Prognostic Factors in CBFB-MYH11 Positive AML: Trisomy 21, AGE, and t(16;16) Are Associated with Inferior Outcome.

The fusion transcript CBFB-MYH11 is the molecular correlate of inv(16)/t(16;16) and strictly associated with FAB subtype M4eo. This subgroup is associated with a favorable prognosis in AML. However, approximately 30% of the patients relapse. Our intention was to examine prognostic factors for the outcome within this subgroup. Therefore 153 CBFB-MYH11 positive AML patients were analyzed. The median age was 52 years (range 18–83), 80 patients were female, 73 were male. In 22 cases AML was therapy-related, in 131 cases a de novo AML was diagnosed. Inv(16) was detected in 138 and t(16;16) in 12 cases. In 3 cases neither inv(16) nor t(16;16) were detectable despite PCR and FISH positivity for CBFB-MYH11 suggesting cryptic rearrangements. The most frequent additional cytogenetic abnormalities were +8 (n=19), +9 (n=3), +21 (n=7), +22 (n=23). Cox regression analysis revealed that advanced age (OS: p=0.026; EFS: p=0.029) and increased CBFB-MYH11/ABL ratio at diagnosis (OS: 0.016, EFS: p=0.064) were associated with a worse prognosis. Using log rank test additional factors influencing survival were detected. These included: t(16;16) vs inv(16) (OS: n=8, censored 4, median 362 days vs n=118, censored 92, median not reached, p=0.018; EFS: n=8, censored 4, median 232 days vs n=118, censored 70, median 918 days, p=0.048) and trisomy 21 vs no additional aberrations (OS: n=6, censored 3, median 435 days vs n=74, censored 59, median not reached, p=0.024; EFS: n=6, censored 2, median 293 d vs n=74, censored 44, median 764 days, p=0.0047). Therapy related AML was associated with worse EFS than de novo AML (n=16, censored 6, median 371 days vs n=112, censored 70, median 1179 days, p=0.0167) and there was a trend towards worse OS (p=0.157 n=16, censored 10, median 764 days vs n=112, censored 88, median not reached). A multivariate analysis including t(16;16), age, CBFB-MYH11/ABL ratio, therapy related AML and +21 as covariates revealed t(16;16) and age as independent factor for OS (p=0.014 and p=0.015, respectively) and age, t(16;16), and +21 as independent factors for EFS (p=0.047, p=0.013, and p=0.016, respectively). There was no evidence that the additional aberrations +22 or +8 had an influence on survival. Taken together our data suggest that t(16;16) as compared to inv(16), trisomy 21 and age are associated with worse prognosis in patients with CBFB-MYH11 positive AML.