Annals of Oncology Advance Access originally published online on February 13, 2008
Annals of Oncology 2008 19(5):898-902; doi:10.1093/annonc/mdm606
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
gynecologic tumors |
Does body mass index affect progression-free or overall survival in patients with ovarian cancer? Results from SCOTROC I trial
1 Department of Medical Oncology, Beatson West of Scotland Cancer Centre, Glasgow
2 Cancer Research UK Clinical Trials Unit, Beatson West of Scotland Cancer Centre, Glasgow, UK
3 Department of Medical Oncology, University of Queensland, Brisbane, Australia
4 Section of Medicine, The Institute of Cancer Research, London, UK
* Correspondence to: Dr S. V. Barrett, Beatson West of Scotland Cancer Centre, 1053 Great Western Road, Glasgow G12 0YN, UK. Tel: +44-141-301-7000; Fax: +44-141-301-7124; E-mail: sophie.barrett{at}nhs.net
 |
Abstract |
|---|
 Top Abstract introductionpatients and methodsresultsdiscussionAcknowledgementsReferences |
|---|
Background: Previous studies have indicated an association between obesity and poor survival in several tumour types, including ovarian cancer. We sought to test the hypothesis that obesity reduces survival in a large, well-characterised and relatively homogeneous cohort of ovarian cancer patients.
Patients and methods: The relationship between body mass index (BMI) and overall survival (OS) and progression-free survival (PFS) in 1067 patients participating in the Scottish Randomised Trial in Ovarian Cancer I trial was assessed. All patients received first-line carboplatin/taxane chemotherapy. The dose of carboplatin was determined by a measured glomerular filtration rate (GFR), ensuring accurate dosing in all categories of BMI and the dose of taxane was not capped. Patients were assigned to one of four categories: underweight (BMI < 18.5), ideal weight (BMI 18.5–24.9), overweight (BMI 25–29.9) or obese (BMI 
30).
Results: There were neither statistically significant differences in PFS or OS between these four groups nor were there any differences in taxane or carboplatin dose intensity. Furthermore, there was no association between BMI and tumour stage or grade at presentation, or completeness of debulking surgery.
Conclusions: Obese patients with epithelial ovarian cancer do not have a poorer prognosis, provided that they receive optimal doses of chemotherapy based on measured GFR and actual body weight.
Key words: body mass index, ovarian cancer, survival
|
introduction |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionAcknowledgementsReferences |
|---|
The incidence of obesity is increasing in the developed world. Obesity is associated with an increased risk of malignancy, contributing to 14%–20% of cancer-related mortality [1] but an association between obesity and cancer survival is less clear. Although obese women are more likely to develop endometrial cancer, there is evidence that they have improved survival [2] and, in ovarian cancer, one study identifying increased body mass index (BMI) as an independent negative prognostic factor for disease-free survival (DFS) and overall survival (OS) [3] and a second study reported an association between BMI > 25 kg/m2 and reduced survival [4].
These studies are, however, relatively small and non-homogeneous. We have retrospectively investigated the association between BMI and progression-free survival (PFS) and OS in women with ovarian cancer or peritoneal cancer treated on the Scottish Randomised Trial in Ovarian Cancer (SCOTROC) I study. This trial is a large, multicentre, international, phase III randomised trial of docetaxel–carboplatin versus paclitaxel–carboplatin as first-line chemotherapy which recruited 1077 patients, therefore allowing analysis of a large number of well-characterised patients treated according to the same protocol. In addition to survival, a pre-planned analysis of BMI and tumour stage/grade, carboplatin and taxane dose intensity and extent of debulking surgery was carried out.
|
patients and methods |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionAcknowledgementsReferences |
|---|
The SCOTROC I trial is a randomised phase III study, coordinated by the Scottish Gynaecological Cancer Trials Group, to compare efficacy, tolerability and quality of life of docetaxel (Taxotere®, Sanofi-Aventis, Guildford, UK)–carboplatin with paclitaxel (Taxol® Bristol-Myres Squibb, Uxbridge, UK)–carboplatin as initial chemotherapy for stages Ic–IV ovarian or peritoneal carcinoma.
patients
In all, 1077 patients from 83 international centres were recruited from October 1998 to May 2000. Patients gave written informed consent to participate and the trial had full multicentre ethics committee approval. Separate approval was granted for the analysis of BMI. Patients received docetaxel–carboplatin (n = 537) or paclitaxel–carboplatin (n = 538) (two patients were not treated). Chemotherapy doses were docetaxel 75 mg/m2 1-h infusion or paclitaxel 175 mg/m2 3-h infusion, followed by carboplatin area under the curve (AUC) 5 1-h infusion, to a maximum of six cycles. The primary end point of the study was PFS. Trial results have been published [5].
chemotherapy dose calculation
Carboplatin dose was derived using the Calvert formula [6]. Investigations required before study entry included measurement of glomerular filtration rate (GFR) by plasma clearance of 51Cr-ethylenediaminotetraacetic acid (EDTA) [7], 99mTc-diethylenetriaminepentaacetic acid (99mTc-DTPA) or 24-h urine collection for creatinine clearance. Estimations of GFR were not permitted. Taxane dose was calculated using body surface area (BSA), with no dose capping carried out.
statistical analysis
BMI was calculated from the height and weight recorded at first treatment using the formula: weight (kg)/height2 (m2). Patients were assigned to one of four categories: underweight (BMI < 18.5), ideal weight (BMI 18.5–24.9), overweight (BMI 25–29.9) and obese (BMI 30). Survival was compared between these four groups using the log-rank test. Dependence of BMI distribution on tumour stage, grade and completeness of debulking surgery was assessed using the Kruskal–Wallis test. Distribution of dose intensity and percentage of total intended dose was compared with the BMI categories using the Kruskal–Wallis test.
|
results |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionAcknowledgementsReferences |
|---|
patient characteristics
Overall clinico-pathological characteristics of patients in the SCOTROC I trial are demonstrated and a breakdown in each of the BMI groups are shown in Tables 1 and 2. A histogram demonstrating the distribution of BMI in trial patients is shown in Figure 1.
|
|
|
GFR calculation
GFR was calculated by clearance of 51Cr-EDTA in 456 patients (42.4%), clearance of 99mTc-DTPA in 159 patients (14.8%) and 24-h urine collection in 460 patients (42.8%). Information was unavailable for two patients.
PFS and OS
Analysis is restricted to 1067 patients who received taxane treatment and have recorded BMI. Median PFS was 14.7 months [95% confidence interval (CI) 11.6–17.9, n = 59], in the underweight group; 14.7 months (95% CI 13.3–16.1, n = 576), in the ideal weight group; 14.7 months (95% CI 12.6–16.8, n = 303), in the overweight group and 16.6 months (95% CI 11.8–21.4, n = 129), in the obese group. There was no statistically significant association between BMI category and PFS (log-rank test statistic 2.31, P = 0.51, n = 1067).
There was no statistically significant association between BMI category and OS (log-rank test statistic 6.33, P = 0.10, n = 1067), with median OS of 32.9 months (95% CI 23.5–42.4, n = 59), in the underweight group; median not attained in the ideal weight group (n = 576); 30.1 months (95% CI 25.4–34.7, n = 303), in the overweight group and 34.3 months (95% CI 26.9–41.7, n = 129), in the obese group. Kaplan–Meier survival curves demonstrating PFS and OS in all four BMI groups are shown in Figure 2.
|
dose intensity and intended dose
In the SCOTROC I study overall, there were no statistically significant differences between the two arms for dose intensity or cumulative dose [5]. In the current study, there was neither a statistically significant difference in taxane dose intensity between each of the BMI categories (P = 0.120, n = 1067) or carboplatin dose intensity (P = 0.578, n = 1067) nor a statistically significant difference between total intended taxane dose (P = 0.217, n = 1067) or total intended carboplatin dose (P = 0.722, n = 1067). Boxplots of total intended taxane and carboplatin dose are shown in Figure 3. If there was >10% weight loss when compared with baseline weight during the study, leading to a significant change in BSA, taxane dose was to be reduced accordingly. Overall, 95 patients (0% in the underweight group, 8% in the ideal weight group, 12% in the overweight group and 9% in the obese group) had weight loss of this magnitude during treatment.
|
association between BMI and tumour stage, grade or debulking surgery
There was no association between BMI and International Federation of Gynecology and Obstetrics tumour stage (P = 0.190, n = 1076), tumour grade (P = 0.280, n = 1076) or extent of debulking surgery (P = 0.194, n = 1076).
|
discussion |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionAcknowledgementsReferences |
|---|
Obesity is associated with poor outcome in several tumour types and a recent study reported an association between high BMI and poor survival in ovarian cancer [3]. The study was, however, relatively small and it may have been confounded by lower dose intensity in obese patients. The SCOTROC I trial offers an opportunity to investigate the impact of BMI on outcome in patients with stages Ic–IV epithelial ovarian or peritoneal cancer, all treated using the same protocol. We do, however, acknowledge that analysis of the relationship between BMI and survival was not one of the original aims of the trial. Table 2 demonstrates no difference between clinico-pathological characteristics in each of the BMI groups, except for a difference between allocated treatment in the underweight group (36% received paclitaxel/carboplatin and 64% received docetaxel/carboplatin). Although the study was designed to investigate the differences between paclitaxel and docetaxel in ovarian cancer management, the study demonstrated no statistically significant difference in PFS or OS between the two groups [5] and the difference in allocated treatment in the underweight group is, therefore, unlikely to be of significance.
We found no difference between BMI category and PFS or OS in our study. Furthermore, there was no association between BMI and tumour stage or grade, or completeness of debulking surgery, indicating that these were not confounding factors. Our findings do not support previous reports of a link between obesity and poorer prognosis in ovarian cancer patients. We indicate that this difference is because of more accurate dose calculations in our study.
Carboplatin–paclitaxel is considered standard chemotherapy in ovarian cancer. Carboplatin dose is traditionally calculated using targeted area under the plasma carboplatin concentration–time curve (AUC), rather than using BSA [8]. Using AUC to calculate carboplatin dose correlates with response rates in ovarian cancer [9]. Dose is determined using the Calvert formula, a calculation incorporating the GFR. The most accurate methods of measuring GFR are by analysing plasma clearance of 51Cr-EDTA [7] or 99mTc-DTPA following i.v. injection or by 24-h urine creatinine clearance [6, 10] but these approaches are time consuming and expensive. Several formulae have, therefore, been derived which estimate GFR, including the Cockcroft–Gault [11] and Wright [12] formulae.
Several studies have evaluated estimations of GFR and compared them with the gold standard of 51Cr-EDTA clearance and have concluded that these formulae may underestimate GFR when compared with 51Cr-EDTA clearance [13, 14]. When analysing the influence of age and BMI on estimates of GFR in obese patients, none of the formulae offered a reliable estimation of GFR [15] and in patients with gynaecological cancer, the accuracy of the calculations became more unreliable at the extremes of BSA [16]. It is possible that, in contrast to SCOTROC I, where the eligibility criteria included calculation of GFR by one of three established and reproducible methods, other studies have used varying methodology, which could have underestimated the carboplatin dose. If this underestimation of dose is particularly marked in the obese population, then this could negatively influence survival data in this group, falsely giving the impression that ovarian cancer patients with a high BMI have a poorer prognosis.
A second reason why obese ovarian cancer patients might appear to have a poorer outcome is because chemotherapy doses in overweight patients are frequently capped in routine practice. The previous study of 216 patients with ovarian or peritoneal cancer which identified increasing BMI as an independent negative prognostic factor for DFS and OS [3] also reported a statistically significant difference in chemotherapy dose between the ideal body weight and obese groups. Although one study in breast cancer suggested that BMI is an independent prognostic factor for OS [18], others have reported a significantly worse prognosis for patients with higher BMI who were treated on the basis of their ideal body weight [17], indicating that chemotherapy dosing using ideal body weight may compromise treatment efficacy and adversely affect survival.
Drug elimination may also differ between ideal weight and obese patients. A recent study in patients with solid tumours receiving paclitaxel reported that, for every increase in BSA of 0.2 m2, there is a 9% increase in drug elimination [19]. This finding further supports the argument that obese patients should receive chemotherapy based on actual body weight. There was no indication of chemotherapy dose capping in the SCOTROC I trial, as there were no statistically different differences in either taxane or carboplatin dose intensity between each of the four BMI categories.
In conclusion, we have found no association between BMI and outcome in patients with ovarian or peritoneal cancer. We recommend both accurate measurement of GFR before commencing combination chemotherapy and also that patients should receive chemotherapy doses based on actual body weight, rather than ideal body weight.
|
Acknowledgements |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionAcknowledgementsReferences |
|---|
The SCOTROC-1 study was supported by grants from Aventis Pharmaceuticals (now Sanofi-Aventis). The authors acknowledge the support of the Scottish Gynaecological Cancer Trials Group, Cancer Research UK, which funds the Clinical Trials Unit in Glasgow and also the clinicians and patients who participated in the SCOTROC-1 trial.
Received for publication October 4, 2007. Revision received December 18, 2007. Accepted for publication December 19, 2007.
|
References |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionAcknowledgementsReferences |
|---|
1. Calle EE, Rodriguez C, Walker-Thurmond K, et al. Overweight, obesity, and mortality from cancer in a prospectively studied cohort of U.S. adults. N Engl J Med (2003) 348(17):1625–1638.
2. Temkin SM, Pezzullo JC, Hellmann M, et al. Is body mass index an independent risk factor of survival among patients with endometrial cancer? Am J Clin Oncol (2007) 30(1):8–14.[CrossRef][Web of Science][Medline]
3. Pavelka JC, Brown RS, Karlan BY, et al. Effect of obesity on survival in epithelial ovarian cancer. Cancer (2006) 107(7):1520–1524.[CrossRef][Medline]
4. Zhang M, Xie X, Lee AH, et al. Body mass index in relation to ovarian cancer survival. Cancer Epidemiol Biomarkers Prev (2005) 14(5):1307–1310.
5. Vasey PA, Jayson GC, Gordon A, et al. Phase III randomized trial of docetaxel-carboplatin versus paclitaxel-carboplatin as first-line chemotherapy for ovarian carcinoma. J Natl Cancer Inst (2004) 96(22):1682–1691.
6. Calvert AH, Newell DR, Gumbrell LA, et al. Carboplatin dosage: prospective evaluation of a simple formula based on renal function. J Clin Oncol (1989) 7(11):1748–1756.[Abstract]
7. Chantler C, Garnett ES, Parsons V, et al. Glomerular filtration rate measurement in man by the single injection methods using 51Cr-EDTA. Clin Sci (1969) 37(1):169–180.[Web of Science][Medline]
8. Duffull SB, Robinson BA. Clinical pharmacokinetics and dose optimisation of carboplatin. Clin Pharmacokinet (1997) 33(3):161–183.[Web of Science][Medline]
9. Jodrell DI, Egorin MJ, Canetta RM, et al. Relationships between carboplatin exposure and tumor response and toxicity in patients with ovarian cancer. J Clin Oncol (1992) 10(4):520–528.
10. Martino G, Frusciante V, Varraso A, et al. Efficacy of 51Cr-EDTA clearance to tailor a carboplatin therapeutic regimen in ovarian cancer patients. Anticancer Res (1999) 19(6C):5587–5591.[Web of Science][Medline]
11. Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron (1976) 16(1):31–41.[Web of Science][Medline]
12. Wright JG, Boddy AV, Highley M, et al. Estimation of glomerular filtration rate in cancer patients. Br J Cancer (2001) 84(4):452–459.[CrossRef][Web of Science][Medline]
13. Jodrell DI. Formula-based dosing for carboplatin. Eur J Cancer (1999) 35(9):1299–1301.[CrossRef][Web of Science][Medline]
14. Calvert AH. A review of the pharmacokinetics and pharmacodynamics of combination carboplatin/paclitaxel. Semin Oncol (1997) 24(1 Suppl 2):S2-85–S2-90.
15. Verhave JC, Fesler P, Ribstein J, et al. Estimation of renal function in subjects with normal serum creatinine levels: influence of age and body mass index. Am J Kidney Dis (2005) 46(2):233–241.[CrossRef][Web of Science][Medline]
16. Nagao S, Fujiwara K, Imafuku N, et al. Difference of carboplatin clearance estimated by the Cockroft-Gault, Jelliffe, modified-Jelliffe, Wright or Chatelut formula. Gynecol Oncol (2005) 99(2):327–333.[CrossRef][Web of Science][Medline]
17. Rosner GL, Hargis JB, Hollis DR, et al. Relationship between toxicity and obesity in women receiving adjuvant chemotherapy for breast cancer: results from cancer and leukemia group B study 8541. J Clin Oncol (1996) 14(11):3000–3008.[Abstract]
18. Berclaz G, Li S, Price KN, et al. Body mass index as a prognostic feature in operable breast cancer: the International Breast Cancer Study Group experience. Ann Oncol (2004) 15(6):875–884.
19. Joerger M, Huitema AD, van den Bongard DH, et al. Quantitative effect of gender, age, liver function, and body size on the population pharmacokinetics of paclitaxel in patients with solid tumors. Clin Cancer Res (2006) 12((7 Pt 1)):2150–2157.
CiteULike 
Connotea 
Del.icio.us What's this?
Related articles in Ann Oncol:
- In this issue
Ann Oncol 2008 19: 827.[Extract] [FREE Full Text]
This article has been cited by other articles:
|
|
 |
|
  S. R. McWhinney, R. M. Goldberg, and H. L. McLeod Platinum neurotoxicity pharmacogenetics Mol. Cancer Ther., January 1, 2009; 8(1): 10 - 16. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||