- Email: [email protected]
A Population-Based Study of Factors Affecting the Use of Radiotherapy for Endometrial Cancer
Clinical Oncology 24 (2012) e113ee124 Contents lists available at SciVerse ScienceDirect
Clinical Oncology journal homepage: www.clinicaloncologyonline.net
Original Article
A Population-Based Study of Factors Affecting the Use of Radiotherapy for Endometrial Cancer T.P. Hanna *y, H. Richardson *z, Y. Peng *x, W. Kong *, J. Zhang-Salomons *, W.J. Mackillop *y * Division
of Cancer Care and Epidemiology, Queen’s University Cancer Research Institute, Kingston, Ontario, Canada Cancer Centre of Southeastern Ontario, Kingston, Ontario, Canada z National Cancer Institute of Canada, Clinical Trials Group, Kingston, Ontario, Canada x Department of Mathematics and Statistics, Queen’s University, Kingston, Ontario, Canada y
Received 8 December 2010; received in revised form 14 December 2011; accepted 27 January 2012
Abstract Aims: To describe the use of adjuvant radiotherapy for endometrial cancer in Ontario, and identify factors associated with its use, and to determine whether variation in the use of radiation is associated with differences in survival. Materials and methods: This was a retrospective, population-based, cohort study of all patients who had a hysterectomy for endometrial cancer in Ontario between 1992 and 2003. We used multiple logistic regression to identify health system-related factors associated with the use of radiotherapy, while controlling for disease- and patient-related factors. Survival and cancer cause-specific survival were compared among regions of the province with higher and lower rates of use of radiotherapy. Results: The study population included a total of 9411 women with a median age of 63 years. Overall, 26.2% received adjuvant radiotherapy. Patients living further from regional cancer centres were slightly less likely to receive radiation (P ¼ 0.02). Patients who had their surgery during longer prevailing waiting times for radiotherapy were less likely to receive radiation (P ¼ 0.04). The use of radiotherapy varied widely from 18.0 to 34.3% among the catchment areas of provincial radiotherapy centres (P < 0.0001). In the overall population, there was no difference in survival among regions with higher and lower rates of use of radiotherapy. However, in the subgroup of cases with clear cell and serous carcinomas, both overall survival and cancer cause-specific survival were significantly lower in regions with lower rates of use of radiotherapy (P < 0.05). This difference remained significant after controlling for other factors (P < 0.05; hazard ratio 1.43; 95% confidence limits 1.06e1.93). Conclusions: Health system-related factors unrelated to patients’ needs affect the use of adjuvant radiotherapy in Ontario. Lower rates of use of adjuvant radiotherapy are associated with lower rates of survival in patients with serous and clear cell carcinomas. Ó 2012 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved. Key words: Endometrial neoplasms; health services accessibility; health services research; outcome assessment (healthcare); radiotherapy; survival rate
Introduction It is estimated that between 34 to 58% of all patients with endometrial cancer will have an indication for external beam radiation at some point in the course of their disease [1]. There is no doubt that adjuvant radiotherapy reduces the risk of pelvic failure after surgery for early endometrial cancer and there is some evidence that it improves survival in high-risk stage I cases [2,3]. Randomised trials have, however, failed to show a survival advantage from Author for correspondence: W.J. Mackillop, Division of Cancer Care and Epidemiology, Queen’s University Cancer Research Institute, 10 Stuart St. Level 2, Kingston, ON, Canada K7L 3N6. Tel: þ613-533-6895; Fax: þ613-533-6794. E-mail address: [email protected] (W.J. Mackillop).
radiotherapy in intermediate-risk stage I patients and the role of postoperative radiotherapy in this large group remains controversial [4e7]. A guideline published by Ontario’s Program in Evidence-Based Care recommends adjuvant radiotherapy for stage I patients with a high risk of recurrence (stage IC, grade 3), and recommends against radiotherapy in those at low risk of recurrence (stage IA, IB, grades 1 and 2) [8]. This guideline states that radiotherapy ’is a reasonable treatment option’ for patients at intermediate risk of recurrence (stage IA, IB, grade 3 or stage IC, grades 1 and 2), and recommends that the decisions about radiotherapy in this situation should be based on consideration of the risks and benefits in the individual patient [8]. Given the continuing scientific controversies about the appropriate role of radiotherapy in early endometrial
0936-6555/$36.00 Ó 2012 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.clon.2012.01.007
e114
T.P. Hanna et al. / Clinical Oncology 24 (2012) e113ee124
cancer, it is not surprising that variations in the beliefs of individual doctors and variations in access to radiotherapy have been shown to influence the use of adjuvant radiotherapy in this situation [9]. Previous studies have shown variations in the use of radiotherapy in a number of clinical situations, and have identified potential barriers to access to radiotherapy in Ontario [10e12]. Older patients, residents of poorer communities, and those who reside further from a radiotherapy centre, have consistently been shown to be less likely than others to receive radiotherapy. Moreover, in some situations, there are large inter-regional variations in the use of radiotherapy, which are unexplained by variations in case mix [10,11]. It is not known whether such factors affect the use of postoperative radiotherapy for endometrial cancer, or whether variations in the use of radiation in this context affect outcomes at the population level. These questions can only be answered by populationbased outcome studies [13]. The first objective of this study was to describe the use of postoperative radiation for endometrial cancer in Ontario between 1992 and 2003, and to identify potential barriers to the use of radiotherapy in this context. The generally accepted indications for the use of radiotherapy at that time were as described in Ontario’s current practice guideline [8], but we hypothesised that the use of postoperative radiotherapy might also vary with factors affecting access to radiotherapy, such as distance of patient’s residence to the nearest radiotherapy centre, or the prevailing waiting list for radiotherapy. We also hypothesised that there might be differences in the use of postoperative radiotherapy between patients who have their surgery in hospitals with specialist expertise in gynaecological oncology and those managed in less specialised general hospitals. The second objective of this study was to explore the relationship between rates of use of radiotherapy and survival in the general population, particularly in the context of high-risk histologies. Unfortunately, we were unable to measure rates of pelvic recurrence in this study because the relevant information was not available at the population level.
Materials and Methods Ontario’s Health System for the Management of Gynaecological Malignancies Ontario’s one-tier, publically funded health care system provides cancer surgery, radiotherapy and hospital care, without direct charge, for all Ontario residents. There is no parallel private sector that offers these services. Both gynaecological oncologists and general gynaecologists are involved in the care of patients with endometrial cancer in Ontario. Gynaecological oncologists are usually confined to large hospitals in urban areas. Radiotherapy is available only at provincial cancer centres located in Ontario’s larger cities. At the time of this study, there were nine radiotherapy centres in Ontario. There were no formal guidelines for the use of radiotherapy in endometrial cancer in Ontario
during the study period. We believe, however, that the general approach to the use of radiotherapy in stage I endometrial cancer in that era is accurately reflected in the guideline published not long afterwards by Ontario’s Program in Evidence-Based Care. This was summarised above and is readily available online [8]. There are still no formal provincial guidelines for stage II and III cases, or for those with clear cell or serous histologies, but in most cases these characteristics would probably have been considered as indications for radiotherapy in that era. Some radiotherapy centres also have their own in-house treatment guidelines, but ultimately all decisions about the use of radiotherapy were, and still are, made by individual radiation oncologists and patients. Study Population This study was a retrospective, population-based study of all patients diagnosed with endometrial cancer in Ontario between 1992 and 2003, who underwent a simple or radical hysterectomy. Only patients with adenocarcinoma, clear cell carcinoma or serous carcinoma by ICD0 criteria were included (adenocarcinoma 8140/3, 8210/3, 8380/3, 8570/3, clear cell carcinoma 8005/3, 8310/3, and serous carcinoma 8460/3, 8461/3). Patients with any previous malignancy were excluded. Patients known to have distant metastases at diagnosis and those whose first radiation after their primary surgery was described as palliative were excluded. Cases of endometrial cancer diagnosed by autopsy or death certificate were excluded. Patients dying within 90 days of surgery were excluded because they had not lived long enough to ascertain whether they would have received postoperative radiation, had they lived longer. Sources of Data Ontario Cancer Registry The Ontario Cancer Registry (OCR) is a passive, population-based registry that collects administrative data about all cancer diagnosed in the province and identifies individual cases through a process of probabilistic linkage. The OCR is known to be at least 95% complete for all sites combined [14,15]. The OCR record includes: date of birth; postal code; Ministry of Health (MOH) residence code; date of cancer diagnosis; histological diagnosis in ICD0 morphology code; and primary site in ICD-9, ICD-9-CM or ICD-10 topography code. The OCR database is regularly linked to the provincial and national death registries to provide information about date and cause of death. Figure 1 describes the sources of additional administrative data that were linked to the OCR to describe the management of endometrial carcinoma, as described previously [10,11]. Canadian Institute of Health Information The Canadian Institute of Health Information (CIHI) Discharge Abstract database includes information about all hospital admissions in Ontario, including: admission and discharge dates; diagnosis most responsible for admission
T.P. Hanna et al. / Clinical Oncology 24 (2012) e113ee124
e115
diagnosed in 2003. For that period, the fifth edition of UICC TNM was used, which corresponds precisely to the 1988 FIGO Staging System. In this study, pathological TNM was used to assign stage group in most cases, and clinical TNM was used in the remainder. Neighbourhood income quintile The neighbourhood income quintile was used as an ecological measure of each patient’s socioeconomic status. Information about the median household income from Statistics Canada was linked to each case in the OCR. The linkage was from census enumeration area to the patient’s postal code, or from census subdivision to the patient’s MOH residence code, as described previously [16].
Fig. 1. The Endometrial Cancer Database.
and other diagnoses, in ICD-9 or ICD-10 code; and surgical or other procedures coded using the Canadian Classification of Diagnostic, Therapeutic and Surgical Procedures (CCP) code. Radiotherapy database This database was assembled at the Queen’s University Cancer Research Institute Division of Cancer Care and Epidemiology with permission from each of the nine provincial cancer centres operating during the period under study. These centres have kept electronic radiation therapy records in uniform format since 1982. The database includes records of: anatomic site irradiated, type of radiation; treatment; dose; number of fractions; and dates of all treatments. Over the period of this study, radiotherapy for almost all residents of Ontario was provided exclusively by nine provincial cancer centres. However, residents of the most westerly county of Kenora were usually referred for radiotherapy to Winnipeg in the neighbouring province of Manitoba from which radiotherapy records were not available. The rate of radiotherapy in Kenora was therefore not reported (see below). Definition of Variables Used in the Analysis Disease characteristics Neither tumour stage nor grade was available for most patients in this cohort. Clinical and pathologic TNM stage are now routinely collected and compiled at all provincial cancer centres, but this practice was only uniformly adopted after 2002. There is still no information about stage available for patients who are not seen at a cancer centre. In this study, information about stage was, therefore, only available for patients seen at a cancer centre who were
Comorbidity Comorbidity was scored using the Elixhauser comorbidity score, based on all diagnoses in the CIHI record of the patient’s admission for hysterectomy. The Elixhauser score is a validated scoring system for comorbidity data extracted from electronic records [17e19]. Its performance characteristics have been compared with the Deyo adaption of the Charlson score and found to be superior in predicting inhospital mortality for cancer patients and other patient populations [17e20]. Classification of surgical procedures Patients were subdivided using CIHI procedure data into those receiving a simple hysterectomy (CCP codes 80.3, 80.4) or radical hysterectomy (CCP codes 80.5, 80.6). CCP codes for surgical staging of lymph nodes were 52.19, 52.2, 52.4, 52.41, 52.44, 52.49, 52.81 and 52.89. These had to occur during the patient’s admission for hysterectomy for endometrial cancer to be accepted as a surgical staging of lymph nodes. CCP procedure code 66.82 was used to identify cases of peritoneal biopsy, the definition of which includes biopsy of mesentery, omentum or peritoneal implant and does not include peritoneal washings. Classification of hospitals Hospitals that had both expertise in gynaecological oncology and a radiotherapy centre on site were classified as comprehensive gynaecological oncology centres (CGOCs). We used the 1 April 2006 Ontario Ministry of Health and Long Term Care Master Numbering System to identify hospital amalgamations and name changes [21]. The most responsible radiotherapy centre was determined for each hospital in the province. This was defined as the radiotherapy centre at which most endometrial cancer patients diagnosed at that hospital received adjuvant radiotherapy. If hospitals had less than 20 patients who received radiation and less than 70% of these received radiation at the same cancer centre, the most responsible radiotherapy centre was defined using information on radiation use for other common cancer surgeries using CCP surgical codes specified by Bardell et al. [22]. If no patients from a given hospital had ever received adjuvant radiation for any indication, the closest cancer centre was assigned as the most responsible cancer centre.
e116
T.P. Hanna et al. / Clinical Oncology 24 (2012) e113ee124
Prevailing waiting time for radiotherapy The concept of prevailing waiting time for radiotherapy was first described by Huang [23]. For the purposes of this study, it was defined as the median time from surgery to the start of adjuvant radiation for cancer at the most responsible radiotherapy centre, at the time of the index patient’s surgery. Specifically, the prevailing waiting time attributed to each case in this study was the median waiting time for adjuvant radiotherapy observed at the most responsible radiotherapy centre over a period starting 4 months before her surgery and ending 2 months after it. Distance to most responsible radiotherapy centre This was defined as the linear distance between the geometric centre of the postal code, or MOH residence code, within which the patient resided, and the most responsible radiotherapy centre. Adjuvant radiotherapy was defined here as radiotherapy to the pelvis given with curative or adjuvant intent within 6 months of a hysterectomy for endometrial cancer. The 6 month cut-off was chosen using available information on timing of adjuvant radiation and treatment of recurrence available for a representative subset of the data. External beam radiation and brachytherapy were both included as forms of adjuvant radiation for the purpose of this study. Overall survival and cancer-specific survival Overall survival and cancer-specific survival were described by the KaplaneMeier method. Equality of strata was tested using the Wilcoxon test. Factors associated with survival were evaluated in a Cox model. Local control could not be assessed in this study because information about local recurrence is not routinely collected at the population level in Ontario. Statistical Analysis We used SAS version 9.1 for statistical analysis. Relationships between independent variables and the use of radiation were assessed using chi-squared statistics. Exploration of confounding and effect modification was carried out using stratified contingency tables and logistic regression. The strength of association between independent variables and the use of radiation was estimated using logistic regression. Both crude and adjusted odds ratios were computed to examine the probability of radiation use for each factor of interest. The Hosmer and Lemeshow Goodness-of-Fit Test was used to assess the degree of fit of the fixed effects model and the Likelihood Ratio Test was used to evaluate the predictive value of the model. Multi-level multiple logistic regression was used to assess and control for the potential effects of data clustering by operating hospital.
Results Patient Characteristics In total, 15 291 patients with endometrial cancer were diagnosed in Ontario between 1 January 1992 and 31
December 2003. Of these, 2314 were excluded because they did not have a hysterectomy (some of the excluded cases also lacked a microscopic diagnosis and a few were known to the registry from death certificate only). Also excluded were: 936 patients who had a previous malignancy; 447 who underwent initial palliative radiotherapy, 410 patients who died within 90 days of surgery; and 1773 with a uterine sarcoma or other histology. The remaining 9411 patients comprised the study population; all these cases had been confirmed microscopically and were known to the registry while they were alive. About 28% of the patients had their initial surgery at CGOCs and 72% had their surgery at other general hospitals. Table 1 shows that the age distribution of patients differed very little between CGOCs and other hospitals, but there was a slightly higher proportion of patients with high-risk histologies at the CGOCs. Table 1 also shows that surgical staging of lymph nodes and peritoneal biopsies were more frequently done at the CGOCs than in other general hospitals. Table 1 also shows the stage distribution of the subset of 505 cases seen at a cancer centre in 2003. Note that the stage distribution differs little between patients who had their surgery at a CGOC and those who had their surgery at other hospitals. Factors Associated with the Use of Radiotherapy The results of our analysis of factors associated with the use of postoperative radiotherapy are described in Tables 2 and 3. Table 2 provides the rates of use of radiotherapy for subgroups defined by patient-related, disease-related and treatment-related variables. Table 3 provides rates for groups defined by health system-related variables. These tables also include the results of a fixed effect, multiple logistic regression. The data were a good fit with the fixed effects model according to the Hosmer and Lemeshow Goodness-of-Fit Test (P ¼ 0.7932), and the Likelihood Ratio Test (P < 0.0001) confirmed the predictive value of the model. Multilevel modelling was also done to assess and account for the potential effects of data clustering by operating hospital, but this approach did not alter our main findings or conclusions and, therefore, these data are not presented. Patient-related Factors Just over one-quarter of all patients in the study population received adjuvant radiotherapy (26.2%). Table 2 shows that the use of radiation increased with increasing age up to the age of 80 years. The radiotherapy rate increased significantly from 16.7% in patients under 40 years to 30.4% in patients aged 71e80 years (P < 0.0001), and this trend remained significant in the multivariate analysis. Table 2 shows that there was no discernible association between the use of radiation and the median household income in the neighbourhood where the patient lived. About two-thirds of patients were classified as having no comorbidity using the Elixhauser score. Table 2 shows that there was no significant decrease in the use of radiation
T.P. Hanna et al. / Clinical Oncology 24 (2012) e113ee124
e117
Table 1 Characteristics of patients managed at comprehensive gynaecological oncology centres (CGOCs) and other hospitals Characteristics Age (years) 60 61e70 >70 High-risk Histologyy (%)* Surgical lymph node staging (%)* Peritoneal biopsy (%)* Radiation* Stage (%) for subset of 505 patients with complete stage informationz IA IB IC IIA IIB IIIA IIIB IIIC IVA IVB * y z
Surgery at CGOC (n ¼ 2605)
Surgery at other hospitals (n ¼ 6806)
All cases combined (n ¼ 9411)
42.7% 29.6% 27.7% 10.2% 30.1% 9.8% 29.3% Surgery at CGOC (n ¼ 172) 18.0% 34.9% 20.3% 4.7% 8.7% 5.8% 0.6% 5.2% 1.2% 0.6%
41.0% 30.7% 28.2% 4.0% 4.8% 6.5% 24.9% Surgery at other hospitals (n ¼ 333) 12.3% 47.4% 22.2% 5.1% 6.6% 3.0% 0.9% 1.5% 0.3% 0.6%
41.5% 30.4% 28.1% 5.7% 11.8% 7.4% 26.2% All cases combined (n ¼ 505) 14.3% 43.2% 21.6% 5.0% 7.3% 4.0% 0.8% 2.8% 0.6% 0.6%
P < 0.0001. Serous or clear cell carcinoma. FIGO 1988 stage distribution for subset of patients with complete stage information.
with increasing levels of comorbidity in either the univariate or the multivariate analysis.
who had a peritoneal biopsy, and those who had a radical hysterectomy, were all significantly more likely than others to have postoperative radiotherapy.
Disease-related Factors and Use of Radiation Most patients in the study population (94.3%) were classified as having an adenocarcinoma, 4.1% had a serous carcinoma and 1.5% had a clear cell carcinoma. Table 2 shows that rates of use of radiotherapy in the clear cell (50.3%) and serous carcinomas (46.4%) were significantly higher than the rate of 24.9% observed in the adenocarcinomas (P < 0.0001), and these differences remained significant in the multivariate analysis. We were unable to report on the use of radiotherapy by stage in the overall population because this information was only available on a minority of patients (see above). However, in the staged subgroup of 505 cases, adjuvant radiotherapy rates were as follows: stage IA, 1.4%; stage IB, 11.9%; stage IC, 68.2%; stage IIA, 84.0%; stage IIB, 83.8%; stage III, 91.9%. Treatment-related Factors Surgical staging of lymph nodes, as defined above, was carried out in 11.8% of patients, but Figure 2 shows that rates of surgical staging varied widely across the province. The proportion of patients who were surgically staged in each county ranged from 0 to 51.9%, with a median 7.9% and an interquartile range of 4.7e16.6%. The rate of surgical staging at CGOC hospitals (30.1%) was significantly higher than the rate observed at other hospitals (4.8%). Table 2 shows that patients who had surgical staging of lymph nodes, those
Health System-related Factors Figure 3 illustrates that the proportion of patients who received adjuvant radiation varied widely among hospitals in Ontario with a median of 24.3% and an interquartile range of 17.2e30.7%. Rates of radiotherapy varied widely among CGOCs, and also among non-CGOC hospitals. Table 3 shows that the rate of use of radiotherapy among patients who had their surgery at a CGOC (29.3%) was significantly higher than among patients who had surgery at other hospitals (25.0%). Table 3 shows that, when patientrelated, disease-related and surgery-related factors were included in a multivariate analysis together with other health system-related factors, no association was found between having surgery at a CGOC and the use of radiotherapy (odds ratio 1.00, 95% confidence limits 0.89e1.13). To explore why the apparent association between surgery at a CGOC and the use of radiotherapy disappeared in the multivariate analysis, we assessed the effect of removing other variables from the model. When the surgery-related variables (type of hysterectomy, lymph node staging and peritoneal biopsy) were removed from the model, surgery at a CGOC became borderline significant (odds ratio 1.12, 95% confidence limits 1.00e1.26), and when histology was also removed, the association became stronger (odds ratio 1.22, 95% confidence limits 1.09e1.36). We also carried out stratified analyses on cases that had surgical lymph node
e118
T.P. Hanna et al. / Clinical Oncology 24 (2012) e113ee124
Table 2 Patient-related, disease-related and treatment-related factors affecting use of adjuvant radiation in Ontario Patient factors
Patient age (years)y 40 41e50 51e60 61e70 71e80 >80 Neighbourhood income quintile 1 (Poorest) 2 3 4 5 (Richest) Elixhauser comorbidity 0 1 2 3 4
Number
Use of radiation (%)
Use of radiation multivariate logistic regression estimate odds ratio (95% confidence limit)*
221 893 2791 2863 2096 547
16.74 19.60 24.58 27.84 30.44 23.40
Reference 1.23 (0.83, 1.66 (1.15, 1.92 (1.32, 2.16 (1.49, 1.46 (0.97,
1950 2102 1904 1778 1668
26.36 25.69 27.84 24.75 26.02
Reference 0.98 (0.85,1.13) 1.09 (0.94,1.26) 0.93 (0.80,1.09) 0.98 (0.83,1.15)
6587 1771 730 234 89
25.84 28.18 25.34 25.21 17.98
0.96 (0.90,1.02)z
8876 145 390
24.86 50.34 46.41
Reference 2.71 (1.93, 3.80) 2.19 (1.77, 2.71)
8303 1108
25.04 34.48
Reference 1.45 (1.23, 1.70)
9249 162
25.89 40.74
Reference 1.66 (1.18, 2.33)
8715 696
25.44 35.06
Reference 1.48 (1.24, 1.76)
1.83) 2.41) 2.77) 3.13) 2.21)
Disease- and treatment-related factors Histologyy Adenocarcinoma Clear cell Serous Surgical stagingy None Staged Hysterectomy typey Simple hysterectomy Radical hysterectomy Peritoneal biopsyy No Yes
* Fixed effects multivariate logistic regression controlling for patient age, neighbourhood income quintile, Elixhauser comorbidity score, histology, surgical staging, hysterectomy type, peritoneal biopsy, hospital type, prevailing waiting time at the most responsible cancer centre, distance from residence to most responsible cancer centre, year of diagnosis and most responsible cancer centre. y P < 0.0001 in univariate analysis. z Elixhauser comorbidity was modelled as a continuous variable. The odds ratio reported corresponds to one unit increase in comorbidity.
staging and those that did not. Among the subgroup that had lymph node staging, patients who had their surgery at a CGOC were more likely to receive radiation than those who had their surgery in other hospitals (37.0% versus 28.3%, P < 0.01), and this association was significant in the multivariate logistic regression (odds ratio 1.53, 95% confidence limits 1.12e2.09). Among those who did not have surgical lymph node staging there was no significant difference in the use of radiation between patients who had their surgery at a CGOC and those who had surgery in other hospitals, in either the univariate or the multivariate analysis (26.0% versus 24.8%, P ¼ 0.30, odds ratio 0.94, 95% confidence limits 0.82e1.07). There was a small but significant inverse relationship between the prevailing waiting time for radiotherapy and the chance of receiving postoperative radiotherapy
(Table 3). When surgery was carried out during prevailing waiting times of 78 days, patients were less likely to receive radiation after controlling for other factors in multivariate logistic regression (odds ratio 0.853, 95% confidence limits 0.730e0.996). The median distance from the patient’s home to the most responsible radiotherapy centre was 26.9 km (interquartile range, 8.5e68.8 km). Table 3 shows that lower rates of use of radiotherapy were observed among patients who lived further from a radiotherapy centre. The magnitude of this association was minimal except in the small group who lived more than 180 km from a radiotherapy centre, of whom 18.8% received radiotherapy, compared with 27.2% of those who lived within 15 km of their radiotherapy centre. The association was barely statistically significant in the multivariate analysis, but the
T.P. Hanna et al. / Clinical Oncology 24 (2012) e113ee124
e119
Table 3 System-related factors affecting use of adjuvant radiation in Ontario System factors
Number
Use of radiation (%)
Hospital typey Other hospital 6806 24.95 Comprehensive gynaecological oncology centre 2605 29.29 Prevailing waiting time at the most responsible cancer centre (days) <78 2326 26.74 78 7085 25.96 Distance to most responsible cancer centre (km) 0.0e15.0 3629 27.20 16e60 3027 25.27 61e120 1723 26.70 121e180 599 25.21 >180 337 23.74 NA 96 18.75 Year of diagnosis 1992 650 28.62 1993 715 29.65 1994 709 26.80 1995 670 28.36 1996 766 24.67 1997 809 24.47 1998 778 25.84 1999 809 26.70 2000 884 23.30 2001 850 26.82 2002 880 24.32 2003 891 25.93 Most responsible cancer centrey A 2494 26.06 B 845 27.57 C 1494 18.01 D 715 26.43 E 1416 34.25 F 1610 25.71 G 144 27.08 H 243 26.34 I 450 26.22
Use of radiation multivariate logistic regression estimate odds ratio (95% confidence limit)* Reference 1.00 (0.89, 1.13) Reference 0.853 (0.730, 0.996) Reference 0.99 (0.88, 0.85 (0.74, 0.76 (0.61, 0.76 (0.56, 0.61 (0.36,
1.11) 0.99) 0.94) 1.02) 1.06)
Reference 1.05 (0.83, 0.91 (0.71, 0.94 (0.73, 0.76 (0.60, 0.79 (0.62, 0.89 (0.70, 0.90 (0.71, 0.78 (0.61, 0.99 (0.78, 0.79 (0.62, 0.88 (0.69,
1.34) 1.16) 1.19) 0.97) 1.01) 1.14) 1.15) 1.00) 1.26) 1.01) 1.12)
Reference 0.89 (0.73, 0.56 (0.47, 1.04 (0.83, 1.51 (1.29, 0.93 (0.79, 1.00 (0.67, 0.94 (0.69, 1.12 (0.85,
1.09) 0.66) 1.30) 1.76) 1.08) 1.51) 1.29) 1.46)
* Fixed effects multivariate logistic regression controlling for patient age, neighbourhood income quintile, Elixhauser comorbidity score, histology, surgical staging, hysterectomy type, peritoneal biopsy, hospital type, prevailing waiting time at the most responsible cancer centre, distance from residence to most responsible cancer centre, year of diagnosis and most responsible cancer centre. y P < 0.0001 in univariate analysis.
relationship was consistent over the whole range of distances studied, suggesting that this association is probably real. The proportion of patients who received radiation varied among the catchment areas of the provincial cancer centres (median 26.3%). The rates observed in the catchment areas of most centres were close to 26%, but one centre had a significantly lower rate of 18.0%, and another centre had a significantly higher rate of 34.3%, and these differences remained significant in the multivariate analysis (Table 3). Figure 4 illustrates geographical variations in the use of radiotherapy at county level. The proportion of patients who received radiotherapy varied widely from 13 to 40.3% with a median rate of 25.6% and an interquartile range of 22.0e30.4%. Table 3 also shows that there was no significant
change in the rate of use of radiotherapy over the 12 year period of this study. Survival Figure 5 compares survival among residents of counties with high, medium and low rates of use of radiotherapy. Figure 5A illustrates overall survival for the entire cohort and shows that there was no discernible difference in survival among residents of counties with high, medium and low rates of use of radiotherapy. Figure 5B shows that, in the overall cohort, there was no difference in cancer cause-specific survival among residents of counties with high, medium and low rates of use of radiotherapy. These findings are summarised in the top section of Table 4, which
e120
T.P. Hanna et al. / Clinical Oncology 24 (2012) e113ee124
Fig. 2. Geographical variations in the proportion of patients with endometrial cancer who had surgical staging of lymph nodes. The map illustrates variations in the proportion of endometrial cancer patients who underwent surgical lymph node staging among Ontario counties. Counties were grouped into quintiles based on their rate of surgical staging. Circles indicate locations of radiotherapy centres.
also shows that there was no significant difference in overall survival among these groups when they were compared in a multivariate Cox analysis that controlled for other prognostic factors. Figure 5C and D provide similar data for the subgroup of patients with the higher risk, clear cell and serous histologies. For this subgroup of cases, both
overall survival and cancer cause-specific survival were worse among residents of counties where the lowest rates of use of radiotherapy were observed. Table 4 summarises these findings and also shows that overall survival remained significantly worse in counties with a low rate of use of radiotherapy, when other factors associated with survival had been controlled for in a Cox analysis. A secondary analysis of the subgroup of 505 staged cases from the 2002e2003 period showed that there was no significant difference in stage distribution between the regions with low rates of radiotherapy and other regions of the province (P ¼ 0.43).
Discussion
Fig. 3. Variations in the use of adjuvant radiotherapy for endometrial cancer among Ontario hospitals. The dot plot shows the proportion of patients who received adjuvant radiotherapy at every Ontario hospital that operated on more than 60 cases of endometrial cancer over the study period. The bars show the 95% confidence limits. Comprehensive gynaecological oncology centres are shown as black dots, and other hospitals are shown as grey dots.
In this study, we identified a number of health systemrelated factors associated with the use of adjuvant radiation for endometrial cancer in Ontario. For example, patients who had their surgery where longer radiotherapy waiting times prevailed, and those who lived farther from a radiotherapy centre, were less likely to receive radiation. Although our study was confined to Ontario, we suspect that similar factors may affect the use of radiotherapy in this disease, particularly in other publicly funded health care systems. The observed differences in the use of radiotherapy may be important because we also found that lower rates of radiotherapy were significantly associated with lower survival in a subgroup of high-risk patients in whom
T.P. Hanna et al. / Clinical Oncology 24 (2012) e113ee124
e121
Fig. 4. Geographical variations in the proportion of patients with endometrial cancer who received adjuvant radiation. The map illustrates variations in the proportion of patients with endometrial cancer who received adjuvant radiotherapy among Ontario counties. Counties were grouped into quintiles based on their rate of use of adjuvant radiotherapy. Circles indicate locations of radiotherapy centres.
postoperative radiation might be expected to improve survival (serous and clear cell carcinoma). This study shares the strengths and weaknesses of many other registry-based, outcome studies. The OCR is an
accurate source of basic diagnostic information about almost every incident case of cancer in Ontario, and it also provides reliable information about vital status and date of death. The administrative databases of CIHI and CCO are
Fig. 5. Comparison of survival among regions with higher and lower rates of use of adjuvant radiotherapy. The graphs illustrate the survival of patients with endometrial cancer who reside in regions of Ontario with high (:), medium (C), and low rates (-) of use of adjuvant radiotherapy. For the purpose of this analysis, counties were assigned to tertiles based on rates of use of radiotherapy. (A) Overall survival for all cases combined. (B) Cancer cause-specific survival for all cases combined. (C) Overall survival for patients with high-risk histologies. (D) Cancer causespecific survival for patients with high-risk histologies.
e122
T.P. Hanna et al. / Clinical Oncology 24 (2012) e113ee124
Table 4 Variations in survival by rates of radiation use in Ontario Tertile (% radiation use)
n
5-year overall survival (95% confidence limits)
Hazard ratio for overall survival (95% confidence limits)
5-year cancer-specific survival (95% confidence limits)
Hazard ratio for cancer-specific survival (95% confidence limits)
Total cohort Low-use counties (<23.1%) Medium-use counties (23.1e29.2%) High-use counties (>29.2%)
9369 2399 4522
* 86.7% (85.3%, 88.0%) 87.4% (86.4%, 88.3%)
0.92 (0.82, 1.03) 0.93 (0.84, 1.03)
90.4% (89.1%, 91.5%) 90.9% (90.0%, 91.7%)
0.92 (0.78, 1.09) 0.90 (0.78. 1.05)
2448
85.8% (84.3%, 87.1%)
Reference
89.4% (88.1%, 90.6%)
Reference
High-risk histology Low-use counties (<40.0%) Medium-use counties (40.0e52.9%) High-use counties (>52.9%)
531 121 178
* 44.6% (35.6%, 53.2%) 58.4% (50.8%, 65.3%)
1.43 (1.06, 1.93) 1.10 (0.83, 1.44)
* 46.6% (37.0%, 55.6%) 61.1% (53.4%, 67.9%)
1.52 (1.09, 2.12) 1.12 (0.82, 1.53)
232
59.5% (52.9%, 65.5%)
Reference
63.5% (56.7%, 69.5%)
Reference
For the total cohort, stepwise selection identified a Cox model controlling for age, neighbourhood income quintile, Elixhauser comorbidity, histology, hysterectomy type, peritoneal biopsy, prevailing waiting time and distance to the most responsible cancer centre. For high-risk histologies, the Cox model identified by stepwise selection controlled for age, surgical staging and hysterectomy type. * Differences between groups were statistically significant (Wilcoxon test for equality of strata P < 0.05).
a fairly accurate source of basic information about the use of surgery and radiotherapy in almost every patient. Therefore, the linkage of these province-wide databases allowed us to provide a simplified description of the treatment and outcome of endometrial cancer in a large population. Our study was therefore free of the problem of referral bias inherent to hospital-based case series, and had great statistical power to detect inter-regional differences in treatment and outcome. However, the absence of any information about tumour grade and the absence of complete information about stage made it impossible to control for these important factors in our analyses of treatment and outcome. Nonetheless, there is no a priori reason to suspect that there might be large inter-regional variations in the distributions of tumour grade or stage, and the limited evidence available suggests that these characteristics vary little across the province [24]. It is therefore improbable that the variations in the use of radiotherapy reported here are explained by variations in case mix. Lack of the details of the care path in the individual case also made it impossible to determine if a patient was not treated with radiotherapy because she was not referred to a radiation oncologist or because radiotherapy was not recommended by her radiation oncologist. Thus, we were unable to determine the extent to which gynaecologists and radiation oncologists were responsible for the observed variations in the use of radiotherapy across the province. Administrative databases provide no information about the patients’ preferences or their role in treatment decisions and we cannot comment directly on the extent to which rates of use of radiotherapy are influenced by patients’ wishes. Our reliance on administrative data also precluded assessment of rates of pelvic failure, because none of the available databases provided any direct information about the date or site of recurrence. In reality, this outcome is not consistently evaluated or reliably reported in routine practice, and is difficult even to measure in clinical trials. Whether and when a patient is found to have a pelvic
recurrence depends on the type and frequency of follow-up investigations. We doubt if it will ever be possible to ensure sufficiently consistent follow-up in the community at large to permit pelvic failure rates to be reliably reported in population-based studies. Overall, we found that 26% of women who had a hysterectomy for endometrial cancer received adjuvant radiation therapy. Delaney et al. [1] estimated, in a population with a surgical staging rate similar to Ontario’s, that 58% of women with endometrial cancer will develop an indication for radiotherapy at some point in the course of the illness. In estimating the appropriate rate of use of radiotherapy, these investigators included indications for subsequent treatment for recurrence and metastasis, and excluded indications for brachytherapy [1]. Thus, we cannot directly compare our observed rate with their estimate of the appropriate rate. However, given that relatively few patients with endometrial cancer go on to develop recurrence or metastases, the rate we have observed seems to fall far short of the Delaney et al. [1] estimate of the appropriate rate. Moreover, although the rate of use of adjuvant radiotherapy varied widely across Ontario, the observed rate did not approach the estimated appropriate rate in any part of the province. Nonetheless, we hesitate to conclude that radiotherapy is under-utilised in Ontario, because the indications for radiotherapy in this disease remain controversial and the evidentiary basis of the evidence-based rate is not very strong. An alternative approach to establishing the ‘right’ rate of radiotherapy is to set a ‘benchmark’ based on the rates of use of radiotherapy observed in circumstances where access to radiotherapy is not constrained, and decisions about its use are made in a multidisciplinary setting [25]. However, we found that the rate of use of radiotherapy in this disease varies widely even among the CGOCs, where patients have ready access to radiotherapy and gynaecological and radiation oncologists work closely with one another. In fact, the variation in rates observed in these more sophisticated hospitals was just as great as it was
T.P. Hanna et al. / Clinical Oncology 24 (2012) e113ee124
among other hospitals. This makes it impossible to identify a robust benchmark rate for the use of radiotherapy in endometrial cancer, as we have done successfully in several other diseases [25e27]. The results of our study suggest that longer prevailing waiting times for adjuvant radiotherapy may be associated with lower rates of use of adjuvant radiation for endometrial cancer. This may be due to lower rates of referral, or to less frequent prescription of radiation after referral when prevailing waiting times are longer. Prevailing waiting times have been associated with variations in the use of radiation in some, but not all studies. Zhang-Salomons et al. [28] found that elderly patients were found less likely to receive post-lumpectomy radiation for breast cancer during longer prevailing waiting times for radiotherapy in Ontario. However, Huang [23] found no association between prevailing waiting times for radiotherapy and the choice of radiation versus surgery for prostate cancer. Prevailing waiting times may also affect the way radiotherapy is delivered; Kong et al. [12] found that a longer prevailing waiting time was associated with the use of shorter fractionation schemes for bone metastases. In keeping with the results of the relevant clinical trials, we found no association between the rate of use of postoperative radiotherapy and survival in the overall study population. Regrettably, for reasons discussed above, we were unable to determine whether lower rates of radiotherapy were associated with higher rates of pelvic recurrence, as predicted by the results of clinical trials. However, in a subpopulation of higher risk patients with clear cell or serous histologies, we observed worse survival among residents of regions with lower rates of use of radiotherapy. The association between low radiotherapy use and poorer survival in this subgroup was clinically and statistically significant in both the univariate and multivariate analyses. Although the stage distribution observed in regions with low rates of use of radiotherapy was similar to that observed in other regions of the province, it remains possible that the observed differences in survival reflect differences in case mix rather than differences in treatment. The association between radiation use and survival might have been confounded by the use of adjuvant chemotherapy, but we think this is unlikely. The period under study (1992e2003) was before the publication of the commonly cited 2007 study by Hogberg et al. [29] that suggested a small benefit in disease-free survival with chemotherapy in patients with high-risk endometrial cancer. Even if chemotherapy had been widely used between 1992 and 2003, it would have to be used much more in regions with high radiation use for any confounding to occur, and there is no reason to suspect this. Finally, if the evidence for an overall survival benefit of adjuvant chemotherapy for high-risk endometrial cancer is limited, it is even more so specifically for clear cell and serous carcinoma [30,31].
Conclusions This study provides evidence that factors unrelated to patient’s needs influenced the use of adjuvant radiation for
e123
endometrial cancer in Ontario between 1992 and 2003, and that lower rates of use of radiation were associated with poorer survival among a subset of the patients who would be expected to benefit most from radiation.
Acknowledgements The authors wish to acknowledge the support of the Canadian Institute of Health Research (CIHR) Fellowship in Policy and Health Services Research, Queen’s University and CIHR Transdisciplinary Training Program in Cancer Research Fellowship, CIHR New Emerging Team (NET) Grant, Cancer Care Ontario (CCO) and the Ontario Cancer Registry (OCR).
References [1] Delaney G, Jacob S, Barton M. Estimation of an optimal radiotherapy utilization rate for gynecologic carcinoma: part II e carcinoma of the endometrium. Cancer 2004;101:682e692. [2] Johnson N, Cornes P. Survival and recurrent disease after postoperative radiotherapy for early endometrial cancer: systematic review and meta-analysis. BJOG 2007;114:1313e1320. [3] Lee CM, Szabo A, Shrieve DC, et al. Frequency and effect of adjuvant radiation therapy among women with stage I endometrial adenocarcinoma. JAMA 2006;295:389e397. [4] Blake P, Swart AM, Orton J, et al. Adjuvant external beam radiotherapy in the treatment of endometrial cancer (MRC ASTEC and NCIC CTG EN.5 randomised trials): pooled trial results, systematic review, and meta-analysis. Lancet 2009; 373:137e146. [5] Creutzberg CL, van Putten WL, Koper PC, et al. Surgery and postoperative radiotherapy versus surgery alone for patients with stage-1 endometrial carcinoma: multicentre randomised trial. PORTEC Study Group. Post Operative Radiation Therapy in Endometrial Carcinoma. Lancet 2000;355:1404e1411. [6] Keys HM, Roberts JA, Brunetto VL, et al. A phase III trial of surgery with or without adjunctive external pelvic radiation therapy in intermediate risk endometrial adenocarcinoma: a Gynecologic Oncology Group study. Gynecol Oncol 2004;92:744e751. [7] Nout RA, Smit VT, Putter H, et al. Vaginal brachytherapy versus pelvic external beam radiotherapy for patients with endometrial cancer of high-intermediate risk (PORTEC-2): an open-label, non-inferiority, randomised trial. Lancet 2010; 375:816e823. [8] Lukka H, Chambers A, Fyles A, et al. Adjuvant radiotherapy in women with stage I endometrial cancer: a clinical practice guideline 2006. Available at: https://www.cancercare.on.ca/ common/pages/UserFile.aspx?serverId¼6&path¼/File%20Data base/CCO%20Files/PEBC/pebc4-10f.pdf. [9] Kwon JS, Carey MS, Cook EF, et al. Patterns of practice and outcomes in intermediate- and high-risk stage I and II endometrial cancer: a population-based study. Int J Gynecol Cancer 2007;17:433e440. [10] Paszat LF, Mackillop WJ, Groome PA, et al. Radiotherapy for breast cancer in Ontario: rate variation associated with region, age and income. Clin Invest Med 1998;21:125e134. [11] Huang J, Zhou S, Groome P, et al. Factors affecting the use of palliative radiotherapy in Ontario. J Clin Oncol 2001;19: 137e144. [12] Kong W, Zhang-Salomons J, Hanna TP, et al. A populationbased study of the fractionation of palliative radiotherapy
e124
[13]
[14]
[15]
[16]
[17] [18]
[19]
[20]
[21]
[22]
[23]
T.P. Hanna et al. / Clinical Oncology 24 (2012) e113ee124 for bone metastasis in Ontario. Int J Radiat Oncol Biol Phys 2007;69:1209e1217. Booth CM, Mackillop WJ. Translating new medical therapies into societal benefit: the role of population-based outcome studies. JAMA 2008;300:2177e2179. Robles SC, Marrett LD, Clarke EA, et al. An application of captureerecapture methods to the estimation of completeness of cancer registration. J Clin Epidemiol 1988;41:495e501. Clarke EA, Marrett LD, Kreiger N. Cancer registration in Ontario: a computer approach. IARC Scientific Publications 1991:246e257. Mackillop WJ, Zhang-Salomons J, Groome PA, et al. Socioeconomic status and cancer survival in Ontario. J Clin Oncol 1997;15:1680. Elixhauser A, Steiner C, Harris DR, et al. Comorbidity measures for use with administrative data. Med Care 1998;36:8e27. Southern DA, Quan H, Ghali WA. Comparison of the Elixhauser and Charlson/Deyo methods of comorbidity measurement in administrative data. Med Care 2004;42: 355e360. Stukenborg GJ, Wagner DP, Connors Jr AF. Comparison of the performance of two comorbidity measures, with and without information from prior hospitalizations. Med Care 2001;39: 727e739. de Groot V, Beckerman H, Lankhorst GJ, et al. How to measure comorbidity. A critical review of available methods. J Clin Epidemiol 2003;56:221e229. Ministry of Health and Long Term Care Finance & Information Management Branch Information Management Unit (editors). Ministry of Health and Long Term Care Master Numbering System (CIHI Institution Numbers). Ontario, 2006. Bardell T, Belliveau P, Kong W, et al. Waiting times for cancer surgery in Ontario: 1984e2000. Clin Oncol (R Coll Radiol) 2006;18:401e409. Huang M. The influence of waiting time on treatment choices among prostate cancer patients. Queen’s University, Department of Community Health and Epidemiology, 2006: 120.
[24] Kwon JS, Carey MS, Cook EF, et al. Are there regional differences in gynecologic cancer outcomes in the context of a single-payer, publicly-funded health care system? A population-based study. Can J Public Health 2008;99: 221e226. [25] Barbera L, Zhang-Salomons J, Huang J, Tyldesley S, Mackillop WJ. Defining the need for radiotherapy for lung cancer in the general population: a criterion-based, benchmarking approach. Medical Care 2003;41:1074e1085. [26] Kerba M, Miao Q, Zhang-Salomons J, Mackillop WJ. Defining the need for breast cancer radiotherapy in the general population: a criterion-based benchmarking approach. Clin Oncol 2007;19:481e489. [27] Kerba M, Miao Q, Zhang-Salomons J, Mackillop W. Defining the need for prostate cancer radiotherapy in the general population: a criterion-based benchmarking approach. Clin Oncol 2010;22:801e809. [28] Zhang-Salomons J, Fourodi F, Huang J, et al. Health system effects on the use of radiotherapy following surgery for breast cancer. Abstr Acad Health Serv Res Health Policy Meet 2002;19:45. [29] Hogberg T, Rosenberg P, Kristensen G, et al. A randomized phase-III study on adjuvant treatment with radiation (RT) chemotherapy (CT) in early-stage high-risk endometrial cancer (NSGO-EC-9501/EORTC 55991). J Clin Oncol 2007;25: 5503. [30] Hogberg T, Rosenberg P, Kristensen G, et al. A randomized phase-III study on adjuvant treatment with radiation (RT) chemotherapy (CT) in early-stage high-risk endometrial cancer (NSGO-EC-9501/EORTC 55991). Online oral presentation. Available at: http://www.asco.org/ASCOv2/MultiMedia/ VirtualþMeeting?&vmview¼vm_session_presentations_ view&confID¼47&sessionID¼2213. [31] Randall ME, Filiaci VL, Muss H, et al. Randomized phase III trial of whole-abdominal irradiation versus doxorubicin and cisplatin chemotherapy in advanced endometrial carcinoma: a Gynecologic Oncology Group Study. J Clin Oncol 2006;24: 36e44.
Clinical Oncology journal homepage: www.clinicaloncologyonline.net
Original Article
A Population-Based Study of Factors Affecting the Use of Radiotherapy for Endometrial Cancer T.P. Hanna *y, H. Richardson *z, Y. Peng *x, W. Kong *, J. Zhang-Salomons *, W.J. Mackillop *y * Division
of Cancer Care and Epidemiology, Queen’s University Cancer Research Institute, Kingston, Ontario, Canada Cancer Centre of Southeastern Ontario, Kingston, Ontario, Canada z National Cancer Institute of Canada, Clinical Trials Group, Kingston, Ontario, Canada x Department of Mathematics and Statistics, Queen’s University, Kingston, Ontario, Canada y
Received 8 December 2010; received in revised form 14 December 2011; accepted 27 January 2012
Abstract Aims: To describe the use of adjuvant radiotherapy for endometrial cancer in Ontario, and identify factors associated with its use, and to determine whether variation in the use of radiation is associated with differences in survival. Materials and methods: This was a retrospective, population-based, cohort study of all patients who had a hysterectomy for endometrial cancer in Ontario between 1992 and 2003. We used multiple logistic regression to identify health system-related factors associated with the use of radiotherapy, while controlling for disease- and patient-related factors. Survival and cancer cause-specific survival were compared among regions of the province with higher and lower rates of use of radiotherapy. Results: The study population included a total of 9411 women with a median age of 63 years. Overall, 26.2% received adjuvant radiotherapy. Patients living further from regional cancer centres were slightly less likely to receive radiation (P ¼ 0.02). Patients who had their surgery during longer prevailing waiting times for radiotherapy were less likely to receive radiation (P ¼ 0.04). The use of radiotherapy varied widely from 18.0 to 34.3% among the catchment areas of provincial radiotherapy centres (P < 0.0001). In the overall population, there was no difference in survival among regions with higher and lower rates of use of radiotherapy. However, in the subgroup of cases with clear cell and serous carcinomas, both overall survival and cancer cause-specific survival were significantly lower in regions with lower rates of use of radiotherapy (P < 0.05). This difference remained significant after controlling for other factors (P < 0.05; hazard ratio 1.43; 95% confidence limits 1.06e1.93). Conclusions: Health system-related factors unrelated to patients’ needs affect the use of adjuvant radiotherapy in Ontario. Lower rates of use of adjuvant radiotherapy are associated with lower rates of survival in patients with serous and clear cell carcinomas. Ó 2012 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved. Key words: Endometrial neoplasms; health services accessibility; health services research; outcome assessment (healthcare); radiotherapy; survival rate
Introduction It is estimated that between 34 to 58% of all patients with endometrial cancer will have an indication for external beam radiation at some point in the course of their disease [1]. There is no doubt that adjuvant radiotherapy reduces the risk of pelvic failure after surgery for early endometrial cancer and there is some evidence that it improves survival in high-risk stage I cases [2,3]. Randomised trials have, however, failed to show a survival advantage from Author for correspondence: W.J. Mackillop, Division of Cancer Care and Epidemiology, Queen’s University Cancer Research Institute, 10 Stuart St. Level 2, Kingston, ON, Canada K7L 3N6. Tel: þ613-533-6895; Fax: þ613-533-6794. E-mail address: [email protected] (W.J. Mackillop).
radiotherapy in intermediate-risk stage I patients and the role of postoperative radiotherapy in this large group remains controversial [4e7]. A guideline published by Ontario’s Program in Evidence-Based Care recommends adjuvant radiotherapy for stage I patients with a high risk of recurrence (stage IC, grade 3), and recommends against radiotherapy in those at low risk of recurrence (stage IA, IB, grades 1 and 2) [8]. This guideline states that radiotherapy ’is a reasonable treatment option’ for patients at intermediate risk of recurrence (stage IA, IB, grade 3 or stage IC, grades 1 and 2), and recommends that the decisions about radiotherapy in this situation should be based on consideration of the risks and benefits in the individual patient [8]. Given the continuing scientific controversies about the appropriate role of radiotherapy in early endometrial
0936-6555/$36.00 Ó 2012 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.clon.2012.01.007
e114
T.P. Hanna et al. / Clinical Oncology 24 (2012) e113ee124
cancer, it is not surprising that variations in the beliefs of individual doctors and variations in access to radiotherapy have been shown to influence the use of adjuvant radiotherapy in this situation [9]. Previous studies have shown variations in the use of radiotherapy in a number of clinical situations, and have identified potential barriers to access to radiotherapy in Ontario [10e12]. Older patients, residents of poorer communities, and those who reside further from a radiotherapy centre, have consistently been shown to be less likely than others to receive radiotherapy. Moreover, in some situations, there are large inter-regional variations in the use of radiotherapy, which are unexplained by variations in case mix [10,11]. It is not known whether such factors affect the use of postoperative radiotherapy for endometrial cancer, or whether variations in the use of radiation in this context affect outcomes at the population level. These questions can only be answered by populationbased outcome studies [13]. The first objective of this study was to describe the use of postoperative radiation for endometrial cancer in Ontario between 1992 and 2003, and to identify potential barriers to the use of radiotherapy in this context. The generally accepted indications for the use of radiotherapy at that time were as described in Ontario’s current practice guideline [8], but we hypothesised that the use of postoperative radiotherapy might also vary with factors affecting access to radiotherapy, such as distance of patient’s residence to the nearest radiotherapy centre, or the prevailing waiting list for radiotherapy. We also hypothesised that there might be differences in the use of postoperative radiotherapy between patients who have their surgery in hospitals with specialist expertise in gynaecological oncology and those managed in less specialised general hospitals. The second objective of this study was to explore the relationship between rates of use of radiotherapy and survival in the general population, particularly in the context of high-risk histologies. Unfortunately, we were unable to measure rates of pelvic recurrence in this study because the relevant information was not available at the population level.
Materials and Methods Ontario’s Health System for the Management of Gynaecological Malignancies Ontario’s one-tier, publically funded health care system provides cancer surgery, radiotherapy and hospital care, without direct charge, for all Ontario residents. There is no parallel private sector that offers these services. Both gynaecological oncologists and general gynaecologists are involved in the care of patients with endometrial cancer in Ontario. Gynaecological oncologists are usually confined to large hospitals in urban areas. Radiotherapy is available only at provincial cancer centres located in Ontario’s larger cities. At the time of this study, there were nine radiotherapy centres in Ontario. There were no formal guidelines for the use of radiotherapy in endometrial cancer in Ontario
during the study period. We believe, however, that the general approach to the use of radiotherapy in stage I endometrial cancer in that era is accurately reflected in the guideline published not long afterwards by Ontario’s Program in Evidence-Based Care. This was summarised above and is readily available online [8]. There are still no formal provincial guidelines for stage II and III cases, or for those with clear cell or serous histologies, but in most cases these characteristics would probably have been considered as indications for radiotherapy in that era. Some radiotherapy centres also have their own in-house treatment guidelines, but ultimately all decisions about the use of radiotherapy were, and still are, made by individual radiation oncologists and patients. Study Population This study was a retrospective, population-based study of all patients diagnosed with endometrial cancer in Ontario between 1992 and 2003, who underwent a simple or radical hysterectomy. Only patients with adenocarcinoma, clear cell carcinoma or serous carcinoma by ICD0 criteria were included (adenocarcinoma 8140/3, 8210/3, 8380/3, 8570/3, clear cell carcinoma 8005/3, 8310/3, and serous carcinoma 8460/3, 8461/3). Patients with any previous malignancy were excluded. Patients known to have distant metastases at diagnosis and those whose first radiation after their primary surgery was described as palliative were excluded. Cases of endometrial cancer diagnosed by autopsy or death certificate were excluded. Patients dying within 90 days of surgery were excluded because they had not lived long enough to ascertain whether they would have received postoperative radiation, had they lived longer. Sources of Data Ontario Cancer Registry The Ontario Cancer Registry (OCR) is a passive, population-based registry that collects administrative data about all cancer diagnosed in the province and identifies individual cases through a process of probabilistic linkage. The OCR is known to be at least 95% complete for all sites combined [14,15]. The OCR record includes: date of birth; postal code; Ministry of Health (MOH) residence code; date of cancer diagnosis; histological diagnosis in ICD0 morphology code; and primary site in ICD-9, ICD-9-CM or ICD-10 topography code. The OCR database is regularly linked to the provincial and national death registries to provide information about date and cause of death. Figure 1 describes the sources of additional administrative data that were linked to the OCR to describe the management of endometrial carcinoma, as described previously [10,11]. Canadian Institute of Health Information The Canadian Institute of Health Information (CIHI) Discharge Abstract database includes information about all hospital admissions in Ontario, including: admission and discharge dates; diagnosis most responsible for admission
T.P. Hanna et al. / Clinical Oncology 24 (2012) e113ee124
e115
diagnosed in 2003. For that period, the fifth edition of UICC TNM was used, which corresponds precisely to the 1988 FIGO Staging System. In this study, pathological TNM was used to assign stage group in most cases, and clinical TNM was used in the remainder. Neighbourhood income quintile The neighbourhood income quintile was used as an ecological measure of each patient’s socioeconomic status. Information about the median household income from Statistics Canada was linked to each case in the OCR. The linkage was from census enumeration area to the patient’s postal code, or from census subdivision to the patient’s MOH residence code, as described previously [16].
Fig. 1. The Endometrial Cancer Database.
and other diagnoses, in ICD-9 or ICD-10 code; and surgical or other procedures coded using the Canadian Classification of Diagnostic, Therapeutic and Surgical Procedures (CCP) code. Radiotherapy database This database was assembled at the Queen’s University Cancer Research Institute Division of Cancer Care and Epidemiology with permission from each of the nine provincial cancer centres operating during the period under study. These centres have kept electronic radiation therapy records in uniform format since 1982. The database includes records of: anatomic site irradiated, type of radiation; treatment; dose; number of fractions; and dates of all treatments. Over the period of this study, radiotherapy for almost all residents of Ontario was provided exclusively by nine provincial cancer centres. However, residents of the most westerly county of Kenora were usually referred for radiotherapy to Winnipeg in the neighbouring province of Manitoba from which radiotherapy records were not available. The rate of radiotherapy in Kenora was therefore not reported (see below). Definition of Variables Used in the Analysis Disease characteristics Neither tumour stage nor grade was available for most patients in this cohort. Clinical and pathologic TNM stage are now routinely collected and compiled at all provincial cancer centres, but this practice was only uniformly adopted after 2002. There is still no information about stage available for patients who are not seen at a cancer centre. In this study, information about stage was, therefore, only available for patients seen at a cancer centre who were
Comorbidity Comorbidity was scored using the Elixhauser comorbidity score, based on all diagnoses in the CIHI record of the patient’s admission for hysterectomy. The Elixhauser score is a validated scoring system for comorbidity data extracted from electronic records [17e19]. Its performance characteristics have been compared with the Deyo adaption of the Charlson score and found to be superior in predicting inhospital mortality for cancer patients and other patient populations [17e20]. Classification of surgical procedures Patients were subdivided using CIHI procedure data into those receiving a simple hysterectomy (CCP codes 80.3, 80.4) or radical hysterectomy (CCP codes 80.5, 80.6). CCP codes for surgical staging of lymph nodes were 52.19, 52.2, 52.4, 52.41, 52.44, 52.49, 52.81 and 52.89. These had to occur during the patient’s admission for hysterectomy for endometrial cancer to be accepted as a surgical staging of lymph nodes. CCP procedure code 66.82 was used to identify cases of peritoneal biopsy, the definition of which includes biopsy of mesentery, omentum or peritoneal implant and does not include peritoneal washings. Classification of hospitals Hospitals that had both expertise in gynaecological oncology and a radiotherapy centre on site were classified as comprehensive gynaecological oncology centres (CGOCs). We used the 1 April 2006 Ontario Ministry of Health and Long Term Care Master Numbering System to identify hospital amalgamations and name changes [21]. The most responsible radiotherapy centre was determined for each hospital in the province. This was defined as the radiotherapy centre at which most endometrial cancer patients diagnosed at that hospital received adjuvant radiotherapy. If hospitals had less than 20 patients who received radiation and less than 70% of these received radiation at the same cancer centre, the most responsible radiotherapy centre was defined using information on radiation use for other common cancer surgeries using CCP surgical codes specified by Bardell et al. [22]. If no patients from a given hospital had ever received adjuvant radiation for any indication, the closest cancer centre was assigned as the most responsible cancer centre.
e116
T.P. Hanna et al. / Clinical Oncology 24 (2012) e113ee124
Prevailing waiting time for radiotherapy The concept of prevailing waiting time for radiotherapy was first described by Huang [23]. For the purposes of this study, it was defined as the median time from surgery to the start of adjuvant radiation for cancer at the most responsible radiotherapy centre, at the time of the index patient’s surgery. Specifically, the prevailing waiting time attributed to each case in this study was the median waiting time for adjuvant radiotherapy observed at the most responsible radiotherapy centre over a period starting 4 months before her surgery and ending 2 months after it. Distance to most responsible radiotherapy centre This was defined as the linear distance between the geometric centre of the postal code, or MOH residence code, within which the patient resided, and the most responsible radiotherapy centre. Adjuvant radiotherapy was defined here as radiotherapy to the pelvis given with curative or adjuvant intent within 6 months of a hysterectomy for endometrial cancer. The 6 month cut-off was chosen using available information on timing of adjuvant radiation and treatment of recurrence available for a representative subset of the data. External beam radiation and brachytherapy were both included as forms of adjuvant radiation for the purpose of this study. Overall survival and cancer-specific survival Overall survival and cancer-specific survival were described by the KaplaneMeier method. Equality of strata was tested using the Wilcoxon test. Factors associated with survival were evaluated in a Cox model. Local control could not be assessed in this study because information about local recurrence is not routinely collected at the population level in Ontario. Statistical Analysis We used SAS version 9.1 for statistical analysis. Relationships between independent variables and the use of radiation were assessed using chi-squared statistics. Exploration of confounding and effect modification was carried out using stratified contingency tables and logistic regression. The strength of association between independent variables and the use of radiation was estimated using logistic regression. Both crude and adjusted odds ratios were computed to examine the probability of radiation use for each factor of interest. The Hosmer and Lemeshow Goodness-of-Fit Test was used to assess the degree of fit of the fixed effects model and the Likelihood Ratio Test was used to evaluate the predictive value of the model. Multi-level multiple logistic regression was used to assess and control for the potential effects of data clustering by operating hospital.
Results Patient Characteristics In total, 15 291 patients with endometrial cancer were diagnosed in Ontario between 1 January 1992 and 31
December 2003. Of these, 2314 were excluded because they did not have a hysterectomy (some of the excluded cases also lacked a microscopic diagnosis and a few were known to the registry from death certificate only). Also excluded were: 936 patients who had a previous malignancy; 447 who underwent initial palliative radiotherapy, 410 patients who died within 90 days of surgery; and 1773 with a uterine sarcoma or other histology. The remaining 9411 patients comprised the study population; all these cases had been confirmed microscopically and were known to the registry while they were alive. About 28% of the patients had their initial surgery at CGOCs and 72% had their surgery at other general hospitals. Table 1 shows that the age distribution of patients differed very little between CGOCs and other hospitals, but there was a slightly higher proportion of patients with high-risk histologies at the CGOCs. Table 1 also shows that surgical staging of lymph nodes and peritoneal biopsies were more frequently done at the CGOCs than in other general hospitals. Table 1 also shows the stage distribution of the subset of 505 cases seen at a cancer centre in 2003. Note that the stage distribution differs little between patients who had their surgery at a CGOC and those who had their surgery at other hospitals. Factors Associated with the Use of Radiotherapy The results of our analysis of factors associated with the use of postoperative radiotherapy are described in Tables 2 and 3. Table 2 provides the rates of use of radiotherapy for subgroups defined by patient-related, disease-related and treatment-related variables. Table 3 provides rates for groups defined by health system-related variables. These tables also include the results of a fixed effect, multiple logistic regression. The data were a good fit with the fixed effects model according to the Hosmer and Lemeshow Goodness-of-Fit Test (P ¼ 0.7932), and the Likelihood Ratio Test (P < 0.0001) confirmed the predictive value of the model. Multilevel modelling was also done to assess and account for the potential effects of data clustering by operating hospital, but this approach did not alter our main findings or conclusions and, therefore, these data are not presented. Patient-related Factors Just over one-quarter of all patients in the study population received adjuvant radiotherapy (26.2%). Table 2 shows that the use of radiation increased with increasing age up to the age of 80 years. The radiotherapy rate increased significantly from 16.7% in patients under 40 years to 30.4% in patients aged 71e80 years (P < 0.0001), and this trend remained significant in the multivariate analysis. Table 2 shows that there was no discernible association between the use of radiation and the median household income in the neighbourhood where the patient lived. About two-thirds of patients were classified as having no comorbidity using the Elixhauser score. Table 2 shows that there was no significant decrease in the use of radiation
T.P. Hanna et al. / Clinical Oncology 24 (2012) e113ee124
e117
Table 1 Characteristics of patients managed at comprehensive gynaecological oncology centres (CGOCs) and other hospitals Characteristics Age (years) 60 61e70 >70 High-risk Histologyy (%)* Surgical lymph node staging (%)* Peritoneal biopsy (%)* Radiation* Stage (%) for subset of 505 patients with complete stage informationz IA IB IC IIA IIB IIIA IIIB IIIC IVA IVB * y z
Surgery at CGOC (n ¼ 2605)
Surgery at other hospitals (n ¼ 6806)
All cases combined (n ¼ 9411)
42.7% 29.6% 27.7% 10.2% 30.1% 9.8% 29.3% Surgery at CGOC (n ¼ 172) 18.0% 34.9% 20.3% 4.7% 8.7% 5.8% 0.6% 5.2% 1.2% 0.6%
41.0% 30.7% 28.2% 4.0% 4.8% 6.5% 24.9% Surgery at other hospitals (n ¼ 333) 12.3% 47.4% 22.2% 5.1% 6.6% 3.0% 0.9% 1.5% 0.3% 0.6%
41.5% 30.4% 28.1% 5.7% 11.8% 7.4% 26.2% All cases combined (n ¼ 505) 14.3% 43.2% 21.6% 5.0% 7.3% 4.0% 0.8% 2.8% 0.6% 0.6%
P < 0.0001. Serous or clear cell carcinoma. FIGO 1988 stage distribution for subset of patients with complete stage information.
with increasing levels of comorbidity in either the univariate or the multivariate analysis.
who had a peritoneal biopsy, and those who had a radical hysterectomy, were all significantly more likely than others to have postoperative radiotherapy.
Disease-related Factors and Use of Radiation Most patients in the study population (94.3%) were classified as having an adenocarcinoma, 4.1% had a serous carcinoma and 1.5% had a clear cell carcinoma. Table 2 shows that rates of use of radiotherapy in the clear cell (50.3%) and serous carcinomas (46.4%) were significantly higher than the rate of 24.9% observed in the adenocarcinomas (P < 0.0001), and these differences remained significant in the multivariate analysis. We were unable to report on the use of radiotherapy by stage in the overall population because this information was only available on a minority of patients (see above). However, in the staged subgroup of 505 cases, adjuvant radiotherapy rates were as follows: stage IA, 1.4%; stage IB, 11.9%; stage IC, 68.2%; stage IIA, 84.0%; stage IIB, 83.8%; stage III, 91.9%. Treatment-related Factors Surgical staging of lymph nodes, as defined above, was carried out in 11.8% of patients, but Figure 2 shows that rates of surgical staging varied widely across the province. The proportion of patients who were surgically staged in each county ranged from 0 to 51.9%, with a median 7.9% and an interquartile range of 4.7e16.6%. The rate of surgical staging at CGOC hospitals (30.1%) was significantly higher than the rate observed at other hospitals (4.8%). Table 2 shows that patients who had surgical staging of lymph nodes, those
Health System-related Factors Figure 3 illustrates that the proportion of patients who received adjuvant radiation varied widely among hospitals in Ontario with a median of 24.3% and an interquartile range of 17.2e30.7%. Rates of radiotherapy varied widely among CGOCs, and also among non-CGOC hospitals. Table 3 shows that the rate of use of radiotherapy among patients who had their surgery at a CGOC (29.3%) was significantly higher than among patients who had surgery at other hospitals (25.0%). Table 3 shows that, when patientrelated, disease-related and surgery-related factors were included in a multivariate analysis together with other health system-related factors, no association was found between having surgery at a CGOC and the use of radiotherapy (odds ratio 1.00, 95% confidence limits 0.89e1.13). To explore why the apparent association between surgery at a CGOC and the use of radiotherapy disappeared in the multivariate analysis, we assessed the effect of removing other variables from the model. When the surgery-related variables (type of hysterectomy, lymph node staging and peritoneal biopsy) were removed from the model, surgery at a CGOC became borderline significant (odds ratio 1.12, 95% confidence limits 1.00e1.26), and when histology was also removed, the association became stronger (odds ratio 1.22, 95% confidence limits 1.09e1.36). We also carried out stratified analyses on cases that had surgical lymph node
e118
T.P. Hanna et al. / Clinical Oncology 24 (2012) e113ee124
Table 2 Patient-related, disease-related and treatment-related factors affecting use of adjuvant radiation in Ontario Patient factors
Patient age (years)y 40 41e50 51e60 61e70 71e80 >80 Neighbourhood income quintile 1 (Poorest) 2 3 4 5 (Richest) Elixhauser comorbidity 0 1 2 3 4
Number
Use of radiation (%)
Use of radiation multivariate logistic regression estimate odds ratio (95% confidence limit)*
221 893 2791 2863 2096 547
16.74 19.60 24.58 27.84 30.44 23.40
Reference 1.23 (0.83, 1.66 (1.15, 1.92 (1.32, 2.16 (1.49, 1.46 (0.97,
1950 2102 1904 1778 1668
26.36 25.69 27.84 24.75 26.02
Reference 0.98 (0.85,1.13) 1.09 (0.94,1.26) 0.93 (0.80,1.09) 0.98 (0.83,1.15)
6587 1771 730 234 89
25.84 28.18 25.34 25.21 17.98
0.96 (0.90,1.02)z
8876 145 390
24.86 50.34 46.41
Reference 2.71 (1.93, 3.80) 2.19 (1.77, 2.71)
8303 1108
25.04 34.48
Reference 1.45 (1.23, 1.70)
9249 162
25.89 40.74
Reference 1.66 (1.18, 2.33)
8715 696
25.44 35.06
Reference 1.48 (1.24, 1.76)
1.83) 2.41) 2.77) 3.13) 2.21)
Disease- and treatment-related factors Histologyy Adenocarcinoma Clear cell Serous Surgical stagingy None Staged Hysterectomy typey Simple hysterectomy Radical hysterectomy Peritoneal biopsyy No Yes
* Fixed effects multivariate logistic regression controlling for patient age, neighbourhood income quintile, Elixhauser comorbidity score, histology, surgical staging, hysterectomy type, peritoneal biopsy, hospital type, prevailing waiting time at the most responsible cancer centre, distance from residence to most responsible cancer centre, year of diagnosis and most responsible cancer centre. y P < 0.0001 in univariate analysis. z Elixhauser comorbidity was modelled as a continuous variable. The odds ratio reported corresponds to one unit increase in comorbidity.
staging and those that did not. Among the subgroup that had lymph node staging, patients who had their surgery at a CGOC were more likely to receive radiation than those who had their surgery in other hospitals (37.0% versus 28.3%, P < 0.01), and this association was significant in the multivariate logistic regression (odds ratio 1.53, 95% confidence limits 1.12e2.09). Among those who did not have surgical lymph node staging there was no significant difference in the use of radiation between patients who had their surgery at a CGOC and those who had surgery in other hospitals, in either the univariate or the multivariate analysis (26.0% versus 24.8%, P ¼ 0.30, odds ratio 0.94, 95% confidence limits 0.82e1.07). There was a small but significant inverse relationship between the prevailing waiting time for radiotherapy and the chance of receiving postoperative radiotherapy
(Table 3). When surgery was carried out during prevailing waiting times of 78 days, patients were less likely to receive radiation after controlling for other factors in multivariate logistic regression (odds ratio 0.853, 95% confidence limits 0.730e0.996). The median distance from the patient’s home to the most responsible radiotherapy centre was 26.9 km (interquartile range, 8.5e68.8 km). Table 3 shows that lower rates of use of radiotherapy were observed among patients who lived further from a radiotherapy centre. The magnitude of this association was minimal except in the small group who lived more than 180 km from a radiotherapy centre, of whom 18.8% received radiotherapy, compared with 27.2% of those who lived within 15 km of their radiotherapy centre. The association was barely statistically significant in the multivariate analysis, but the
T.P. Hanna et al. / Clinical Oncology 24 (2012) e113ee124
e119
Table 3 System-related factors affecting use of adjuvant radiation in Ontario System factors
Number
Use of radiation (%)
Hospital typey Other hospital 6806 24.95 Comprehensive gynaecological oncology centre 2605 29.29 Prevailing waiting time at the most responsible cancer centre (days) <78 2326 26.74 78 7085 25.96 Distance to most responsible cancer centre (km) 0.0e15.0 3629 27.20 16e60 3027 25.27 61e120 1723 26.70 121e180 599 25.21 >180 337 23.74 NA 96 18.75 Year of diagnosis 1992 650 28.62 1993 715 29.65 1994 709 26.80 1995 670 28.36 1996 766 24.67 1997 809 24.47 1998 778 25.84 1999 809 26.70 2000 884 23.30 2001 850 26.82 2002 880 24.32 2003 891 25.93 Most responsible cancer centrey A 2494 26.06 B 845 27.57 C 1494 18.01 D 715 26.43 E 1416 34.25 F 1610 25.71 G 144 27.08 H 243 26.34 I 450 26.22
Use of radiation multivariate logistic regression estimate odds ratio (95% confidence limit)* Reference 1.00 (0.89, 1.13) Reference 0.853 (0.730, 0.996) Reference 0.99 (0.88, 0.85 (0.74, 0.76 (0.61, 0.76 (0.56, 0.61 (0.36,
1.11) 0.99) 0.94) 1.02) 1.06)
Reference 1.05 (0.83, 0.91 (0.71, 0.94 (0.73, 0.76 (0.60, 0.79 (0.62, 0.89 (0.70, 0.90 (0.71, 0.78 (0.61, 0.99 (0.78, 0.79 (0.62, 0.88 (0.69,
1.34) 1.16) 1.19) 0.97) 1.01) 1.14) 1.15) 1.00) 1.26) 1.01) 1.12)
Reference 0.89 (0.73, 0.56 (0.47, 1.04 (0.83, 1.51 (1.29, 0.93 (0.79, 1.00 (0.67, 0.94 (0.69, 1.12 (0.85,
1.09) 0.66) 1.30) 1.76) 1.08) 1.51) 1.29) 1.46)
* Fixed effects multivariate logistic regression controlling for patient age, neighbourhood income quintile, Elixhauser comorbidity score, histology, surgical staging, hysterectomy type, peritoneal biopsy, hospital type, prevailing waiting time at the most responsible cancer centre, distance from residence to most responsible cancer centre, year of diagnosis and most responsible cancer centre. y P < 0.0001 in univariate analysis.
relationship was consistent over the whole range of distances studied, suggesting that this association is probably real. The proportion of patients who received radiation varied among the catchment areas of the provincial cancer centres (median 26.3%). The rates observed in the catchment areas of most centres were close to 26%, but one centre had a significantly lower rate of 18.0%, and another centre had a significantly higher rate of 34.3%, and these differences remained significant in the multivariate analysis (Table 3). Figure 4 illustrates geographical variations in the use of radiotherapy at county level. The proportion of patients who received radiotherapy varied widely from 13 to 40.3% with a median rate of 25.6% and an interquartile range of 22.0e30.4%. Table 3 also shows that there was no significant
change in the rate of use of radiotherapy over the 12 year period of this study. Survival Figure 5 compares survival among residents of counties with high, medium and low rates of use of radiotherapy. Figure 5A illustrates overall survival for the entire cohort and shows that there was no discernible difference in survival among residents of counties with high, medium and low rates of use of radiotherapy. Figure 5B shows that, in the overall cohort, there was no difference in cancer cause-specific survival among residents of counties with high, medium and low rates of use of radiotherapy. These findings are summarised in the top section of Table 4, which
e120
T.P. Hanna et al. / Clinical Oncology 24 (2012) e113ee124
Fig. 2. Geographical variations in the proportion of patients with endometrial cancer who had surgical staging of lymph nodes. The map illustrates variations in the proportion of endometrial cancer patients who underwent surgical lymph node staging among Ontario counties. Counties were grouped into quintiles based on their rate of surgical staging. Circles indicate locations of radiotherapy centres.
also shows that there was no significant difference in overall survival among these groups when they were compared in a multivariate Cox analysis that controlled for other prognostic factors. Figure 5C and D provide similar data for the subgroup of patients with the higher risk, clear cell and serous histologies. For this subgroup of cases, both
overall survival and cancer cause-specific survival were worse among residents of counties where the lowest rates of use of radiotherapy were observed. Table 4 summarises these findings and also shows that overall survival remained significantly worse in counties with a low rate of use of radiotherapy, when other factors associated with survival had been controlled for in a Cox analysis. A secondary analysis of the subgroup of 505 staged cases from the 2002e2003 period showed that there was no significant difference in stage distribution between the regions with low rates of radiotherapy and other regions of the province (P ¼ 0.43).
Discussion
Fig. 3. Variations in the use of adjuvant radiotherapy for endometrial cancer among Ontario hospitals. The dot plot shows the proportion of patients who received adjuvant radiotherapy at every Ontario hospital that operated on more than 60 cases of endometrial cancer over the study period. The bars show the 95% confidence limits. Comprehensive gynaecological oncology centres are shown as black dots, and other hospitals are shown as grey dots.
In this study, we identified a number of health systemrelated factors associated with the use of adjuvant radiation for endometrial cancer in Ontario. For example, patients who had their surgery where longer radiotherapy waiting times prevailed, and those who lived farther from a radiotherapy centre, were less likely to receive radiation. Although our study was confined to Ontario, we suspect that similar factors may affect the use of radiotherapy in this disease, particularly in other publicly funded health care systems. The observed differences in the use of radiotherapy may be important because we also found that lower rates of radiotherapy were significantly associated with lower survival in a subgroup of high-risk patients in whom
T.P. Hanna et al. / Clinical Oncology 24 (2012) e113ee124
e121
Fig. 4. Geographical variations in the proportion of patients with endometrial cancer who received adjuvant radiation. The map illustrates variations in the proportion of patients with endometrial cancer who received adjuvant radiotherapy among Ontario counties. Counties were grouped into quintiles based on their rate of use of adjuvant radiotherapy. Circles indicate locations of radiotherapy centres.
postoperative radiation might be expected to improve survival (serous and clear cell carcinoma). This study shares the strengths and weaknesses of many other registry-based, outcome studies. The OCR is an
accurate source of basic diagnostic information about almost every incident case of cancer in Ontario, and it also provides reliable information about vital status and date of death. The administrative databases of CIHI and CCO are
Fig. 5. Comparison of survival among regions with higher and lower rates of use of adjuvant radiotherapy. The graphs illustrate the survival of patients with endometrial cancer who reside in regions of Ontario with high (:), medium (C), and low rates (-) of use of adjuvant radiotherapy. For the purpose of this analysis, counties were assigned to tertiles based on rates of use of radiotherapy. (A) Overall survival for all cases combined. (B) Cancer cause-specific survival for all cases combined. (C) Overall survival for patients with high-risk histologies. (D) Cancer causespecific survival for patients with high-risk histologies.
e122
T.P. Hanna et al. / Clinical Oncology 24 (2012) e113ee124
Table 4 Variations in survival by rates of radiation use in Ontario Tertile (% radiation use)
n
5-year overall survival (95% confidence limits)
Hazard ratio for overall survival (95% confidence limits)
5-year cancer-specific survival (95% confidence limits)
Hazard ratio for cancer-specific survival (95% confidence limits)
Total cohort Low-use counties (<23.1%) Medium-use counties (23.1e29.2%) High-use counties (>29.2%)
9369 2399 4522
* 86.7% (85.3%, 88.0%) 87.4% (86.4%, 88.3%)
0.92 (0.82, 1.03) 0.93 (0.84, 1.03)
90.4% (89.1%, 91.5%) 90.9% (90.0%, 91.7%)
0.92 (0.78, 1.09) 0.90 (0.78. 1.05)
2448
85.8% (84.3%, 87.1%)
Reference
89.4% (88.1%, 90.6%)
Reference
High-risk histology Low-use counties (<40.0%) Medium-use counties (40.0e52.9%) High-use counties (>52.9%)
531 121 178
* 44.6% (35.6%, 53.2%) 58.4% (50.8%, 65.3%)
1.43 (1.06, 1.93) 1.10 (0.83, 1.44)
* 46.6% (37.0%, 55.6%) 61.1% (53.4%, 67.9%)
1.52 (1.09, 2.12) 1.12 (0.82, 1.53)
232
59.5% (52.9%, 65.5%)
Reference
63.5% (56.7%, 69.5%)
Reference
For the total cohort, stepwise selection identified a Cox model controlling for age, neighbourhood income quintile, Elixhauser comorbidity, histology, hysterectomy type, peritoneal biopsy, prevailing waiting time and distance to the most responsible cancer centre. For high-risk histologies, the Cox model identified by stepwise selection controlled for age, surgical staging and hysterectomy type. * Differences between groups were statistically significant (Wilcoxon test for equality of strata P < 0.05).
a fairly accurate source of basic information about the use of surgery and radiotherapy in almost every patient. Therefore, the linkage of these province-wide databases allowed us to provide a simplified description of the treatment and outcome of endometrial cancer in a large population. Our study was therefore free of the problem of referral bias inherent to hospital-based case series, and had great statistical power to detect inter-regional differences in treatment and outcome. However, the absence of any information about tumour grade and the absence of complete information about stage made it impossible to control for these important factors in our analyses of treatment and outcome. Nonetheless, there is no a priori reason to suspect that there might be large inter-regional variations in the distributions of tumour grade or stage, and the limited evidence available suggests that these characteristics vary little across the province [24]. It is therefore improbable that the variations in the use of radiotherapy reported here are explained by variations in case mix. Lack of the details of the care path in the individual case also made it impossible to determine if a patient was not treated with radiotherapy because she was not referred to a radiation oncologist or because radiotherapy was not recommended by her radiation oncologist. Thus, we were unable to determine the extent to which gynaecologists and radiation oncologists were responsible for the observed variations in the use of radiotherapy across the province. Administrative databases provide no information about the patients’ preferences or their role in treatment decisions and we cannot comment directly on the extent to which rates of use of radiotherapy are influenced by patients’ wishes. Our reliance on administrative data also precluded assessment of rates of pelvic failure, because none of the available databases provided any direct information about the date or site of recurrence. In reality, this outcome is not consistently evaluated or reliably reported in routine practice, and is difficult even to measure in clinical trials. Whether and when a patient is found to have a pelvic
recurrence depends on the type and frequency of follow-up investigations. We doubt if it will ever be possible to ensure sufficiently consistent follow-up in the community at large to permit pelvic failure rates to be reliably reported in population-based studies. Overall, we found that 26% of women who had a hysterectomy for endometrial cancer received adjuvant radiation therapy. Delaney et al. [1] estimated, in a population with a surgical staging rate similar to Ontario’s, that 58% of women with endometrial cancer will develop an indication for radiotherapy at some point in the course of the illness. In estimating the appropriate rate of use of radiotherapy, these investigators included indications for subsequent treatment for recurrence and metastasis, and excluded indications for brachytherapy [1]. Thus, we cannot directly compare our observed rate with their estimate of the appropriate rate. However, given that relatively few patients with endometrial cancer go on to develop recurrence or metastases, the rate we have observed seems to fall far short of the Delaney et al. [1] estimate of the appropriate rate. Moreover, although the rate of use of adjuvant radiotherapy varied widely across Ontario, the observed rate did not approach the estimated appropriate rate in any part of the province. Nonetheless, we hesitate to conclude that radiotherapy is under-utilised in Ontario, because the indications for radiotherapy in this disease remain controversial and the evidentiary basis of the evidence-based rate is not very strong. An alternative approach to establishing the ‘right’ rate of radiotherapy is to set a ‘benchmark’ based on the rates of use of radiotherapy observed in circumstances where access to radiotherapy is not constrained, and decisions about its use are made in a multidisciplinary setting [25]. However, we found that the rate of use of radiotherapy in this disease varies widely even among the CGOCs, where patients have ready access to radiotherapy and gynaecological and radiation oncologists work closely with one another. In fact, the variation in rates observed in these more sophisticated hospitals was just as great as it was
T.P. Hanna et al. / Clinical Oncology 24 (2012) e113ee124
among other hospitals. This makes it impossible to identify a robust benchmark rate for the use of radiotherapy in endometrial cancer, as we have done successfully in several other diseases [25e27]. The results of our study suggest that longer prevailing waiting times for adjuvant radiotherapy may be associated with lower rates of use of adjuvant radiation for endometrial cancer. This may be due to lower rates of referral, or to less frequent prescription of radiation after referral when prevailing waiting times are longer. Prevailing waiting times have been associated with variations in the use of radiation in some, but not all studies. Zhang-Salomons et al. [28] found that elderly patients were found less likely to receive post-lumpectomy radiation for breast cancer during longer prevailing waiting times for radiotherapy in Ontario. However, Huang [23] found no association between prevailing waiting times for radiotherapy and the choice of radiation versus surgery for prostate cancer. Prevailing waiting times may also affect the way radiotherapy is delivered; Kong et al. [12] found that a longer prevailing waiting time was associated with the use of shorter fractionation schemes for bone metastases. In keeping with the results of the relevant clinical trials, we found no association between the rate of use of postoperative radiotherapy and survival in the overall study population. Regrettably, for reasons discussed above, we were unable to determine whether lower rates of radiotherapy were associated with higher rates of pelvic recurrence, as predicted by the results of clinical trials. However, in a subpopulation of higher risk patients with clear cell or serous histologies, we observed worse survival among residents of regions with lower rates of use of radiotherapy. The association between low radiotherapy use and poorer survival in this subgroup was clinically and statistically significant in both the univariate and multivariate analyses. Although the stage distribution observed in regions with low rates of use of radiotherapy was similar to that observed in other regions of the province, it remains possible that the observed differences in survival reflect differences in case mix rather than differences in treatment. The association between radiation use and survival might have been confounded by the use of adjuvant chemotherapy, but we think this is unlikely. The period under study (1992e2003) was before the publication of the commonly cited 2007 study by Hogberg et al. [29] that suggested a small benefit in disease-free survival with chemotherapy in patients with high-risk endometrial cancer. Even if chemotherapy had been widely used between 1992 and 2003, it would have to be used much more in regions with high radiation use for any confounding to occur, and there is no reason to suspect this. Finally, if the evidence for an overall survival benefit of adjuvant chemotherapy for high-risk endometrial cancer is limited, it is even more so specifically for clear cell and serous carcinoma [30,31].
Conclusions This study provides evidence that factors unrelated to patient’s needs influenced the use of adjuvant radiation for
e123
endometrial cancer in Ontario between 1992 and 2003, and that lower rates of use of radiation were associated with poorer survival among a subset of the patients who would be expected to benefit most from radiation.
Acknowledgements The authors wish to acknowledge the support of the Canadian Institute of Health Research (CIHR) Fellowship in Policy and Health Services Research, Queen’s University and CIHR Transdisciplinary Training Program in Cancer Research Fellowship, CIHR New Emerging Team (NET) Grant, Cancer Care Ontario (CCO) and the Ontario Cancer Registry (OCR).
References [1] Delaney G, Jacob S, Barton M. Estimation of an optimal radiotherapy utilization rate for gynecologic carcinoma: part II e carcinoma of the endometrium. Cancer 2004;101:682e692. [2] Johnson N, Cornes P. Survival and recurrent disease after postoperative radiotherapy for early endometrial cancer: systematic review and meta-analysis. BJOG 2007;114:1313e1320. [3] Lee CM, Szabo A, Shrieve DC, et al. Frequency and effect of adjuvant radiation therapy among women with stage I endometrial adenocarcinoma. JAMA 2006;295:389e397. [4] Blake P, Swart AM, Orton J, et al. Adjuvant external beam radiotherapy in the treatment of endometrial cancer (MRC ASTEC and NCIC CTG EN.5 randomised trials): pooled trial results, systematic review, and meta-analysis. Lancet 2009; 373:137e146. [5] Creutzberg CL, van Putten WL, Koper PC, et al. Surgery and postoperative radiotherapy versus surgery alone for patients with stage-1 endometrial carcinoma: multicentre randomised trial. PORTEC Study Group. Post Operative Radiation Therapy in Endometrial Carcinoma. Lancet 2000;355:1404e1411. [6] Keys HM, Roberts JA, Brunetto VL, et al. A phase III trial of surgery with or without adjunctive external pelvic radiation therapy in intermediate risk endometrial adenocarcinoma: a Gynecologic Oncology Group study. Gynecol Oncol 2004;92:744e751. [7] Nout RA, Smit VT, Putter H, et al. Vaginal brachytherapy versus pelvic external beam radiotherapy for patients with endometrial cancer of high-intermediate risk (PORTEC-2): an open-label, non-inferiority, randomised trial. Lancet 2010; 375:816e823. [8] Lukka H, Chambers A, Fyles A, et al. Adjuvant radiotherapy in women with stage I endometrial cancer: a clinical practice guideline 2006. Available at: https://www.cancercare.on.ca/ common/pages/UserFile.aspx?serverId¼6&path¼/File%20Data base/CCO%20Files/PEBC/pebc4-10f.pdf. [9] Kwon JS, Carey MS, Cook EF, et al. Patterns of practice and outcomes in intermediate- and high-risk stage I and II endometrial cancer: a population-based study. Int J Gynecol Cancer 2007;17:433e440. [10] Paszat LF, Mackillop WJ, Groome PA, et al. Radiotherapy for breast cancer in Ontario: rate variation associated with region, age and income. Clin Invest Med 1998;21:125e134. [11] Huang J, Zhou S, Groome P, et al. Factors affecting the use of palliative radiotherapy in Ontario. J Clin Oncol 2001;19: 137e144. [12] Kong W, Zhang-Salomons J, Hanna TP, et al. A populationbased study of the fractionation of palliative radiotherapy
e124
[13]
[14]
[15]
[16]
[17] [18]
[19]
[20]
[21]
[22]
[23]
T.P. Hanna et al. / Clinical Oncology 24 (2012) e113ee124 for bone metastasis in Ontario. Int J Radiat Oncol Biol Phys 2007;69:1209e1217. Booth CM, Mackillop WJ. Translating new medical therapies into societal benefit: the role of population-based outcome studies. JAMA 2008;300:2177e2179. Robles SC, Marrett LD, Clarke EA, et al. An application of captureerecapture methods to the estimation of completeness of cancer registration. J Clin Epidemiol 1988;41:495e501. Clarke EA, Marrett LD, Kreiger N. Cancer registration in Ontario: a computer approach. IARC Scientific Publications 1991:246e257. Mackillop WJ, Zhang-Salomons J, Groome PA, et al. Socioeconomic status and cancer survival in Ontario. J Clin Oncol 1997;15:1680. Elixhauser A, Steiner C, Harris DR, et al. Comorbidity measures for use with administrative data. Med Care 1998;36:8e27. Southern DA, Quan H, Ghali WA. Comparison of the Elixhauser and Charlson/Deyo methods of comorbidity measurement in administrative data. Med Care 2004;42: 355e360. Stukenborg GJ, Wagner DP, Connors Jr AF. Comparison of the performance of two comorbidity measures, with and without information from prior hospitalizations. Med Care 2001;39: 727e739. de Groot V, Beckerman H, Lankhorst GJ, et al. How to measure comorbidity. A critical review of available methods. J Clin Epidemiol 2003;56:221e229. Ministry of Health and Long Term Care Finance & Information Management Branch Information Management Unit (editors). Ministry of Health and Long Term Care Master Numbering System (CIHI Institution Numbers). Ontario, 2006. Bardell T, Belliveau P, Kong W, et al. Waiting times for cancer surgery in Ontario: 1984e2000. Clin Oncol (R Coll Radiol) 2006;18:401e409. Huang M. The influence of waiting time on treatment choices among prostate cancer patients. Queen’s University, Department of Community Health and Epidemiology, 2006: 120.
[24] Kwon JS, Carey MS, Cook EF, et al. Are there regional differences in gynecologic cancer outcomes in the context of a single-payer, publicly-funded health care system? A population-based study. Can J Public Health 2008;99: 221e226. [25] Barbera L, Zhang-Salomons J, Huang J, Tyldesley S, Mackillop WJ. Defining the need for radiotherapy for lung cancer in the general population: a criterion-based, benchmarking approach. Medical Care 2003;41:1074e1085. [26] Kerba M, Miao Q, Zhang-Salomons J, Mackillop WJ. Defining the need for breast cancer radiotherapy in the general population: a criterion-based benchmarking approach. Clin Oncol 2007;19:481e489. [27] Kerba M, Miao Q, Zhang-Salomons J, Mackillop W. Defining the need for prostate cancer radiotherapy in the general population: a criterion-based benchmarking approach. Clin Oncol 2010;22:801e809. [28] Zhang-Salomons J, Fourodi F, Huang J, et al. Health system effects on the use of radiotherapy following surgery for breast cancer. Abstr Acad Health Serv Res Health Policy Meet 2002;19:45. [29] Hogberg T, Rosenberg P, Kristensen G, et al. A randomized phase-III study on adjuvant treatment with radiation (RT) chemotherapy (CT) in early-stage high-risk endometrial cancer (NSGO-EC-9501/EORTC 55991). J Clin Oncol 2007;25: 5503. [30] Hogberg T, Rosenberg P, Kristensen G, et al. A randomized phase-III study on adjuvant treatment with radiation (RT) chemotherapy (CT) in early-stage high-risk endometrial cancer (NSGO-EC-9501/EORTC 55991). Online oral presentation. Available at: http://www.asco.org/ASCOv2/MultiMedia/ VirtualþMeeting?&vmview¼vm_session_presentations_ view&confID¼47&sessionID¼2213. [31] Randall ME, Filiaci VL, Muss H, et al. Randomized phase III trial of whole-abdominal irradiation versus doxorubicin and cisplatin chemotherapy in advanced endometrial carcinoma: a Gynecologic Oncology Group Study. J Clin Oncol 2006;24: 36e44.