- Email: [email protected]
Impact of Body Mass Index on Outcomes After Conformal Radiotherapy in Patients With Prostate Cancer
Int. J. Radiation Oncology Biol. Phys., Vol. 81, No. 1, pp. 16–22, 2011 Copyright Ó 2011 Elsevier Inc. Printed in the USA. All rights reserved 0360-3016/$–see front matter
doi:10.1016/j.ijrobp.2010.05.059
CLINICAL INVESTIGATION
Prostate
IMPACT OF BODY MASS INDEX ON OUTCOMES AFTER CONFORMAL RADIOTHERAPY IN PATIENTS WITH PROSTATE CANCER HANS GEINITZ, M.D.,* REINHARD THAMM, M.D.,* TOBIAS MUELLER, M.D.,* KERSTIN JESS, M.D.,* FRANK B. ZIMMERMANN, M.D.,y MICHAEL MOLLS, M.D.,* AND CARSTEN NIEDER, M.D.zx *Department of Radiation Oncology, Klinikum rechts der Isar der Technischen Universita¨t Mu¨nchen, Munich, Germany; yDepartment of Radiation Oncology, Universita¨tsspital Basel, Basel, Switzerland; zDepartment of Oncology and Palliative Medicine, Nordland Hospital, Bodø, Norway; and xFaculty of Medicine, University of Tromsø, Tromsø, Norway Purpose: Several retrospective analyses have suggested that obese men with prostate cancer treated with external beam radiotherapy (EBRT) have outcomes inferior to those of normal-weight men. However, a recently presented analysis for the first time challenged this association between body mass index (BMI) and treatment failure. It is therefore important to provide further data on this issue. Methods and Materials: This was a retrospective analysis of 564 men treated with risk-adapted conformal EBRTat a single institution. Low-risk patients received EBRT alone, and the other patients received EBRT plus endocrine treatment. In addition, high-risk patients were treated to higher EBRT doses (74 Gy). A rectal balloon catheter for internal immobilization, which can be identified on portal images, was used in 261 patients (46%). Thus, localization did not rely on bony landmarks alone in these cases. Results: The median BMI was 26, and 15% of patients had BMI $30. Neither univariate nor multivariate analyses detected any significant impact of BMI on biochemical relapse, prostate cancer–specific survival, or overall survival. The 5-year biochemical relapse rate was 21% and prostate cancerspecific survival 96%. Conclusions: The present analysis of a large cohort of consecutively treated patients suggests that efforts to reduce prostate movement and geographic miss might result in comparable outcomes in obese and normal-weight patients. Ó 2011 Elsevier Inc. Conformal radiotherapy, Prostate cancer, Body mass index, Biochemical recurrence, Prognostic factors.
clinical understaging might be an important concern in obese prostate cancer patients. Consistent with prior studies, Loeb et al. found an inverse relationship between obesity and serum prostate specific antigen (PSA) (2). Other authors found associations between diabetes and lower PSA levels (3). However the magnitude of the difference was small. Thus Loeb et al. suggested that adjusting PSA for BMI does not appear to be warranted. Obese patients treated with radical prostatectomy were significantly more likely to harbor high-grade disease, had increased tumor volumes, and had higher rates of positive surgical margins (4). Other data indicate that higher BMI is associated with adverse pathologic findings and is a strong independent predictor of biochemical recurrence after radical prostatectomy (5). These results led to the hypothesis that inherent differences may exist in the biological properties of prostate cancer in obese men compared with normalweight men. However, not all studies confirmed that BMI
INTRODUCTION Obese patients, typically defined by a high body mass index (BMI), with clinically localized prostate cancer might have inferior treatment outcomes than their nonobese counterparts. Among other things, obesity causes endocrine and microenvironmental changes, which could eventually lead to increased tumor aggressiveness. In the Physicians’ Health Study, prediagnostic plasma concentrations of adiponectin and leptin, adipose-derived hormones that play a key role in energy intake and energy expenditure including appetite and metabolism, were measured (1). Although leptin was not found to be related to prostate cancer risk or mortality, higher prediagnostic adiponectin concentrations predisposed men to a lower risk of developing high-grade prostate cancer and to a lower risk of subsequently dying from the cancer. Adiponectin, which is inversely correlated to BMI, has been shown to inhibit prostate cancer cell growth. Besides inherent differences in biology, diagnostic issues such as
Acknowledgment—This study was supported by a grant from the Deutsche Krebshilfe. Received Feb 1, 2010, and in revised form April 7, 2010. Accepted for publication May 6, 2010.
Reprint requests to: Hans Geinitz, M.D., Klinik und Poliklinik fu¨r Strahlentherapie der Technischen Universita¨t Mu¨nchen, Ismaninger Straße 22, 81675 Mu¨nchen, Germany. Tel: ++49 89 41405412; Fax: ++49 89 41404587; E-mail: [email protected] Conflict of interest: none. 16
Body mass index and prostate cancer d H. GEINITZ et al.
contributes significantly to models predicting prostate cancer–specific mortality after radical prostatectomy (6) or predicts biochemical recurrence (7–9). In a large database of 7,274 men with clinically localized prostate cancer from the Cancer of the Prostate Strategic Urological Research Endeavor, no relationship between BMI and cancerspecific or overall survival was found (10). Therefore the debate around potential differences in biological properties continues. In men with castration-resistant metastatic prostate cancer, BMI was not significantly associated with overall survival or PSA declines (11, 12). Higher BMI was associated with better overall and progression-free survival in patients with androgen-dependent metastatic prostate cancer (12). In another study of men with metastatic, castration-resistant prostate cancer, an elevated BMI appeared to have a protective effect against overall mortality and prostate cancer–specific mortality (13). The authors hypothesized that a higher BMI may reflect differences in cancer biology such as the lack of cachexia-producing substances. In men with clinically localized prostate cancer treated with brachytherapy, BMI had no significant impact on biochemical recurrence (14–16), cancer-specific survival, and overall survival (15, 16). These findings are in sharp contrast to the results of the first five studies on external beam radiotherapy (EBRT), which have suggested inferior outcomes in obese patients (17–21). However, a recently presented analysis challenges this association between BMI and treatment failure (22). It is therefore important to provide further data on this issue and to compare the similarities and differences between the EBRT studies. The present analysis relates to a large cohort of consecutively treated patients and suggests that efforts to increase the precision of radiotherapy delivery might result in comparable outcomes in obese and normal-weight patients.
17
METHODS AND MATERIALS This retrospective analysis includes 564 men with clinically localized prostate cancer treated with 3-dimensional computed tomography (CT)–based conformal EBRT at the University Hospital of the Technische Universita¨t in Munich, Germany. The patients were treated between March 1994 and December 2002. Their characteristics are shown in Table 1. Weight and height before the start of EBRT were routinely recorded in the patient charts and used to calculate the BMI. The treatment concept has been described earlier (23–25). Dose was prescribed according to the International Commission on Radiation Units and Measurements (ICRU) 50 guidelines. The 95% isodose encompassed the planning target volume (PTV) and the maximum dose did not exceed 107% of the prescribed dose. Dose per fraction was 1.8 or 2 Gy. Both CT and magnetic resonance imaging (MRI) scans were available to define the clinical target volume (CTV). No image fusion was available. Low-risk patients (at our center, defined as T1/T2a and G1 or G2 (Gleason scores 2–6) and pretreatment PSA # 10 ng/ml) were treated with 70 Gy to a prostate-only CTV. Before 2000 it was our policy to treat these patients with a dose of 66–70 Gy. Intermediate-risk patients (T1/T2 and G3 [Gleason scores 7–10] and/or pretreatment PSA >10 ng/ml and # 20 ng/ml) were treated with 70 Gy to the prostate and the base of the seminal vesicles. High risk patients (T3 or pretreatment PSA >20 ng/ml) were treated with 74 Gy to the prostate and base of the seminal vesicles. If grossly involved on treatment planning MRI, the CTV included the seminal vesicles completely. Before 2000, intermediate and high risk patients received radiation treatment to the entire seminal vesicles up to a dose of 50 Gy, followed by a boost to the prostate only of 20 Gy. None of the patients received treatment to the pelvic lymph nodes. The margins added to the CTV to create the PTV were 1.2 cm in the dorsal direction and 1.5 cm in all other directions for patients treated before 2000. Beginning in 2000, patients were treated within a Phase II multicenter trial (23, 26) with a rectal balloon catheter for internal immobilisation. Overall, 261 patients (46%) were treated with this internal immobilization device. Safety margins in these cases were 1.0 cm in all directions. However, patients who
Table 1. Patient characteristics Parameter
All (n = 564)
BMI <25 kg/m2 (n = 177)
BMI $25 and <30 kg/m2 (n = 302)
BMI $30 kg/m2 (n = 85)
Median follow-up (mo), range Median age (y), range Median initial PSA (ng/ml), range Percent T1, T2, T3, T4 stage Percent Gleason score #6, 7, 8–10* Percent WHO histologic Grade 1, 2, 3 Percent EBRT dose <70, 70, 74 Gy Percent rectal balloon Percent neoadjuvant HT Median duration of HT (mo) Percent BMI <18.5 kg/m2 Percent BMI 18.5–25 kg/m2 Percent BMI 25–30 kg/m2 Percent BMI 30–35 kg/m2 Percent BMI $35 kg/m2 Median BMI, range
51, 16–128 71, 51–88 11, 0.8–171 18, 58, 24, 1 63, 29, 9 14, 73, 14 13, 80, 7 46 84 5.0 0.4 31.1 53.5 12.7 2.3 26, 18–44
50, 19–124 72, 52–86 10, 0.8–121 18, 55, 24, 2 62, 30, 8 12, 75, 13 15, 79, 6 51 86 5.2 1.1 98.9 0 0 0 24, 18–24.9
53, 16–128 70, 52–84 11, 0.75–171 20, 58, 22, 0 62, 29, 9 13, 73, 13 15, 78, 7 42 81 5.0 0 0 100 0 0 27, 25–29.9
46, 20–127 70, 51–88# 13, 1–66 9, 65, 26, 0 64, 26, 9 18, 65, 17 5, 84, 11y 51 89 5.4 0 0 0 83.5 16.5 32, 30–44
Abbreviations: BMI = body mass index; EBRT = external beam radiotherapy; HT = hormonal therapy; PSA = prostate-specific antigen. * A total of 113 patients were without information; only the World Health Organization (WHO) grade is known in these cases. y p < 0.05 for comparison between groups (Kruskal–Wallis test).
18
I. J. Radiation Oncology d Biology d Physics
Volume 81, Number 1, 2011
received 74 Gy had reduced safety margins of 0.5 cm in the dorsal direction for the initial four fractions of 2 Gy. Portal imaging (two orthogonal planes) before each of these four fractions was used to correct the position of the isocenter relative to the rectal balloon catheter. Additional images were taken once weekly or, if the previous images resulted in changes $5 mm, before the next fraction. In patients with smaller initial deviations, the time interval between portal images could be increased during the treatment course. All patients were treated with 6- to 15-MV photons from a linear accelerator via four or five individually shaped treatment fields. Treatment planning was carried out using the HELAX TMS planning system (Nucletron, Veenendaal, The Netherlands, 529 patients) and the Siemens AXIOM planning system (Siemens Erlangen, Germany; 35 patients). Intermediateand high-risk patients were offered neoadjuvant hormonal therapy for 3 to 6 months before EBRT and concomitant to EBRT (antiandrogen plus luteinizing hormone-releasing hormone [LHRH] agonist or LHRH agonist alone).
Biochemical recurrence Biochemical recurrence was defined according to the RTOGASTRO Phoenix consensus, i.e., a rise by $2 ng/ml above the nadir PSA (27). Additional treatment after biochemical failure was at the discretion of the patient’s urologist. No uniform criteria for further intervention were applied.
Statistical analysis For comparison of dichotomous variables the Chi square test and Fisher’s exact test, where applicable, were employed and for continuous variables the Mann–Whitney U Test and the Kruskal–Wallis test. For estimates of biochemical relapse–free rates (bNED), prostate cancer–specific and overall survival from initiation of EBRT the Kaplan–Meier method was used. Comparison between groups was carried out using the log-rank test. For multivariate prediction of bNED and other endpoints, Cox regression analysis (forward stepwise data selection method) was used. Age, PSA, and radiation dose were entered as continuous variables. Endocrine treatment, T stage, and histological grade were entered as noncontinuous variables. BMI was entered as a continuous or noncontinuous (<25 kg/m2 vs. 25 to <30 kg/m2 vs. $30 kg/m2) variable. To compute hazard ratios for all entered variables, Cox regression analysis was carried out also in a saturated model. Histological grade was used rather than Gleason score because patients treated in the earliest phase of the study had no information on Gleason score available. Significance was set to 5%. All tests were carried out two-sided.
Fig. 1. Actuarial biochemical relapse–free rates (bNED) in men with body mass index <25 kg/m2 (blue) vs. 25 to <30 kg/m2 (green) vs. $30 kg/m2 (grey), p = 0.37.
5-year prostate cancer–specific survival was 99%, 94%, and 96%, respectively (p = 0.74) (Fig. 2). The 5-year overall survival was 92%, 84%, and 87%, respectively (p = 0.67) (Fig. 3). Distant metastases developed in 5%, 7%, and 8% of the patients, respectively (p = 0.85). Patient age differed between the three groups, with patients in the low BMI group (<25 kg/m2) being a little older (p = 0.004; Table 1). Followup appeared to be shorter in the group with a BMI $30 kg/ m2, but the difference reached only borderline significance (p = 0.051). No statistically significant differences were found regarding histological grade, Gleason score, initial PSA, T-stage, and use of endocrine treatment among the three BMI groups (Table 1). However, patients with BMI $30 kg/m2 received radiation to significantly higher total doses (Kruskal–Wallis test, p = 0.021; 11% received 74
RESULTS The majority of patients were classified as low risk (17%) or intermediate risk (76%). As shown in Table 1, the typical patient was treated with 70 Gy and almost 6 months of neoadjuvant and concurrent endocrine treatment, and had T2 disease with Gleason score 6 and initial PSA of approximately 11 ng/ml. The median follow-up was 51 months. The median BMI was 26 kg/m2. Of the patients, 31% had a BMI of <25 kg/m2, 54% a BMI of $25 to <30 kg/m2, and 15% a BMI of $30 kg/m2. There was no statistically significant difference in bNED, prostate cancer–specific survival, and overall survival for men with a BMI <25 vs. $25 to <30 vs. $30 (log-rank test, univariate). BNED at 5 years was 84%, 77%, and 75% for the three groups (p = 0.37) (Fig. 1). The
Fig. 2. Actuarial prostate cancer–specific survival in men with body mass index <25 kg/m2 (blue) vs. 25 to <30 kg/m2 (green) vs. $30 kg/m2 (grey) (p = 0.74).
Body mass index and prostate cancer d H. GEINITZ et al.
19
Table 3. Results of multivariate analysis for prostate cancer– specific survival
Fig. 3. Actuarial overall survival in men with body mass index (BMI) <25 kg/m2 (blue) vs. 25 to <30 kg/m2 (green) vs. $30 kg/ m2 (grey) (p = 0.67).
Gy) as a result of risk group assignment. In addition, a larger proportion of men with a BMI $30 kg/m2 had rectal balloon catheters inserted for internal immobilization as compared with patients with a BMI $25 to <30 kg/m2 (51% vs. 42%, difference not statistically significant). In multivariate analyses, BMI as a continuous or noncontinuous (<25 kg/m2, 25 to < 30 kg/m2, > 30 kg/m2) variable had no significant impact on bNED, prostate cancer–specific survival, and overall survival. Regarding bNED, initial PSA (p < 0.001) and histological grade (p = 0.003) were statistically significant. Regarding prostate cancer–specific survival, only histological grade was statistically significant (p = 0.029). Regarding overall survival, only age reached statistical significance (p = 0.027, Tables 2–4). Table 2. Results of multivariate analysis for biochemical relapse-free rate (bNED)
Parameter
Hazard ratio
95% CI
p Value
Age, continuous Hormonal therapy, yes vs. no T-stage T2 vs. T1 T3/T4 vs. T1 WHO grade Grade 2 vs. grade 1 Grade 3 vs. grade 1 EBRT dose, continuous Initial PSA, continuous BMI 25 to <30 kg/m2 vs. <25 kg/m2 $30 kg/m2 vs. <25 kg/m2
1.043 0.725
0.970–1.122 0.215–2.450
0.254 0.605
0.738 2.034
0.142–3.833 0.376–10.99
1.657 4.712 0.998 1.015
0.205–13.37 0.513–43.31 0.811–1.227 1.002–1.028
1.669
0.515–5.408
0.214* 0.718 0.410 0.137* 0.635 0.171 0.982 0.029 0.694* 0.393
1.418
0.279–7.198
0.673
Abbreviations: BMI = body mass index; CI = confidence interval; EBRT = external beam radiotherapy; PSA = prostate-specific antigen; WHO = world health organization. Data in boldface type: p < 0.05. * Likelihood ratio test.
DISCUSSION The present retrospective study showed excellent 5-year prostate cancer–specific survival and low rates of distant metastases in a group of patients treated with risk-adapted EBRT and endocrine treatment. Biochemical relapse occurred in 21% of patients at 5 years. This figure is comparable to the results of several other studies recently reviewed elsewhere (28). Current strategies of radiation dose escalation attempt to reduce the biochemical relapse rate. Many patients (46%) were treated in the context of a well-defined Phase II study protocol. One might criticize the lack of information on Gleason score in some patients, the slight variations in Table 4. Results of multivariate analysis for overall survival
Parameter
Hazard ratio
95% CI
p Value
Parameter/overall survival
Hazard ratio
Age, continuous Hormonal therapy, yes vs. no T-stage T2 vs. T1 T3/T4 vs. T1 WHO grade Grade 2 vs. grade 1 Grade 3 vs. grade 1 EBRT dose, continuous Initial PSA, continuous BMI 25 to <30 kg/m2 vs. <25 kg/m2 $30 kg/m2 vs. <25 kg/m2
0.985 1.004
0.957–1.014 0.591–1.706
0.318 0.988
1.043 0.645
1.363 1.781
0.724–2.568 0.884–3.587
4.190 6.662 0.931 1.018
1.520–11.55 2.226–19.94 0.864–1.003 1.012–1.025
1.272
0.842–1.963
0.242* 0.338 0.106 0.003* 0.006 0.001 0.059 <0.001 0.457* 0.277
0.980
0.507–1.897
0.953
Age, continuous Hormonal therapy, yes vs. no T-stage T2 vs. T1 T3/T4 vs. T1 WHO grade Grade 2 vs. grade 1 Grade 3 vs. grade 1 EBRT dose, continuous Initial PSA, continuous BMI 25 to <30 kg/m2 vs. <25 kg/m2 $30 kg/m2 vs. <25 kg/m2
Abbreviations: BMI = body mass index; EBRT = external beam radiotherapy; PSA = prostate-specific antigen; WHO = world health organization. Data in boldface type: p < 0.05. * Likelihood ratio test.
95% CI
p Value
1.005–1.082 0.027 0.376–1.106 0.111
1.289
0.461* 0.645–2.545 0.479 0.389–2.107 0.818 0.274* 0.458–1.758 0.752 0.648–3.549 0.338 0.915–1.105 0.915 0.996–1.017 0.209 0.636* 0.744–2.235 0.365
1.307
0.614–2.781 0.488
1.282 0.906 0.897 1.516 1.005 1.006
Abbreviations: BMI = body mass index; CI = confidence interval; EBRT = external beam radiotherapy; PSA = prostate-specific antigen; WHO = world health organization. Data in boldface type: p < 0.05. * Likelihood ratio test.
I. J. Radiation Oncology d Biology d Physics
20
length of endocrine treatment, or the definition of risk groups. However, these issues have no confounding influence on our main point of interest, i.e., the potential impact of BMI on outcomes. When interpreting the results, one should note some potential limitations of the study, i.e., that only 15% of patients had BMI $30 kg/m2, that very few patients had BMI >35 kg/m2, and that retrospective studies typically provide hints and hypotheses rather than definitive conclusions. Nevertheless, we believe that the present negative result can be explained by certain treatment-related factors, which will be discussed below, and that it is not simply a false-negative result. Five previous studies, all retrospective in nature, suggested that patients with higher BMI treated with EBRT might have inferior outcomes (17–21). It should be noted that BMI was not the most important predictor of treatment failure (looking at hazard ratios and p values of the multivariate
Volume 81, Number 1, 2011
models). The absolute difference in PSA recurrence was only 3.6% in the study by Stroup et al. (21). It was proposed that technical difficulties with EBRT might explain inferior outcomes. Data from 117 men treated with image-guided radiation therapy (IGRT) suggested strong correlations between BMI and standard deviations of daily shifts in the left–right direction (29). Monitoring of 3 morbidly obese patients with gold marker seeds with daily electronic portal imaging indicated that setup error exceeding 20 mm was present in 20% of treatment days (30). Patient repositioning was considered mandatory if the magnitude of error was >4 mm in any one direction. Patient position correction was necessary on 80% of treatment days. Error in the left–right direction was the major cause of position correction. Daily setup in these patients might be difficult because skin tattoos could display large shifts in relation to bony structures. The aperture limit of CT scanners might cause compromises in patient
Table 5. Overview of previous studies Author, year (Ref)
Patients
King et al. 2009 (31)
n = 90, 1984–2004, median follow-up 44 months
Efstathiou et al., 2007 (17)
n = 99, 1995–2001, median follow-up 83 months
Palma et al., 2007 (20)
n = 706, 1994–2001, median follow-up 91 months
Stroup et al., 2007 (21)
n = 1,868, 1989–2003, median follow-up 43 months
Efstathiou et al., 2007 (19)
n = 788, 1985–1992, median follow-up 97 months
Strom et al., 2006 (18)
n = 873, 1988-2001, median follow-up 96 months
Showalter et al., 2009 (22)
n = 201, 2001–2006, median follow-up 36 months
Treatment Salvage EBRT after prostatectomy, with or without pelvic nodes, endocrine therapy in 61% EBRT 70 Gy plus endocrine treatment
Results
Open questions/remarks
BMI evaluated as continuous Impact of differences in field and categorical variable, size and endocrine therapy significantly associated as confounding factors is with PSA failure difficult to estimate.
BMI evaluated as continuous Secondary analysis of variable, significantly a prospective trial with well associated with PSA defined treatment protocol, failure radiation dose and margins. Avoids certain weaknesses of the other studies. EBRT, median 66 Gy, no BMI evaluated as categorical No information on treatment endocrine treatment variable, significantly details, e.g., target volume associated with PSA definition and dose failure and prostate prescription. cancer–specific mortality EBRT, median 68.4 Gy, BMI evaluated as continuous Did the nine sites contributing endocrine therapy in and categorical variable, patients have uniform 27% significantly associated radiation dose prescriptions with PSA failure and margin widths? EBRT to varying doses BMI evaluated as continuous Impact of differences in including postoperative and categorical variable, disease stage, field size and therapy, with or without sign. associated with endocrine therapy as goserelin prostate cancer–specific confounding factors is mortality difficult to estimate. PSA was not mandatory; thus the interaction of PSA and BMI could not be assessed and corrected. EBRT to varying doses, no BMI evaluated as continuous Impact of changes in EBRT technique and dose as endocrine treatment and categorical variable, confounding factors is sign. associated with PSA difficult to estimate. failure and clinical failure Patients with BMI $35 had the highest median PSA value and lowest percentage of Gleason 2–6 tumors. EBRT 65–79 Gy, median BMI evaluated as continuous Abstract-only publication. 74 Gy and categorical variable, not associated with PSA failure
Abbreviations: BMI = body mass index; EBRT = external beam radiotherapy; HT = hormonal therapy; PSA = prostate-specific antigen.
Body mass index and prostate cancer d H. GEINITZ et al.
position, which will disappear during daily treatment on the wider table in the treatment room. Given these findings, it is not unreasonable to expect inferior outcomes in obese men treated with EBRT and standard margins without measures to avoid set-up errors and geographical miss. The same hypothesis might apply to men treated with salvage EBRT for biochemical relapse after prostatectomy (31). We have summarized the findings of all previous studies in Table 5. Some general remarks on the weaknesses of retrospective studies including our own have to be made, although the authors made considerable efforts to account for potential sources of bias. All studies included more or less inhomogeneous patient groups. Even sophisticated multivariate analyses might not correct for all potential confounding factors. If technical problems such as geographical miss result in insufficient dose to the CTV, information on field size, margin width, frequency of portal imaging or other efforts to detect deviations, and level of action to correct deviations, etc., are crucial to correctly interpret the results. However, no detailed information on all of these parameters is provided in the previous studies. Imbalances in EBRT dose or use of endocrine treatment might have a profound impact on the results. If patients with higher BMI received lower EBRT doses or no endocrine treatment, these factors might explain inferior outcomes. Endocrine therapy might compensate for administration of lower radiation doses resulting from geographical miss. However, one of the positive studies relates to data from a prospective clinical trial where all patients had received endocrine treatment and where protocoldefined target volumes and margins were applied (17). Recruitment occurred between 1995 and 2001, i.e., largely before modern IGRT technology was used. The design of this study reduces the likelihood of a false-positive finding. It therefore supports the hypothesis that technical reasons and organ motion might cause setup errors and eventually inferior biochemical control. In the current era of IGRT with more knowledge on the daily repositioning errors in obese ptients, the impact of BMI on outcome is expected to diminish. The results of the brachytherapy studies (14–16), which did not show that BMI predicts for inferior outcome, also are in line with the
21
hypothesis that accurate dose delivery or omission of setup errors counteracts obesity-related problems. In our own study, internal immobilization devices, which are better surrogates of the prostate position than bony landmarks and reduce prostate movement (26, 32), were used in 46% of patients. In addition, patients who received 74 Gy had portal imaging before each of the first four fractions to correct the position of the isocenter relative to the rectal balloon catheter. A larger proportion of our patients with BMI $30 kg/m2 were treated to higher radiation doses and had rectal balloon catheters inserted. We believe that these measures, although not fully equivalent to the IGRT technology available now and not applied in all cases, might explain the fact that patients with higher BMI had the same good outcomes as patients with lower BMI. As mentioned in the Introduction, several authors have suggested that inherent differences might exist in the biological properties of prostate cancer in obese men compared with normal-weight men. In the study by Palma et al., obese men were diagnosed with prostate cancer at an earlier age than normal-weight and overweight men (20). Furthermore, obesity is known to cause changes in serum levels of testosterone, estradiol, insulin, insulin-like growth factors, adiponectin, and leptin (33). These changes might influence growth conditions and microenvironment predisposing to more aggressive cancers. However the true magnitude of the impact of such differences on prostate cancer–specific survival after adequate treatment continues to be debated. Despite lower pretreatment serum testosterone levels, obese men might be more resistant to the testosteronedepleting effects of androgen-suppressing therapy (20, 34). It might also be true that obesity and associated comorbidity reduce the likelihood of initiating salvage treatment, thereby resulting in inferior outcomes after prostate cancer relapse. Treatment of obese patients is therefore more challenging than treatment of normal-weight patients. The present study suggests that obesity is not a contraindication to riskadapted EBRT as long as attempts are made to actually deliver the intended radiation dose to the CTV. Whether this is also true in severely obese patients with BMI >35 kg/m2 requires additional studies in larger groups of patients.
REFERENCES 1. Li H, Stampfer MJ, Mucci L, et al. A 25-year prospective study of plasma adiponectin and leptin concentrations and prostate cancer risk and survival. Clin Chem 2010;56:34–43. 2. Loeb S, Carter HB, Schaeffer EM, et al. Should prostate specific antigen be adjusted for body mass index? Data from the Baltimore Longitudinal Study of Aging. J Urol 2009;182:2646– 2651. 3. Mu¨ller H, Raum E, Rothenbacher D, et al. Association of diabetes and body mass index with levels of prostate-specific antigen: Implications for correction of prostate-specific antigen cutoff values? Cancer Epidemiol Biomarkers Prev 2009;18:1350– 1356. 4. Freedland SJ, Ban˜ez LL, Sun LL, et al. Obese men have highergrade and larger tumors: An analysis of the Duke Prostate Center Database. Prostate Cancer Prostatic Dis 2009;12:259–263.
5. Magheli A, Rais-Bahrami S, Trock BJ, et al. Impact of body mass index on biochemical recurrence rates after radical prostatectomy: An analysis utilizing propensity score matching. Urology 2008;72:1246–1251. 6. Stephenson AJ, Kattan MW, Eastham JA, et al. Prostate cancerspecific mortality after radical prostatectomy for patients treated in the prostate-specific antigen era. J Clin Oncol 2009;27:4300– 4305. 7. van Roermund JG, van Basten JP, Kiemeney LA, et al. Impact of obesity on surgical outcomes following open radical prostatectomy. Urol Int 2009;82:256–261. 8. van Roermund JG, Kok DE, Wildhagen MF, et al. Body mass index as a prognostic marker for biochemical recurrence in Dutch men treated with radical prostatectomy. BJU Int 2009; 104:321–325.
22
I. J. Radiation Oncology d Biology d Physics
9. Motamedinia P, Korets R, Spencer BA, et al. Body mass index trends and role of obesity in predicting outcome after radical prostatectomy. Urology 2008;72:1106–1110. 10. Davies BJ, Smaldone MC, Sadetsky N, et al. The impact of obesity on overall and cancer specific survival in men with prostate cancer. J Urol 2009;182:112–117. 11. Armstrong AJ, Halabi S, de Wit R, et al. The relationship of body mass index and serum testosterone with disease outcomes in men with castration-resistant metastatic prostate cancer. Prostate Cancer Prostatic Dis 2009;12:88–93. 12. Montgomery RB, Goldman B, Tangen CM, et al. Association of body mass index with response and survival in men with metastatic prostate cancer: Southwest Oncology Group trials 8894 and 9916. J Urol 2007;178:1946–1951. 13. Halabi S, Ou SS, Vogelzang NJ, Small EJ. Inverse correlation between body mass index and clinical outcomes in men with advanced castration-recurrent prostate cancer. Cancer 2007;110: 1478–1484. 14. Efstathiou JA, Skowronski RY, Coen JJ, et al. Body mass index and prostate-specific antigen failure following brachytherapy for localized prostate cancer. Int J Radiat Oncol Biol Phys 2008;71:1302–1308. 15. van Roermund JG, Hinnen KA, Battermann JJ, et al. Body mass index is not a prognostic marker for prostate-specific antigen failure and survival in Dutch men treated with brachytherapy. BJU Int 2010;105:42–48. 16. Merrick GS, Galbreath RW, Butler WM, et al. Obesity is not predictive of overall survival following permanent prostate brachytherapy. Am J Clin Oncol 2007;30:588–596. 17. Efstathiou JA, Chen MH, Renshaw AA, et al. Influence of body mass index on prostate-specific antigen failure after androgen suppression and radiation therapy for localized prostate cancer. Cancer 2007;109:1493–1498. 18. Strom SS, Kamat AM, Gruschkus SK, et al. Influence of obesity on biochemical and clinical failure after external-beam radiotherapy for localized prostate cancer. Cancer 2006;107:631– 639. 19. Efstathiou JA, Bae K, Shipley WU, et al. Obesity and mortality in men with locally advanced prostate cancer: Analysis of RTOG 85-31. Cancer 2007;110:2691–2699. 20. Palma D, Pickles T, Tyldesley S. Prostate Cohort Outcomes Initiative. Obesity as a predictor of biochemical recurrence and survival after radiation therapy for prostate cancer. BJU Int 2007;100:315–319. 21. Stroup SP, Cullen J, Auge BK, et al. Effect of obesity on prostate-specific antigen recurrence after radiation therapy for localized prostate cancer as measured by the 2006 Radiation Therapy Oncology Group-American Society for Therapeutic Radiation and Oncology (RTOG-ASTRO) Phoenix consensus definition. Cancer 2007;110:1003–1009. 22. Showalter TN, Lawrence YR, Xu X, et al. The influence of obesity on toxicity and biochemical control after external beam ra-
Volume 81, Number 1, 2011
23.
24. 25. 26.
27.
28.
29.
30.
31.
32.
33.
34.
diation therapy for prostate cancer (Abstract). Int J Radiat Oncol Biol Phys 2009;75:S351–S352. Goldner G, Bombosch V, Geinitz H, et al. Moderate riskadapted dose escalation with three-dimensional conformal radiotherapy of localized prostate cancer from 70 to 74 Gy: First report on 5-year morbidity and biochemical control from a prospective Austrian-German multicenter phase II trial. Strahlenther Onkol 2009;185:94–100. Geinitz H, Zimmermann F, Thamm R, et al. Late rectal symptoms and quality of life after conformal radiation therapy for prostate cancer. Radiother Oncol 2006;79:341–347. Geinitz H, Zimmermann F, Thamm R, et al. 3-D conformal radiation therapy for prostate cancer in elderly patients. Radiother Oncol 2005;76:27–34. Wachter S, Gerstner N, Dorner D, et al. The influence of a rectal balloon tube as internal immobilization device on variations of volumes and dose-volume histograms during treatment course of conformal radiotherapy for prostate cancer. Int J Radiat Oncol Biol Phys 2002;52:91–100. Roach M III, Hanks G, Thames H Jr., et al. Defining biochemical failure following radiotherapy with or without hormonal therapy in men with clinically localized prostate cancer: Recommendations of the RTOG-ASTRO Phoenix Consensus Conference. Int J Radiat Oncol Biol Phys 2006;65:965–974. Viani GA, Stefano EJ, Afonso SL. Higher-than-conventional radiation doses in localized prostate cancer treatment: A metaanalysis of randomized, controlled trials. Int J Radiat Oncol Biol Phys 2009;74:1405–1418. Wong JR, Gao Z, Merrick S, et al. Potential for higher treatment failure in obese patients: Correlation of elevated body mass index and increased daily prostate deviations from the radiation beam isocenters in an analysis of 1,465 computed tomographic images. Int J Radiat Oncol Biol Phys 2009;75:49–55. Millender LE, Aubin M, Pouliot J, et al. Daily electronic portal imaging for morbidly obese men undergoing radiotherapy for localized prostate cancer. Int J Radiat Oncol Biol Phys 2004; 59:6–10. King CR, Spiotto MT, Kapp DS. Obesity and risk of biochemical failure for patients receiving salvage radiotherapy after prostatectomy. Int J Radiat Oncol Biol Phys 2009;73:1017– 1022. Michalski JM, Roach M 3rd, Merrick G, et al. ACR appropriateness criteria on external beam radiation therapy treatment planning for clinically localized prostate cancer. Int J Radiat Oncol Biol Phys 2009;74:667–672. Skolarus TA, Wolin KY, Grubb RL 3rd. The effect of body mass index on PSA levels and the development, screening and treatment of prostate cancer. Nat Clin Pract Urol 2007;4: 605–614. Smith R. Obesity and sex steroids during gonadotropinreleasing hormone agonist treatment for prostate cancer. Clin Cancer Res 2007;13:241–245.
doi:10.1016/j.ijrobp.2010.05.059
CLINICAL INVESTIGATION
Prostate
IMPACT OF BODY MASS INDEX ON OUTCOMES AFTER CONFORMAL RADIOTHERAPY IN PATIENTS WITH PROSTATE CANCER HANS GEINITZ, M.D.,* REINHARD THAMM, M.D.,* TOBIAS MUELLER, M.D.,* KERSTIN JESS, M.D.,* FRANK B. ZIMMERMANN, M.D.,y MICHAEL MOLLS, M.D.,* AND CARSTEN NIEDER, M.D.zx *Department of Radiation Oncology, Klinikum rechts der Isar der Technischen Universita¨t Mu¨nchen, Munich, Germany; yDepartment of Radiation Oncology, Universita¨tsspital Basel, Basel, Switzerland; zDepartment of Oncology and Palliative Medicine, Nordland Hospital, Bodø, Norway; and xFaculty of Medicine, University of Tromsø, Tromsø, Norway Purpose: Several retrospective analyses have suggested that obese men with prostate cancer treated with external beam radiotherapy (EBRT) have outcomes inferior to those of normal-weight men. However, a recently presented analysis for the first time challenged this association between body mass index (BMI) and treatment failure. It is therefore important to provide further data on this issue. Methods and Materials: This was a retrospective analysis of 564 men treated with risk-adapted conformal EBRTat a single institution. Low-risk patients received EBRT alone, and the other patients received EBRT plus endocrine treatment. In addition, high-risk patients were treated to higher EBRT doses (74 Gy). A rectal balloon catheter for internal immobilization, which can be identified on portal images, was used in 261 patients (46%). Thus, localization did not rely on bony landmarks alone in these cases. Results: The median BMI was 26, and 15% of patients had BMI $30. Neither univariate nor multivariate analyses detected any significant impact of BMI on biochemical relapse, prostate cancer–specific survival, or overall survival. The 5-year biochemical relapse rate was 21% and prostate cancerspecific survival 96%. Conclusions: The present analysis of a large cohort of consecutively treated patients suggests that efforts to reduce prostate movement and geographic miss might result in comparable outcomes in obese and normal-weight patients. Ó 2011 Elsevier Inc. Conformal radiotherapy, Prostate cancer, Body mass index, Biochemical recurrence, Prognostic factors.
clinical understaging might be an important concern in obese prostate cancer patients. Consistent with prior studies, Loeb et al. found an inverse relationship between obesity and serum prostate specific antigen (PSA) (2). Other authors found associations between diabetes and lower PSA levels (3). However the magnitude of the difference was small. Thus Loeb et al. suggested that adjusting PSA for BMI does not appear to be warranted. Obese patients treated with radical prostatectomy were significantly more likely to harbor high-grade disease, had increased tumor volumes, and had higher rates of positive surgical margins (4). Other data indicate that higher BMI is associated with adverse pathologic findings and is a strong independent predictor of biochemical recurrence after radical prostatectomy (5). These results led to the hypothesis that inherent differences may exist in the biological properties of prostate cancer in obese men compared with normalweight men. However, not all studies confirmed that BMI
INTRODUCTION Obese patients, typically defined by a high body mass index (BMI), with clinically localized prostate cancer might have inferior treatment outcomes than their nonobese counterparts. Among other things, obesity causes endocrine and microenvironmental changes, which could eventually lead to increased tumor aggressiveness. In the Physicians’ Health Study, prediagnostic plasma concentrations of adiponectin and leptin, adipose-derived hormones that play a key role in energy intake and energy expenditure including appetite and metabolism, were measured (1). Although leptin was not found to be related to prostate cancer risk or mortality, higher prediagnostic adiponectin concentrations predisposed men to a lower risk of developing high-grade prostate cancer and to a lower risk of subsequently dying from the cancer. Adiponectin, which is inversely correlated to BMI, has been shown to inhibit prostate cancer cell growth. Besides inherent differences in biology, diagnostic issues such as
Acknowledgment—This study was supported by a grant from the Deutsche Krebshilfe. Received Feb 1, 2010, and in revised form April 7, 2010. Accepted for publication May 6, 2010.
Reprint requests to: Hans Geinitz, M.D., Klinik und Poliklinik fu¨r Strahlentherapie der Technischen Universita¨t Mu¨nchen, Ismaninger Straße 22, 81675 Mu¨nchen, Germany. Tel: ++49 89 41405412; Fax: ++49 89 41404587; E-mail: [email protected] Conflict of interest: none. 16
Body mass index and prostate cancer d H. GEINITZ et al.
contributes significantly to models predicting prostate cancer–specific mortality after radical prostatectomy (6) or predicts biochemical recurrence (7–9). In a large database of 7,274 men with clinically localized prostate cancer from the Cancer of the Prostate Strategic Urological Research Endeavor, no relationship between BMI and cancerspecific or overall survival was found (10). Therefore the debate around potential differences in biological properties continues. In men with castration-resistant metastatic prostate cancer, BMI was not significantly associated with overall survival or PSA declines (11, 12). Higher BMI was associated with better overall and progression-free survival in patients with androgen-dependent metastatic prostate cancer (12). In another study of men with metastatic, castration-resistant prostate cancer, an elevated BMI appeared to have a protective effect against overall mortality and prostate cancer–specific mortality (13). The authors hypothesized that a higher BMI may reflect differences in cancer biology such as the lack of cachexia-producing substances. In men with clinically localized prostate cancer treated with brachytherapy, BMI had no significant impact on biochemical recurrence (14–16), cancer-specific survival, and overall survival (15, 16). These findings are in sharp contrast to the results of the first five studies on external beam radiotherapy (EBRT), which have suggested inferior outcomes in obese patients (17–21). However, a recently presented analysis challenges this association between BMI and treatment failure (22). It is therefore important to provide further data on this issue and to compare the similarities and differences between the EBRT studies. The present analysis relates to a large cohort of consecutively treated patients and suggests that efforts to increase the precision of radiotherapy delivery might result in comparable outcomes in obese and normal-weight patients.
17
METHODS AND MATERIALS This retrospective analysis includes 564 men with clinically localized prostate cancer treated with 3-dimensional computed tomography (CT)–based conformal EBRT at the University Hospital of the Technische Universita¨t in Munich, Germany. The patients were treated between March 1994 and December 2002. Their characteristics are shown in Table 1. Weight and height before the start of EBRT were routinely recorded in the patient charts and used to calculate the BMI. The treatment concept has been described earlier (23–25). Dose was prescribed according to the International Commission on Radiation Units and Measurements (ICRU) 50 guidelines. The 95% isodose encompassed the planning target volume (PTV) and the maximum dose did not exceed 107% of the prescribed dose. Dose per fraction was 1.8 or 2 Gy. Both CT and magnetic resonance imaging (MRI) scans were available to define the clinical target volume (CTV). No image fusion was available. Low-risk patients (at our center, defined as T1/T2a and G1 or G2 (Gleason scores 2–6) and pretreatment PSA # 10 ng/ml) were treated with 70 Gy to a prostate-only CTV. Before 2000 it was our policy to treat these patients with a dose of 66–70 Gy. Intermediate-risk patients (T1/T2 and G3 [Gleason scores 7–10] and/or pretreatment PSA >10 ng/ml and # 20 ng/ml) were treated with 70 Gy to the prostate and the base of the seminal vesicles. High risk patients (T3 or pretreatment PSA >20 ng/ml) were treated with 74 Gy to the prostate and base of the seminal vesicles. If grossly involved on treatment planning MRI, the CTV included the seminal vesicles completely. Before 2000, intermediate and high risk patients received radiation treatment to the entire seminal vesicles up to a dose of 50 Gy, followed by a boost to the prostate only of 20 Gy. None of the patients received treatment to the pelvic lymph nodes. The margins added to the CTV to create the PTV were 1.2 cm in the dorsal direction and 1.5 cm in all other directions for patients treated before 2000. Beginning in 2000, patients were treated within a Phase II multicenter trial (23, 26) with a rectal balloon catheter for internal immobilisation. Overall, 261 patients (46%) were treated with this internal immobilization device. Safety margins in these cases were 1.0 cm in all directions. However, patients who
Table 1. Patient characteristics Parameter
All (n = 564)
BMI <25 kg/m2 (n = 177)
BMI $25 and <30 kg/m2 (n = 302)
BMI $30 kg/m2 (n = 85)
Median follow-up (mo), range Median age (y), range Median initial PSA (ng/ml), range Percent T1, T2, T3, T4 stage Percent Gleason score #6, 7, 8–10* Percent WHO histologic Grade 1, 2, 3 Percent EBRT dose <70, 70, 74 Gy Percent rectal balloon Percent neoadjuvant HT Median duration of HT (mo) Percent BMI <18.5 kg/m2 Percent BMI 18.5–25 kg/m2 Percent BMI 25–30 kg/m2 Percent BMI 30–35 kg/m2 Percent BMI $35 kg/m2 Median BMI, range
51, 16–128 71, 51–88 11, 0.8–171 18, 58, 24, 1 63, 29, 9 14, 73, 14 13, 80, 7 46 84 5.0 0.4 31.1 53.5 12.7 2.3 26, 18–44
50, 19–124 72, 52–86 10, 0.8–121 18, 55, 24, 2 62, 30, 8 12, 75, 13 15, 79, 6 51 86 5.2 1.1 98.9 0 0 0 24, 18–24.9
53, 16–128 70, 52–84 11, 0.75–171 20, 58, 22, 0 62, 29, 9 13, 73, 13 15, 78, 7 42 81 5.0 0 0 100 0 0 27, 25–29.9
46, 20–127 70, 51–88# 13, 1–66 9, 65, 26, 0 64, 26, 9 18, 65, 17 5, 84, 11y 51 89 5.4 0 0 0 83.5 16.5 32, 30–44
Abbreviations: BMI = body mass index; EBRT = external beam radiotherapy; HT = hormonal therapy; PSA = prostate-specific antigen. * A total of 113 patients were without information; only the World Health Organization (WHO) grade is known in these cases. y p < 0.05 for comparison between groups (Kruskal–Wallis test).
18
I. J. Radiation Oncology d Biology d Physics
Volume 81, Number 1, 2011
received 74 Gy had reduced safety margins of 0.5 cm in the dorsal direction for the initial four fractions of 2 Gy. Portal imaging (two orthogonal planes) before each of these four fractions was used to correct the position of the isocenter relative to the rectal balloon catheter. Additional images were taken once weekly or, if the previous images resulted in changes $5 mm, before the next fraction. In patients with smaller initial deviations, the time interval between portal images could be increased during the treatment course. All patients were treated with 6- to 15-MV photons from a linear accelerator via four or five individually shaped treatment fields. Treatment planning was carried out using the HELAX TMS planning system (Nucletron, Veenendaal, The Netherlands, 529 patients) and the Siemens AXIOM planning system (Siemens Erlangen, Germany; 35 patients). Intermediateand high-risk patients were offered neoadjuvant hormonal therapy for 3 to 6 months before EBRT and concomitant to EBRT (antiandrogen plus luteinizing hormone-releasing hormone [LHRH] agonist or LHRH agonist alone).
Biochemical recurrence Biochemical recurrence was defined according to the RTOGASTRO Phoenix consensus, i.e., a rise by $2 ng/ml above the nadir PSA (27). Additional treatment after biochemical failure was at the discretion of the patient’s urologist. No uniform criteria for further intervention were applied.
Statistical analysis For comparison of dichotomous variables the Chi square test and Fisher’s exact test, where applicable, were employed and for continuous variables the Mann–Whitney U Test and the Kruskal–Wallis test. For estimates of biochemical relapse–free rates (bNED), prostate cancer–specific and overall survival from initiation of EBRT the Kaplan–Meier method was used. Comparison between groups was carried out using the log-rank test. For multivariate prediction of bNED and other endpoints, Cox regression analysis (forward stepwise data selection method) was used. Age, PSA, and radiation dose were entered as continuous variables. Endocrine treatment, T stage, and histological grade were entered as noncontinuous variables. BMI was entered as a continuous or noncontinuous (<25 kg/m2 vs. 25 to <30 kg/m2 vs. $30 kg/m2) variable. To compute hazard ratios for all entered variables, Cox regression analysis was carried out also in a saturated model. Histological grade was used rather than Gleason score because patients treated in the earliest phase of the study had no information on Gleason score available. Significance was set to 5%. All tests were carried out two-sided.
Fig. 1. Actuarial biochemical relapse–free rates (bNED) in men with body mass index <25 kg/m2 (blue) vs. 25 to <30 kg/m2 (green) vs. $30 kg/m2 (grey), p = 0.37.
5-year prostate cancer–specific survival was 99%, 94%, and 96%, respectively (p = 0.74) (Fig. 2). The 5-year overall survival was 92%, 84%, and 87%, respectively (p = 0.67) (Fig. 3). Distant metastases developed in 5%, 7%, and 8% of the patients, respectively (p = 0.85). Patient age differed between the three groups, with patients in the low BMI group (<25 kg/m2) being a little older (p = 0.004; Table 1). Followup appeared to be shorter in the group with a BMI $30 kg/ m2, but the difference reached only borderline significance (p = 0.051). No statistically significant differences were found regarding histological grade, Gleason score, initial PSA, T-stage, and use of endocrine treatment among the three BMI groups (Table 1). However, patients with BMI $30 kg/m2 received radiation to significantly higher total doses (Kruskal–Wallis test, p = 0.021; 11% received 74
RESULTS The majority of patients were classified as low risk (17%) or intermediate risk (76%). As shown in Table 1, the typical patient was treated with 70 Gy and almost 6 months of neoadjuvant and concurrent endocrine treatment, and had T2 disease with Gleason score 6 and initial PSA of approximately 11 ng/ml. The median follow-up was 51 months. The median BMI was 26 kg/m2. Of the patients, 31% had a BMI of <25 kg/m2, 54% a BMI of $25 to <30 kg/m2, and 15% a BMI of $30 kg/m2. There was no statistically significant difference in bNED, prostate cancer–specific survival, and overall survival for men with a BMI <25 vs. $25 to <30 vs. $30 (log-rank test, univariate). BNED at 5 years was 84%, 77%, and 75% for the three groups (p = 0.37) (Fig. 1). The
Fig. 2. Actuarial prostate cancer–specific survival in men with body mass index <25 kg/m2 (blue) vs. 25 to <30 kg/m2 (green) vs. $30 kg/m2 (grey) (p = 0.74).
Body mass index and prostate cancer d H. GEINITZ et al.
19
Table 3. Results of multivariate analysis for prostate cancer– specific survival
Fig. 3. Actuarial overall survival in men with body mass index (BMI) <25 kg/m2 (blue) vs. 25 to <30 kg/m2 (green) vs. $30 kg/ m2 (grey) (p = 0.67).
Gy) as a result of risk group assignment. In addition, a larger proportion of men with a BMI $30 kg/m2 had rectal balloon catheters inserted for internal immobilization as compared with patients with a BMI $25 to <30 kg/m2 (51% vs. 42%, difference not statistically significant). In multivariate analyses, BMI as a continuous or noncontinuous (<25 kg/m2, 25 to < 30 kg/m2, > 30 kg/m2) variable had no significant impact on bNED, prostate cancer–specific survival, and overall survival. Regarding bNED, initial PSA (p < 0.001) and histological grade (p = 0.003) were statistically significant. Regarding prostate cancer–specific survival, only histological grade was statistically significant (p = 0.029). Regarding overall survival, only age reached statistical significance (p = 0.027, Tables 2–4). Table 2. Results of multivariate analysis for biochemical relapse-free rate (bNED)
Parameter
Hazard ratio
95% CI
p Value
Age, continuous Hormonal therapy, yes vs. no T-stage T2 vs. T1 T3/T4 vs. T1 WHO grade Grade 2 vs. grade 1 Grade 3 vs. grade 1 EBRT dose, continuous Initial PSA, continuous BMI 25 to <30 kg/m2 vs. <25 kg/m2 $30 kg/m2 vs. <25 kg/m2
1.043 0.725
0.970–1.122 0.215–2.450
0.254 0.605
0.738 2.034
0.142–3.833 0.376–10.99
1.657 4.712 0.998 1.015
0.205–13.37 0.513–43.31 0.811–1.227 1.002–1.028
1.669
0.515–5.408
0.214* 0.718 0.410 0.137* 0.635 0.171 0.982 0.029 0.694* 0.393
1.418
0.279–7.198
0.673
Abbreviations: BMI = body mass index; CI = confidence interval; EBRT = external beam radiotherapy; PSA = prostate-specific antigen; WHO = world health organization. Data in boldface type: p < 0.05. * Likelihood ratio test.
DISCUSSION The present retrospective study showed excellent 5-year prostate cancer–specific survival and low rates of distant metastases in a group of patients treated with risk-adapted EBRT and endocrine treatment. Biochemical relapse occurred in 21% of patients at 5 years. This figure is comparable to the results of several other studies recently reviewed elsewhere (28). Current strategies of radiation dose escalation attempt to reduce the biochemical relapse rate. Many patients (46%) were treated in the context of a well-defined Phase II study protocol. One might criticize the lack of information on Gleason score in some patients, the slight variations in Table 4. Results of multivariate analysis for overall survival
Parameter
Hazard ratio
95% CI
p Value
Parameter/overall survival
Hazard ratio
Age, continuous Hormonal therapy, yes vs. no T-stage T2 vs. T1 T3/T4 vs. T1 WHO grade Grade 2 vs. grade 1 Grade 3 vs. grade 1 EBRT dose, continuous Initial PSA, continuous BMI 25 to <30 kg/m2 vs. <25 kg/m2 $30 kg/m2 vs. <25 kg/m2
0.985 1.004
0.957–1.014 0.591–1.706
0.318 0.988
1.043 0.645
1.363 1.781
0.724–2.568 0.884–3.587
4.190 6.662 0.931 1.018
1.520–11.55 2.226–19.94 0.864–1.003 1.012–1.025
1.272
0.842–1.963
0.242* 0.338 0.106 0.003* 0.006 0.001 0.059 <0.001 0.457* 0.277
0.980
0.507–1.897
0.953
Age, continuous Hormonal therapy, yes vs. no T-stage T2 vs. T1 T3/T4 vs. T1 WHO grade Grade 2 vs. grade 1 Grade 3 vs. grade 1 EBRT dose, continuous Initial PSA, continuous BMI 25 to <30 kg/m2 vs. <25 kg/m2 $30 kg/m2 vs. <25 kg/m2
Abbreviations: BMI = body mass index; EBRT = external beam radiotherapy; PSA = prostate-specific antigen; WHO = world health organization. Data in boldface type: p < 0.05. * Likelihood ratio test.
95% CI
p Value
1.005–1.082 0.027 0.376–1.106 0.111
1.289
0.461* 0.645–2.545 0.479 0.389–2.107 0.818 0.274* 0.458–1.758 0.752 0.648–3.549 0.338 0.915–1.105 0.915 0.996–1.017 0.209 0.636* 0.744–2.235 0.365
1.307
0.614–2.781 0.488
1.282 0.906 0.897 1.516 1.005 1.006
Abbreviations: BMI = body mass index; CI = confidence interval; EBRT = external beam radiotherapy; PSA = prostate-specific antigen; WHO = world health organization. Data in boldface type: p < 0.05. * Likelihood ratio test.
I. J. Radiation Oncology d Biology d Physics
20
length of endocrine treatment, or the definition of risk groups. However, these issues have no confounding influence on our main point of interest, i.e., the potential impact of BMI on outcomes. When interpreting the results, one should note some potential limitations of the study, i.e., that only 15% of patients had BMI $30 kg/m2, that very few patients had BMI >35 kg/m2, and that retrospective studies typically provide hints and hypotheses rather than definitive conclusions. Nevertheless, we believe that the present negative result can be explained by certain treatment-related factors, which will be discussed below, and that it is not simply a false-negative result. Five previous studies, all retrospective in nature, suggested that patients with higher BMI treated with EBRT might have inferior outcomes (17–21). It should be noted that BMI was not the most important predictor of treatment failure (looking at hazard ratios and p values of the multivariate
Volume 81, Number 1, 2011
models). The absolute difference in PSA recurrence was only 3.6% in the study by Stroup et al. (21). It was proposed that technical difficulties with EBRT might explain inferior outcomes. Data from 117 men treated with image-guided radiation therapy (IGRT) suggested strong correlations between BMI and standard deviations of daily shifts in the left–right direction (29). Monitoring of 3 morbidly obese patients with gold marker seeds with daily electronic portal imaging indicated that setup error exceeding 20 mm was present in 20% of treatment days (30). Patient repositioning was considered mandatory if the magnitude of error was >4 mm in any one direction. Patient position correction was necessary on 80% of treatment days. Error in the left–right direction was the major cause of position correction. Daily setup in these patients might be difficult because skin tattoos could display large shifts in relation to bony structures. The aperture limit of CT scanners might cause compromises in patient
Table 5. Overview of previous studies Author, year (Ref)
Patients
King et al. 2009 (31)
n = 90, 1984–2004, median follow-up 44 months
Efstathiou et al., 2007 (17)
n = 99, 1995–2001, median follow-up 83 months
Palma et al., 2007 (20)
n = 706, 1994–2001, median follow-up 91 months
Stroup et al., 2007 (21)
n = 1,868, 1989–2003, median follow-up 43 months
Efstathiou et al., 2007 (19)
n = 788, 1985–1992, median follow-up 97 months
Strom et al., 2006 (18)
n = 873, 1988-2001, median follow-up 96 months
Showalter et al., 2009 (22)
n = 201, 2001–2006, median follow-up 36 months
Treatment Salvage EBRT after prostatectomy, with or without pelvic nodes, endocrine therapy in 61% EBRT 70 Gy plus endocrine treatment
Results
Open questions/remarks
BMI evaluated as continuous Impact of differences in field and categorical variable, size and endocrine therapy significantly associated as confounding factors is with PSA failure difficult to estimate.
BMI evaluated as continuous Secondary analysis of variable, significantly a prospective trial with well associated with PSA defined treatment protocol, failure radiation dose and margins. Avoids certain weaknesses of the other studies. EBRT, median 66 Gy, no BMI evaluated as categorical No information on treatment endocrine treatment variable, significantly details, e.g., target volume associated with PSA definition and dose failure and prostate prescription. cancer–specific mortality EBRT, median 68.4 Gy, BMI evaluated as continuous Did the nine sites contributing endocrine therapy in and categorical variable, patients have uniform 27% significantly associated radiation dose prescriptions with PSA failure and margin widths? EBRT to varying doses BMI evaluated as continuous Impact of differences in including postoperative and categorical variable, disease stage, field size and therapy, with or without sign. associated with endocrine therapy as goserelin prostate cancer–specific confounding factors is mortality difficult to estimate. PSA was not mandatory; thus the interaction of PSA and BMI could not be assessed and corrected. EBRT to varying doses, no BMI evaluated as continuous Impact of changes in EBRT technique and dose as endocrine treatment and categorical variable, confounding factors is sign. associated with PSA difficult to estimate. failure and clinical failure Patients with BMI $35 had the highest median PSA value and lowest percentage of Gleason 2–6 tumors. EBRT 65–79 Gy, median BMI evaluated as continuous Abstract-only publication. 74 Gy and categorical variable, not associated with PSA failure
Abbreviations: BMI = body mass index; EBRT = external beam radiotherapy; HT = hormonal therapy; PSA = prostate-specific antigen.
Body mass index and prostate cancer d H. GEINITZ et al.
position, which will disappear during daily treatment on the wider table in the treatment room. Given these findings, it is not unreasonable to expect inferior outcomes in obese men treated with EBRT and standard margins without measures to avoid set-up errors and geographical miss. The same hypothesis might apply to men treated with salvage EBRT for biochemical relapse after prostatectomy (31). We have summarized the findings of all previous studies in Table 5. Some general remarks on the weaknesses of retrospective studies including our own have to be made, although the authors made considerable efforts to account for potential sources of bias. All studies included more or less inhomogeneous patient groups. Even sophisticated multivariate analyses might not correct for all potential confounding factors. If technical problems such as geographical miss result in insufficient dose to the CTV, information on field size, margin width, frequency of portal imaging or other efforts to detect deviations, and level of action to correct deviations, etc., are crucial to correctly interpret the results. However, no detailed information on all of these parameters is provided in the previous studies. Imbalances in EBRT dose or use of endocrine treatment might have a profound impact on the results. If patients with higher BMI received lower EBRT doses or no endocrine treatment, these factors might explain inferior outcomes. Endocrine therapy might compensate for administration of lower radiation doses resulting from geographical miss. However, one of the positive studies relates to data from a prospective clinical trial where all patients had received endocrine treatment and where protocoldefined target volumes and margins were applied (17). Recruitment occurred between 1995 and 2001, i.e., largely before modern IGRT technology was used. The design of this study reduces the likelihood of a false-positive finding. It therefore supports the hypothesis that technical reasons and organ motion might cause setup errors and eventually inferior biochemical control. In the current era of IGRT with more knowledge on the daily repositioning errors in obese ptients, the impact of BMI on outcome is expected to diminish. The results of the brachytherapy studies (14–16), which did not show that BMI predicts for inferior outcome, also are in line with the
21
hypothesis that accurate dose delivery or omission of setup errors counteracts obesity-related problems. In our own study, internal immobilization devices, which are better surrogates of the prostate position than bony landmarks and reduce prostate movement (26, 32), were used in 46% of patients. In addition, patients who received 74 Gy had portal imaging before each of the first four fractions to correct the position of the isocenter relative to the rectal balloon catheter. A larger proportion of our patients with BMI $30 kg/m2 were treated to higher radiation doses and had rectal balloon catheters inserted. We believe that these measures, although not fully equivalent to the IGRT technology available now and not applied in all cases, might explain the fact that patients with higher BMI had the same good outcomes as patients with lower BMI. As mentioned in the Introduction, several authors have suggested that inherent differences might exist in the biological properties of prostate cancer in obese men compared with normal-weight men. In the study by Palma et al., obese men were diagnosed with prostate cancer at an earlier age than normal-weight and overweight men (20). Furthermore, obesity is known to cause changes in serum levels of testosterone, estradiol, insulin, insulin-like growth factors, adiponectin, and leptin (33). These changes might influence growth conditions and microenvironment predisposing to more aggressive cancers. However the true magnitude of the impact of such differences on prostate cancer–specific survival after adequate treatment continues to be debated. Despite lower pretreatment serum testosterone levels, obese men might be more resistant to the testosteronedepleting effects of androgen-suppressing therapy (20, 34). It might also be true that obesity and associated comorbidity reduce the likelihood of initiating salvage treatment, thereby resulting in inferior outcomes after prostate cancer relapse. Treatment of obese patients is therefore more challenging than treatment of normal-weight patients. The present study suggests that obesity is not a contraindication to riskadapted EBRT as long as attempts are made to actually deliver the intended radiation dose to the CTV. Whether this is also true in severely obese patients with BMI >35 kg/m2 requires additional studies in larger groups of patients.
REFERENCES 1. Li H, Stampfer MJ, Mucci L, et al. A 25-year prospective study of plasma adiponectin and leptin concentrations and prostate cancer risk and survival. Clin Chem 2010;56:34–43. 2. Loeb S, Carter HB, Schaeffer EM, et al. Should prostate specific antigen be adjusted for body mass index? Data from the Baltimore Longitudinal Study of Aging. J Urol 2009;182:2646– 2651. 3. Mu¨ller H, Raum E, Rothenbacher D, et al. Association of diabetes and body mass index with levels of prostate-specific antigen: Implications for correction of prostate-specific antigen cutoff values? Cancer Epidemiol Biomarkers Prev 2009;18:1350– 1356. 4. Freedland SJ, Ban˜ez LL, Sun LL, et al. Obese men have highergrade and larger tumors: An analysis of the Duke Prostate Center Database. Prostate Cancer Prostatic Dis 2009;12:259–263.
5. Magheli A, Rais-Bahrami S, Trock BJ, et al. Impact of body mass index on biochemical recurrence rates after radical prostatectomy: An analysis utilizing propensity score matching. Urology 2008;72:1246–1251. 6. Stephenson AJ, Kattan MW, Eastham JA, et al. Prostate cancerspecific mortality after radical prostatectomy for patients treated in the prostate-specific antigen era. J Clin Oncol 2009;27:4300– 4305. 7. van Roermund JG, van Basten JP, Kiemeney LA, et al. Impact of obesity on surgical outcomes following open radical prostatectomy. Urol Int 2009;82:256–261. 8. van Roermund JG, Kok DE, Wildhagen MF, et al. Body mass index as a prognostic marker for biochemical recurrence in Dutch men treated with radical prostatectomy. BJU Int 2009; 104:321–325.
22
I. J. Radiation Oncology d Biology d Physics
9. Motamedinia P, Korets R, Spencer BA, et al. Body mass index trends and role of obesity in predicting outcome after radical prostatectomy. Urology 2008;72:1106–1110. 10. Davies BJ, Smaldone MC, Sadetsky N, et al. The impact of obesity on overall and cancer specific survival in men with prostate cancer. J Urol 2009;182:112–117. 11. Armstrong AJ, Halabi S, de Wit R, et al. The relationship of body mass index and serum testosterone with disease outcomes in men with castration-resistant metastatic prostate cancer. Prostate Cancer Prostatic Dis 2009;12:88–93. 12. Montgomery RB, Goldman B, Tangen CM, et al. Association of body mass index with response and survival in men with metastatic prostate cancer: Southwest Oncology Group trials 8894 and 9916. J Urol 2007;178:1946–1951. 13. Halabi S, Ou SS, Vogelzang NJ, Small EJ. Inverse correlation between body mass index and clinical outcomes in men with advanced castration-recurrent prostate cancer. Cancer 2007;110: 1478–1484. 14. Efstathiou JA, Skowronski RY, Coen JJ, et al. Body mass index and prostate-specific antigen failure following brachytherapy for localized prostate cancer. Int J Radiat Oncol Biol Phys 2008;71:1302–1308. 15. van Roermund JG, Hinnen KA, Battermann JJ, et al. Body mass index is not a prognostic marker for prostate-specific antigen failure and survival in Dutch men treated with brachytherapy. BJU Int 2010;105:42–48. 16. Merrick GS, Galbreath RW, Butler WM, et al. Obesity is not predictive of overall survival following permanent prostate brachytherapy. Am J Clin Oncol 2007;30:588–596. 17. Efstathiou JA, Chen MH, Renshaw AA, et al. Influence of body mass index on prostate-specific antigen failure after androgen suppression and radiation therapy for localized prostate cancer. Cancer 2007;109:1493–1498. 18. Strom SS, Kamat AM, Gruschkus SK, et al. Influence of obesity on biochemical and clinical failure after external-beam radiotherapy for localized prostate cancer. Cancer 2006;107:631– 639. 19. Efstathiou JA, Bae K, Shipley WU, et al. Obesity and mortality in men with locally advanced prostate cancer: Analysis of RTOG 85-31. Cancer 2007;110:2691–2699. 20. Palma D, Pickles T, Tyldesley S. Prostate Cohort Outcomes Initiative. Obesity as a predictor of biochemical recurrence and survival after radiation therapy for prostate cancer. BJU Int 2007;100:315–319. 21. Stroup SP, Cullen J, Auge BK, et al. Effect of obesity on prostate-specific antigen recurrence after radiation therapy for localized prostate cancer as measured by the 2006 Radiation Therapy Oncology Group-American Society for Therapeutic Radiation and Oncology (RTOG-ASTRO) Phoenix consensus definition. Cancer 2007;110:1003–1009. 22. Showalter TN, Lawrence YR, Xu X, et al. The influence of obesity on toxicity and biochemical control after external beam ra-
Volume 81, Number 1, 2011
23.
24. 25. 26.
27.
28.
29.
30.
31.
32.
33.
34.
diation therapy for prostate cancer (Abstract). Int J Radiat Oncol Biol Phys 2009;75:S351–S352. Goldner G, Bombosch V, Geinitz H, et al. Moderate riskadapted dose escalation with three-dimensional conformal radiotherapy of localized prostate cancer from 70 to 74 Gy: First report on 5-year morbidity and biochemical control from a prospective Austrian-German multicenter phase II trial. Strahlenther Onkol 2009;185:94–100. Geinitz H, Zimmermann F, Thamm R, et al. Late rectal symptoms and quality of life after conformal radiation therapy for prostate cancer. Radiother Oncol 2006;79:341–347. Geinitz H, Zimmermann F, Thamm R, et al. 3-D conformal radiation therapy for prostate cancer in elderly patients. Radiother Oncol 2005;76:27–34. Wachter S, Gerstner N, Dorner D, et al. The influence of a rectal balloon tube as internal immobilization device on variations of volumes and dose-volume histograms during treatment course of conformal radiotherapy for prostate cancer. Int J Radiat Oncol Biol Phys 2002;52:91–100. Roach M III, Hanks G, Thames H Jr., et al. Defining biochemical failure following radiotherapy with or without hormonal therapy in men with clinically localized prostate cancer: Recommendations of the RTOG-ASTRO Phoenix Consensus Conference. Int J Radiat Oncol Biol Phys 2006;65:965–974. Viani GA, Stefano EJ, Afonso SL. Higher-than-conventional radiation doses in localized prostate cancer treatment: A metaanalysis of randomized, controlled trials. Int J Radiat Oncol Biol Phys 2009;74:1405–1418. Wong JR, Gao Z, Merrick S, et al. Potential for higher treatment failure in obese patients: Correlation of elevated body mass index and increased daily prostate deviations from the radiation beam isocenters in an analysis of 1,465 computed tomographic images. Int J Radiat Oncol Biol Phys 2009;75:49–55. Millender LE, Aubin M, Pouliot J, et al. Daily electronic portal imaging for morbidly obese men undergoing radiotherapy for localized prostate cancer. Int J Radiat Oncol Biol Phys 2004; 59:6–10. King CR, Spiotto MT, Kapp DS. Obesity and risk of biochemical failure for patients receiving salvage radiotherapy after prostatectomy. Int J Radiat Oncol Biol Phys 2009;73:1017– 1022. Michalski JM, Roach M 3rd, Merrick G, et al. ACR appropriateness criteria on external beam radiation therapy treatment planning for clinically localized prostate cancer. Int J Radiat Oncol Biol Phys 2009;74:667–672. Skolarus TA, Wolin KY, Grubb RL 3rd. The effect of body mass index on PSA levels and the development, screening and treatment of prostate cancer. Nat Clin Pract Urol 2007;4: 605–614. Smith R. Obesity and sex steroids during gonadotropinreleasing hormone agonist treatment for prostate cancer. Clin Cancer Res 2007;13:241–245.