With the advent of TKIs, significant improvement has been made in the survival outcome of CML patients. Many prognostic factors have in identified in chronic phase CML patients. However, prognostic factors in blastic phase (BP) and accelerated phase (AP) of CML patients have not been previously investigated, especially in the era of TKIs.
We have analyzed CML patients either been diagnosed in or progressed into BP or AP from 2000 to 2010 enrolled into the clinical trials with different TKIs such as imatinib, nilotinib, dasatinib, and bosutinib at MD Anderson Cancer Center. All demographic, clinicopathologic variables and medication data including TKIs and other medication were collected at the time of the diagnosis of BP or AP. Primary outcomes; progression free survival (PFS) and overall survival (OS) were defined as time from diagnosis of either BP or AP to treatment failure and death, respectively.
Among a total of 260 CML patients, mean age was 51.4 (SD 14.1). Males were 59%. Caucasians were 72%; African-Americans, 12%; and Hispanics, 14%. Only 17% reached complete hematologic remission and 62% eventually died during the follow up (mean 4.1 yrs, SD 4.0 yrs). 26% were in BP and 74% in AP. 38 patients were in BP (n=9) or AP (n=29) at initial diagnosis. Mean time from initial diagnosis in chronic phase to BP or AP was 4.1 yrs (0 to 23 yrs). Among BP, 17% had lymphoid BP and 84% myeloid BP. Cytogenetic analysis at diagnosis demonstrated 38% had only Philadelphia chromosome (Ph) present while 62% had other chromosomal abnormalities besides Ph. 30% had received one or more prior TKI by the time of transformation. 64% started on monotherapy imatinib (no prior TKI), 15% nilotinib (86% prior TKI), 14% dasatinib (93% prior TKI), and 3% bosutinib (100% prior TKI); 3.5% received imatinib combined with other agents (57% prior monotherapy with imatinib). Median PFS was 0.7 yr and median OS, 2.4 yrs.
Factors linked with worse PFS were higher bone marrow (BM) blast % (HR=1.01, p<0.001), higher peripheral blood (PB) blast % (HR=1.01, p<0.001), higher PB basophil % (HR=1.01, p=0.02), lower PB polymononuclear cells (HR=0.98, p<0.001), lower hemoglobin (HR=0.91, p=0.02), lower platelet counts (HR=0.99, p=0.045), and higher number of prior TKI treatments (HR=1.52, p<0.001). In AP, compared to other TKIs, imatinib was associated with superior PFS (HR=0.63, p=0.03) while nilotinib was linked with inferior PFS (HR=1.81, p=0.04). Dasatinib (HR=1.17, p=0.57) and bosutinib (HR=1.85, p=0.18) did not show difference in PFS. In BP, there was no difference in OS by TKIs. Concurrent use of ACE inhibitors, ARBs, statins, aspirin, or metformin with TKIs was not associated with change in PFS. In multivariate analysis adjusting for the above variables, only prior TKI use was associated with worse PFS (adjusted HR [AHR]=1.57, p=0.001). Imatinib (AHR=0.70, p=0.36) nor dasatinib (AHR=0.67, p=0.11) were no longer associated with favorable or unfavorable PFS.
Factors associated with inferior OS were old age (HR=1.02, p<0.001), bigger spleen size (HR=1.03, p=0.001), higher BM blast % (HR=1.02, p<0.001), higher PB blast % (HR=1.02, p<0.001), lower PB polymononuclear cells (HR=0.98, p<0.001), lower hemoglobin (HR=0.80, p<0.001), lower platelet counts (HR=0.99, p=0.008), higher LDH (HR=1.0002, p<0.001), presence of other cytogenetic abnormality besides Ph (HR=1.45, p=0.03),and higher number of prior TKI treatments (HR=1.64, p<0.001). In AP, imatinib was associated with more favorable OS (HR=0.65, p=0.07) compared to others (nilotinib: HR=1.28, p=0.47; dasatinib: HR=1.70, p=0.077; bosutinib: HR=1.09, p=0.88). In BP, there was no difference in OS by TKIs. Concurrent use of ACE inhibitors, ARBs, statins, aspirin, or metformin with TKIs was not associated with change in OS. In multivariate analysis adjusting for the above variables, only age (AHR=1.01, p=0.002) and BM blast% (AHR=1.02, p=0.005) were associated with worse OS. Imatinib was no longer associated with favorable outcome (AHR=1.27, p=0.57).
In conclusion, we identified spleen size, blast counts, and number of prior TKI use as major prognostic factors in CML patients in BC or AP. Imatinib was associated with favorable outcome. This is likely due to the fact that patients treated with imatinib had received no prior TKIs while most patients treated with 2nd generation TKI had. Further studies are required to validate our findings in a larger and prospective cohort.
Kantarjian:Novartis: Research Funding; Bristol Myers Squibb: Research Funding; Ariad: Research Funding; Pfizer: Research Funding. Jabbour:Bristol Myers Squibb: Honoraria; Pfizer: Honoraria; Novartis: Honoraria. Cortes:Pfizer: Consultancy, Research Funding; Bristol Myers Squibb: Consultancy, Research Funding; Novartis: Consultancy, Research Funding; Ariad: Consultancy, Research Funding.
Asterisk with author names denotes non-ASH members.