ELDERLY WOMEN WITH DIFFUSE LARGE B-CELL LYMPHOMA (DLBCL) TREATED WITH RITUXIMAB BASED CHEMOTHERAPY DO NOT HAVE BETTER OUTCOME COMPARED TO MEN

ABSSUB-6033

Background: The DLBCL patient’s outcome has been significantly improved by rituximab introduction (Coiffier, 2002 ). The most powerful and used prognostic system remains IPI  designed in pre- rituximab era (Shipp 1993), although some modifications have to be studied - i.e. R-IPI (Sehn 2007), NCCN-IPI (Zhou 2014). The recent publications suggest that women could have different outcome compared to men when treated with rituximab either as part of the induction (Pfreundschuh 2013) or maintenance (Jaeger 2013, Gisselbrecht 2012). The different rituximab clearance in older women compared to other gender-age subgroups could be one of the possible explanations (Muller 2012), although the data is not consistently  explained by this factor only.

Aims: Evaluate the impact of different prognostic factors including IPI, gender, bulk and BMI in unselected population of DLBCL patients treated with anthracycline based chemotherapy combined with rituximab.

Methods: The unselected cohort of consecutive 611 patients with DLBCL (primary CNS lymphoma were excluded) treated with R-CHOP-like chemotherapy at General University Hospital in Prague between 2001-2012 has been retrospectively analyzed. IPI and its components, gender, bulk (7.5 cm) and BMI were the factors tested. Statistical analysis was performed by Cox regression model and supported by univariate analysis using Kaplan-Maier estimation model.

Patients: The median age was 60 years, 51.6 % men, 48.4 % women. IPI subgroups: good IPI (low or low intermediate risk) 48.8 %, poor IPI (high or high intermediate risk) 46.6 %, unknown IPI 4.6 %. R-CHOP regimen was the most frequent (77.1 %). With median follow up of 4.5 year, the median PFS and OS wasn't reached. The 5-year PFS and OS were 67.0 % ± 2.1 %, 72.6 % ± 2.0 % respectively.

Results: The following factors had significant prognostic impact on PFS and partially on OS: IPI (concurrently significant impact on OS), bulk (significant impact on OS has not been generally confirmed in multivariate analysis), but not the gender or BMI. These factors remain independent in multivariate analysis. When we analyzed younger and older population (60 years cut off) separately, for younger patients the IPI and bulky remain the independent prognostic features, however for elderly only the IPI remains the independent prognostic factor. We did not find linear relationship between BMI and patient's outcome, but for proper evaluation of this factor further analysis ought to be done. We tested separately the gender influence in elderly population of good and poor IPI subgroups. Surprisingly, the data revealed  that in good IPI subgroups men (median PFS not reached)  have better outcome compared to women (median PFS 6.2 years) - p= 0.0347, whereas OS wasn't significantly different, and  also that there is no difference in poor IPI risk subgroups.

Summary/Conclusion: We are not able to confirm the findings that elderly women have better outcome compared to men in large series of DLBCL patients. Based on these data we are not strongly convinced that it is worthwhile to change the rituximab treatment strategy according to gender in routine clinical practice.

This work was supported by PRVOUK 27.

Keywords: Diffuse large B cell lymphoma, DLBCL, Gender, Rituximab

http://www.ehaweb.org/