Abstract
Introduction
After treatment response, cytogenetics and molecular aberrations are the most important prognostic factors in children with de novo acute myeloid leukemia (AML). However, little is known about the impact of cytogenetics at relapse. This international retrospective study aimed to provide insight into the prognostic impact of cytogenetic profiles and the role of karyotype changes from diagnosis to relapse in pediatric patients with relapsed AML.
Methods
Cytogenetic reports from patients registered to the Relapsed AML 2001/01 Study and diagnosed between 2001 and 2010 were centrally reviewed and classified by two independent researchers plus a cytogenetic expert. Patients with refractory or relapsed AML and available cytogenetics at relapse were included in order to assess the prognostic impact of different cytogenetic subgroups at relapse. Patients with karyotypes available at both diagnosis and relapse were included in order to study the impact of karyotype changes. Recurrent cytogenetic aberrations present in ≥5 patients defined the subgroups. Changes at relapse were categorized as: no change, gain, loss, both gain and loss, or structural other aberration(s). Primary endpoints were the probabilities of event-free survival (pEFS) and overall survival (pOS). Univariate analyses were conducted using chi-square tests, binary univariate logistic regression or Kaplan Meier estimates with a log-rank test. Multivariable Cox regression analyses were conducted to evaluate the independent impact of cytogenetic profiles and karyotype changes at relapse. For these analyses, cytogenetic subgroups were regrouped into good risk (GR) cytogenetics [t(8;21)(q22;q22) or inv(16)/t(16;16)(p13.1;q22)] or "other".
Results
Of the 569 registered patients, 402 patients (71%) had available cytogenetic information at relapse. Frequently detected aberrations at relapse were t(8;21) (n=60, 15%) and inv(16)/t(16;16) (n=24, 6%), both indicating a relatively good prognosis. Although patient numbers were small (n=5), t(6;9)(p23;q34) also had a relatively good outcome. Monosomy 7/7q-, t(9;11)(p22;q23), t(10;11)(p12;q23) and complex karyotypes had a relatively poor outcome. Figure 1 shows the Kaplan Meier curves of the investigated cytogenetic subgroups with corresponding patient numbers.
In total, 306 patients (54%) with available karyotypes at both diagnosis and relapse were included to study cytogenetic changes. Patients with any change (n=148, 48%) had inferior outcome compared to patients without changes (3-year pEFS 21% [SE, 3.4%] versus 39% [SE, 3.9%]; overall P<0.01). Patients with both loss and gain or structural other aberrations had the worst outcome at the univariate level.
After multivariable adjustment (final models including cytogenetics at relapse, time to relapse </≥ 1 year and type of change), having both gain and loss of aberrations remained associated with inferior pEFS and pOS (Hazard Ratio [HR] 2.10 [95% confidence interval (CI) 1.17-3.76] and HR 2.18 [95% CI 1.19-3.99] respectively). Having structural other aberrations at relapse was also associated with inferior pEFS (HR 1.81 [95% CI 1.03-3.18]).
GR cytogenetics at relapse were significantly associated with better early treatment response. Subsequently, response to treatment at day 28 (Creutzig et al. Haematologica 2014) was an important mediator. If this prognostic parameter was included in the models, the effect of changes diminished, but GR cytogenetics at relapse remained an important prognostic factor associated with superior outcome (pEFS: HR 0.53 [95% CI 0.35-0.81], pOS: HR 0.51 [95% CI 0.32-0.80]).
Conclusion
Together with early treatment response, the cytogenetic profile at relapse is an important prognostic factor. In particular t(8;21) and inv(16)/t(16;16) at relapse were associated with a favorable outcome. Furthermore, cytogenetic changes between diagnosis and relapse were associated with inferior outcome. Future studies should explore the mechanism(s) of these changes, being either clonal evolution or clonal selection.
Interpretation of our results is hampered by the retrospective design of the study, small subgroup numbers and missing cytogenetic and molecular data. Nonetheless, these findings suggest that assessing cytogenetics at time of relapse is of high importance, and can be used for risk group adapted treatment.
Reinhardt:Boehringer Ingelheim: Membership on an entity's Board of Directors or advisory committees; Pfizer: Membership on an entity's Board of Directors or advisory committees; Jazz Pharma: Other: Travel Accomodation; Celgene: Membership on an entity's Board of Directors or advisory committees; Celgene: Research Funding.
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal