Abstract 1067

Elderly patients with acute myeloid leukemia are heterogenous group with poor outcome. All age, biological status and co-morbidities limit applicability of intensive chemotherapy. The PALG elaborated original system allowing stratification of patients aged >60 years to three groups with different therapeutic approach. Altogether 537 patients with newly diagnosed AML and median age 70 years (range 60–93) were classified as 1) ‘fit’ (n=163): age 60–79y, ECOG 0–2, proper liver and kidney function, without comorbidities, 2) ‘unfit’ (n=210): age >60 years, ECOG 0–2, normal liver and kidney function, comorbidities allowed, 3) ‘frail’ (n=164): ECOG 3–4. According to PALG 1/2005 protocol ‘fit’ patients were treated similarly as younger adults with daunorubicin (DNR, 3 days) + cytarabine (AraC, 7 days) +/− cladribine, followed by DNR + AraC consolidation and maintenance. ‘Unfit’ patients received either two courses of AraC+DNR (2+5) or AraC (5 days) + thioguanine + methotrexate, followed by manitenance. ‘Frail’ patients were considered for palliative cytoreduction and supportive care.


Complete remission (CR) rate was 35% for ‘fit’, 22% for ‘unfit’ and 0% for ‘frail’ patients. Median survival in the respective groups equaled 39 weeks, 26 w., and 14 w., while the probability of survival at 1 year was 39%, 27% and 10%. The rate of early (up to 8 weeks) mortality was 31%, 24% and 31%, respectively. In the Cox model the only factor independently affecting the risk of overall mortality in both ‘fit’ and x‘unfit’ group was serum LDH above upper quartile (HR=2, p=0.005 for ‘fit’, HR=1.65, p=0.006 for ‘unfit’). Among ‘frail’ patients the risk of mortality was increased in patients with performance status ECOG>2 (HR=1.85, p=0.0008), initial WBC >8.5×10e9/L (HR=1.65, p=0.006), and bone marrow blasts >58% (HR=1.8, p=0.001).

We conclude that the proposed stratification system is feasible for elderly AML patients and represets a model for further developments of individualized therapeutic approaches. Survival of patients in whom remission induction therapy may be applied depends on initial tumor burden as reflected by high serum LDH level. The outcome of patients referred for palliative treatment depends additionally on initial performance status. In contrast, neither age nor karyotype were found to independently affect outcome in this study.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution