Dose Interruption/Reduction Of Tyrosine Kinase Inhibitors In The First 3 Months Of Treatment of CML Is Associated With Inferior Early Molecular Responses and Predicts For An Increased Likelihood Of Discontinuation Of The 1st Line Agent

In CML, rapid reductions in tumor load, as defined by at least 65% Ph-negativity and/or a RT-qPCR <10%^IS at 3 mths, are associated with an improved probability of CCyR and better overall and progression free survivals. However some 15-20% of patients experience cytopenias shortly after starting TKI treatment and are managed by drug interruptions and/or dose reduction. The dilemma is whether these early periods of altered treatment should be considered in the interpretation of the 3 mth results. We investigated the impact of missing days of therapy and average dosing over the first 3 mths on the 3 mth RT-qPCR levels, CCyR at 12 mths and ability to remain on study, in patients treated in a phase III randomized study of imatinib (IM) versus dasatinib (DA) in newly diagnosed patients. RT-qPCR results at 3 mths were available for 585 (IM-292, DA-293) of 632 patients who completed 3 mths of therapy. Patients were divided according to their randomized drug and the amount missed: those who did not miss any days (IM0 [243], DA0 [211]), those who missed 1-14 days (IM1-14 [38], DA1-14 [37]) and those who missed >14 days (IM>14 [11], DA>14 [45]). More patients on DA missed days of dosing (28%) than on IM (17%) with a median number of missed days for DA of 16 (range 1-62) compared to 12.5 (range 1-42) for IM p=0.008. Achievement of a RT-qPCR <10%^IS at 3 mths for IM0, IM1-14 and IM>14 was 78.6%, 63.2% and 63.5% p=0.033 and 93.8%, 91.9% and 77.8% for DA-0, DA1-14 and DA>14 respectively p=0.001. Predictably the chance of a RT-qPCR <10%^IS at 3 mths is higher with DA than IM but DA is less well tolerated in the early months. RT-qPCR <10%^IS at 3 mths occurred in 78.7% and 93. 8% of patients who took more than 95% of the standard doses of IM and DA respectively in contrast to 60% (IM) and 84% (DA) in patients who missed more than 20% of the prescribed doses. Thus missing drug or dose reductions are associated with a reduced chance of achieving a RT-qPCR <10%^IS at 3 mths, although the effect is not seen for missing <14 days of DA, and is less marked for reduced average dosing of DA, suggesting that the higher potency of DA compensates for the impact of missing a few days of drug or dose reduction. 107/632 (17%) patients had discontinued the study by 12 mths and RT-qPCR results are available for all remaining patients. Using RT-qPCR <1% (MR2) as a surrogate for CCyR, the 12 mth MR2 rates for patients on IM were 78%, 78% and 90% in the IM0, IM1-14 and IM>14 cohorts p=0.5, and 96%, 88.6% and 79.5% p=0.026 for the DA0, DA1-4 and DA>14 cohorts, confirming superiority of DA over IM for MR2 and the potential impact of missing early doses of DA. Whether this is related to inadequate dosing or whether failure to tolerate the recommended dose is indicative of higher risk disease is not yet clear. We then studied whether missing drug in the first 3 months predicts the ability to stay on trial. The proportion of patients able to take the study drug consistently through mths 3-12 were 91%, 71% and 57% in the IM0/DA0, IM1-14/DA1-14 and IM>14/DA>14 groups respectively. This correlated with discontinuing the study (treatment failure) for 12.5%, 38.1% and 35.7% of patients in the IM0, IM1-14 and IM>14 cohorts and 4.4%, 12.8% and 18.8% in the equivalent DA cohorts. Thus the ability to tolerate daily drug in the first 3 mths predicts future tolerability and efficacy over the subsequent 9 mths and long-term compliance with the first line therapy. RT-qPCR monitoring was provided long-term for all patients entered into the study irrespective of their continuation in the study, so we were able to study the achievement of MR3 at 12 mths in an intention to treat analysis in all patients. With respect to average dosing over the first 3 mths MR3 was seen in 50%, 52.8% and 54.9% p=0.17 of patients who took >95%, 80-95% and <80% of the prescribed dose of IM and 68.1% (>95%), 59.4% (80-95%) and 53.1% (<80%) p=0.038 for the equivalent doses of DA. Similar results were obtained using the number of days of missed drug (data not shown), suggesting that early failure to tolerate IM as 1^st line does not impact on subsequent responses as effective alternative therapy is available for the majority of patients. In contract failure to tolerate DA is associated with a reduced chance of MR3 at 12 mths. Patients who experience drug cessation/reduction of either drug in the first 3 mths are less likely to obtain RT-qPCR <10%^IS at 3 mths and are more likely to require a change of drug in the longer-term and therefore require close observation during the first year.