome compared to ALK- ALCL, but ALK+ ALCL is also associated with young age and other low risk features and not all studies have found ALK-expression to be an independent prognostic factor. In this population-based study, we aimed at analyzing the outcome and risk factors for survival in a bi-national cohort of patients with systemic ALCL.
All adult (>18 years) patients with systemic ALCL in the Swedish and Danish Lymphoma Registries diagnosed between 2000 and 2010 were included in the study. Primary cutaneous ALCL cases were excluded. The diagnosis of ALCL was established in routine care and no study-specific pathology review was performed.
A total of 371 patients (ALK+ ALCL n=122) were identified, representing 1.3% of all lymphomas, through both national registries. ALK-status was missing in 33 patients (ALK u ALCL). The median follow-up was 7.2 years. ALK+ patients were younger than ALK- patients (median age 40 versus 66 years, p<0.001).
In all, 209 patients died (ALK+ n=32, ALK- n=151, ALK u n=26) and among the 328 patients with available relapse data, 118 patients experienced relapse or progression (ALK+ n=20, ALK- n= 83, ALK u n=15). The 5-year overall and progression-free survival (OS and PFS, respectively) were 78% and 64% in ALK+ ALCL, 37% and 32% in ALK- ALCL and 27% and 25% in ALK u ALCL.
Data on primary treatment was available in 341 out of 371 patients (92%). The majority of patients (n=278, 82%) was treated with cyclophosphamide, doxorubicin, vincristine and prednisone (CHOP) or CHOP plus etoposide (CHOEP). Up-front autologous stem cell transplantation (ASCT) was performed in 38 patients with ALK- ALCL and in 6 patients with ALK+ ALCL. Most ALK- ALCL patients undergoing up-front ASCT consolidation received CHOEP as induction treatment.
Age had a profound impact on survival and based on the Kaplan-Meier estimates the age cut-offs described for the National Comprehensive Cancer Network-International Prognostic Index (NCCN-IPI) were used. All features, including treatment with CHOP compared to CHOEP, that were associated with survival at the level of p<0.1 in univariable analysis were tested in a multivariable model. The only independent risk factors in the multivariable analysis were treatment with CHOEP, which was associated with better OS (HR 0.48 95% CI 0.32-0.74, p=0.001), and increasing NCCN-IPI score (HR [for each increment] 1.6 95% CI 1.5-1.8, p<0.001), which was associated with inferior OS.
A separate multivariable risk factor analysis for OS was performed in patients treated with CHOEP (N=108). In this analysis, age (HR 2.9 95% CI 1.5-5.3, p=0.001), ALK-negativity (HR 2.6 95% CI 1.2-6.0, p=0.020) and elevated LDH (HR 2.1 95% CI 1.0-4.3, p=0.047) were independently associated to worse OS. Assigning 0,1 and 2 points for age <40, 40-60 and 60-75 respectively, ALK negativity 1 point and elevated LDH 1 point, we created a score that identified 4 groups with significantly different OS. Patients with a score of 3 or 4 had a similar OS, and were thus combined.
This population-based study based on two national registries reports the outcome of the largest cohort of adult ALCL patients published so far. Our study confirms the favourable outcome of ALK+ ALCL patients and the association with low-risk features. The addition of etoposide to CHOP was independently associated with a superior OS, and when adjusting for this treatment modification, the impact of ALK-expression on OS was mitigated.
We also performed a separate risk factor analysis in the group of patients receiving CHOEP treatment. Age, ALK-negativity and elevated LDH were independent risk factors for OS in this group and were assembled in a proposed novel score, which could represent a useful tool in future management strategies in ALCL.
Our data supports that the addition of etoposide to CHOP, if tolerated, is an important component in the treatment of ALCL and that the impact of ALK-expression on outcome is affected by treatment. Based on multivariable risk factor analysis in CHOEP treated patients, we propose a novel ALCL-specific score for future validation in independent cohorts.
Relander:Respiratorius: Patents & Royalties: valproate for DLBCL.
Asterisk with author names denotes non-ASH members.