Response and Overall Survival Is Impaired in Obese Patients with Acute Myelogenous Leukemia

Dose calculation of chemotherapeutic agents is mainly based on body surface area (BSA). Historically, in many institutions, doses are not adapted to a BSA above 2 sqm, although recent data suggested that this relative dose reduction in patients above 2 sqm is associated with a worse outcome. Obesity is a widespread and increasing phenomenon in the developed countries and is mainly defined by the body mass index (BMI). It is known, that the risk to develop haematological or solid malignancies is increased in obese patients – and, additionally, weight influences outcome in several tumours. Therefore, in our study, we analyzed the prognostic impact of obesity in newly diagnosed AML regarding the response to the first cycle of induction therapy and overall survival.

We identified 145 patients with newly diagnosed AML who were treated with induction therapy containing cytarabine and an anthracycline in our institution. Clinical data including several laboratory parameters associated with nutritional status (cholesterol, triglycerides, protein, C-reactive protein, albumin, lymphocyte count, transferrin, pseudo-cholinesterase, fT3, b-type natriuretic peptide), long-term medication with statins, chemotherapy dosing as well as response and overall survival have been assessed retrospectively from the institution's database. In our institution, all patients with a high BSA received a chemotherapy-dose calculated with a cut-off of not more than 2 sqm.

In our cohort, median BMI was 25.2 kg/sqm (range 17.0 – 48.1). Seventeen patients had a BMI above 31 kg/sqm, 128 below. The median BSA of all patients was 1.83 sqm (range 1.49 – 2.40). In 41 patients, chemotherapy doses have been adjusted owing to a BSA of more than 2 sqm. We included cytogenetic risk group, BMI, BSA above 2.0 sqm, weight, long-term medication with statins and laboratory parameters in our univariate and multivariate analysis. Only cytogenetic risk group (p=0.001), triglycerides (p=0.008) and the BMI (p=0.032) were independent risk factors for overall survival. Univariate analysis showed similar results. Patients with a BMI >31 showed a significantly worse response (PR + CR) on first induction therapy (47.1% vs. 74.8%, p=0.018) and a shorter median survival (11.2 vs. 25.1 months, p=0.004). Both BMI-groups showed the same distribution of well-known risk factors including age, cytogenetic risk groups and secondary AML.

In our cohort, a high BMI was associated with poorer response and impaired overall survival, whereas BSA was not. This leads to the conclusion, that the adverse effect may be mediated by obesity itself and not caused by underdosing of chemotherapy. Although an effect of BMI is known from several solid tumours, the reason remains unclear. Possible explanations include altered metabolisation and production of growth factors in adipose tissue (possibly indicated by elevated triglycerides), impaired or accelerated hepatic drug activation and/or metabolisation or impact on immune function. This study is limited by the retrospective single-center design and the relatively small patient number. Nevertheless, the data clearly confirmed other well established risk factors like the cytogenetic risk group supporting the validity of this approach. The study results suggest BMI as an independent risk factor for AML.

No relevant conflicts of interest to declare.