Abstract Abstract 693 Background: Accurate and early response prediction in CML patients treated with imatinib (IM) will facilitate timely therapeutic intervention in those patients predicted to respond poorly. The response of CP-CML patients to imatinib can be predicted by the BCR-ABL level at 3 months (early molecular response – EMR) with a level of ≥ 10% (IS) (EMR failure) associated with inferior outcomes. Response can also be predicted by OCT-1 activity (OA); a measure of active transport of IM into target leukemic cells. In the ALLG TIDEL II study patients were treated with 600mg of IM upfront, and dose escalate (800mg) if IM blood levels are <1000ng/ml at day 21. Patients were also dose escalated or switched to nilotinib for failure to achieve, or loss of, specified time dependent molecular milestones, including EMR failure. The aims of this study were to (1) assess the predictive value of EMR (2) assess the impact of factors which likely contribute to EMR such as dose intensity and OA, and (3) assess factors that modify the risk profile within the cohort who achieve EMR. Results: 210 patients were enrolled to TIDEL II. For this analysis 200 patients were analysed based on the availability of complete OA, day 21 blood levels and 3 month BCR-ABL% data. EMR was associated with a significantly higher rate of major molecular response (MMR) compared to that of patients with EMR failure (n=25/200) at both 12 (70% vs 20% p<0.001) and 24 months (82% vs 24% p<0.001), and a significantly higher event free survival (EFS)^ (83% vs 60%, p=0.013). Similarly, patients with high OA achieved MMR at a higher rate than those with low OA (12 mo 73% vs 53% p=0.003 and 24 mo 82% vs 66% p<0.001), and had significantly higher EFS (89% vs 71% p=0.011). IM drug levels at day 21 were significantly lower in patients with EMR failure (median 1050 vs 1620ng/ml: p=0.004). Levels <1000ng/ml were associated with a significantly higher rate of EMR failure (26% compared with 9% if levels >1000ng/ml p<0.001). Importantly, the negative impact of low IM levels was confined to patients with low OA. EMR failure was no different in patients with high OA regardless of drug level (17% vs 9% p>0.05), whereas 35% of patients with low OA and IM levels <1000 ng/ml had EMR failure compared to 11% in patients with low OA and satisfactory drug levels. (p=0.008). This suggests that IM dose and the intracellular level of drug achieved (OA) are key determinants of the 3 month response. 22/25 patients with EMR failure remained on study at 6 months, 21 of whom were either dose increased or switched to nilotinib. Four of 8 patients with EMR failure and high OA achieved 1% BCR-ABL (IS) by 6 months, compared to 1/14 (7%) of patients with low OA and EMR failure. This suggests that early intervention in patients with EMR failure may not be effective in patients with low OA. Interestingly, the risk of events for patients with EMR failure, compared to those who achieve EMR but have low OA was equally high (EFS; 60% v 74% p=0.280). Importantly, 5/6 (83%) patients who transformed had low OA however 3/5 had achieved EMR. Similarly, the risk of mutation development was significantly associated with low OA, and not predicted by EMR (Table 1). While no patient with EMR failure achieved MR4.5 by 24 months, there was a significant difference in the proportion of patients achieving MR4.5, based on OA, within the cohort of patients who achieve EMR (p=0.003) (Table 1). Conclusion: Despite early intervention to increase dose intensity in cases of low drug levels or EMR failure (dose escalation and/or switch to nilotinib), OA, drug levels, and EMR all remain strong predictors of outcome. Importantly the achievement of EMR does not represent a safe haven for patients with low OA. These data suggest that optimising outcomes in high-risk CML cases may not be achievable even with early proactive intervention. Hence, the use of OA and other potential biomarkers at diagnosis may better identify patients with high-risk CML in whom therapy, other than currently available TKIs, may be required. Disclosures: White: Novartis Oncology: Honoraria, Research Funding; BMS: Research Funding; CSL: Research Funding. Yeung:Novartis Pharmaceuticals: Honoraria, Research Funding; Bristol-Myers Squibb: Honoraria, Research Funding. Grigg:Novartis: Membership on an entity's Board of Directors or advisory committees, Research Funding. Hughes:BMS: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Novartis: Membership on an entity's Board of Directors or advisory committees, Research Funding; Ariad: Membership on an entity's Board of Directors or advisory committees; CSL: Research Funding.