Death In Complete Remission Among Patients with Acute Myeloid Leukemia: A Preventable Problem?

Significant advances in the treatment of patients with acute myelogenous leukemia (AML) and high risk myelodysplastic syndrome (HR-MDS) have improved their outcome. Mortality among patients with newly diagnosed AML and HR-MDS is mainly attributed to persistent disease (i.e., failure to achieve or loss of complete remission [CR]), infectious and other complications during induction chemotherapy. However, a small but significant proportion of patients die in CR. The aim of this study was to investigate the cause of death in AML and HR-MDS patients in CR and to analyze the associated risk factors.

Retrospective review of medical records of patients with newly diagnosed AML, APL and HR-MDS who received induction chemotherapy at the Department of Leukemia at MD Anderson from January 2000 to December 2009.

During the study period, 2156 patients were treated. One thousand one hundred fifty patients achieved CR for an overall CR rate of 53%. Among them, 114 patients (10%) died in CR. The median time from achievement of CR to death was 5.3 months (range, 0.2 – 79). There was a decline in the rate of death in CR over the 10 years of analysis reported (p=0.010). [Table 1]

Information about the causes of death in 35 patients (31%) was not available. The most frequent causes of death in the remaining 79 patients were infections (27%); SCT-related complications (19%); relapse of prior malignancy (8%); hemorrhage (4%); multi-organ failure (4%); and others (9%). Among patients who died in CR, 105 (92%) had AML and 9 high-risk MDS, 60% were female, and the median age was 64 years (range 21–86). Forty-two patients (37%) had history of a prior malignancy, 22% had received previous chemotherapy (for other malignancies), and 20% prior radiotherapy. Sixty-eight percent received high-dose cytarabine-containing regimen for induction therapy, 92% achieved CR after one cycle of chemotherapy, with a median of 33 days (range 21–152) to achievement of CR. Nineteen percent underwent stem cell transplantation while in CR. In comparison to patients who did not die in CR, patients who died in remission were significantly older at the time of diagnosis [64 vs. 57 years, p <.001]; were more likely to have history of prior malignancy, chemotherapy or radiotherapy [37% vs.21%, p<0.001; 22% vs.11%, p<0.001; 20% vs. 9%, p<0.001, respectively]; had worse performance status at the time of diagnosis [26% vs. 14%, p = 0.004]; and were more likely to have undergone stem cell transplantation (SCT) while in CR [19% vs. 8%, p<0.001]. Sixty-three percent of the patients who died in CR were 60 years or older. Among patients age 60 years or older, the probability of death in CR was 14% compared to 7% for patients younger than 60 (p<0.001).

Multivariate logistic regression analysis confirmed that older age (p<0.000), prior malignancy (p<0.001), poor performance status (p<0.001) and SCT while in CR (p<0.000) were independently associated with the probability of dying in CR. Risk factors for dying from infections included only older age (p<0.003) and poor performance status (p, 0.05).

Death in CR affects a significant number of patients with AML, with older patients with poor performance status having the highest probability of dying in CR, particularly from infections. Special attention should be put to these patients to minimize their risk of death in CR to improve their long term outcome.

No relevant conflicts of interest to declare.