Abstract Abstract 3298 Poster Board III-186 Background Nilotinib is a potent and highly selective BCR-ABL kinase inhibitor, approved for the treatment of patients with Ph+ CML in CP or AP who are resistant or intolerant to imatinib therapy. After 24 months (mos) of follow-up of imatinib-resistant and -intolerant patients treated with nilotinib in the pivotal phase II study, achievement of major cytogenetic response (MCyR) was demonstrated in 59% of patients and complete cytogenetic response rates (CCyR) in 44% of patients. The purpose of the analysis was to identify baseline disease factors or time-dependent response endpoints that predict for progression-free survival (PFS). Methods Imatinib-resistant and -intolerant CML-CP patients (N = 321) were treated with nilotinib 400 mg twice daily. A Cox regression model was performed on 21 baseline disease factors or response endpoints. The effect of these factors on PFS was evaluated. Progression was defined as death or discontinuation due to disease progression or progression to accelerated phase or blastic phases. Q-PCR data for BCR-ABL were not included. Patients with T315I mutation at study entry were excluded (n=6). Results Ten factors were included in the multivariate analysis that were all identified as significant factors in the initial univariate analysis (at a significance level of 0.05): minor CyR (≤ 65% Ph+ metaphases) at or before 6 mos, MCyR at or before 12 mos, baseline complete hematologic response (CHR), highest prior imatinib dose (< 600 mg/day), absence of baseline mutations, absence of mutations with low sensitivity (E255K/V, Y253H, F359C/V), intolerance to imatinib (vs resistance), baseline hemoglobin (Hb; continuous variable), baseline basophilia (continuous variable), and baseline percent Philadelphia positive (Ph+) metaphases (continuous variable). One other marginally significant factor was included in the multivariate analysis (prior response to CML therapy). Following the multivariate analysis of these 11 factors, only 4 were found to be independently significant, including: MCyR at or before 12 mos (hazard ratio, HR = 0.359, P < 0.0001); baseline hemoglobin ' 120 g/L (HR = 0.985, P = .0291); baseline basophils < 4% (HR = 1.070, P = .0128) and the presence of baseline mutations with low sensitivity to nilotinib (E255K/V, Y253H, F359C/V) (HR = 6.097, P = <.0001). The four binary factors (baseline basophilia ≥ 4%, baseline Hb < 120 g/L, low sensitivity mutations and MCyR at or within 12 mos nilotinib therapy) were then combined to create a single prognostic score with values between 0 (best) and 4 (worst). Based on these scores, patients were stratified into 5 prognostic subsets with the best group having none of the 3 unfavorable disease risk factors and had MCyR at 12 mos while patients in the worst group had all of the 3 unfavorable disease risk factors and no MCyR at 12 mos. The 24 mo PFS rates for patients based on prognostic scores are shown in Table 1. As expected, PFS rates decreased with increasing risk scores. When only considering patients with at least 12 months follow-up the respective 24 mo PFS rates are 90%, 79%, 67%, and 37% (no patients with prognostic score of 4), for patients with prognostic scores of 0, 1, 2, and 3, respectively. The multivariate analysis did not find that at least minor CyR or PCyR responses at or within 6 mos of initiation of nilotinib significantly influenced 24 mo PFS compared to these other factors. Conclusion The preliminary results of this multivariate analysis of 21 disease or response factors identified 4 factors that predict for PFS in imatinib-resistant and -intolerant patients receiving nilotinib therapy in this large dataset. The achievement of MCyR by 12 mos, lack of baseline mutations with low sensitivity to nilotinib, baseline Hb > 120 g/L, and baseline basophilia (< 4%) were all associated with the most favorable PFS on nilotinib therapy. These results suggest that even in the presence of poor disease characteristics, nilotinib therapy provides patients with clinical benefit as measured by PFS. This developed risk model may be useful to determine the benefit of continued nilotinib therapy versus allogeneic transplant. Disclosures Jabbour: Bristol-Myers Squibb: Speakers Bureau; Novartis: Speakers Bureau. Giles:Novartis: Consultancy, Research Funding; BMS: Research Funding; Merck: Research Funding; Clavis: Research Funding. Bhalla:Novartis: Honoraria, Research Funding; Merck: Honoraria. Pinilla-Ibarz:Novartis: Membership on an entity's Board of Directors or advisory committees, Research Funding, Speakers Bureau. Larson:Novartis: Consultancy, Honoraria, Research Funding. Gattermann:Novartis, Celgene: Honoraria, Participation in Advisory Boards on deferasirox clinical trials, Research Funding. Ottmann:Novartis: Consultancy, Honoraria, Research Funding, Speakers Bureau; Bristol-Myers Squibb: Consultancy, Honoraria, Research Funding. Wang:Novartis: Employment. Woodman:Novartis: Employment. Baccarani:Novartis Pharma: Consultancy, Honoraria, Research Funding, Speakers Bureau; Bristol-Mayer Squibb: Consultancy, Honoraria, Research Funding, Speakers Bureau. le Coutre:Novartis: Honoraria, Research Funding; BMS: Honoraria. Cortes:Novartis: Honoraria, Research Funding, Speakers Bureau; BMS: Honoraria, Research Funding, Speakers Bureau; Wyeth: Honoraria, Research Funding, Speakers Bureau.