Search

6545 Background: RIC allo-HCT has reduced transplant-related mortality (TRM) in patients with relapsed NHL. However, relapses do occur despite potential graft vs. lymphoma (GVL) effect. We hypothesized that adding Zevalin to Flu/Mel may improve disease control and reduce relapse post allo-HCT. Methods: Patients received In- Zevalin on day -21 followed by 90Y- Zevalin 0.4mCi/kg on day -14, flu 25 mg/m2 daily on days -9 to -5 and mel 140 mg/m2 on day -4. Rituximab (R) level was measured on Day -22 and -15 and if the level was ≥ 10 μg/ml, R was not given prior to In-Zevalin or 90Y- Zevalin to enhance biodistribution. Tacrolimus and sirolimus was used for GVHD prophylaxis. Between 10/2007 and 11/2011, 31 were treated. Median age 55 (range 27-67), median regimen =3 (range 2-8). Median time from diagnosis to HCT was 20 mo (range 5-105). Histology: DLBCL (including transformed lymphoma)=14 (45%), MCL=7 (23%), FL=5 (16%) and SLL/CLL=5 (16%). Disease status at HCT: 1CR=7, Relapse=9, ≥2CR=5, primary refractory =10. Fifteen had chemoresistant and 19 had FDG PET+ at HCT. Donors: sib=13, URD=18. Results: All patients engrafted with the median time to ANC ≥500 and platelet ≥ 20,000 was 14 (range 10-28) and 13.5 days (range 11-28), respectively. There were 10 deaths from disease progression (2) infection (5) GVHD (1) and multi-organ failure (2). TRM at day +100 and at 1 yr was 6.5% and 17%, Five with DLBCL relapsed between 3-7 mos. Grade II-IV acute GVHD was 65%, Grade III-IV was 16%, chronic GVHD was 65%. Fifteen became PET- at day +100 while 4 remained PET+ and relapsed. Twenty-one are alive at a median followup of 24 mos (range 2-46). The 2 yrs OS and PFS was 65% (95% CI, 51-75%) and 57% (95% CI, 46-67%), respectively. Univariate analysis identified disease status predict for OS and PFS while histology predict for PFS. Conclusions: This study demonstrates the feasibility of adding Zevalin to flu/mel in the allo-HCT setting for B-cell NHL and suggest that this approach could be used to provide early disease control before GVL effect takes place. Innovative approaches should be explored in DLBCL.

6547 Background: Gene expression profiling has improved risk stratification of diffuse large B-cell lymphoma (DLBCL) in the first line setting as shown by DNA microarray analysis and immunohistochemistry (IHC). 5-year overall survival (OS) for germinal center B-cell-like (GCB) DLBCL is significantly better than that for non-GCB DLBCL in newly diagnosed patients. However, in the relapsed/refractory setting, autologous stem cell transplant (ASCT) appeared to overcome the poor prognosis of cell of origin as defined by IHC using CD10, Bcl6 and MUM1 and there was no difference in survival between GCB and non-GCB groups. Ibritumomab tiuxetan (Z)-BEAM conditioning regimen for ASCT in DLBCL has produced promising response rates and progression free survival (PFS). The role of poor risk molecular markers in predicting outcome with the novel conditioning regimen Z-BEAM is unknown. Methods: We evaluated cell of origin (GCB, non-GCB) using IHC in 36 patients undergoing ASCT with Z-BEAM conditioning from 2002 to 2010. Median age of the patients was 54.5 years. There were 12 females and 24 males. 6 patients were in 1st CR/PR ( 17%), 12 had induction failure disease (33%), 9 were in 1st relapse (25%), 1 in 2nd relapse (3%) and 8 were in ≥2nd CR (22%). 27 patients had chemosensitive disease (75%) and all had prior rituximab exposure. Median number of prior treatments was 2. IPI at transplant was mainly low risk (78%). 22 patients had GCB (61%) and 14 (39%) had non-GCB DLBCL. Results: Median follow up is 25.6 months. 12 patients relapsed (33%) and 9 died (25%). The major cause of death is relapse/disease progression. At 2 years, median OS is 79% and median PFS is 64%. No significant difference in 2 year OS and PFS is noted between the GCB and non-GCB groups (p=0.07 and 0.06 respectively). Conclusions: Z-BEAM conditioning for ASCT may overcome the poor prognostic effect of non-GCB DLBCL in relapsed/refractory disease. Further evaluation of cell of origin using new markers like GCET1, LMO2 and FOXP1 could confirm these results as there seems to be better concordance of these markers with DNA microarray analysis.