We have reported that S-phase kinase-associated protein 2 (Skp2) expression in tumor cells is an unfavorable prognostic factor in diffuse large B cell lymphoma (DLBCL) with CHOP-R. Therapeutic strategies other than CHOP-R should be needed for DLBCL patients exhibiting a high Skp2 expression at the time of diagnosis. High dose chemotherapy followed by autologous peripheral blood stem cell transplantation (APBSCT) is mainly adapted for DLBCL patients in high and high intermediate risk groups in IPI. However, the definite beneficial evidence for APBSCT in DLBCL remains unclear.In the present study, we investigated the clinical significance of Skp2 expression in the patients with DLBCL treated with CHOP-R plus high-dose chemotherapy followed by APBSCT. We retrospectively analyzed the outcomes of 93 patients (age range: 14-65). The patients were newly diagnosed as having DLBCL from 2002 to 2012, and were treated with either CHOP-R plus upfront APBSCT (n=31) or CHOP-R (n=62) in the19 hospitals in Kyushu, Japan. All patients had high and high intermediate risk in IPI. All biopsy specimens were immunohistopathologically reconfirmed by one pathologist with expertise before entering into this study. The median follow-up time was 2.3 y. Survival analyses were performed using the Kaplan-Meier method and the Cox proportional Hazard model with inverse probability of treatment weight (IPTW).In background of the patients, age was younger in transplant group (mean age 52) than in non-transplant group (mean age 59). CR rate was higher in non-transplant (30.7%) than in transplant (12.9%). Sex, stage, PS, LDH extranodal lesion and IPI showed no difference in both group.Overall survival (OS) for 3 years were 77.0% and 59% in transplant group and in non-transplant group (P=0.077), respectively. Progression-free survival (PFS) for 3 years, 65.7% and 53.2% in transplant and non-transplant group (P=0.233), respectively. In Skp2 high expression (positive rate >40%) group (n=37), 3y-OS was 44.9% and 16.4% in transplant (n=13) and non-transplant (n=24) group (P=0.065), respectively. 3y-PFS was 40.3% and 7.5% in transplant and non-transplant (P=0.032), respectively (Fig A). However, in low Skp2 expression group (n=56), OS was 100% and 92.7% (P=0.254), PFS was 83.6% and 82.6% (P=0.945)(Fig B)in transplant (n=18) and non-transplant (n=38) group, respectively. In propensity score analysis using center effect, age, extranodal lesion, IPI and CR rate as logistic regression model, transplant group showed the excellent benefit in OS ( OR= 0.469, 95%CI =0.266-0.825, P=0.009) and PFS (OR=0.456, 95%CI=0.255-0.815, P=0.008) in Skp2 high expression group (n=37).P value for transplant x Skp2 interaction was P=0.643 in OS and P=0.020 in PFS.In conclusion, Skp2 was a good biomarker for indication of ABSCT for high risk DLBCL patients. In low expression of Skp2, patients should not be treated with high dose chemotherapy followed by APBSCT, even though in patients with high risk in IPI. However, ABSCT have some advantage in DLBCL patients with high expression of Skp2. Disclosures:No relevant conflicts of interest to declare.