Impact of Partial Response Versus Complete Response before RIC/RTC Allogeneic Stem Cell Transplantation on Survival in a Cohort of 115 Chemosensitive Advanced NHL

Allogeneic Hematopoietic stem cell transplantation (Allo-HSCT) is a curative option for advanced Non Hodgkin Lymphoma (NHL). Reduced intensity conditioning or reduced toxicity conditioning (RIC/RTC) allows improving survival and decreasing non-relapse mortality. Nevertheless it is well known that status disease before transplant is highly related to survival. Indeed, patients in non-responsive disease (ie stable disease or progressive disease (SD/PD)) have very poor outcomes compared to patients achieving a response. We addressed the question whether in responsive patients the level of response (partial response (PD) vs. complete response (CR)) had a profound impact on outcome in the context of RIC/RTC transplantation.

Inclusion criteria for this bi-centric retrospective study were 1) first Allo-HSCT 2) aggressive NHL including Diffuse Large B-cell Lymphoma (DLBCL), Mantle-cell Lymphoma (MCL) and nodal T-cell NHL 3) Matched related donor (MRD), 8/8 or 10/10 matched unrelated donor (MUD) and Haploidentical related donor (HRD) 4) conditioning including RIC or RTC. All patients in this cohort are considered as intermediate risk group in the DRI classification (Armand et al, Blood, 2012). Extranodal T-cell NHL, NK-cell NHL, Cord Blood Unit transplant, partially matched (HLA 7/8 or 9/10) unrelated donor, myeloablative conditioning and patients in SD/PD before transplantation were not included in this study. We used Cheson criteria (Cheson et al, JCO, 2007) to evaluate the disease status before Allo-HSCT and we focused on patients with PET positivity using Deauville criteria before transplantation.

115 patients included between 06.1999 and 09.2013. Diagnoses were DLBCL, MCL and nodal T-cell NHL in 40%, 30% and 30% respectively. Median age was 54 years old. Median line of treatments before transplantation was 3 (1-5), 81% of patients received an autologous SCT (31% in a tandem auto/allo and 50% relapsed after auto SCT). Donor types were MRD, MUD and HRD in 57,4%, 25,1%, and 57,4% of patients respectively. Conditioning was for 86% of patients a RIC and for 14% a RTC generally including Thiothepa.

Patients were in CR in 80 cases (69.6%) and in PR in 35 cases (31.4%) according to Cheson criteria. 5-years PFS and OS for patients in CR were 61.1% (49,1-76) and 64,3% (52,6-78,6) respectively, whereas 5-years PFS and OS for patients in PR were 38,7% (21,9-68,4) and 38,1% (21,5-67,6) respectively with a significant difference (p=0,04 and p=0,05 respectively). Among this cohort, 80 patients (33 DLBCL, 19 MCL and 28 nodal T-cell NHL) had a PET TDM before Allo-HSCT and 11 patients had a positive PET using Deauville criteria. PFS and OS for patients with PET negative were 65.5% and 65,8% respectively and PFS and OS for PET positive patients were 46,8% and 45,5% (p=0,32 and 0,33 respectively). In multivariate analysis including histology, type of donor, type of conditioning, age and status disease at transplantation, the only factor having a negative impact both on PFS and OS were disease status according to Cheson criteria: PR at transplantation is associated with a lower PFS an OS with a Hazard Ratio of 2 (p= 0,04).

In conclusion: for aggressive NHL, PR before Allo-HSCT evaluated with Cheson Criteria increases by 2 fold the risk of death or relapse in comparison with CR, indicating that either better response should be searched or transplant procedure should be adjusted to this risk.