- Email: [email protected]
Poorer outcome in Polynesian patients with prostate cancer treated with definitive conformational radiation therapy
Radiotherapy and Oncology 101 (2011) 502–507
Contents lists available at ScienceDirect
Radiotherapy and Oncology journal homepage: www.thegreenjournal.com
Prostate cancer radiotherapy
Poorer outcome in Polynesian patients with prostate cancer treated with definitive conformational radiation therapy Antonin Levy a, Cyrus Chargari a,b, Gonzague Desrez c, Stéphane Leroux d, Mary Jane Sneyd e, Pierre Mozer f, Eva Comperat g, Loïc Feuvret a, Philippe Lang a, Stéphane Lopez a, Avi Assouline a, Charles Hemery a, Jean Jacques Mazeron a, Jean Marc Simon a,⇑ a
Department of Radiation Oncology, Pitie-Salpetriere University Hospital; b Department of Radiation Oncology, Val-De-Grace Hospital, Paris, France; c Department of Urology, BP 21 491, Papeete, French Polynesia; d Department of Urology, Centre Hospitalier Mamao, Papeete, French Polynesia; e Hugh Adam Cancer Epidemiology Unit, University of Otago, Dunedin, New Zealand; f Department of Urology; and g Academic Pathology Department, Pitie-Salpetriere University Hospital, Paris, France
a r t i c l e
i n f o
Article history: Received 11 February 2011 Received in revised form 19 May 2011 Accepted 26 May 2011 Available online 30 June 2011 Keywords: Prostate cancer Biochemical relapse Ethnicity Polynesia
a b s t r a c t Purpose: To compare freedom from biochemical failure (FFBF) of French Polynesian (FP) and Native European (NE) prostate cancer patients after definitive conformal radiotherapy (RT). Patients and methods: Data were reviewed from medical records of 152 consecutive patients (46 FP and 106 NE) with clinically localised prostate cancer treated with definitive RT. Neoadjuvant androgen deprivation therapy (ADT) was used in 22% of cases. Definition for biochemical failure was a rise by 2 ng/mL or more above the nadir prostate-specific antigen (PSA) level. The median follow-up was 34 months. Results: In comparison to NE patients, FP patients were younger (p = 0.002) with a higher low-risk proportion (p = 0.06). Probability of 5-year FFBF was 77% in the NE cohort and 58.0% in the FP cohort (p = 0.017). Univariate analysis showed that FP ethnicity was associated with worse prognosis in highrisk tumours (p = 0.004). Cox multivariate analysis showed that factors associated with FFBF were risk category (p < 0.017), and FP origin (p = 0.03), independently of ADT and radiation dose. Conclusion: FP ethnicity was an independent prognostic factor for biochemical relapse after definitive conformal RT for prostate cancer. Ó 2011 Elsevier Ireland Ltd. All rights reserved. Radiotherapy and Oncology 101 (2011) 502–507
French Polynesia is a set of 48 islands spread over five archipelagos, about 6000 km east of Australia. The population of French Polynesia was 256,000 in 2007, with three ethnic groups: Polynesians (78%), Europeans (12%), and Asians (10%). Medical coverage is good on the largest islands but there is no Radiation Oncology Department in French Polynesia, and patients requiring radiotherapy (RT) are transferred to France. The French Polynesian health care system is similar to that in France, with complete public health coverage, managed by the Social Welfare Fund. Treatment of cancer is free of charge, and that includes diagnostic tests, surgery, RT, drug treatments, follow-up, and transportation from place of residence to place of treatment. Poorer outcome and advanced disease at diagnosis have already been described for the Ma¯ori male population, part of the Pacific Islands population [1,2]. It is still not clear whether prostate cancers are biologically more virulent in indigenous men or whether the mortality rates simply reflect differential access to early diagnosis (including screening) and treatment. Classical prognostic factors ⇑ Corresponding author. Address: Pitie Salpetriere Hospital, Assistance Publique – Hôpitaux de Paris, Paris VI University, 47-83, Boulevard de l’Hôpital, 75651 Paris cedex 13, France. E-mail address: [email protected] (J.M. Simon). 0167-8140/$ - see front matter Ó 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.radonc.2011.05.075
such as clinical and/or pathological stage, Gleason score and prostate-specific antigen (PSA) levels have been clearly identified for the survival of prostate cancer [3,4]; however, there is scant outcome data for the FP population. The purpose of this study was to compare freedom from biochemical failure (FFBF) of French Polynesian (FP) and Native European (NF) prostate cancer patients after definitive conformal RT. Material and methods Study design and patient populations We conducted a retrospective study of 264 consecutive patients with localised prostate adenocarcinoma, treated in our institution with conformal RT between April 1999 and July 2007. Thirty-six patients (9 FP and 27 NE) were excluded because they received irradiation after radical prostatectomy, and 76 patients (0 FP and 76 NE) were excluded because they received pelvic irradiation (Fig. 1). Therefore, 152 patients were included: 46 were FP and 106 were NE, 14 of whom were living in French Polynesia. All patient medical records were discussed at our institution’s Multidisciplinary Meeting for the Management of Urologic Malignancies, after which, patients were transferred from French
A. Levy et al. / Radiotherapy and Oncology 101 (2011) 502–507
503
guidelines [7]. Acute toxicity and late toxicity were defined as reported toxicity that occurred within, or after 90 days after RT completion, respectively. Response assessment
Fig. 1. Flowchart. Abbreviations: FP, French Polynesian; NE, Native European; RT, Radiotherapy; SV, seminal vesicles.
Polynesia to Paris. After treatment, FP patients returned to Polynesia and were followed by their urologists (GD and SL). Staging All patients had a physical examination, including digital rectal examination, PSA determination, and ultrasound-guided transrectal prostate biopsy with Gleason score histological grading. All FP pathological samples were analysed at the Polynesian medical centre by an experienced pathologist trained for 15 years in France. All patients underwent pelvic computed tomography; those with PSA levels >10 ng/mL underwent a bone scan. Other staging modalities such as magnetic resonance imaging of the prostate or pelvis were performed at the discretion of the attending physician. Diagnostic lymphadenectomy was performed for patients with risk >10% of nodal involvement according to the Partin tables [5]. All patients with histologically proven positive lymph nodes were excluded from our study. Staging was performed in accordance with the 2002 American Joint Committee on Cancer Staging System [6]. Treatment characteristics All patients were treated with three-dimensional conformational external-beam RT. Radiation was delivered in 2.0-Gy daily fractions, using 18-MV photons. Conformal RT delivered 46 Gy with a four-field technique to a planning target volume 1 (PTV1) of a 1-cm margin around the prostate and seminal vesicles in three dimensions, except for the rectal-prostate interface, where a 0.5cm margin was used. Then an additional irradiation of 20 Gy to 34 Gy was delivered to PTV2 that included only the prostate with the same margins as PTV1. No patients received pelvic node irradiation. Twenty-five percent of patients received 70 Gy, 25% received 74 Gy, and 33% received 76 Gy. All patients were given specific instructions before each fraction, i.e., empty the rectum and bladder one hour before CT planning, and before each irradiation session. Positioning quality control was carried out with portal imaging during the three days before the first irradiation session and then weekly thereafter. Thirty-three patients in the intermediate and high-risk categories (22%: 3 FP and 22 NE) received neoadjuvant and/or concurrent androgen deprivation therapy (ADT) (gonadotropin-releasing hormone agonist) for 6–24 months. Toxicity assessment Toxicity was recorded using the National Cancer Institute’s Common Terminology Criteria for Adverse Events, version 3.0
The primary end point was freedom from biochemical failure (FFBF) defined using the RTOG-ASTRO Phoenix Consensus Conference criteria, i.e., a rise by 2 ng/mL or more above the nadir PSA level [8]. Other end points included biochemical progression free survival (BPFS), which included biochemical failure, distant metastasis, salvage ADT, and death from any cause (dates of death were obtained from the city of birth). Because of insufficient data, outcome in terms of distant metastases or overall survival could not be analysed. PSA assays were repeated 6–8 weeks after the completion of RT, and patients returned for follow-up visits, which included a clinical examination and a PSA test every four months the first year and every six months thereafter, until end point (September 2010). All PSA tests were performed in the same laboratory (either in France or in French Polynesia) for any given patient. The median follow-up time was 34 months from the first day of RT to the last measured PSA. Statistical analysis The association between ethnic origin and clinicopathological parameters was analysed using the Chi-square test for categorical variables (Yates’ correction was used when a count was smaller than 5); the independent t test was used for continuous variables. Univariate analyses of FFBF and BPFS were performed using the Kaplan–Meier method and log-rank test. Univariate analysis was performed on ethnic origin, tumour stage, Gleason score, pre-treatment PSA levels, radiation dose, and use of ADT. Patients were also grouped according to prognostic risk categories: low-risk patients had PSA levels 610 ng/mL, Gleason score 66, and AJCC category T1c or T2a disease; intermediate-risk patients had PSA levels of between 10 ng/mL and 20 ng/mL, Gleason score of 7, and/or AJCC category T2b disease; high-risk patients had PSA levels of more than 20 ng/mL, Gleason score of 8–10, and/or AJCC category T2c to T3a disease [3]. The Cox proportional hazards model was used for multivariate analysis to determine prognostic factors for FFBF and BPFS. Multivariate analysis was performed according to risk categories, ethnic origin, and ADT. The year of radiotherapy was included as a covariate to adjust for the possible effect of a learning curve. Statistical analyses were performed using SAS software, version 8.2 (SAS Institute Inc, Cary, NC, USA). For all tests, a two-sided p < 0.05 was considered statistically significant. The review board at our institution approved this study. The study was conducted in accordance with the Helsinki Declaration of 1975, revised in 2000. Written consent was not obtained from the participants because this was a retrospective review of existing patient data. Results Patient characteristics FP men were younger than NE men: median age was 65 (51–73) and 69 (52–82) years, respectively (p < 0.002). FP patients tended to have more favourable tumours than NE men: proportion of patients with a low-risk prognosis disease was 30% and 14%, respectively (p = 0.06). No statistically significant differences in mean PSA values at presentation or Gleason score were observed between the FP and NE cohorts. The most frequent comorbidities were cardiovascular diseases and diabetes, with a similar proportion in the
504
Outcome in Polynesian patients with prostate cancer
two populations. Patient characteristics are summarised by ethnic origin in Table 1.
100
Prrobability of FFBF(%)
90
Treatment characteristics Three (7%) FP and 22 (21%) NE patients received neo- and/or adjuvant ADT (p = 0.053; Table 1). Twenty-three (50%) FP and 42 (40%) NE men had a lymphadenectomy before RT. None of them showed evidence of nodal invasion. Median RT doses were similar in the two groups: 73 Gy (36–76) for FP patients and 74 Gy (66–80) for NE patients (p = 0.9), delivered over a median of 59 days.
80 70 60 50
Median follow-up from the first day of RT was 34 months (range 2–103 months). A total of 24 events for FFBF (11 and 13, respectively, for FP and NE groups) and 28 events for BPFS (13 and 15, respectively, for FP and NE groups), including four deaths, were reported. The 5-year actuarial FFBF and BPFS for the entire population were 71.6%, and 67.1%, respectively. By risk category, 5-year FFBF was 92.3% for low-risk patients, 84.4% for intermediate-risk patients, and 53.5% for high-risk patients (p = 0.007).
French Polynesian
30
p 0 017 p=0.017
20 10 0
Number at risk Native European French Polynesian
Treatment outcome
Native European
40
0
1
2
3
4
5
106 46
92 38
71 28
45 14
28 10
15 5
Time (Years)
Fig. 2. Freedom from biochemical failure (FFBF) in French Polynesian cohort and Native European cohort I bars indicate 95% confidence intervals.
Five-year BPFS was 77.5%, 73.4%, and 53.4% for low-, intermediate-, and high-risk patients (p = 0.046), respectively. The probability of 5-year FFBF was 58% for FP patients versus 77% for NE patients (p = 0.017, Fig. 2). Fig. 2 shows earlier biochemical
Table 1 Patient characteristics according to ethnic origin. Native European No. of Patients (%)
French Polynesian No. of Patients (%)
All
p
Number of patients Age (years) Median Range
106 (70)
46 (30)
152 (100)
69 52–82
65 51–73
68 51–73
0.002
Tumour stage T1c T2a T2b–T2c T3a–T3b
28 26 27 25
(26) (25) (25) (24)
17 (37) 14 (30) 6 (13) 9 (20)
45 40 33 34
0.25
Gleason score <6 7 >8
47 (44) 55 (52) 4 (4)
22 (48) 23 (50) 1 (2)
69 (45) 78 (51) 5 (4)
0.8
Pre-treatment PSA concentration (ng/mL) Median Range
12.0 1.3–93.5
10.1 4.6–60.0
11.5 1.3–93.5
0.7
PSA level (ng/mL) 0–10 >10–20 >20
46 (43) 31 (29) 29 (28)
24 (52) 10 (22) 12 (26)
70 (46) 41 (27) 41 (27)
0.6
Prognostic risk category Low-risk Intermediate-risk High-risk
15 (14) 49 (46) 42 (40)
14 (30) 16 (35) 16 (35)
29 (19) 65 (43) 58 (38)
0.06
Method of diagnosis Biopsy TURP
102 (96) 4 (4)
44 (96) 2 (4)
146 (96) 6 (4)
0.8
Pelvic lymphadenectomy No Yes
64 (60) 42 (40)
23 (50) 23 (50)
87 (57) 65 (43)
0.2
Comorbidities CV disease Diabetes
20 (19) 7 (7)
8 (17) 3 (7)
28 (18) 10 (7)
0.8 1.0
Adjuvant androgen deprivation No Yes
84 (79) 22 (21)
43 (93) 3 (7)
127 (84) 25 (16)
0.053
Radiation dose (Gy) <74 P74
37 (35) 69 (65)
16 (35) 30 (65)
53 (35) 99 (65)
0.9
TURP: Transurethral resection of the prostate. CV: cardiovascular.
(30) (26) (22) (22)
505
A. Levy et al. / Radiotherapy and Oncology 101 (2011) 502–507 Table 2 Univariate analysis for factors associated with FFBF and BPFS, according to ethnic origin. Variables
5-Year FFBF (%)
5-Year BPFS (%)
NE
FP
p
NE
FP
p
Overall
77
58
0.017
72
56
0.03
Tumour stage T1c–T2a T2b–T3
79 75
71 38
0.5 0.004
75 67
68 38
0.4 0.002
Gleason score 66 P7
80 77
74 45
0.6 0.003
80 64
70 45
0.4 0.011
Pre-treatment PSA concentration (ng/mL) <20 85 P20 48
69 27
0.2 0.014
77 48
65 27
0.4 0.014
Prognostic risk category Low-risk Intermediate-risk High-risk
67 72 40
0.8 0.2 0.004
86 73 61
61 72 40
0.4 0.6 0.004
86 88 61
Abbreviations: FFBF, freedom from biochemical failure; BPFS, biochemical progression free survival.
relapses for FP patients, occurring during the first two years of follow-up after treatment. The probability of 5-year BPFS was 57.9% for FP patients versus 71.5% for NE patients (p = 0.033). Age at presentation, use of ADT, and radiation dose were not predictive factors for either FFBF or BPFS. Table 2 shows univariate analyses of probability of 5-year FFBF and probability of 5-year BPFS according to ethnic origin. Compared to NE patients, probabilities of 5-year FFBF were lower for FP patients with T2b-T3 stage (p = 0.004), or with a Gleason score of 7 or more (p = 0.003), or with a PSA level at presentation of more than 20 (p = 0.014). When classified in three prognostic risk category classes, and compared to NE patients, FP ethnicity was correlated with a significantly lower FFBF in the high-risk group (p = 0.004), whereas there were no significant differences between FP and NE patients in the low- and intermediate-risk groups. Similar results were observed for BPFS (Table 2). For multivariate analysis, factors associated with decreased FFBF were risk category (high-risk, p < 0.017), and FP origin (p = 0.03), adjusting for year of radiotherapy (Table 3). Neither age at presentation nor ADT nor radiation doses were independent prognostic factors. Factors associated with decreased BPFS for multivariate analysis included risk category (high-risk, p < 0.013), and FP origin (p = 0.016; Table 3). Toxicity No differences in acute and late genitourinary or in gastrointestinal toxicities were observed between the two populations. Two Table 3 Multivariate analysis for factors associated with FFBF and BPFS. Variable
FFBF
BPFS
HR
95%CI
p
HR
95%CI
p
Prognostic risk category Low-risk Intermediate-risk High-risk Year of Radiotherapy
1.0 2.0 6.3 0.3
0.4–9.9 1.4–28.2 0.1–0.8
0.38 0.017 0.01
1.0 2.0 4.2 0.4
0.5–7.4 1.2–14.9 0.2–0.8
0.3 0.027 0.018
Androgen deprivation No Yes
1.0 0.9
0.2–4.1
0.8
1.0 1.1
0.3–3.9
0.9
Ethnic origin Native European French Polynesian
1.0 2.5
1.1–5.7
0.03
1.0 2.2
1.1–4.8
0.04
Abbreviations: FFBF, freedom from biochemical failure; BPFS, biochemical progression-free survival.
NE patients experienced Grade 3 or 4 genitourinary toxicity. Five patients (1 FP and 4 NE) experienced late Grade 3 gastrointestinal toxicity, and required argon plasma coagulation. No Grade 4 gastrointestinal toxicity was observed. Discussion We retrospectively compared two cohorts of patients (FP and NE) with localised prostate adenocarcinoma, treated in the same department with the same 3D conformal RT technique. The probability of 5-year FFBF was 58% for the FP patients versus 77% for the NE patients (p = 0.017). After adjustment for known prognostic factors and adjuvant ADT, FP patients had a 2.5-fold lower probability of FFBF than NE patients (Table 3). Various studies have shown differences in survival between Pacific and Caucasian patients. A study on Native Hawaiian health compared data collected in the year 2000 with data from 1982 and 1990. The overall life expectancy of Native Hawaiians was shorter than other ethnic groups residing in Hawaii. Native Hawaiians had a higher prevalence of hypertension, diabetes, and asthma than other ethnic groups, and had higher rates of smoking, excessive alcohol consumption, and being overweight [9]. In Hawaii, survival rates of Hawaiian patients with breast, lung, and colon cancer were worse than those of European patients living in Hawaii [10–12]. In New Zealand, prostate cancer is the second most common cancer among Ma¯ori men and the second leading cause of cancer death [13]. From 1996 to 2001, the mortality/incidence ratio was 29% among Ma¯ori and 15% among non-Ma¯ori men. Ma¯ori men were 16% less likely than non-Ma¯ori men to be diagnosed with prostate cancer, but 60% more likely to die from it. Ma¯ori men were significantly more likely to be diagnosed at a later stage of disease spread than non-Ma¯ori men. Once diagnosed with prostate cancer, Ma¯ori men were more than twice as likely as non-Ma¯ori men to die from their cancer (HR = 2.3, 95% CI [1.9–2.8], p < 0.0001). Authors estimated that more than half of this survival disparity could be attributed to differences in stage at diagnosis. The socioeconomic status of the Ma¯ori and Pacific People (low income, low employment rates, and low educational achievement) probably also has an effect on mortality [14–16]. In addition to socioeconomic factors and lifestyle differences, biological factors may contribute to this discrepancy [17]. For example, androgen and androgen receptor pathways have long been associated with prostate growth, and ethnic differences have been found among variants of the genes of the enzymes involved in androgen biosynthesis and metabolism [18,19]. Ethnic differences have also
506
Outcome in Polynesian patients with prostate cancer
been found among levels of expression and CAG repeat length of the androgen receptor, and critical molecular alterations may occur, resulting in racial disparity [20]. Interethnic differences have been observed in growth factors and their receptors, such as the epidermal growth factor receptor (EGFR) [21–23], which promote cancer cell growth, and this may also explain some of the disparity. Nevertheless, there is considerable controversy concerning the existence of genetic differences among races. As Pearce et al. stated there is much confusion between the concepts of race, ethnicity, and genetics [24]. To date, no significant genetic differences have been found between races, especially with regard to genes that may affect health. However, ethnic differences in health are strongly determined by historical, cultural, and socio-economic factors, and especially by differences in access to health care. Low access to health care could cause delays in diagnosis and affect stages of disease at diagnosis. Age and associated comorbidities could influence treatment choices. All these issues could play a role in the survival differences observed. These differences can disappear when there is equity in access to health care, as evidenced in the study carried out by the U.S. Department of Defense [25,26]. While some authors have reported poorer outcomes for lowrisk patients, our findings indicate poorer outcomes with high-risk patients. Sneyd et al. recently observed that higher mortality rates from prostate cancer for Ma¯ori and Pacific men were due to a higher risk of dying from prostate cancer during the first year of followup after treatment [27]. Compared to Europeans, Ma¯ori men had a significantly increased risk of dying from low-grade prostate cancer (HR = 1.6, 95%CI [1.2–2.1]), and Pacific men had a significantly increased risk of dying from low-grade prostate cancer (HR = 2.8, 95%CI [1.1–6.8]). Observations made in our series are consistent with these reports: a greater proportion of FP patients were more likely to have low-risk prostate cancer (Table 1). We also observed that the biochemical relapses of FP patients appeared earlier, occurring during the first two years of follow-up after treatment, whereas those of NE patients occurred after the third year (Fig. 2). This information could be relevant for tumour aggressiveness [28,29]. Indeed, Pound and al. observed in a population of patients with prostate carcinoma, that the time to development of distant metastases was shorter for patients with a PSA elevation in less than 2 years following surgery, compared with those with a PSA elevation in more than 2 years. In our study, FP patients had biochemical relapses within the first two years after RT. Although patients did not have prostate biopsies at the time of early rising PSA, these biochemical relapses could be explained by local recurrences. Other biological explanations of tumour aggressiveness are possible, and one of these could be that of the RT procedure. As we did not use cone beam CT for daily patient positioning verification during RT, significant prostate displacement could have occurred between treatments. Although an empty rectum was required for the CT planning, rectal distension during RT courses could cause geographic misses and thus decrease the probability of biochemical control [30]. We provided specific instructions at the first consultation, such as emptying the rectum and bladder one hour before CT planning, and before each session of irradiation. Perhaps these instructions were not carefully followed in the low-literacy populations because they could not understand many of the words that are used in clinical and investigative settings [31]. Many of these words are used in the long-established written protocols given to patients. We must rigorously investigate whether or not the words used in our procedures are actually understood by low-literacy FP patients. This study had some limitations because it was retrospective. FP men were younger than NE men, and our FP study population was a sample selected from the entire FP population with prostate cancer. We hypothesise that younger patients were more likely to be detected than older patients, and that some patients with diag-
nosed prostate carcinoma dropped out because of the long journey from a remote island in the middle of the Pacific Ocean to France to receive RT. Nonetheless, age at presentation was not found to be an independent prognostic factor for outcome. In recent large trials less aggressive cases of prostate cancer were reported to be in older men (i.e., >70 years), independent of other clinical features [32]. In our study, FP men were younger than NE men, median age was 65 and 69 years respectively (p < 0.002). Nevertheless median age <70 years in both groups may partially explain the absence of significance of this well-established prognostic factor. In our series, surprisingly, FP patients had a low rate of comorbidities (Table 1). It is possible that FP patients with serious illnesses were not screened, or were not selected for the trip. This kind of selection bias may have affected results of the good-prognosis group of FP patients. Another limitation was that the Gleason scores were not reviewed centrally. However, an experienced pathologist trained in France for 15 years analysed all FP pathological samples. Furthermore, there was a borderline significant difference in use of ADT between the two ethnic groups: the NE group received more ADT. Nevertheless both groups were statically comparable in terms of ADT, so that we believe that the difference observed in terms of outcome cannot be attributed to simply differences between the two groups. However, our data are retrospective and there was lack of information regarding the exact timing of treatment for the 25 patients who received ADT. Consequently, it cannot be excluded that a marginal difference between both groups in terms of endocrine therapy delivery could have partially contributed to the difference in outcome. These limitations highlight that only well-conducted prospective analyses can accurately investigate the true impact of ethnicity in prostate cancer. Conclusion In this retrospective study, we observed that FP ethnicity was an independent prognostic factor of FFBF and BPFS after conformal radiotherapy for prostate adenocarcinoma. These findings support the need for further investigation of factors that compromise the RT efficacy for FP patients with prostate cancer. We must be trained on how best to communicate with low-literacy FP patients. Also, the confirmation of these results could provide further evidence for proposing more aggressive treatment with ADT in association with RT for FP patients with high-risk prostate cancer. Conflict of interest statement None declared. Acknowledgement Authors thank Lois Rose for editing. References [1] Chuba PJ, Moughan J, Forman JD, Owen J, Hanks G. The 1989 patterns of care study for prostate cancer: five-year outcomes. Int J Radiat Oncol Biol Phys 2001;50:325–34. [2] Goggins WB, Wong GK. Poor survival for US Pacific Islander cancer patients: evidence from the surveillance, epidemiology, and end results database: 1991 to 2004. J Clin Oncol 2007;25:5738–41. [3] D’Amico AV, Whittington R, Malkowicz SB, et al. Biochemical outcome after radical prostatectomy, external beam radiation therapy, or interstitial radiation therapy for clinically localized prostate cancer. JAMA 1998;280:969–74. [4] Walsh PC, DeWeese TL, Eisenberger MA. Localized prostate cancer. N Engl J Med 2007;357:2696–705. [5] Partin AW, Kattan MW, Subong EN, et al. Combination of prostate-specific antigen, clinical stage, and Gleason score to predict pathological stage of localized prostate cancer. A multi-institutional update. JAMA 1997;277:1445–51.
A. Levy et al. / Radiotherapy and Oncology 101 (2011) 502–507 [6] Page DL, Fleming ID, Fritz A, Balch CM, Haller DG, Morrow M. American joint committee on cancer national cancer institute (US). In: Greene FL, editor. AJCC cancer staging manual. New York: Springer; 2002. p. 309–13. [7] Trotti A, Colevas AD, Setser A, et al. CTCAE v3.0: development of a comprehensive grading system for the adverse effects of cancer treatment. Semin Radiat Oncol 2003;13:176–81. [8] Roach 3rd M, Hanks G, Thames Jr H, 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;65:965–74. [9] Johnson DB, Oyama N, LeMarchand L, Wilkens L. Native Hawaiians mortality, morbidity, and lifestyle: comparing data from 1982, 1990, and 2000. Pac Health Dialog 2004;11:120–30. [10] Meng L, Maskarinec G, Wilkens L. Ethnic differences and factors related to breast cancer survival in Hawaii. Int J Epidemiol 1997;26:1151–8. [11] Nomura A, Kolonel L, Rellahan W, Lee J, Wegner E. Racial survival patterns for lung cancer in Hawaii. Cancer 1981;48:1265–71. [12] Wegner EL, Kolonel LN, Nomura AM, Lee J. Racial and socioeconomic status differences in survival of colorectal cancer patients in Hawaii. Cancer 1982;49:2208–16. [13] Robson B, Purdie G, Cormack D. Unequal Impact: Ma¯ori and non-Ma¯ori cancer statistics 1996–2001. In: Health Mo, editor, Wellington; 2006. [14] Blakely T, Fawcett J, Hunt D, Wilson N. What is the contribution of smoking and socioeconomic position to ethnic inequalities in mortality in New Zealand? Lancet 2006;368:44–52. [15] Dachs GU, Currie MJ, McKenzie F, et al. Cancer disparities in indigenous Polynesian populations: Maori, Native Hawaiians, and Pacific people. Lancet Oncol 2008;9:473–84. [16] Jeffreys M, Stevanovic V, Tobias M, et al. Ethnic inequalities in cancer survival in New Zealand: linkage study. Am J Public Health 2005;95:834–7. [17] Hatcher D, Daniels G, Osman I, Lee P. Molecular mechanisms involving prostate cancer racial disparity. Am J Transl Res 2009;1:235–48. [18] Ntais C, Polycarpou A, Ioannidis JP. Association of the CYP17 gene polymorphism with the risk of prostate cancer: a meta-analysis. Cancer Epidemiol Biomarkers Prev 2003;12:120–6. [19] Reichardt JK, Makridakis N, Henderson BE, Yu MC, Pike MC, Ross RK. Genetic variability of the human SRD5A2 gene: implications for prostate cancer risk. Cancer Res 1995;55:3973–5.
507
[20] Deka R, Guangyun S, Wiest J, et al. Patterns of instability of expanded CAG repeats at the ERDA1 locus in general populations. Am J Hum Genet 1999;65:192–8. [21] Di Lorenzo G, Tortora G, D’Armiento FP, et al. Expression of epidermal growth factor receptor correlates with disease relapse and progression to androgenindependence in human prostate cancer. Clin Cancer Res 2002;8:3438–44. [22] Douglas DA, Zhong H, Ro JY, et al. Novel mutations of epidermal growth factor receptor in localized prostate cancer. Front Biosci 2006;11:2518–25. [23] Shuch B, Mikhail M, Satagopan J, et al. Racial disparity of epidermal growth factor receptor expression in prostate cancer. J Clin Oncol 2004;22:4725–9. [24] Pearce N, Foliaki S, Sporle A, Cunningham C. Genetics, race, ethnicity, and health. BMJ 2004;328:1070–2. [25] Johnstone PA, Kane CJ, Sun L, et al. Effect of race on biochemical disease-free outcome in patients with prostate cancer treated with definitive radiation therapy in an equal-access health care system: radiation oncology report of the department of defense center for prostate disease research. Radiology 2002;225:420–6. [26] Optenberg SA, Thompson IM, Friedrichs P, Wojcik B, Stein CR, Kramer B. Race, treatment, and long-term survival from prostate cancer in an equal-access medical care delivery system. JAMA 1995;274:1599–605. [27] Sneyd MJ. Ethnic differences in prostate cancer survival in New Zealand: a national study. Cancer Causes Control 2008;19:993–9. [28] Pound CR, Partin AW, Eisenberger MA, Chan DW, Pearson JD, Walsh PC. Natural history of progression after PSA elevation following radical prostatectomy. JAMA 1999;281:1642–5. [29] Simmons MN, Stephenson AJ, Klein EA. Natural history of biochemical recurrence after radical prostatectomy: risk assessment for secondary therapy. Eur Urol 2007;51:1175–84. [30] de Crevoisier R, Tucker SL, Dong L, et al. Increased risk of biochemical and local failure in patients with distended rectum on the planning CT for prostate cancer radiotherapy. Int J Radiat Oncol Biol Phys 2005;62:965–73. [31] Kilbridge KL, Fraser G, Krahn M, et al. Lack of comprehension of common prostate cancer terms in an underserved population. J Clin Oncol 2009;27:2015–21. [32] Hamstra DA, Bae K, Pilepich MV, et al. Older age predicts decreased metastasis and prostate cancer-specific death for men treated with radiation therapy: meta-analysis of radiation therapy oncology group trials. Int J Radiat Oncol Biol Phys 2011.
Contents lists available at ScienceDirect
Radiotherapy and Oncology journal homepage: www.thegreenjournal.com
Prostate cancer radiotherapy
Poorer outcome in Polynesian patients with prostate cancer treated with definitive conformational radiation therapy Antonin Levy a, Cyrus Chargari a,b, Gonzague Desrez c, Stéphane Leroux d, Mary Jane Sneyd e, Pierre Mozer f, Eva Comperat g, Loïc Feuvret a, Philippe Lang a, Stéphane Lopez a, Avi Assouline a, Charles Hemery a, Jean Jacques Mazeron a, Jean Marc Simon a,⇑ a
Department of Radiation Oncology, Pitie-Salpetriere University Hospital; b Department of Radiation Oncology, Val-De-Grace Hospital, Paris, France; c Department of Urology, BP 21 491, Papeete, French Polynesia; d Department of Urology, Centre Hospitalier Mamao, Papeete, French Polynesia; e Hugh Adam Cancer Epidemiology Unit, University of Otago, Dunedin, New Zealand; f Department of Urology; and g Academic Pathology Department, Pitie-Salpetriere University Hospital, Paris, France
a r t i c l e
i n f o
Article history: Received 11 February 2011 Received in revised form 19 May 2011 Accepted 26 May 2011 Available online 30 June 2011 Keywords: Prostate cancer Biochemical relapse Ethnicity Polynesia
a b s t r a c t Purpose: To compare freedom from biochemical failure (FFBF) of French Polynesian (FP) and Native European (NE) prostate cancer patients after definitive conformal radiotherapy (RT). Patients and methods: Data were reviewed from medical records of 152 consecutive patients (46 FP and 106 NE) with clinically localised prostate cancer treated with definitive RT. Neoadjuvant androgen deprivation therapy (ADT) was used in 22% of cases. Definition for biochemical failure was a rise by 2 ng/mL or more above the nadir prostate-specific antigen (PSA) level. The median follow-up was 34 months. Results: In comparison to NE patients, FP patients were younger (p = 0.002) with a higher low-risk proportion (p = 0.06). Probability of 5-year FFBF was 77% in the NE cohort and 58.0% in the FP cohort (p = 0.017). Univariate analysis showed that FP ethnicity was associated with worse prognosis in highrisk tumours (p = 0.004). Cox multivariate analysis showed that factors associated with FFBF were risk category (p < 0.017), and FP origin (p = 0.03), independently of ADT and radiation dose. Conclusion: FP ethnicity was an independent prognostic factor for biochemical relapse after definitive conformal RT for prostate cancer. Ó 2011 Elsevier Ireland Ltd. All rights reserved. Radiotherapy and Oncology 101 (2011) 502–507
French Polynesia is a set of 48 islands spread over five archipelagos, about 6000 km east of Australia. The population of French Polynesia was 256,000 in 2007, with three ethnic groups: Polynesians (78%), Europeans (12%), and Asians (10%). Medical coverage is good on the largest islands but there is no Radiation Oncology Department in French Polynesia, and patients requiring radiotherapy (RT) are transferred to France. The French Polynesian health care system is similar to that in France, with complete public health coverage, managed by the Social Welfare Fund. Treatment of cancer is free of charge, and that includes diagnostic tests, surgery, RT, drug treatments, follow-up, and transportation from place of residence to place of treatment. Poorer outcome and advanced disease at diagnosis have already been described for the Ma¯ori male population, part of the Pacific Islands population [1,2]. It is still not clear whether prostate cancers are biologically more virulent in indigenous men or whether the mortality rates simply reflect differential access to early diagnosis (including screening) and treatment. Classical prognostic factors ⇑ Corresponding author. Address: Pitie Salpetriere Hospital, Assistance Publique – Hôpitaux de Paris, Paris VI University, 47-83, Boulevard de l’Hôpital, 75651 Paris cedex 13, France. E-mail address: [email protected] (J.M. Simon). 0167-8140/$ - see front matter Ó 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.radonc.2011.05.075
such as clinical and/or pathological stage, Gleason score and prostate-specific antigen (PSA) levels have been clearly identified for the survival of prostate cancer [3,4]; however, there is scant outcome data for the FP population. The purpose of this study was to compare freedom from biochemical failure (FFBF) of French Polynesian (FP) and Native European (NF) prostate cancer patients after definitive conformal RT. Material and methods Study design and patient populations We conducted a retrospective study of 264 consecutive patients with localised prostate adenocarcinoma, treated in our institution with conformal RT between April 1999 and July 2007. Thirty-six patients (9 FP and 27 NE) were excluded because they received irradiation after radical prostatectomy, and 76 patients (0 FP and 76 NE) were excluded because they received pelvic irradiation (Fig. 1). Therefore, 152 patients were included: 46 were FP and 106 were NE, 14 of whom were living in French Polynesia. All patient medical records were discussed at our institution’s Multidisciplinary Meeting for the Management of Urologic Malignancies, after which, patients were transferred from French
A. Levy et al. / Radiotherapy and Oncology 101 (2011) 502–507
503
guidelines [7]. Acute toxicity and late toxicity were defined as reported toxicity that occurred within, or after 90 days after RT completion, respectively. Response assessment
Fig. 1. Flowchart. Abbreviations: FP, French Polynesian; NE, Native European; RT, Radiotherapy; SV, seminal vesicles.
Polynesia to Paris. After treatment, FP patients returned to Polynesia and were followed by their urologists (GD and SL). Staging All patients had a physical examination, including digital rectal examination, PSA determination, and ultrasound-guided transrectal prostate biopsy with Gleason score histological grading. All FP pathological samples were analysed at the Polynesian medical centre by an experienced pathologist trained for 15 years in France. All patients underwent pelvic computed tomography; those with PSA levels >10 ng/mL underwent a bone scan. Other staging modalities such as magnetic resonance imaging of the prostate or pelvis were performed at the discretion of the attending physician. Diagnostic lymphadenectomy was performed for patients with risk >10% of nodal involvement according to the Partin tables [5]. All patients with histologically proven positive lymph nodes were excluded from our study. Staging was performed in accordance with the 2002 American Joint Committee on Cancer Staging System [6]. Treatment characteristics All patients were treated with three-dimensional conformational external-beam RT. Radiation was delivered in 2.0-Gy daily fractions, using 18-MV photons. Conformal RT delivered 46 Gy with a four-field technique to a planning target volume 1 (PTV1) of a 1-cm margin around the prostate and seminal vesicles in three dimensions, except for the rectal-prostate interface, where a 0.5cm margin was used. Then an additional irradiation of 20 Gy to 34 Gy was delivered to PTV2 that included only the prostate with the same margins as PTV1. No patients received pelvic node irradiation. Twenty-five percent of patients received 70 Gy, 25% received 74 Gy, and 33% received 76 Gy. All patients were given specific instructions before each fraction, i.e., empty the rectum and bladder one hour before CT planning, and before each irradiation session. Positioning quality control was carried out with portal imaging during the three days before the first irradiation session and then weekly thereafter. Thirty-three patients in the intermediate and high-risk categories (22%: 3 FP and 22 NE) received neoadjuvant and/or concurrent androgen deprivation therapy (ADT) (gonadotropin-releasing hormone agonist) for 6–24 months. Toxicity assessment Toxicity was recorded using the National Cancer Institute’s Common Terminology Criteria for Adverse Events, version 3.0
The primary end point was freedom from biochemical failure (FFBF) defined using the RTOG-ASTRO Phoenix Consensus Conference criteria, i.e., a rise by 2 ng/mL or more above the nadir PSA level [8]. Other end points included biochemical progression free survival (BPFS), which included biochemical failure, distant metastasis, salvage ADT, and death from any cause (dates of death were obtained from the city of birth). Because of insufficient data, outcome in terms of distant metastases or overall survival could not be analysed. PSA assays were repeated 6–8 weeks after the completion of RT, and patients returned for follow-up visits, which included a clinical examination and a PSA test every four months the first year and every six months thereafter, until end point (September 2010). All PSA tests were performed in the same laboratory (either in France or in French Polynesia) for any given patient. The median follow-up time was 34 months from the first day of RT to the last measured PSA. Statistical analysis The association between ethnic origin and clinicopathological parameters was analysed using the Chi-square test for categorical variables (Yates’ correction was used when a count was smaller than 5); the independent t test was used for continuous variables. Univariate analyses of FFBF and BPFS were performed using the Kaplan–Meier method and log-rank test. Univariate analysis was performed on ethnic origin, tumour stage, Gleason score, pre-treatment PSA levels, radiation dose, and use of ADT. Patients were also grouped according to prognostic risk categories: low-risk patients had PSA levels 610 ng/mL, Gleason score 66, and AJCC category T1c or T2a disease; intermediate-risk patients had PSA levels of between 10 ng/mL and 20 ng/mL, Gleason score of 7, and/or AJCC category T2b disease; high-risk patients had PSA levels of more than 20 ng/mL, Gleason score of 8–10, and/or AJCC category T2c to T3a disease [3]. The Cox proportional hazards model was used for multivariate analysis to determine prognostic factors for FFBF and BPFS. Multivariate analysis was performed according to risk categories, ethnic origin, and ADT. The year of radiotherapy was included as a covariate to adjust for the possible effect of a learning curve. Statistical analyses were performed using SAS software, version 8.2 (SAS Institute Inc, Cary, NC, USA). For all tests, a two-sided p < 0.05 was considered statistically significant. The review board at our institution approved this study. The study was conducted in accordance with the Helsinki Declaration of 1975, revised in 2000. Written consent was not obtained from the participants because this was a retrospective review of existing patient data. Results Patient characteristics FP men were younger than NE men: median age was 65 (51–73) and 69 (52–82) years, respectively (p < 0.002). FP patients tended to have more favourable tumours than NE men: proportion of patients with a low-risk prognosis disease was 30% and 14%, respectively (p = 0.06). No statistically significant differences in mean PSA values at presentation or Gleason score were observed between the FP and NE cohorts. The most frequent comorbidities were cardiovascular diseases and diabetes, with a similar proportion in the
504
Outcome in Polynesian patients with prostate cancer
two populations. Patient characteristics are summarised by ethnic origin in Table 1.
100
Prrobability of FFBF(%)
90
Treatment characteristics Three (7%) FP and 22 (21%) NE patients received neo- and/or adjuvant ADT (p = 0.053; Table 1). Twenty-three (50%) FP and 42 (40%) NE men had a lymphadenectomy before RT. None of them showed evidence of nodal invasion. Median RT doses were similar in the two groups: 73 Gy (36–76) for FP patients and 74 Gy (66–80) for NE patients (p = 0.9), delivered over a median of 59 days.
80 70 60 50
Median follow-up from the first day of RT was 34 months (range 2–103 months). A total of 24 events for FFBF (11 and 13, respectively, for FP and NE groups) and 28 events for BPFS (13 and 15, respectively, for FP and NE groups), including four deaths, were reported. The 5-year actuarial FFBF and BPFS for the entire population were 71.6%, and 67.1%, respectively. By risk category, 5-year FFBF was 92.3% for low-risk patients, 84.4% for intermediate-risk patients, and 53.5% for high-risk patients (p = 0.007).
French Polynesian
30
p 0 017 p=0.017
20 10 0
Number at risk Native European French Polynesian
Treatment outcome
Native European
40
0
1
2
3
4
5
106 46
92 38
71 28
45 14
28 10
15 5
Time (Years)
Fig. 2. Freedom from biochemical failure (FFBF) in French Polynesian cohort and Native European cohort I bars indicate 95% confidence intervals.
Five-year BPFS was 77.5%, 73.4%, and 53.4% for low-, intermediate-, and high-risk patients (p = 0.046), respectively. The probability of 5-year FFBF was 58% for FP patients versus 77% for NE patients (p = 0.017, Fig. 2). Fig. 2 shows earlier biochemical
Table 1 Patient characteristics according to ethnic origin. Native European No. of Patients (%)
French Polynesian No. of Patients (%)
All
p
Number of patients Age (years) Median Range
106 (70)
46 (30)
152 (100)
69 52–82
65 51–73
68 51–73
0.002
Tumour stage T1c T2a T2b–T2c T3a–T3b
28 26 27 25
(26) (25) (25) (24)
17 (37) 14 (30) 6 (13) 9 (20)
45 40 33 34
0.25
Gleason score <6 7 >8
47 (44) 55 (52) 4 (4)
22 (48) 23 (50) 1 (2)
69 (45) 78 (51) 5 (4)
0.8
Pre-treatment PSA concentration (ng/mL) Median Range
12.0 1.3–93.5
10.1 4.6–60.0
11.5 1.3–93.5
0.7
PSA level (ng/mL) 0–10 >10–20 >20
46 (43) 31 (29) 29 (28)
24 (52) 10 (22) 12 (26)
70 (46) 41 (27) 41 (27)
0.6
Prognostic risk category Low-risk Intermediate-risk High-risk
15 (14) 49 (46) 42 (40)
14 (30) 16 (35) 16 (35)
29 (19) 65 (43) 58 (38)
0.06
Method of diagnosis Biopsy TURP
102 (96) 4 (4)
44 (96) 2 (4)
146 (96) 6 (4)
0.8
Pelvic lymphadenectomy No Yes
64 (60) 42 (40)
23 (50) 23 (50)
87 (57) 65 (43)
0.2
Comorbidities CV disease Diabetes
20 (19) 7 (7)
8 (17) 3 (7)
28 (18) 10 (7)
0.8 1.0
Adjuvant androgen deprivation No Yes
84 (79) 22 (21)
43 (93) 3 (7)
127 (84) 25 (16)
0.053
Radiation dose (Gy) <74 P74
37 (35) 69 (65)
16 (35) 30 (65)
53 (35) 99 (65)
0.9
TURP: Transurethral resection of the prostate. CV: cardiovascular.
(30) (26) (22) (22)
505
A. Levy et al. / Radiotherapy and Oncology 101 (2011) 502–507 Table 2 Univariate analysis for factors associated with FFBF and BPFS, according to ethnic origin. Variables
5-Year FFBF (%)
5-Year BPFS (%)
NE
FP
p
NE
FP
p
Overall
77
58
0.017
72
56
0.03
Tumour stage T1c–T2a T2b–T3
79 75
71 38
0.5 0.004
75 67
68 38
0.4 0.002
Gleason score 66 P7
80 77
74 45
0.6 0.003
80 64
70 45
0.4 0.011
Pre-treatment PSA concentration (ng/mL) <20 85 P20 48
69 27
0.2 0.014
77 48
65 27
0.4 0.014
Prognostic risk category Low-risk Intermediate-risk High-risk
67 72 40
0.8 0.2 0.004
86 73 61
61 72 40
0.4 0.6 0.004
86 88 61
Abbreviations: FFBF, freedom from biochemical failure; BPFS, biochemical progression free survival.
relapses for FP patients, occurring during the first two years of follow-up after treatment. The probability of 5-year BPFS was 57.9% for FP patients versus 71.5% for NE patients (p = 0.033). Age at presentation, use of ADT, and radiation dose were not predictive factors for either FFBF or BPFS. Table 2 shows univariate analyses of probability of 5-year FFBF and probability of 5-year BPFS according to ethnic origin. Compared to NE patients, probabilities of 5-year FFBF were lower for FP patients with T2b-T3 stage (p = 0.004), or with a Gleason score of 7 or more (p = 0.003), or with a PSA level at presentation of more than 20 (p = 0.014). When classified in three prognostic risk category classes, and compared to NE patients, FP ethnicity was correlated with a significantly lower FFBF in the high-risk group (p = 0.004), whereas there were no significant differences between FP and NE patients in the low- and intermediate-risk groups. Similar results were observed for BPFS (Table 2). For multivariate analysis, factors associated with decreased FFBF were risk category (high-risk, p < 0.017), and FP origin (p = 0.03), adjusting for year of radiotherapy (Table 3). Neither age at presentation nor ADT nor radiation doses were independent prognostic factors. Factors associated with decreased BPFS for multivariate analysis included risk category (high-risk, p < 0.013), and FP origin (p = 0.016; Table 3). Toxicity No differences in acute and late genitourinary or in gastrointestinal toxicities were observed between the two populations. Two Table 3 Multivariate analysis for factors associated with FFBF and BPFS. Variable
FFBF
BPFS
HR
95%CI
p
HR
95%CI
p
Prognostic risk category Low-risk Intermediate-risk High-risk Year of Radiotherapy
1.0 2.0 6.3 0.3
0.4–9.9 1.4–28.2 0.1–0.8
0.38 0.017 0.01
1.0 2.0 4.2 0.4
0.5–7.4 1.2–14.9 0.2–0.8
0.3 0.027 0.018
Androgen deprivation No Yes
1.0 0.9
0.2–4.1
0.8
1.0 1.1
0.3–3.9
0.9
Ethnic origin Native European French Polynesian
1.0 2.5
1.1–5.7
0.03
1.0 2.2
1.1–4.8
0.04
Abbreviations: FFBF, freedom from biochemical failure; BPFS, biochemical progression-free survival.
NE patients experienced Grade 3 or 4 genitourinary toxicity. Five patients (1 FP and 4 NE) experienced late Grade 3 gastrointestinal toxicity, and required argon plasma coagulation. No Grade 4 gastrointestinal toxicity was observed. Discussion We retrospectively compared two cohorts of patients (FP and NE) with localised prostate adenocarcinoma, treated in the same department with the same 3D conformal RT technique. The probability of 5-year FFBF was 58% for the FP patients versus 77% for the NE patients (p = 0.017). After adjustment for known prognostic factors and adjuvant ADT, FP patients had a 2.5-fold lower probability of FFBF than NE patients (Table 3). Various studies have shown differences in survival between Pacific and Caucasian patients. A study on Native Hawaiian health compared data collected in the year 2000 with data from 1982 and 1990. The overall life expectancy of Native Hawaiians was shorter than other ethnic groups residing in Hawaii. Native Hawaiians had a higher prevalence of hypertension, diabetes, and asthma than other ethnic groups, and had higher rates of smoking, excessive alcohol consumption, and being overweight [9]. In Hawaii, survival rates of Hawaiian patients with breast, lung, and colon cancer were worse than those of European patients living in Hawaii [10–12]. In New Zealand, prostate cancer is the second most common cancer among Ma¯ori men and the second leading cause of cancer death [13]. From 1996 to 2001, the mortality/incidence ratio was 29% among Ma¯ori and 15% among non-Ma¯ori men. Ma¯ori men were 16% less likely than non-Ma¯ori men to be diagnosed with prostate cancer, but 60% more likely to die from it. Ma¯ori men were significantly more likely to be diagnosed at a later stage of disease spread than non-Ma¯ori men. Once diagnosed with prostate cancer, Ma¯ori men were more than twice as likely as non-Ma¯ori men to die from their cancer (HR = 2.3, 95% CI [1.9–2.8], p < 0.0001). Authors estimated that more than half of this survival disparity could be attributed to differences in stage at diagnosis. The socioeconomic status of the Ma¯ori and Pacific People (low income, low employment rates, and low educational achievement) probably also has an effect on mortality [14–16]. In addition to socioeconomic factors and lifestyle differences, biological factors may contribute to this discrepancy [17]. For example, androgen and androgen receptor pathways have long been associated with prostate growth, and ethnic differences have been found among variants of the genes of the enzymes involved in androgen biosynthesis and metabolism [18,19]. Ethnic differences have also
506
Outcome in Polynesian patients with prostate cancer
been found among levels of expression and CAG repeat length of the androgen receptor, and critical molecular alterations may occur, resulting in racial disparity [20]. Interethnic differences have been observed in growth factors and their receptors, such as the epidermal growth factor receptor (EGFR) [21–23], which promote cancer cell growth, and this may also explain some of the disparity. Nevertheless, there is considerable controversy concerning the existence of genetic differences among races. As Pearce et al. stated there is much confusion between the concepts of race, ethnicity, and genetics [24]. To date, no significant genetic differences have been found between races, especially with regard to genes that may affect health. However, ethnic differences in health are strongly determined by historical, cultural, and socio-economic factors, and especially by differences in access to health care. Low access to health care could cause delays in diagnosis and affect stages of disease at diagnosis. Age and associated comorbidities could influence treatment choices. All these issues could play a role in the survival differences observed. These differences can disappear when there is equity in access to health care, as evidenced in the study carried out by the U.S. Department of Defense [25,26]. While some authors have reported poorer outcomes for lowrisk patients, our findings indicate poorer outcomes with high-risk patients. Sneyd et al. recently observed that higher mortality rates from prostate cancer for Ma¯ori and Pacific men were due to a higher risk of dying from prostate cancer during the first year of followup after treatment [27]. Compared to Europeans, Ma¯ori men had a significantly increased risk of dying from low-grade prostate cancer (HR = 1.6, 95%CI [1.2–2.1]), and Pacific men had a significantly increased risk of dying from low-grade prostate cancer (HR = 2.8, 95%CI [1.1–6.8]). Observations made in our series are consistent with these reports: a greater proportion of FP patients were more likely to have low-risk prostate cancer (Table 1). We also observed that the biochemical relapses of FP patients appeared earlier, occurring during the first two years of follow-up after treatment, whereas those of NE patients occurred after the third year (Fig. 2). This information could be relevant for tumour aggressiveness [28,29]. Indeed, Pound and al. observed in a population of patients with prostate carcinoma, that the time to development of distant metastases was shorter for patients with a PSA elevation in less than 2 years following surgery, compared with those with a PSA elevation in more than 2 years. In our study, FP patients had biochemical relapses within the first two years after RT. Although patients did not have prostate biopsies at the time of early rising PSA, these biochemical relapses could be explained by local recurrences. Other biological explanations of tumour aggressiveness are possible, and one of these could be that of the RT procedure. As we did not use cone beam CT for daily patient positioning verification during RT, significant prostate displacement could have occurred between treatments. Although an empty rectum was required for the CT planning, rectal distension during RT courses could cause geographic misses and thus decrease the probability of biochemical control [30]. We provided specific instructions at the first consultation, such as emptying the rectum and bladder one hour before CT planning, and before each session of irradiation. Perhaps these instructions were not carefully followed in the low-literacy populations because they could not understand many of the words that are used in clinical and investigative settings [31]. Many of these words are used in the long-established written protocols given to patients. We must rigorously investigate whether or not the words used in our procedures are actually understood by low-literacy FP patients. This study had some limitations because it was retrospective. FP men were younger than NE men, and our FP study population was a sample selected from the entire FP population with prostate cancer. We hypothesise that younger patients were more likely to be detected than older patients, and that some patients with diag-
nosed prostate carcinoma dropped out because of the long journey from a remote island in the middle of the Pacific Ocean to France to receive RT. Nonetheless, age at presentation was not found to be an independent prognostic factor for outcome. In recent large trials less aggressive cases of prostate cancer were reported to be in older men (i.e., >70 years), independent of other clinical features [32]. In our study, FP men were younger than NE men, median age was 65 and 69 years respectively (p < 0.002). Nevertheless median age <70 years in both groups may partially explain the absence of significance of this well-established prognostic factor. In our series, surprisingly, FP patients had a low rate of comorbidities (Table 1). It is possible that FP patients with serious illnesses were not screened, or were not selected for the trip. This kind of selection bias may have affected results of the good-prognosis group of FP patients. Another limitation was that the Gleason scores were not reviewed centrally. However, an experienced pathologist trained in France for 15 years analysed all FP pathological samples. Furthermore, there was a borderline significant difference in use of ADT between the two ethnic groups: the NE group received more ADT. Nevertheless both groups were statically comparable in terms of ADT, so that we believe that the difference observed in terms of outcome cannot be attributed to simply differences between the two groups. However, our data are retrospective and there was lack of information regarding the exact timing of treatment for the 25 patients who received ADT. Consequently, it cannot be excluded that a marginal difference between both groups in terms of endocrine therapy delivery could have partially contributed to the difference in outcome. These limitations highlight that only well-conducted prospective analyses can accurately investigate the true impact of ethnicity in prostate cancer. Conclusion In this retrospective study, we observed that FP ethnicity was an independent prognostic factor of FFBF and BPFS after conformal radiotherapy for prostate adenocarcinoma. These findings support the need for further investigation of factors that compromise the RT efficacy for FP patients with prostate cancer. We must be trained on how best to communicate with low-literacy FP patients. Also, the confirmation of these results could provide further evidence for proposing more aggressive treatment with ADT in association with RT for FP patients with high-risk prostate cancer. Conflict of interest statement None declared. Acknowledgement Authors thank Lois Rose for editing. References [1] Chuba PJ, Moughan J, Forman JD, Owen J, Hanks G. The 1989 patterns of care study for prostate cancer: five-year outcomes. Int J Radiat Oncol Biol Phys 2001;50:325–34. [2] Goggins WB, Wong GK. Poor survival for US Pacific Islander cancer patients: evidence from the surveillance, epidemiology, and end results database: 1991 to 2004. J Clin Oncol 2007;25:5738–41. [3] D’Amico AV, Whittington R, Malkowicz SB, et al. Biochemical outcome after radical prostatectomy, external beam radiation therapy, or interstitial radiation therapy for clinically localized prostate cancer. JAMA 1998;280:969–74. [4] Walsh PC, DeWeese TL, Eisenberger MA. Localized prostate cancer. N Engl J Med 2007;357:2696–705. [5] Partin AW, Kattan MW, Subong EN, et al. Combination of prostate-specific antigen, clinical stage, and Gleason score to predict pathological stage of localized prostate cancer. A multi-institutional update. JAMA 1997;277:1445–51.
A. Levy et al. / Radiotherapy and Oncology 101 (2011) 502–507 [6] Page DL, Fleming ID, Fritz A, Balch CM, Haller DG, Morrow M. American joint committee on cancer national cancer institute (US). In: Greene FL, editor. AJCC cancer staging manual. New York: Springer; 2002. p. 309–13. [7] Trotti A, Colevas AD, Setser A, et al. CTCAE v3.0: development of a comprehensive grading system for the adverse effects of cancer treatment. Semin Radiat Oncol 2003;13:176–81. [8] Roach 3rd M, Hanks G, Thames Jr H, 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;65:965–74. [9] Johnson DB, Oyama N, LeMarchand L, Wilkens L. Native Hawaiians mortality, morbidity, and lifestyle: comparing data from 1982, 1990, and 2000. Pac Health Dialog 2004;11:120–30. [10] Meng L, Maskarinec G, Wilkens L. Ethnic differences and factors related to breast cancer survival in Hawaii. Int J Epidemiol 1997;26:1151–8. [11] Nomura A, Kolonel L, Rellahan W, Lee J, Wegner E. Racial survival patterns for lung cancer in Hawaii. Cancer 1981;48:1265–71. [12] Wegner EL, Kolonel LN, Nomura AM, Lee J. Racial and socioeconomic status differences in survival of colorectal cancer patients in Hawaii. Cancer 1982;49:2208–16. [13] Robson B, Purdie G, Cormack D. Unequal Impact: Ma¯ori and non-Ma¯ori cancer statistics 1996–2001. In: Health Mo, editor, Wellington; 2006. [14] Blakely T, Fawcett J, Hunt D, Wilson N. What is the contribution of smoking and socioeconomic position to ethnic inequalities in mortality in New Zealand? Lancet 2006;368:44–52. [15] Dachs GU, Currie MJ, McKenzie F, et al. Cancer disparities in indigenous Polynesian populations: Maori, Native Hawaiians, and Pacific people. Lancet Oncol 2008;9:473–84. [16] Jeffreys M, Stevanovic V, Tobias M, et al. Ethnic inequalities in cancer survival in New Zealand: linkage study. Am J Public Health 2005;95:834–7. [17] Hatcher D, Daniels G, Osman I, Lee P. Molecular mechanisms involving prostate cancer racial disparity. Am J Transl Res 2009;1:235–48. [18] Ntais C, Polycarpou A, Ioannidis JP. Association of the CYP17 gene polymorphism with the risk of prostate cancer: a meta-analysis. Cancer Epidemiol Biomarkers Prev 2003;12:120–6. [19] Reichardt JK, Makridakis N, Henderson BE, Yu MC, Pike MC, Ross RK. Genetic variability of the human SRD5A2 gene: implications for prostate cancer risk. Cancer Res 1995;55:3973–5.
507
[20] Deka R, Guangyun S, Wiest J, et al. Patterns of instability of expanded CAG repeats at the ERDA1 locus in general populations. Am J Hum Genet 1999;65:192–8. [21] Di Lorenzo G, Tortora G, D’Armiento FP, et al. Expression of epidermal growth factor receptor correlates with disease relapse and progression to androgenindependence in human prostate cancer. Clin Cancer Res 2002;8:3438–44. [22] Douglas DA, Zhong H, Ro JY, et al. Novel mutations of epidermal growth factor receptor in localized prostate cancer. Front Biosci 2006;11:2518–25. [23] Shuch B, Mikhail M, Satagopan J, et al. Racial disparity of epidermal growth factor receptor expression in prostate cancer. J Clin Oncol 2004;22:4725–9. [24] Pearce N, Foliaki S, Sporle A, Cunningham C. Genetics, race, ethnicity, and health. BMJ 2004;328:1070–2. [25] Johnstone PA, Kane CJ, Sun L, et al. Effect of race on biochemical disease-free outcome in patients with prostate cancer treated with definitive radiation therapy in an equal-access health care system: radiation oncology report of the department of defense center for prostate disease research. Radiology 2002;225:420–6. [26] Optenberg SA, Thompson IM, Friedrichs P, Wojcik B, Stein CR, Kramer B. Race, treatment, and long-term survival from prostate cancer in an equal-access medical care delivery system. JAMA 1995;274:1599–605. [27] Sneyd MJ. Ethnic differences in prostate cancer survival in New Zealand: a national study. Cancer Causes Control 2008;19:993–9. [28] Pound CR, Partin AW, Eisenberger MA, Chan DW, Pearson JD, Walsh PC. Natural history of progression after PSA elevation following radical prostatectomy. JAMA 1999;281:1642–5. [29] Simmons MN, Stephenson AJ, Klein EA. Natural history of biochemical recurrence after radical prostatectomy: risk assessment for secondary therapy. Eur Urol 2007;51:1175–84. [30] de Crevoisier R, Tucker SL, Dong L, et al. Increased risk of biochemical and local failure in patients with distended rectum on the planning CT for prostate cancer radiotherapy. Int J Radiat Oncol Biol Phys 2005;62:965–73. [31] Kilbridge KL, Fraser G, Krahn M, et al. Lack of comprehension of common prostate cancer terms in an underserved population. J Clin Oncol 2009;27:2015–21. [32] Hamstra DA, Bae K, Pilepich MV, et al. Older age predicts decreased metastasis and prostate cancer-specific death for men treated with radiation therapy: meta-analysis of radiation therapy oncology group trials. Int J Radiat Oncol Biol Phys 2011.