Abstract
BACKGROUND: About one third of patients with diffuse large B-cell lymphoma (DLBCL) eventually die from their disease. Thus, it is extremely important to identify powerful and reliable predictive and prognostic markers. In real life patient management, the clinical relevance of the assessment of the Cell of Origin (COO), of MYC and BCL2 protein overexpression, is still not fully clarified. The aim of the study is to evaluate the predictive and prognostic value of COO with Lympho2Cx assay by Nanostring technology and BCL2 and MYC overexpression by immunohistochemistry (IHC), in a real-life context.
METHODS: This is a retrospective multicenter study which recruited 154 DLBCL patients treated with R-containing regimen between June 2010 and December 2016 in Modena, Rome (Italy) and Haifa (Israel). All clinical data were recorded at diagnosis and during follow up, including response assessment and survival outcome. COO was determined on formalin-fixed paraffin-embedded diagnostic specimens either by IHC using a panel of antibodies which included CD10, BCL6, MUM1 (i.e. Hans algorithm), or at RNA level with Lymph2Cx assay by Nanostring technology. Expression of MYC and BCL2 was evaluated by IHC. Event free survival (EFS) is defined as the time from diagnosis to the time of last follow-up, or to one of the following events: any response other than complete remission (CR) at the end of therapy or death from any cause. It is assessed by Kaplan-Meier estimates and groups of risk are compared using the log rank test.
RESULT: We have currently evaluated the data of 60% (95 out of the 154) of the patients enrolled in the study (Tab 1). After a median follow up of 49 months, EFS is 63% (95CI 51-72%).
In univariate analysis, the variables with the greatest impact on the response and on EFS are absolute granulocyte count and BCL2 expression (MYC ongoing). Patients with low IPI showed better survival in comparison with patients with high IPI, but the difference is not statistically significant. By IHC non-GCB subtype was more common than GCB (56% vs. 44%); by Nanostring ABC, GCB and unclassified subtypes were 33%, 50% and 17%, respectively, and K statistics was 0,647, showing a substantial agreement between the results obtained by IHC and by Lympho2Cx assay. No statistically significant differences were observed in EFS among ABC, GCB and unclassified subtypes. However, BCL2 protein overexpression in ABC subtype is associated with shorter EFS.
CONCLUSION: Early retrospective studies showed a survival advantage for GCB-type disease. Clinical trial data evaluating the impact of COO determined by GEP on prognosis have shown inconsistent results with 2 studies observing inferior survival of ABC subtype and 2 German studies showing no significant differences in PFS or OS among COO subtypes. Determination of COO by Lympho2Cx is attractive as it is relatively easy and rapid to perform and potentially applicable to clinical practice. However, these preliminary results do not support, at this point, its use as prognostic factor in clinical practice. The analysis of the remaining 59 patients will help to clarify the role Lympho2Cx assay in the context of real life.
Table 1. Patients baseline characteristics.
Tadmor:ABBVIE: Consultancy; ROCHE: Research Funding; JNJ: Consultancy; NOVARTIS: Consultancy; PFIEZER: Consultancy.
