Contributions
Abstract: PB1825
Type: Publication Only
Background
Achieving deep molecular response, ≥4.5-log reduction (MR4.5; BCR-ABL1 on the International Scale [IS] ≤0.0032%), is one of the important prerequisites for attempting treatment-free remission. Limited information is available on comparative rates of MR4.5 between nilotinib and dasatinib in second-line (2L).
Aims
This study aims to investigate time to achieving MR4.5 and major molecular response (MMR; ≥3-log reduction or ≤0.1% in BCR–ABL1 on IS) in CML-CP patients (pts) treated with nilotinib vs dasatinib in 2L.
Methods
An online physician panel approach was used to recruit oncologists (N=141) globally to conduct a retrospective medical chart audit. Physicians were instructed to select up to 4 pts who met the following criteria via a random letter generation scheme for the first letter of pt’s last name: diagnosed with CML-CP at age ≥18 years, initiated 2L nilotinib or dasatinib between 1/1/11 and 3/1/15, and had ≥12 mos of follow-up data after initiating 1L TKI. Multivariate Cox proportional hazards models accounting for country clustering random effects were used to assess the effect of nilotinib vs dasatinib on time to MR4.5 and MMR, adjusting for age, gender, Sokal risk score at diagnosis, hydroxyurea use before 1L TKI, 1st vs 2nd generation TKI as 1L, and reasons for 1L TKI discontinuation. Adjusted hazard ratios (HR) and 95% confidence intervals (CI) were reported. Adverse events (AEs) were also described.
Results
The study included 236 pts from Australia, Brazil, France, Germany, Italy, and Netherlands treated with nilotinib (N=115[49%]) or dasatinib (N=121[51%]) in 2L. Both groups had a similar mean follow-up of 23 mos, mean age of 57 years, and were 35% female. 8% of 2L nilotinib and 22% of 2L dasatinib pts were treated with the other 2nd generation TKI in 1L (p<0.01). A higher proportion of nilotinib pts had high-risk Sokal score (20.9% vs 11.6%, p=0.05) and received prior hydroxyurea (8.7% vs 3.3%, p=0.08) vs dasatinib. 85% and 11% of 2L nilotinib pts discontinued 1L TKI due to resistance and intolerance, respectively, prior to switching to nilotinib, vs 74% and 22% for 2L dasatinib pts (both p< 0.05). The univariate Cox model showed that nilotinib had a non-significantly higher rate of achieving MR4.5 than dasatinib (32% vs 31% at 24 mos for 2L nilotinib and 2L dasatinib pts, respectively, based on the Kaplan Meier estimator; unadjusted HR=1.09, 95% CI [0.87, 1.38], p=0.46); however, after multivariate adjustment, nilotinib reached a significantly higher rate of achieving MR4.5 (adjusted HR=1.36, 95% CI [1.07, 1.73], p<0.01) than dasatinib. Among those who achieved MR4.5, 45% of nilotinib pts maintained MR4.5 for ≥1 year vs 39% of dasatinib pts (p=0.60). Additionally, high-risk Sokal score (HR=0.31; 95% CI [0.14, 0.72], p<0.01) and resistance to 1L TKI (HR=0.60; 95% CI [0.42, 0.86], p<0.01) were inversely associated with achieving MR4.5. There was no significant difference in MMR achievement between 2L TKI groups. Over 3 times more dasatinib pts experienced pleural and pericardial effusion AEs than nilotinib pts (9.9% vs 2.6%; p=0.02). One nilotinib pt had ischemic heart disease-related AE vs none for the dasatinib group (p=0.49).
Conclusion
This retrospective chart audit study suggests that 2L nilotinib may be associated with a higher rate of MR4.5 than 2L dasatinib in CML-CP. Our results should be taken with caution as this study is susceptible to unmeasured confounding and biases due to its retrospective and observational nature. Rigorous clinical assessment in a prospective setting is needed to conclusively compare rates of patients achieving MR4.5.
Session topic: 8. Chronic myeloid leukemia - Clinical
Keyword(s): Tyrosine kinase inhibitor, Molecular response, Chronic myeloid leukemia
Abstract: PB1825
Type: Publication Only
Background
Achieving deep molecular response, ≥4.5-log reduction (MR4.5; BCR-ABL1 on the International Scale [IS] ≤0.0032%), is one of the important prerequisites for attempting treatment-free remission. Limited information is available on comparative rates of MR4.5 between nilotinib and dasatinib in second-line (2L).
Aims
This study aims to investigate time to achieving MR4.5 and major molecular response (MMR; ≥3-log reduction or ≤0.1% in BCR–ABL1 on IS) in CML-CP patients (pts) treated with nilotinib vs dasatinib in 2L.
Methods
An online physician panel approach was used to recruit oncologists (N=141) globally to conduct a retrospective medical chart audit. Physicians were instructed to select up to 4 pts who met the following criteria via a random letter generation scheme for the first letter of pt’s last name: diagnosed with CML-CP at age ≥18 years, initiated 2L nilotinib or dasatinib between 1/1/11 and 3/1/15, and had ≥12 mos of follow-up data after initiating 1L TKI. Multivariate Cox proportional hazards models accounting for country clustering random effects were used to assess the effect of nilotinib vs dasatinib on time to MR4.5 and MMR, adjusting for age, gender, Sokal risk score at diagnosis, hydroxyurea use before 1L TKI, 1st vs 2nd generation TKI as 1L, and reasons for 1L TKI discontinuation. Adjusted hazard ratios (HR) and 95% confidence intervals (CI) were reported. Adverse events (AEs) were also described.
Results
The study included 236 pts from Australia, Brazil, France, Germany, Italy, and Netherlands treated with nilotinib (N=115[49%]) or dasatinib (N=121[51%]) in 2L. Both groups had a similar mean follow-up of 23 mos, mean age of 57 years, and were 35% female. 8% of 2L nilotinib and 22% of 2L dasatinib pts were treated with the other 2nd generation TKI in 1L (p<0.01). A higher proportion of nilotinib pts had high-risk Sokal score (20.9% vs 11.6%, p=0.05) and received prior hydroxyurea (8.7% vs 3.3%, p=0.08) vs dasatinib. 85% and 11% of 2L nilotinib pts discontinued 1L TKI due to resistance and intolerance, respectively, prior to switching to nilotinib, vs 74% and 22% for 2L dasatinib pts (both p< 0.05). The univariate Cox model showed that nilotinib had a non-significantly higher rate of achieving MR4.5 than dasatinib (32% vs 31% at 24 mos for 2L nilotinib and 2L dasatinib pts, respectively, based on the Kaplan Meier estimator; unadjusted HR=1.09, 95% CI [0.87, 1.38], p=0.46); however, after multivariate adjustment, nilotinib reached a significantly higher rate of achieving MR4.5 (adjusted HR=1.36, 95% CI [1.07, 1.73], p<0.01) than dasatinib. Among those who achieved MR4.5, 45% of nilotinib pts maintained MR4.5 for ≥1 year vs 39% of dasatinib pts (p=0.60). Additionally, high-risk Sokal score (HR=0.31; 95% CI [0.14, 0.72], p<0.01) and resistance to 1L TKI (HR=0.60; 95% CI [0.42, 0.86], p<0.01) were inversely associated with achieving MR4.5. There was no significant difference in MMR achievement between 2L TKI groups. Over 3 times more dasatinib pts experienced pleural and pericardial effusion AEs than nilotinib pts (9.9% vs 2.6%; p=0.02). One nilotinib pt had ischemic heart disease-related AE vs none for the dasatinib group (p=0.49).
Conclusion
This retrospective chart audit study suggests that 2L nilotinib may be associated with a higher rate of MR4.5 than 2L dasatinib in CML-CP. Our results should be taken with caution as this study is susceptible to unmeasured confounding and biases due to its retrospective and observational nature. Rigorous clinical assessment in a prospective setting is needed to conclusively compare rates of patients achieving MR4.5.
Session topic: 8. Chronic myeloid leukemia - Clinical
Keyword(s): Tyrosine kinase inhibitor, Molecular response, Chronic myeloid leukemia