Effect of BMI on outcomes: conflicting results

RADAR

REPORT FROM THE AMERICAN SOCIETY OF CLINICAL ONCOLOGY (ASCO) ANNUAL MEETING, CHICAGO, IL, JUNE 3-7, 2011 - While obesity is a known risk factor for the development of RCC (Calle et al. N Engl J Med 2003; 348: 1625-1638; Samanic et al. Cancer Causes Control 2006; 17: 901-909), recent studies have suggested that a higher body-mass index (BMI) may be associated with improved outcomes in RCC.

Waalkes et al. conducted a retrospective review of 1,338 patients (mean age 62.6 years) undergoing surgery for clear-cell RCC during the period 1991-2005 (Waalkes et al. Cancer Causes Control 2010; 21: 1905-1910; free full text here). Mean follow-up was 5.1 years.

BMI was categorized according to WHO definitions: underweight < 18.5 kg/m²; normal weight 18.5 to < 25 kg/m2; pre-obesity 25 to < 30 kg/m²; and obesity >30 kg/m² (grade I, 30 to 35 kg/m²) The proportion of patients in each category was underweight 1.0%; normal weight 33.2%; pre-obesity 44.3%; and obesity 21.4%. The mean BMI was 27.1.

A lower BMI was significantly associated with a higher tumour grade (p=0.009), but not with tumour stage (p=0.67), nodal status (p=0.09), or visceral metastasis (p=0.06) on univariate analysis. The median tumour-specific survival rate was highest in obese subjects (85.6%, obese grade >II, 74.9%, grade I) versus those with a BMI < 25 (63.8%).

At ASCO 2010, a study at six centres in Canada and the U.S. involving 475 patients treated with first-line sunitinib, sorafenib or bevacizumab reported that the median OS was 32.5 months in obese patients versus 20.6 months in non-obese subjects (Choueiri et al. J Clin Oncol 2010; 28(suppl): abstract 4524). The difference in OS persisted after adjustment for Heng criteria on multivariate analysis (obese vs. non-obese, hazard ratio 0.67). The median time to progression was 12.7 months in the obese group versus 4.9 months in the normal BMI/low body-surface area group (adjusted HR 1.54).

A number of previously published studies have also reported an impact of BMI on outcomes in RCC patients (Haferkamp et al. BJU Int 2008; 101: 1243-1246; Parker et al. Urology 2006; 68: 741-746; Donat et al. J Urol 2006; 175: 46-52; Kamat et al. Urology 2004; 63: 46-50; Schips et al. J Surg Oncol 2004; 88: 57-61; Yu et al. Cancer 1991; 68: 1648-1655).

Two studies at ASCO 2011 on BMI in RCC presented conflicting results. A retrospective analysis of patients with advanced RCC examined the effect of BMI on PFS during treatment with a VEGF-TKI or an mTOR inhibitor (Dhaliwal et al. J Clin Oncol 2011; 29(suppl): abstract e15103). A total of 120 patients receiving 215 treatment courses were evaluable. The median age was 59 years; mean BMI was 27.8 kg/m2. Obese subjects (BMI >30) had a 61% reduced risk of progression compared to subjects with the lowest BMI (< 25); subjects with a BMI 25-30 also had a reduced rate of progression compared to the lowest BMI group (HR 0.58).

However, a new study by Waalkes et al. found that BMI and body-surface area (BSA) do not appear to have prognostic significance (Waalkes et al. J Clin Oncol 2011; 29(suppl): abstract e15162). Two groups were examined: patients with mRCC treated between 1990 and 2005 who received palliative nephrectomy and cytokines (mean age 61.9 years, mean BMI 25.5 kg/m2, mean BSA 1.93 m²); and those treated between 2006 and 2009 with a first-line TKI (mean age 61.0 years, mean BMI 25.7 kg/m2, mean BSA 1.97 m²). There was no evidence that BMI or BSA affected prognosis in either group. Multivariate analysis also failed to show that BMI or BSA was an independent prognostic factor.