Effect of Body Mass Index (BMI) At Diagnosis On Survival Patterns of Multiple Myeloma (MM) Patients in the Veterans Health Administration (VHA).

Two thirds of the adult population in the United States (US) is either overweight or obese based on BMI. Elevated BMI has been associated with an increased risk of death from hematologic malignancies, including MM. This occurs through modification of MM disease incidence, survival after diagnosis, or a combination of the two. Limited data is available on the impact of BMI at diagnosis on mortality in patients with MM. We used a retrospective cohort to evaluate the impact of BMI at diagnosis on survival patterns in MM patients treated at US VHA hospitals.

The VHA Central Cancer Registry was used to identify electronic records of 5,013 patients diagnosed with MM between October 1998 and December 2009. To minimize misclassification bias (remove patients with monoclonal gammopathy and smoldering myeloma) we excluded patients who did not receive therapy within 6 months of diagnosis of MM. Patients without weight and height measurements within 1 month of diagnosis were also excluded, resulting in an analytic cohort of 2,968 patients.

Table 1 demonstrates baseline characteristics of the analytic cohort, including stratification by BMI. Based on BMI at time of diagnosis, Cox modeling showed a reduction in mortality for overweight (BMI 25 to <30) and obese (BMI ≥30) patients, (hazard ratio for death [HR], 0.82; 95% CI: 0.75–0.91 and 0.75; 95% CI: 0.67–0.84, respectively), compared to normal weight patients (BMI 18.5 to <25) after controlling for age and co-morbidities. Underweight (BMI <18.5) was associated with a higher mortality compared to normal weight (HR, 1.64; 95% CI: 1.30–2.08). To examine the potential confounder of disease related weight loss, we obtained weight information one year before diagnosis in a subset of the analytic cohort (n=1,983). Patients who lost more than 10% of their body weight over the year before diagnosis, compared to those who did not, had higher mortality (HR, 1.58; 95% CI: 1.41–1.78). When analyzed by BMI one year before diagnosis, the association between obesity and decreased mortality was lost (HR: 0.93, 95% CI 0.81–1.07), while patients who were overweight had only borderline significance in mortality reduction (HR: 0.87, 95% CI 0.76–0.99).

MM patients who are overweight or obese at the time of diagnosis had decreased mortality compared to those who are normal-weight. In an effort to understand the influence of disease-related weight loss on this observation, we examined weight one year before diagnosis and found the association was no longer present in obese patients and only borderline present in overweight patients. This coupled with the observation that patients who lost 10% or more of their body weight in the year leading up to diagnosis had increased mortality (HR 1.58) suggests that disease-related weight loss is a major driver of the decreased survival seen in patients with a lower BMI at diagnosis. To our knowledge, this is the first study demonstrating that disease related weight loss in the time leading up to diagnosis is associated with decreased survival in MM. The mechanisms by which disease related weight loss drives a poorer prognosis cannot be determined in a population-based study. Understanding the causative mechanisms may improve our understanding of the biology of MM as well as biomarkers associated with decreased overall survival in MM.

Vij:Millennium: Speakers Bureau.