Int Urol Nephrol DOI 10.1007/s11255-014-0729-7
UROLOGY - ORIGINAL PAPER
Low body mass index is associated with adverse oncological outcomes following radical prostatectomy in Korean prostate cancer patients Kyo Chul Koo • Young Eun Yoon • Koon Ho Rha • Byung Ha Chung • Seung Choul Yang Sung Joon Hong
•
Received: 5 March 2014 / Accepted: 23 April 2014 Ó Springer Science+Business Media Dordrecht 2014
Abstract Purpose The purpose of this study was to determine the impact of obesity on clinicopathological features and biochemical recurrence (BCR) following radical prostatectomy (RP) in Korean prostate cancer (PCa) patients. Methods A single-institutional retrospective analysis was performed on 880 PCa patients treated by RP without neoadjuvant therapy between July 2005 and December 2011. Patients were stratified according to body mass index (BMI) standards for Asian populations: obese (BMI C25 kg/m2), overweight (BMI 23–24.9 kg/m2), or normal weight (BMI \23 kg/m2). For analysis, overweight and obese patients were combined (n = 592, BMI C23 kg/ m2) and compared with normal weight patients (n = 288, BMI \23 kg/m2). BCR was defined as prostate-specific antigen (PSA) C0.2 ng/ml following RP. Results Normal weight patients tended to be classified into the higher D’Amico risk category with smaller prostate volumes compared with obese and overweight patients. Normal weight patients had higher pathological Gleason scores and were at higher risk of BCR during the mean follow-up of 58.2 months. This translated to a higher 5-year BCR-free survival rate for obese and overweight patients compared with normal weight patients (77.8 vs. 70.3 %; p = 0.017). On multiple Cox-proportional hazards regression analysis incorporating variables of BMI category, PSA, positive surgical margins, pathological T stage, and Gleason score, higher BMI category remained a
K. C. Koo Y. E. Yoon K. H. Rha B. H. Chung S. C. Yang S. J. Hong (&) Departments of Urology and Urological Science Institute, Yonsei University College of Medicine, 134 Shinchon-dong, Seodaemun-gu, Seoul 120-752, Republic of Korea e-mail: [email protected]
significant predictor of a lower risk of BCR (HR = 0.634, p = 0.028). Conclusions Obese and overweight Korean PCa patients have lower Gleason scores and a reduced risk of BCR compared with normal weight patients. These findings suggest that body fat influences pathological features and oncologic outcomes of PCa. Keywords Body mass index Obesity Prostate cancer Prostatectomy Treatment outcome
Introduction The prevalence and degree of obesity are increasing in Asian countries due to Westernization of diet and lifestyle [1]. However, there still exists a vast difference in body mass index (BMI) distribution between Western and Asian men. In some Asian countries, the average BMI is below 25 kg/m2, which is the cutoff value that the World Health Organization (WHO) uses to define overweight individuals [2]. Prostate cancer (PCa) has shown similar trends. The incidence of PCa varies considerably worldwide, with a greater incidence in Europe and the United States than in Asian countries [3]. In 2008, the age-standardized incidence of PCa in Korea was 22.4 per 100,000, far less than the incidence of 83.8 per 100,000 in the USA. However, by 2011, the Korean incidence had surged to 31.4 per 100,000 [4]. Obesity is a major public health hazard associated with increased risks of cardiac disease, hypertension, and diabetes [2]. It is estimated that more than 30 % of USA men are currently obese, and studies have revealed BMI to be linked to several types of cancers including cancers of the esophagus, colon, liver, kidneys, and pancreas [5].
123
Int Urol Nephrol
However, the relationship between obesity and PCa is still unclear, with large studies from different regions reporting controversial results for incidence, risks of aggressive pathological features, and biochemical recurrence (BCR) following radical prostatectomy (RP) [6–8]. Body fat distribution influences serum androgen level and potential PCa growth factors, and thus, a relationship between BMI and PCa is plausible. Numerous studies have reported consistent associations between dietary fat intake and PCa risk [9]. Therefore, investigating the relationship between BMI and PCa in contemporary, Asian men who are relatively lean compared with Western men may reveal racial disparities in clinicopathological features and risk of BCR and ultimately lead to endocrine modification strategies for a better oncological outcome. The objective of this study was to determine the impact of body fat on clinicopathological PCa features and on the risk of BCR following primary treatment with RP among a cohort of Korean PCa patients who have BMIs that are lower than their typical Western counterparts. The results suggest that body fat is an important factor in PCa development and progression.
institutional electronic medical record database. Clinical data obtained were patient age, BMI, serum prostatespecific antigen (PSA) level, biopsy Gleason score, clinical tumor stage, and prostate volume. Patient height and weight for the calculation of BMI were measured on admission, 1 day before surgery. Pathologic data obtained from RP-PLND specimens included pathological Gleason score, pathological tumor stage, positive surgical margin (PSM) status, extracapsular extension (ECE), LN invasion, specimen weight, tumor volume, and the presence of high-grade prostatic intraepithelial neoplasia (HGPIN), perineural invasion (PNI), and lymphovascular invasion (LVI). Following surgery, PSA measurements of all patients were monitored every 3 months for the first year and then every three to 6 months thereafter. All prostate specimens were serially sectioned, processed, and confirmed by a single uropathologist, with a PSM defined as cancer at the inked margin. Collection of retrospective data of the study was approved by the institutional ethics committee after review of the protocol and procedures employed (20090131-001). Study goals
Materials and methods Patients A retrospective, single-institutional analysis was performed on 1,589 consecutive biopsy-proven PCa patients without distant metastases treated by RP between July 2005 and December 2011. Patients who had received neoadjuvant or adjuvant therapies and those with lymph node (LN) metastases on preoperative imaging were excluded from the cohort. In overall, 880 patients were identified who underwent RP with pelvic lymph node dissection (PLND). BMI was calculated by dividing the weight (kg) by the height squared (m2) at the time of diagnosis. Patients were categorized as obese (C25 kg/m2), overweight 2 (23–24.9 kg/m ), or normal (\23 kg/m2). For analysis, overweight and obese groups were combined (n = 592, BMI C23 kg/m2) and compared with the normal weight group (n = 288, BMI \ 23 kg/m2). This BMI classification was used in accordance with the cutoffs proposed by the International Association for the Study of Obesity (IASO) and the International Obesity Task Force (IOTF), in consideration of the disparity of BMI distribution between Asian-Pacific and Western populations [10]. Assessment of clinicopathological variables Clinical and pathological characteristics of the patients and their oncological outcomes were retrieved from the
123
The primary goal was to explore differences in clinicopathological PCa features according to BMI category. The secondary goal was to determine the impact of BMI on BCR, defined as the first of two or more consecutive increases in PSA [0.2 ng/ml at least 3 months following RP. Statistical analysis Appropriate comparative tests such as the Student’s t test, v2-test, and ANOVA were used to compare clinicopathological variables according to BMI subgroup. BCR-free survival (BCRFS) was estimated and compared using the Kaplan–Meier method and a logrank test. Odds ratios of various adverse pathological outcomes, including high-grade disease (pathological Gleason score C8), ECE, SV invasion, and LN involvement, were estimated for BMI categories using logistic regression analysis. Univariate and multivariate analyses were performed according to Cox-proportional hazards regression models. Variables that were potential predictors for multivariate modeling were selected by univariate analyses and subsequently tested in a stepwise forward conditional manner with entry and retention in the model set at a significance level of 0.05. All tests were two-sided, with statistical significance set at p \ 0.05. Statistical analysis was performed using SPSS version 18 (SPSS Inc., Chicago, IL, USA).
Int Urol Nephrol Table 1 Clinicopathologic features of prostate cancer patients who underwent radical prostatectomy, according to body mass index
BMI (kg/m2) category 18.5–22.9 (N = 288)
p [23 (N = 592)
Preoperative characteristics Age (years)
64.5 ± 7.4
64.1 ± 7.6
0.207
PSA (ng/ml)
10.5 ± 8.7
9.7 ± 8.7
0.227
Clinical tumor stage
0.295
cT1
82 (28.4 %)
cT2
159 (55.3 %)
179 (30.2 %) 334 (56.5 %)
cT3
47 (16.3 %)
79 (13.3 %)
B6
128 (44.5 %)
271 (45.7 %)
7
95 (32.9 %)
194 (32.8 %)
65 (22.6 %)
127 (21.5 %)
Low
28 (9.8 %)
79 (13.4 %)
Intermediate
160 (55.5 %)
361 (60.9 %)
High
100 (34.7 %)
152 (25.7 %)
Prostate volume (gm)
34.2 ± 17.9
37.6 ± 17.3
B6
80 (27.9 %)
207 (35.0 %)
7
156 (54.0 %)
297 (50.1 %)
C8
52 (18.1 %)
88 (14.9 %)
Biopsy Gleason score
C8 D’Amico risk group
0.876
0.011
\0.001
Postoperative characteristics Pathological Gleason score
Data are number (%) and mean (SD) BCR biochemical recurrence, BCRFS BCR-free survival, BMI body mass index, ECE extracapsular extension, PSA prostate-specific antigen, PSM positive surgical margin, SV seminal vesicle
0.041
Lymphovascular invasion
18 (6.3 %)
45 (7.5 %)
0.754
Perineural invasion
147 (51.1 %)
297 (50.2 %)
0.842
PSM
129 (44.8 %)
260 (43.9 %)
0.791
ECE
47 (16.3 %)
78 (13.2 %)
0.231
SV invasion
27 (9.4 %)
46 (7.8 %)
0.435
Lymph node involvement Specimen weight (gm)
26 (8.9 %) 37.7 ± 15.3
42 (7.1 %) 40.9 ± 17.1
0.319 0.011
Tumor volume (cc)
2.45 ± 2.73
2.46 ± 3.97
0.923
BCR
65 (22.4 %)
92 (15.5 %)
0.021
Five-year BCRFS rate (%)
70.3
77.8
0.017
Follow-up period (months)
58.9 ± 15.3
60.2 ± 14.3
0.213
Results BMI and clinicopathological features Clinicopathological features of the patients according to BMI category are presented in Table 1. There were no differences between normal weight patients and overweight and obese patients regarding patient age, PSA, clinical tumor stage, or biopsy Gleason score; however, normal weight patients tended to be classified into a higher D’Amico risk category and to harbor high-grade disease. As expected, normal weight patients had smaller prostate volumes. Pathological features revealed that normal weight patients were more likely to have a higher Gleason score. Logistic regression analyses of BMI category and its association with adverse pathological outcomes showed
that normal weight category is associated with high-grade disease (pathologic Gleason score C8); however, there were no significantly increased risks of ECE, SV invasion, or LN involvement in the normal weight group compared with overweight and obese groups (Table 2). BMI and biochemical recurrence Among the 880 patients, 157 (17.8 %) patients developed BCR during the mean follow-up period of 58.2 months. BCR was observed in 65 (22.4 %) patients in the normal weight group and 92 (15.5 %) patients in the overweight and obese groups (p = 0.021) during the mean follow-up periods of 58.9 and 60.2 months, respectively. As depicted in Fig. 1, this translated to a higher 5-year BCR-free survival rate for overweight and obese patients compared with
123
Int Urol Nephrol Table 2 Logistic regression analyses of BMI category and its association with adverse pathological outcomes Univariate OR
Multivariate 95 % CI
p
OR
95 % CI
p
ECE
0.783
0.525–1.169
0.232
0.721
0.476–1.089
0.223
Pathological Gleason score (C8)
0.719
0.525–0.986
0.041
0.718
0.512–0.991
0.039
SV invasion
0.708
0.517–0.971
0.032
1.513
0.845–2.711
0.711
LN invasion
0.971
0.675–1.393
0.871
0.791
0.518–1.209
0.377
CI confidence interval, OR odds ratio, other abbreviations as in Table 1
Table 3 Cox proportional hazards analysis of factors predicting time to biochemical recurrence following radical prostatectomy HR
95 % CI
p
Age
1.02
1.01–1.05
0.039
PSA
1.05
1.04–1.06
\0.001
Univariate analysis
BMI category ([23 kg/m2)
0.64
0.46–0.89
0.009
Pathologic T stage (CT3)
3.70
2.64–5.20
\0.001
Tumor volume
1.16
1.12–1.19
\0.001
Pathologic Gleason score (C8)
4.54
2.77–7.44
\0.001
PSM
3.23
2.28–4.57
\0.001
Lymphovascular invasion
2.29
1.15–4.03
\0.001
Perineural invasion Lymph node metastasis
2.51 0.72
1.43–4.05 0.36–1.48
\0.001 0.374 \0.001
Multivariate analysisa
Fig. 1 Kaplan–Meier curves for BCR-free survival rate according to BMI category (log-rank p = 0.017)
normal weight patients (77.8 vs. 70.3 %; p = 0.017). When multiple Cox-proportional hazards regression analysis was performed incorporating variables of PSA, BMI category, pathological Gleason score pathological T stage, PSM, and lymphovascular and PNIs, higher BMI category remained a significant predictor of reduced risk of BCR (Table 3).
Discussion Previous studies have reported varying results regarding the association between obesity and PCa. Most of these studies have been conducted in Western countries where obesity is very common. Case–control and cohort studies from the United States have reported higher incidence of PCa in obese men and have shown that those men harbor more aggressive PCa features than their normal weight
123
PSA
1.03
1.02–1.05
BMI category ([23 kg/m2)
0.63
0.46–0.89
0.008
Pathologic Gleason score (C8)
2.73
1.61–4.63
0.009
Pathologic T stage (CT3)
1.67
1.14–2.44
0.009
PSM
2.05
1.42–2.96
\0.001
Lymphovascular invasion
1.94
1.24–3.04
0.003
Perineural invasion
1.65
1.10–2.46
0.015
HR Hazard ratio, other abbreviations as in Tables 1 and 2 a
Forward stepwise conditional method
counterparts [6, 7]. Likewise, data from the CaPSURE database indicate that obese individuals are predisposed to a higher risk of BCR following RP [11]. Some European studies have also reported that BMI is associated with increased risks of PCa and aggressive disease [12], while others have failed to reveal any associations with oncologic outcome [13, 14]. To the best of the author’s knowledge, the present study of a Korean population is the first to report results contrary to studies of Western populations, which have shown that obese individuals are predisposed to more aggressive pathological features and higher BCR rates [7]. A binary analysis performed in the present study comparing the outcomes of normal weight (BMI \ 23 kg/m2) and
Int Urol Nephrol
overweight and obese (BMI C23 kg/m2) men found that significantly higher pathological Gleason score and increased risk of BCR were observed in normal weight men. Moreover, multivariate Cox-proportional hazards analysis revealed lower BMI category as an independent predictor of BCR in addition to PSA, pathologic Gleason score, pathologic T stage, and presence of PSM, LVI, and PNI. The biological mechanisms underlying the association between obesity and PCa are unclear. Potential mechanisms that have been proposed are: increased serum levels of estrogenic compounds or insulin-like growth factor 1 or decreased levels of sex hormone-binding globulins that increase the levels of biologically available testosterone [7, 15]. Studies have shown that estrogen receptors (ERs) may play a dual role in the prostate by acting in favor of or against cell proliferation and malignant transformation of prostate cells [16, 17]. ERa plays an important role in the growth of normal stromal cells and is also implicated in inflammation and carcinogenesis [18]. Conversely, ERb expressed by epithelial cells has been shown to exhibit anti-proliferative and anti-oxidant properties on PCa cells by inactivating ERa [18]. In breast, colon, ovary, and prostate cancers, ERb has been observed to be a tumor suppressor [19]. Obesity is linked to decreased ERa and increased ERb [20]. Obese men have exhibited upregulated ERb protein content in the prostate and have been found to produce estrogens and isoflavones showing a higher binding affinity for ERb than for ERa [16, 21]. Several studies have found that underexpression of ERb is associated with carcinogenesis in PCa [22]. The proposed protective mechanisms include an anti-proliferative effect, promotion of apoptosis via extrinsic pathways in castration-resistant epithelial cells, and increased cell death by upregulation of proapoptotic genes [23]. Although the exact mechanisms are not yet clearly defined, the fact that obese men in this study harbor lower-grade cancers supports the notion that ERb signaling may play a protective role in PCa carcinogenesis and development. Phytoestrogens have been proposed to exert anti-proliferative and pro-apoptotic effects in PCa through activation of ERb signaling [24]. Studies have revealed that phytoestrogens have the capacity to act as ligands to ESRs with a preference to ERb and the capacity to revert cancer cells to a less aggressive phenotype [22, 25]. Phytoestrogens possess a high-binding affinity to ERb and produce anti-androgenic effects in the prostate [22]. Populationbased studies have observed that high intake of phytoestrogens is associated with lower risk of PCa, which supports the notion that ERb may act as a protective factor in PCa development [26]. Diets with higher levels of phytoestrogens may explain the relatively low incidence of
PCa in Asian men compared with Caucasian men and may also account for the differences in epidemiological studies demonstrating variable associations between obesity and PCa. This suggests that the favorable oncological features and outcomes observed in the overweight and obese cohort in the present study are partly due to upregulation of ERb and its potential as a subtype-selective phytoestrogen to play a protective role against aggressive pathological features and BCR. Androgens also play a key role in the induction and progression of PCa [27]. Studies have shown that increased circulating androgen levels are a risk factor for PCa development and that hypogonadism is associated with a 49 % decreased risk of PCa compared with eugonadism [28]. Body fat composition affects circulating concentrations of estrogens, of which the main source is peripheral fatty tissue where aromatase converts androgens into estrogens [22]. Obese men exhibit greater conversion of peripheral androgens into estradiol by aromatase activity, resulting in relatively lower levels of circulating androgen compared with non-obese men, which may explain the prevalence of lower-grade cancers and lower risk of BCR in this study’s overweight and obese cohort [29]. The current study has several limitations. First, the retrospective analysis of patients who underwent RP may have induced a selection bias so that only patients whose performances were adapted to surgical treatment were included in the study cohort. Moreover, all patients were from a single institution and thus may not represent the Korean population as a whole. Despite such limitations and potential biases inherent in any surgical studies, the strength of the present study is that the cohort was of a single race, unlike Western series in which variable ethnic backgrounds are confounders. Second, racial differences in body fat composition and different BMI cutoff values for obesity may explain the contradictory results observed between the current study and previous Western studies. For instance, while the WHO definition for obesity in Western men is a BMI higher than 30 kg/ m2, only a single patient in our overall cohort was classified into this category. Although BMI is currently the most convenient tool for body fat measurement, obesity is defined as excess body fat, not excess weight; therefore, the inability of BMI to distinguish fat mass from lean mass is a limitation. Asian men typically possess higher body fat percentages at a lower BMI compared with Western men, so a given BMI for an Asian man may have completely different biological characteristics than the same BMI for a Western man [30]. These racial differences in body composition and related endocrine modulations may be underlying causes of the differences in pathological and oncological outcomes according to BMI.
123
Int Urol Nephrol
Conclusions Normal weight Korean PCa patients demonstrate higher Gleason scores and increased risk of BCR compared with the overweight and obese counterpart. These findings suggest that body fat influences pathological features and oncologic outcomes of PCa. If these findings are further confirmed in other Asian cohorts, efforts to manipulate ER signaling pathways as tumor suppressors may potentially lead to future endocrine strategies for chemoprevention and treatment of PCa. Conflict of interest
None.
References 1. Ando R, Nagaya T, Hashimoto Y et al (2008) Inverse relationship between obesity and serum prostate-specific antigen level in healthy Japanese men: a hospital-based cross-sectional survey, 2004–2006. Urology 72:561–565 2. World Health Organization (2000) Obesity: preventing and managing the global epidemic. Report of a WHO consultation. World Health Organ Tech Rep Ser 894:1–253 3. Kamangar F, Dores GM, Anderson WF (2006) Patterns of cancer incidence, mortality, and prevalence across five continents: defining priorities to reduce cancer disparities in different geographic regions of the world. J Clin Oncol 24:2137–2150 4. GLOBOCAN (2008) IARC. http://globocan.iarc.fr/. Accessed 11 Aug 2012 5. Tomaszewski JJ, Chen YF, Bertolet M et al (2013) Obesity is not associated with aggressive pathologic features or biochemical recurrence after radical prostatectomy. Urology 81:992–996 6. Freedland SJ, Isaacs WB, Mangold LA et al (2005) Stronger association between obesity and biochemical progression after radical prostatectomy among men treated in the last 10 years. Clin Cancer Res 11:2883–2888 7. Amling CL, Kane CJ, Riffenburgh RH et al (2001) Relationship between obesity and race in predicting adverse pathologic variables in patients undergoing radical prostatectomy. Urology 58:723–728 8. Narita S, Mitsuzuka K, Yoneyama T et al (2013) Impact of body mass index on clinicopathological outcome and biochemical recurrence after radical prostatectomy. Prostate Cancer Prostatic Dis 16:271–276 9. Giovannucci E, Rimm EB, Colditz GA et al (1993) A prospective study of dietary fat and risk of prostate cancer. J Natl Cancer Inst 85:1571–1579 10. WHO Expert Consultation (2004) Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet 363:157–163 11. Bassett WW, Cooperberg MR, Sadetsky N et al (2005) Impact of obesity on prostate cancer recurrence after radical prostatectomy: data from CaPSURE. Urology 66:1060–1065 12. Chun FK, Briganti A, Graefen M et al (2007) Body mass index does not improve the ability to predict biochemical recurrence after radical prostatectomy. Eur J Cancer 43:375–382
123
13. Isbarn H, Jeldres C, Budaus L et al (2009) Effect of body mass index on histopathologic parameters: results of large European contemporary consecutive open radical prostatectomy series. Urology 73:615–619 14. Gallina A, Karakiewicz PI, Hutterer GC et al (2007) Obesity does not predispose to more aggressive prostate cancer either at biopsy or radical prostatectomy in European men. Int J Cancer 121:791–795 15. Freedland SJ, Aronson WJ, Kane CJ et al (2004) Impact of obesity on biochemical control after radical prostatectomy for clinically localized prostate cancer: a report by the Shared Equal Access Regional Cancer Hospital database study group. J Clin Oncol 22:446–453 16. Ribeiro DL, Pinto ME, Rafacho A et al (2012) High-fat diet obesity associated with insulin resistance increases cell proliferation, estrogen receptor, and PI3 K proteins in rat ventral prostate. J Androl 33:854–865 17. Landstrom M, Zhang JX, Hallmans G et al (1998) Inhibitory effects of soy and rye diets on the development of Dunning R3327 prostate adenocarcinoma in rats. Prostate 36:151–161 18. Holt SK, Kwon EM, Fu R et al (2013) Association of variants in estrogen-related pathway genes with prostate cancer risk. Prostate 73:1–10 19. Kennelly R, Kavanagh DO, Hogan AM et al (2008) Oestrogen and the colon: potential mechanisms for cancer prevention. Lancet Oncol 9:385–391 20. Tomicek NJ, Lancaster TS, Korzick DH (2011) Increased estrogen receptor beta in adipose tissue is associated with increased intracellular and reduced circulating adiponectin protein levels in aged female rats. Gend Med 8:325–333 21. Honma N, Yamamoto K, Ohnaka K et al (2013) Estrogen receptor-beta gene polymorphism and colorectal cancer risk: effect modified by body mass index and isoflavone intake. Int J Cancer 132:951–958 22. Thelen P, Wuttke W, Seidlova-Wuttke D (2013) Phytoestrogens selective for the estrogen receptor beta exert anti-androgenic effects in castration resistant prostate cancer. J Steroid Biochem Mol Biol. doi:10.1016/j.jsbmb.2013.06.009 23. Dey P, Strom A, Gustafsson JA (2013) Estrogen receptor beta upregulates FOXO3a and causes induction of apoptosis through PUMA in prostate cancer. Oncogene. doi:10.1038/onc.2013.384 24. Turner JV, Agatonovic-Kustrin S, Glass BD (2007) Molecular aspects of phytoestrogen selective binding at estrogen receptors. J Pharm Sci 96:1879–1885 25. Andres S, Abraham K, Appel KE et al (2011) Risks and benefits of dietary isoflavones for cancer. Crit Rev Toxicol 41:463–506 26. Hedelin M, Ba¨lter KA, Chang ET et al (2006) Dietary intake of phytoestrogens, estrogen receptor-beta polymorphisms and the risk of prostate cancer. Prostate 66:1512–1520 27. Gelmann EP (2002) Molecular biology of the androgen receptor. J Clin Oncol 20:3001–3015 28. Parsons JK, Carter HB, Platz EA et al (2005) Serum testosterone and the risk of prostate cancer: potential implications for testosterone therapy. Cancer Epidemiol Biomarkers Prev 14:2257–2260 29. Wake DJ, Strand M, Rask E et al (2007) Intra-adipose sex steroid metabolism and body fat distribution in idiopathic human obesity. Clin Endocrinol 66:440–446 30. Deurenberg P, Deurenberg-Yap M, Guricci S (2002) Asians are different from Caucasians and from each other in their body mass index/body fat per cent relationship. Obes Rev 3:141–146
UROLOGY - ORIGINAL PAPER
Low body mass index is associated with adverse oncological outcomes following radical prostatectomy in Korean prostate cancer patients Kyo Chul Koo • Young Eun Yoon • Koon Ho Rha • Byung Ha Chung • Seung Choul Yang Sung Joon Hong
•
Received: 5 March 2014 / Accepted: 23 April 2014 Ó Springer Science+Business Media Dordrecht 2014
Abstract Purpose The purpose of this study was to determine the impact of obesity on clinicopathological features and biochemical recurrence (BCR) following radical prostatectomy (RP) in Korean prostate cancer (PCa) patients. Methods A single-institutional retrospective analysis was performed on 880 PCa patients treated by RP without neoadjuvant therapy between July 2005 and December 2011. Patients were stratified according to body mass index (BMI) standards for Asian populations: obese (BMI C25 kg/m2), overweight (BMI 23–24.9 kg/m2), or normal weight (BMI \23 kg/m2). For analysis, overweight and obese patients were combined (n = 592, BMI C23 kg/ m2) and compared with normal weight patients (n = 288, BMI \23 kg/m2). BCR was defined as prostate-specific antigen (PSA) C0.2 ng/ml following RP. Results Normal weight patients tended to be classified into the higher D’Amico risk category with smaller prostate volumes compared with obese and overweight patients. Normal weight patients had higher pathological Gleason scores and were at higher risk of BCR during the mean follow-up of 58.2 months. This translated to a higher 5-year BCR-free survival rate for obese and overweight patients compared with normal weight patients (77.8 vs. 70.3 %; p = 0.017). On multiple Cox-proportional hazards regression analysis incorporating variables of BMI category, PSA, positive surgical margins, pathological T stage, and Gleason score, higher BMI category remained a
K. C. Koo Y. E. Yoon K. H. Rha B. H. Chung S. C. Yang S. J. Hong (&) Departments of Urology and Urological Science Institute, Yonsei University College of Medicine, 134 Shinchon-dong, Seodaemun-gu, Seoul 120-752, Republic of Korea e-mail: [email protected]
significant predictor of a lower risk of BCR (HR = 0.634, p = 0.028). Conclusions Obese and overweight Korean PCa patients have lower Gleason scores and a reduced risk of BCR compared with normal weight patients. These findings suggest that body fat influences pathological features and oncologic outcomes of PCa. Keywords Body mass index Obesity Prostate cancer Prostatectomy Treatment outcome
Introduction The prevalence and degree of obesity are increasing in Asian countries due to Westernization of diet and lifestyle [1]. However, there still exists a vast difference in body mass index (BMI) distribution between Western and Asian men. In some Asian countries, the average BMI is below 25 kg/m2, which is the cutoff value that the World Health Organization (WHO) uses to define overweight individuals [2]. Prostate cancer (PCa) has shown similar trends. The incidence of PCa varies considerably worldwide, with a greater incidence in Europe and the United States than in Asian countries [3]. In 2008, the age-standardized incidence of PCa in Korea was 22.4 per 100,000, far less than the incidence of 83.8 per 100,000 in the USA. However, by 2011, the Korean incidence had surged to 31.4 per 100,000 [4]. Obesity is a major public health hazard associated with increased risks of cardiac disease, hypertension, and diabetes [2]. It is estimated that more than 30 % of USA men are currently obese, and studies have revealed BMI to be linked to several types of cancers including cancers of the esophagus, colon, liver, kidneys, and pancreas [5].
123
Int Urol Nephrol
However, the relationship between obesity and PCa is still unclear, with large studies from different regions reporting controversial results for incidence, risks of aggressive pathological features, and biochemical recurrence (BCR) following radical prostatectomy (RP) [6–8]. Body fat distribution influences serum androgen level and potential PCa growth factors, and thus, a relationship between BMI and PCa is plausible. Numerous studies have reported consistent associations between dietary fat intake and PCa risk [9]. Therefore, investigating the relationship between BMI and PCa in contemporary, Asian men who are relatively lean compared with Western men may reveal racial disparities in clinicopathological features and risk of BCR and ultimately lead to endocrine modification strategies for a better oncological outcome. The objective of this study was to determine the impact of body fat on clinicopathological PCa features and on the risk of BCR following primary treatment with RP among a cohort of Korean PCa patients who have BMIs that are lower than their typical Western counterparts. The results suggest that body fat is an important factor in PCa development and progression.
institutional electronic medical record database. Clinical data obtained were patient age, BMI, serum prostatespecific antigen (PSA) level, biopsy Gleason score, clinical tumor stage, and prostate volume. Patient height and weight for the calculation of BMI were measured on admission, 1 day before surgery. Pathologic data obtained from RP-PLND specimens included pathological Gleason score, pathological tumor stage, positive surgical margin (PSM) status, extracapsular extension (ECE), LN invasion, specimen weight, tumor volume, and the presence of high-grade prostatic intraepithelial neoplasia (HGPIN), perineural invasion (PNI), and lymphovascular invasion (LVI). Following surgery, PSA measurements of all patients were monitored every 3 months for the first year and then every three to 6 months thereafter. All prostate specimens were serially sectioned, processed, and confirmed by a single uropathologist, with a PSM defined as cancer at the inked margin. Collection of retrospective data of the study was approved by the institutional ethics committee after review of the protocol and procedures employed (20090131-001). Study goals
Materials and methods Patients A retrospective, single-institutional analysis was performed on 1,589 consecutive biopsy-proven PCa patients without distant metastases treated by RP between July 2005 and December 2011. Patients who had received neoadjuvant or adjuvant therapies and those with lymph node (LN) metastases on preoperative imaging were excluded from the cohort. In overall, 880 patients were identified who underwent RP with pelvic lymph node dissection (PLND). BMI was calculated by dividing the weight (kg) by the height squared (m2) at the time of diagnosis. Patients were categorized as obese (C25 kg/m2), overweight 2 (23–24.9 kg/m ), or normal (\23 kg/m2). For analysis, overweight and obese groups were combined (n = 592, BMI C23 kg/m2) and compared with the normal weight group (n = 288, BMI \ 23 kg/m2). This BMI classification was used in accordance with the cutoffs proposed by the International Association for the Study of Obesity (IASO) and the International Obesity Task Force (IOTF), in consideration of the disparity of BMI distribution between Asian-Pacific and Western populations [10]. Assessment of clinicopathological variables Clinical and pathological characteristics of the patients and their oncological outcomes were retrieved from the
123
The primary goal was to explore differences in clinicopathological PCa features according to BMI category. The secondary goal was to determine the impact of BMI on BCR, defined as the first of two or more consecutive increases in PSA [0.2 ng/ml at least 3 months following RP. Statistical analysis Appropriate comparative tests such as the Student’s t test, v2-test, and ANOVA were used to compare clinicopathological variables according to BMI subgroup. BCR-free survival (BCRFS) was estimated and compared using the Kaplan–Meier method and a logrank test. Odds ratios of various adverse pathological outcomes, including high-grade disease (pathological Gleason score C8), ECE, SV invasion, and LN involvement, were estimated for BMI categories using logistic regression analysis. Univariate and multivariate analyses were performed according to Cox-proportional hazards regression models. Variables that were potential predictors for multivariate modeling were selected by univariate analyses and subsequently tested in a stepwise forward conditional manner with entry and retention in the model set at a significance level of 0.05. All tests were two-sided, with statistical significance set at p \ 0.05. Statistical analysis was performed using SPSS version 18 (SPSS Inc., Chicago, IL, USA).
Int Urol Nephrol Table 1 Clinicopathologic features of prostate cancer patients who underwent radical prostatectomy, according to body mass index
BMI (kg/m2) category 18.5–22.9 (N = 288)
p [23 (N = 592)
Preoperative characteristics Age (years)
64.5 ± 7.4
64.1 ± 7.6
0.207
PSA (ng/ml)
10.5 ± 8.7
9.7 ± 8.7
0.227
Clinical tumor stage
0.295
cT1
82 (28.4 %)
cT2
159 (55.3 %)
179 (30.2 %) 334 (56.5 %)
cT3
47 (16.3 %)
79 (13.3 %)
B6
128 (44.5 %)
271 (45.7 %)
7
95 (32.9 %)
194 (32.8 %)
65 (22.6 %)
127 (21.5 %)
Low
28 (9.8 %)
79 (13.4 %)
Intermediate
160 (55.5 %)
361 (60.9 %)
High
100 (34.7 %)
152 (25.7 %)
Prostate volume (gm)
34.2 ± 17.9
37.6 ± 17.3
B6
80 (27.9 %)
207 (35.0 %)
7
156 (54.0 %)
297 (50.1 %)
C8
52 (18.1 %)
88 (14.9 %)
Biopsy Gleason score
C8 D’Amico risk group
0.876
0.011
\0.001
Postoperative characteristics Pathological Gleason score
Data are number (%) and mean (SD) BCR biochemical recurrence, BCRFS BCR-free survival, BMI body mass index, ECE extracapsular extension, PSA prostate-specific antigen, PSM positive surgical margin, SV seminal vesicle
0.041
Lymphovascular invasion
18 (6.3 %)
45 (7.5 %)
0.754
Perineural invasion
147 (51.1 %)
297 (50.2 %)
0.842
PSM
129 (44.8 %)
260 (43.9 %)
0.791
ECE
47 (16.3 %)
78 (13.2 %)
0.231
SV invasion
27 (9.4 %)
46 (7.8 %)
0.435
Lymph node involvement Specimen weight (gm)
26 (8.9 %) 37.7 ± 15.3
42 (7.1 %) 40.9 ± 17.1
0.319 0.011
Tumor volume (cc)
2.45 ± 2.73
2.46 ± 3.97
0.923
BCR
65 (22.4 %)
92 (15.5 %)
0.021
Five-year BCRFS rate (%)
70.3
77.8
0.017
Follow-up period (months)
58.9 ± 15.3
60.2 ± 14.3
0.213
Results BMI and clinicopathological features Clinicopathological features of the patients according to BMI category are presented in Table 1. There were no differences between normal weight patients and overweight and obese patients regarding patient age, PSA, clinical tumor stage, or biopsy Gleason score; however, normal weight patients tended to be classified into a higher D’Amico risk category and to harbor high-grade disease. As expected, normal weight patients had smaller prostate volumes. Pathological features revealed that normal weight patients were more likely to have a higher Gleason score. Logistic regression analyses of BMI category and its association with adverse pathological outcomes showed
that normal weight category is associated with high-grade disease (pathologic Gleason score C8); however, there were no significantly increased risks of ECE, SV invasion, or LN involvement in the normal weight group compared with overweight and obese groups (Table 2). BMI and biochemical recurrence Among the 880 patients, 157 (17.8 %) patients developed BCR during the mean follow-up period of 58.2 months. BCR was observed in 65 (22.4 %) patients in the normal weight group and 92 (15.5 %) patients in the overweight and obese groups (p = 0.021) during the mean follow-up periods of 58.9 and 60.2 months, respectively. As depicted in Fig. 1, this translated to a higher 5-year BCR-free survival rate for overweight and obese patients compared with
123
Int Urol Nephrol Table 2 Logistic regression analyses of BMI category and its association with adverse pathological outcomes Univariate OR
Multivariate 95 % CI
p
OR
95 % CI
p
ECE
0.783
0.525–1.169
0.232
0.721
0.476–1.089
0.223
Pathological Gleason score (C8)
0.719
0.525–0.986
0.041
0.718
0.512–0.991
0.039
SV invasion
0.708
0.517–0.971
0.032
1.513
0.845–2.711
0.711
LN invasion
0.971
0.675–1.393
0.871
0.791
0.518–1.209
0.377
CI confidence interval, OR odds ratio, other abbreviations as in Table 1
Table 3 Cox proportional hazards analysis of factors predicting time to biochemical recurrence following radical prostatectomy HR
95 % CI
p
Age
1.02
1.01–1.05
0.039
PSA
1.05
1.04–1.06
\0.001
Univariate analysis
BMI category ([23 kg/m2)
0.64
0.46–0.89
0.009
Pathologic T stage (CT3)
3.70
2.64–5.20
\0.001
Tumor volume
1.16
1.12–1.19
\0.001
Pathologic Gleason score (C8)
4.54
2.77–7.44
\0.001
PSM
3.23
2.28–4.57
\0.001
Lymphovascular invasion
2.29
1.15–4.03
\0.001
Perineural invasion Lymph node metastasis
2.51 0.72
1.43–4.05 0.36–1.48
\0.001 0.374 \0.001
Multivariate analysisa
Fig. 1 Kaplan–Meier curves for BCR-free survival rate according to BMI category (log-rank p = 0.017)
normal weight patients (77.8 vs. 70.3 %; p = 0.017). When multiple Cox-proportional hazards regression analysis was performed incorporating variables of PSA, BMI category, pathological Gleason score pathological T stage, PSM, and lymphovascular and PNIs, higher BMI category remained a significant predictor of reduced risk of BCR (Table 3).
Discussion Previous studies have reported varying results regarding the association between obesity and PCa. Most of these studies have been conducted in Western countries where obesity is very common. Case–control and cohort studies from the United States have reported higher incidence of PCa in obese men and have shown that those men harbor more aggressive PCa features than their normal weight
123
PSA
1.03
1.02–1.05
BMI category ([23 kg/m2)
0.63
0.46–0.89
0.008
Pathologic Gleason score (C8)
2.73
1.61–4.63
0.009
Pathologic T stage (CT3)
1.67
1.14–2.44
0.009
PSM
2.05
1.42–2.96
\0.001
Lymphovascular invasion
1.94
1.24–3.04
0.003
Perineural invasion
1.65
1.10–2.46
0.015
HR Hazard ratio, other abbreviations as in Tables 1 and 2 a
Forward stepwise conditional method
counterparts [6, 7]. Likewise, data from the CaPSURE database indicate that obese individuals are predisposed to a higher risk of BCR following RP [11]. Some European studies have also reported that BMI is associated with increased risks of PCa and aggressive disease [12], while others have failed to reveal any associations with oncologic outcome [13, 14]. To the best of the author’s knowledge, the present study of a Korean population is the first to report results contrary to studies of Western populations, which have shown that obese individuals are predisposed to more aggressive pathological features and higher BCR rates [7]. A binary analysis performed in the present study comparing the outcomes of normal weight (BMI \ 23 kg/m2) and
Int Urol Nephrol
overweight and obese (BMI C23 kg/m2) men found that significantly higher pathological Gleason score and increased risk of BCR were observed in normal weight men. Moreover, multivariate Cox-proportional hazards analysis revealed lower BMI category as an independent predictor of BCR in addition to PSA, pathologic Gleason score, pathologic T stage, and presence of PSM, LVI, and PNI. The biological mechanisms underlying the association between obesity and PCa are unclear. Potential mechanisms that have been proposed are: increased serum levels of estrogenic compounds or insulin-like growth factor 1 or decreased levels of sex hormone-binding globulins that increase the levels of biologically available testosterone [7, 15]. Studies have shown that estrogen receptors (ERs) may play a dual role in the prostate by acting in favor of or against cell proliferation and malignant transformation of prostate cells [16, 17]. ERa plays an important role in the growth of normal stromal cells and is also implicated in inflammation and carcinogenesis [18]. Conversely, ERb expressed by epithelial cells has been shown to exhibit anti-proliferative and anti-oxidant properties on PCa cells by inactivating ERa [18]. In breast, colon, ovary, and prostate cancers, ERb has been observed to be a tumor suppressor [19]. Obesity is linked to decreased ERa and increased ERb [20]. Obese men have exhibited upregulated ERb protein content in the prostate and have been found to produce estrogens and isoflavones showing a higher binding affinity for ERb than for ERa [16, 21]. Several studies have found that underexpression of ERb is associated with carcinogenesis in PCa [22]. The proposed protective mechanisms include an anti-proliferative effect, promotion of apoptosis via extrinsic pathways in castration-resistant epithelial cells, and increased cell death by upregulation of proapoptotic genes [23]. Although the exact mechanisms are not yet clearly defined, the fact that obese men in this study harbor lower-grade cancers supports the notion that ERb signaling may play a protective role in PCa carcinogenesis and development. Phytoestrogens have been proposed to exert anti-proliferative and pro-apoptotic effects in PCa through activation of ERb signaling [24]. Studies have revealed that phytoestrogens have the capacity to act as ligands to ESRs with a preference to ERb and the capacity to revert cancer cells to a less aggressive phenotype [22, 25]. Phytoestrogens possess a high-binding affinity to ERb and produce anti-androgenic effects in the prostate [22]. Populationbased studies have observed that high intake of phytoestrogens is associated with lower risk of PCa, which supports the notion that ERb may act as a protective factor in PCa development [26]. Diets with higher levels of phytoestrogens may explain the relatively low incidence of
PCa in Asian men compared with Caucasian men and may also account for the differences in epidemiological studies demonstrating variable associations between obesity and PCa. This suggests that the favorable oncological features and outcomes observed in the overweight and obese cohort in the present study are partly due to upregulation of ERb and its potential as a subtype-selective phytoestrogen to play a protective role against aggressive pathological features and BCR. Androgens also play a key role in the induction and progression of PCa [27]. Studies have shown that increased circulating androgen levels are a risk factor for PCa development and that hypogonadism is associated with a 49 % decreased risk of PCa compared with eugonadism [28]. Body fat composition affects circulating concentrations of estrogens, of which the main source is peripheral fatty tissue where aromatase converts androgens into estrogens [22]. Obese men exhibit greater conversion of peripheral androgens into estradiol by aromatase activity, resulting in relatively lower levels of circulating androgen compared with non-obese men, which may explain the prevalence of lower-grade cancers and lower risk of BCR in this study’s overweight and obese cohort [29]. The current study has several limitations. First, the retrospective analysis of patients who underwent RP may have induced a selection bias so that only patients whose performances were adapted to surgical treatment were included in the study cohort. Moreover, all patients were from a single institution and thus may not represent the Korean population as a whole. Despite such limitations and potential biases inherent in any surgical studies, the strength of the present study is that the cohort was of a single race, unlike Western series in which variable ethnic backgrounds are confounders. Second, racial differences in body fat composition and different BMI cutoff values for obesity may explain the contradictory results observed between the current study and previous Western studies. For instance, while the WHO definition for obesity in Western men is a BMI higher than 30 kg/ m2, only a single patient in our overall cohort was classified into this category. Although BMI is currently the most convenient tool for body fat measurement, obesity is defined as excess body fat, not excess weight; therefore, the inability of BMI to distinguish fat mass from lean mass is a limitation. Asian men typically possess higher body fat percentages at a lower BMI compared with Western men, so a given BMI for an Asian man may have completely different biological characteristics than the same BMI for a Western man [30]. These racial differences in body composition and related endocrine modulations may be underlying causes of the differences in pathological and oncological outcomes according to BMI.
123
Int Urol Nephrol
Conclusions Normal weight Korean PCa patients demonstrate higher Gleason scores and increased risk of BCR compared with the overweight and obese counterpart. These findings suggest that body fat influences pathological features and oncologic outcomes of PCa. If these findings are further confirmed in other Asian cohorts, efforts to manipulate ER signaling pathways as tumor suppressors may potentially lead to future endocrine strategies for chemoprevention and treatment of PCa. Conflict of interest
None.
References 1. Ando R, Nagaya T, Hashimoto Y et al (2008) Inverse relationship between obesity and serum prostate-specific antigen level in healthy Japanese men: a hospital-based cross-sectional survey, 2004–2006. Urology 72:561–565 2. World Health Organization (2000) Obesity: preventing and managing the global epidemic. Report of a WHO consultation. World Health Organ Tech Rep Ser 894:1–253 3. Kamangar F, Dores GM, Anderson WF (2006) Patterns of cancer incidence, mortality, and prevalence across five continents: defining priorities to reduce cancer disparities in different geographic regions of the world. J Clin Oncol 24:2137–2150 4. GLOBOCAN (2008) IARC. http://globocan.iarc.fr/. Accessed 11 Aug 2012 5. Tomaszewski JJ, Chen YF, Bertolet M et al (2013) Obesity is not associated with aggressive pathologic features or biochemical recurrence after radical prostatectomy. Urology 81:992–996 6. Freedland SJ, Isaacs WB, Mangold LA et al (2005) Stronger association between obesity and biochemical progression after radical prostatectomy among men treated in the last 10 years. Clin Cancer Res 11:2883–2888 7. Amling CL, Kane CJ, Riffenburgh RH et al (2001) Relationship between obesity and race in predicting adverse pathologic variables in patients undergoing radical prostatectomy. Urology 58:723–728 8. Narita S, Mitsuzuka K, Yoneyama T et al (2013) Impact of body mass index on clinicopathological outcome and biochemical recurrence after radical prostatectomy. Prostate Cancer Prostatic Dis 16:271–276 9. Giovannucci E, Rimm EB, Colditz GA et al (1993) A prospective study of dietary fat and risk of prostate cancer. J Natl Cancer Inst 85:1571–1579 10. WHO Expert Consultation (2004) Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet 363:157–163 11. Bassett WW, Cooperberg MR, Sadetsky N et al (2005) Impact of obesity on prostate cancer recurrence after radical prostatectomy: data from CaPSURE. Urology 66:1060–1065 12. Chun FK, Briganti A, Graefen M et al (2007) Body mass index does not improve the ability to predict biochemical recurrence after radical prostatectomy. Eur J Cancer 43:375–382
123
13. Isbarn H, Jeldres C, Budaus L et al (2009) Effect of body mass index on histopathologic parameters: results of large European contemporary consecutive open radical prostatectomy series. Urology 73:615–619 14. Gallina A, Karakiewicz PI, Hutterer GC et al (2007) Obesity does not predispose to more aggressive prostate cancer either at biopsy or radical prostatectomy in European men. Int J Cancer 121:791–795 15. Freedland SJ, Aronson WJ, Kane CJ et al (2004) Impact of obesity on biochemical control after radical prostatectomy for clinically localized prostate cancer: a report by the Shared Equal Access Regional Cancer Hospital database study group. J Clin Oncol 22:446–453 16. Ribeiro DL, Pinto ME, Rafacho A et al (2012) High-fat diet obesity associated with insulin resistance increases cell proliferation, estrogen receptor, and PI3 K proteins in rat ventral prostate. J Androl 33:854–865 17. Landstrom M, Zhang JX, Hallmans G et al (1998) Inhibitory effects of soy and rye diets on the development of Dunning R3327 prostate adenocarcinoma in rats. Prostate 36:151–161 18. Holt SK, Kwon EM, Fu R et al (2013) Association of variants in estrogen-related pathway genes with prostate cancer risk. Prostate 73:1–10 19. Kennelly R, Kavanagh DO, Hogan AM et al (2008) Oestrogen and the colon: potential mechanisms for cancer prevention. Lancet Oncol 9:385–391 20. Tomicek NJ, Lancaster TS, Korzick DH (2011) Increased estrogen receptor beta in adipose tissue is associated with increased intracellular and reduced circulating adiponectin protein levels in aged female rats. Gend Med 8:325–333 21. Honma N, Yamamoto K, Ohnaka K et al (2013) Estrogen receptor-beta gene polymorphism and colorectal cancer risk: effect modified by body mass index and isoflavone intake. Int J Cancer 132:951–958 22. Thelen P, Wuttke W, Seidlova-Wuttke D (2013) Phytoestrogens selective for the estrogen receptor beta exert anti-androgenic effects in castration resistant prostate cancer. J Steroid Biochem Mol Biol. doi:10.1016/j.jsbmb.2013.06.009 23. Dey P, Strom A, Gustafsson JA (2013) Estrogen receptor beta upregulates FOXO3a and causes induction of apoptosis through PUMA in prostate cancer. Oncogene. doi:10.1038/onc.2013.384 24. Turner JV, Agatonovic-Kustrin S, Glass BD (2007) Molecular aspects of phytoestrogen selective binding at estrogen receptors. J Pharm Sci 96:1879–1885 25. Andres S, Abraham K, Appel KE et al (2011) Risks and benefits of dietary isoflavones for cancer. Crit Rev Toxicol 41:463–506 26. Hedelin M, Ba¨lter KA, Chang ET et al (2006) Dietary intake of phytoestrogens, estrogen receptor-beta polymorphisms and the risk of prostate cancer. Prostate 66:1512–1520 27. Gelmann EP (2002) Molecular biology of the androgen receptor. J Clin Oncol 20:3001–3015 28. Parsons JK, Carter HB, Platz EA et al (2005) Serum testosterone and the risk of prostate cancer: potential implications for testosterone therapy. Cancer Epidemiol Biomarkers Prev 14:2257–2260 29. Wake DJ, Strand M, Rask E et al (2007) Intra-adipose sex steroid metabolism and body fat distribution in idiopathic human obesity. Clin Endocrinol 66:440–446 30. Deurenberg P, Deurenberg-Yap M, Guricci S (2002) Asians are different from Caucasians and from each other in their body mass index/body fat per cent relationship. Obes Rev 3:141–146
