Abstract
Objective
To determine the impact of radiation treatment for corpus uteri cancer on the development of late gastrointestinal (GI) and bladder toxicities in older women.
Methods
We used a linked database of the Surveillance, Epidemiology and End Results (SEER) program tumor registry records and Medicare claims to estimate the risk of late radiation toxicities in a population based sample of 23,591 women diagnosed with corpus uteri cancer from 1992-2005. Gastrointestinal (GI) and bladder diagnoses were included up to 60 months after diagnosis. Cox proportional hazard models were used to estimate risk of any late GI or bladder diagnosis with or without hospitalization, due to radiation.
Results
Women treated with radiation had an absolute increase in late GI toxicities of 7.7% and an absolute increase in late bladder toxicities of 3.9%. Most toxicities were not severe. Only 10% of irradiated women with a lower GI diagnosis and less than 1% of irradiated women with a bladder diagnosis, required hospitalization. In the multivariable analysis, advancing age, having chemotherapy, higher comorbidities and prior or early GI and bladder diagnoses were predictive of late toxicities among irradiated women.
Conclusions
Late GI toxicities are more common than previously reported. It is important to analyze late toxicities in population based samples of women with uterine cancer. The results of this and future analyses can be used to identify areas where development of “Survivorship Plans” will remind and encourage women to report symptoms related to potential toxicities.
Keywords: Late toxicities, radiation, uterine cancer, older women, SEER-Medicare database
Introduction
Uterine cancer is the fourth most common cancer affecting women in the United States (US) [1]. The incidence for uterine cancer has been calculated to be 23.3 per 100,000 women [1]. There are two main histological types of uterine cancer: endometrial carcinoma and sarcomas. Sarcomas only represent 5% of all the uterine cancers [2]. Survival rates for uterine cancer overall are high, with a 5-year relative survival rate of about 90% reported for women diagnosed between 1992 and 2001 [3]. Assessment of late side effects related to definitive uterine cancer therapies is very important due to the potential for long-term survival.
Radiotherapy (either brachytherapy or external pelvis radiation) is an integral part of post-operative adjuvant therapy for most patients with high risk features [4, 5]. Other adjuvant therapies like hormonal therapy and chemotherapy are individualized and mostly recommended for more advanced stages or patients with metastases [4].
The gastrointestinal (GI) tract and the urinary bladder are the most common sites of toxicities after radiotherapy [6]. Those toxicities occurring after six months have been labeled late toxicities. Late toxicities affect the quality of life of patients and cause concerns among physicians caring for these patients [7, 8]. Studies have reported that the rates of severe (RTOG/EORTC grades 3 and 4 late radiation morbidity score) GI and bladder toxicities after radiation for endometrial cancer are about 8% [6]. Rates of less severe GI and bladder symptoms however may be significantly higher and require further study [9, 10].
Multimodality therapy of postoperative radiation has been shown to improve disease-free survival with a concurrent increase in GI and bladder toxicity profiles [11, 12]. Enhancement in radiation treatment, planning and delivery, has demonstrated a reduction in potential toxicity to organs at risk [10, 13]. Tumor directed targeted therapy with consideration of dose and fractionation are strategies to reduce toxicity and some investigators advocate the use of medications as radioprotectants as well [10, 14, 15].
Most reports on the frequency and risks of toxicities related to radiotherapy come from clinical trials, case reports and single institution studies. This body of data is representative only of a highly selected group of women and is subject to bias, has a short follow-up interval of the subjects and includes a small number of older women. Population based data is a powerful source of information that can help us determine the prevalence and severity of late toxicities among older women after radiotherapy for uterine cancer. Our study analyzes the incidence of GI and bladder toxicities after radiation therapy in a population-based sample of older women diagnosed with uterine cancer.
Methods
Data Source
The source of data for this study is the SEER-Medicare linked database. The SEER program is a national population based tumor registry that gathers information on incidence of cancer cases. Medicare is the largest health insurer covering 97% of adults over 65 years of age in the US. A link between eligible SEER subjects and their Medicare claims has been established through an agreement between the National Cancer Institute and the Centers for Medicare and Medicaid Services [16]., Persons with cancer diagnosed between 1973 and 2005 have been linked and their complete Medicare claims are available between 1991 and 2005. The match rate of persons aged 65 and older with a cancer diagnosis in SEER was 93% when compared to their Medicare enrollment records. Claims data for this study are available in separate files for hospital services (Medicare Provider Analysis and Review File (MEDPAR)), physician and medical services (Carrier File), and outpatient services (Hospital Outpatient Standard Analytic File (OUTSAF).
Population
Our study population includes women in the SEER-Medicare database with uterine cancer diagnosis over the period 1992 -2005. A total of 59,420 women were identified. To ensure complete information, cancer cases identified from autopsy reports or death certificates alone (132 cases), and women with multiple cancer sites (11,309 cases) were excluded. Women without Medicare part A and part B between 1992-2005 (5,779 cases) and women enrolled in an HMO and women with only Medicare part A over the period 1992-2005 (5,267 cases) were also excluded. The radiation category included women with beam radiation, brachytherapy or combination of both. This resulted in a total sample of 23,591 women.
Demographic characteristics, dates of diagnosis, and disease extension were obtained from the SEER-Medicare PEDSF file for all patients. Socioeconomic status was indirectly measured by Medicaid eligibility [17]. Comorbidity scores were calculated using the Klabunde modification of the Charlson index based on claims data in the year prior to uterine cancer diagnosis [18, 19].
Treatment
Radiation was defined as receiving beam radiation, radioactive implants or a combination of both based on SEER database codes 1-6, or having a Medicare radiation claim 1-6 months after diagnosis of uterine cancer. The specific International Classification of Diseases, 9th revision (ICD-9-CM) codes and Healthcare Common Procedure Coding System (HCPCS) codes used to identify claims were: V580, V661, V671, or procedure codes 9221-9229, or HCPCS codes 77401-77499 and 77750-77799.
Adjuvant treatment with chemotherapy and surgery were used as covariates. Chemotherapy was defined using ICD-9CM DX codes V58.1, V66.2, V67.2 or ICD-9-CM procedure codes 992.5 or HCPCS codes 96400-96549, J9000-J9999, Q0083-Q0085, or revenue center codes 0331, 0332, 0335. Surgery was defined using SEER site specific surgery codes 20,30,35,40,50,60,70,61-64,71,72,80-82.
Study Outcomes
GI and bladder diagnoses were included from 6 to 60 months after diagnosis. These diagnoses were defined by selected ICD-9 and Current Procedural Terminology (CPT) codes. A complete list of codes included in this study is provided in Table 1. We classified diagnoses as follows: 1) “Any gastrointestinal or bladder diagnosis,” defined as relevant GI or bladder diagnoses identified on inpatient (MEDPAR file), outpatient (Outpatient file), or physician (Carrier file) claims; 2) “hospitalizations,” defined as GI or bladder diagnoses serious enough to require hospitalization (MEDPAR file); and 3) “procedures,” defined as GI or bladder procedures or surgical interventions (MEDPAR, Outpatient, and Carrier file). For “any diagnosis” and “procedures,” all relevant claims were included, regardless of ICD-9-CM diagnosis position. For “hospitalizations,” only diagnosis in the first or second positions were examined.
Table 1.
GI and bladder diagnoses and procedure ICD-9 and CPT codes used for study
| GI Diagnoses | ICD-9 codes |
| Ischemia | 557.0, 557.1 |
| Inflammation | 558.1, 558.2, 787.91, 562.0, 562.1, 555.1 |
| Hemorrhage | 578.1, 578.9, 569.3 |
| Stricture or obstruction | 569.2, 560.8, 560.9 |
| Ulcers and perforations | 569.82, 569.83, 569.41 |
| Fistulas | 565, 619.0, 619.1, 619.2, 619.9, 537.4, 569.81, 596.1 |
| GI Procedures | ICD-9 and CPT codes |
| Endoscopies | 45.21, 45.23, 45.24, 48.21, 48.23, 49.21. 44360, 44366, 44376, 44385, 45300-45345, 45355-45387, 46600-46615 |
| Surgical interventions or procedures | 45.00, 45.03, 45.7, 45.8, 46.1, 46.2, 46.75, 48.5, 48.9, 48.7, 49.71. 44140-44160, 44204-44212, 45110-45123, 46.76, 47.92, 49.73. 49.1, 48.73, 48.93, 70.72-70.74. 45000, 45005, 45020, 45800, 45805, 45820, 45825, 46.80, 46.81, 96.22, 96.23. 45500, 45905, 45910, 46700 |
| Bladder Diagnoses | ICD-9 codes |
| Cystitis | 595.1, 595.2, 595.3, 595.82, 595.9 |
| Incontinence | 788.2, 788.30, 788.31, 788.34, 788.37, 788.38 |
| Hemorrhage | 596.7, 599.7 |
| Obstruction | 599.6 |
| Fistulas | 619.0, 619.1, 596.1, 596.2 |
| Bladder Procedures | ICD-9 and CPT codes |
| Cystoscopies | 57.32, 57.33. 52000, 52001, 52005, 52007 |
| Surgical interventions or procedures | 57.6, 57.7, 57.81, 57.85, 57.87, 57.91, 57.92. 51725-51798, 51520, 51525, 51530, 51535, 51550, 51555, 51565, 51570, 51575, 51580, 51585, 51590, 51595, 51596, 51800, 51820, 51960, 51980, 52204, 52214, 52260, 52310, 52315, 53400, 53405, 57.0, 57.1, 57.41, 58.5, 58.6. 51600, 51605, 51610, 51700, 51701-51703, 51736, 51741, 51795, 51797, 51798, 56.84, 57.83, 57.84, 58.43. 44660, 44661, 51900, 51920, 51925, 57320, 57330 |
GI = Gastrointestinal; ICD-9 = International Classification of Diseases, 9th version; CPT = Current Procedural Terminology
Statistical Analyses
Patient characteristics were compared between women who had radiation therapy (n=8,006) and those who did not have radiation (n= 15,585) using chi-square test for categorical variables and student t-test for continuous variables. Adjusted time-to-event curves were estimated from stratified Cox Proportional Hazard Models (n=23, 591). The Cox models created were then used to analyze predictors of GI and bladder toxicities in women who received radiation (n=8,006) and included radiation (yes vs. no), age, tumor stage1, tumor grade, ethnicity, Medicaid eligibility, chemotherapy, surgery, comorbidity index, and diagnosis year.
Using only the group of women diagnosed with uterine cancer between 1992-2002 who had complete 60 month follow-up information (n=8,797), we plotted the percentage of patients with any GI or bladder diagnosis for each month starting 12 months prior to uterine cancer diagnosis through 60 months after diagnosis. Diagnoses were divided in two categories due to the clear differences in the frequency of diagnoses around the time of radiation treatment compared to the frequency of late diagnosis. These categories are consistent with known patterns of toxicity related to radiation [7]. Diagnoses occurring within 6 months of first radiation were labeled “early toxicities” and those occurring between 6 and 60 months after initial radiation were labeled “late toxicities”. Data were plotted and resulting curves were compared with General Linear Model repeated ANOVA in the 6-60 month time period to determine significant differences by treatment group and time.
In order to identify differences by type of GI or bladder diagnosis and type of GI or bladder procedures over time, chi-square tests were used to compare women who received radiation to women who did not receive radiation. All p-values were two-sided. Statistical analyses were performed using the SAS software version 9.2 (SAS Institute, Inc., Cary, NC).
Results
Of the total number of women with uterine cancer included in this study, 8,006 (33.9%) received radiation. As shown in Table 2, women who received radiation were older, were more likely to have uterine cancer diagnosed at more advanced stages, and also were more likely to have poor cancer differentiation (grades 3 and 4).
Table 2.
Patient characteristics by radiation status
| n = 23,591 | Radiation (%) n=8,006 | No Radiation (%) n=15,585 | p-value | |
|---|---|---|---|---|
| Age group | ||||
| ≤ 64 | 1,493 | 5.1 | 7.0 | |
| 65-69 | 4,926 | 21.8 | 20.4 | |
| 70-74 | 5,887 | 26.8 | 24.0 | |
| 75-79 | 5,225 | 23.3 | 21.6 | |
| 80+ | 6,060 | 23.0 | 27.1 | <0.0001 |
| Race group | ||||
| White | 20,161 | 84.6 | 85.9 | |
| Black | 1,765 | 8.2 | 7.1 | |
| Hispanic | 917 | 4.1 | 3.8 | |
| Other | 748 | 3.1 | 3.2 | 0.0087 |
| Marriage status | ||||
| All others | 14,031 | 58.4 | 60.0 | |
| Married | 9,560 | 41.6 | 40.0 | 0.0191 |
| Medicaid Eligible | ||||
| Yes | 5,143 | 20.8 | 22.3 | 0.0067 |
| AJCC Stage | ||||
| Stage 0/I* | 15,433 | 51.4 | 72.6 | |
| Stage II | 2,017 | 16.2 | 4.6 | |
| Stage III | 2,147 | 17.1 | 5.0 | |
| Stage IV | 2,351 | 10.3 | 9.8 | <0.0001 |
| Grade | ||||
| grade1 | 7,265 | 17.2 | 37.7 | |
| grade2 | 6,993 | 32.5 | 28.2 | |
| grade3 | 5,116 | 6.6 | 4.2 | |
| grade4 | 1,173 | 12.4 | 13.2 | <0.0001 |
| Acute Myocardial Infarction | ||||
| Yes | 224 | 0.9 | 1.0 | 0.8852 |
| Diabetes | ||||
| Yes | 4,499 | 20.1 | 18.5 | 0.0032 |
| Peripheral Vascular Disease | ||||
| Yes | 539 | 2.2 | 2.3 | 0.4662 |
| Chemotherapy | ||||
| Yes | 5,642 | 29.8 | 20.9 | <0.0001 |
| Cancer directed surgery | ||||
| Had surgery | 20,648 | 86.9 | 87.9 | 0.0365 |
| Comorbidity Index | ||||
| 0 | 15,624 | 66.3 | 66.2 | |
| 1 | 4,940 | 21.1 | 20.9 | |
| 2 | 1,835 | 7.5 | 7.9 | |
| ≥ 3 | 1,192 | 5.1 | 5.0 | 0.7182 |
Stage 0 and I were combined due to confidentiality issues related to the small number of women in stage 0 in our sample; in the combined stage 0/I group, only 2.1% of women belong to stage 0.
AJCC = American Joint Committee on Cancer ; GI = Gastrointestinal; SEER = Surveillance, Epidemiology and End Results (SEER) Program; The SEER-Medicare database has Medicare claims starting in 1991. For this reason staging for cancer is based on the AJCC 1988 staging system.
Figure 1 shows the percentage of women who had lower GI or bladder diagnoses for each month, starting 12 months before diagnosis and extending to 60 months after uterine cancer diagnosis. Separate curves were depicted for those patients who underwent radiation and those who did not. There were increases in early toxicities for both groups around the time of the cancer diagnosis (Figures 1 a and b). The trend was higher for GI toxicities, approximately 6 months after diagnosis and remained elevated (p<.0001). Late GI toxicities appear to peak around 12 to 24 months after diagnosis. The trend for bladder toxicities is less pronounced but still significant (p<.0001).
Figure 1.
Percentage of patients with a Medicare claim for (A) gastrointestinal (GI) late toxicities and (B) bladder late toxicities, for every month between 12 months before and 60 months after cancer diagnosis, stratified by radiation treatment (n=8,797)
(A) Between radiation group effect: F value=151.6, p<.0001 Time*radiation effect: F value=1.98, p<.0001
(B) Between radiation group effect: F value=20.56, p<.0001 Time*radiation effect: F value=1.30, p=.07
We then determined the percentage of women with specific lower GI and bladder diagnoses from one year before diagnosis until five years after radiation. For these analyses, we limited the patient population to the 8,797 women who had complete follow-up for 5 years (Table 3). Overall, no important differences in toxicity rates were observed prior to uterine cancer diagnosis. Most early GI toxicities were significantly higher for the radiation group compared to the no radiation group. Conversely, no significant differences in early bladder toxicities were observed between the two groups. Women treated with radiation had an absolute increase in late GI toxicities of 7.7% and an absolute increase in late bladder toxicities of 3.9%.
Table 3.
Percentages of the 8,797 women with 60 months of complete follow-up who had GI and bladder diagnoses, hospitalizations and procedures.
| GI Diagnosis | Before Dx Toxicity (12 months before) | Early Toxicity (1-6 months) | Late Toxicity (6-60 months) | ||||
|---|---|---|---|---|---|---|---|
| Count | % | Count | % | Count | % | ||
| Any GI toxicity | |||||||
| Radiation | 381 | 14.0 | 595 | 21.9 | 1653 | 60.8 | |
| No Radiation | 968 | 15.9 | 1062 | 17.5 | 3224 | 53.1 | |
| p | 0.0208 | <0.0001 | <0.0001 | ||||
| Ischemia | |||||||
| Radiation | <11 | <1.0* | <11 | <1.0* | 46 | 1.7 | |
| No Radiation | <11 | <1.0* | 22 | 0.4 | 53 | 0.9 | |
| p | 0.6870 | 0.0165 | 0.0008 | ||||
| Inflammation | |||||||
| Radiation | 159 | 5.8 | 341 | 12.5 | 1148 | 42.2 | |
| No Radiation | 457 | 7.5 | 512 | 8.4 | 2104 | 34.6 | |
| p | 0.0044 | <0.0001 | <0.0001 | ||||
| Hemorrhage | |||||||
| Radiation | 162 | 6 | 133 | 4.9 | 857 | 31.5 | |
| No Radiation | 394 | 6.5 | 300 | 4.9 | 1405 | 23.1 | |
| p | 0.3473 | 0.9251 | <0.0001 | ||||
| Stricture/Obstruction | |||||||
| Radiation | 23 | 0.8 | 107 | 3.9 | 336 | 12.4 | |
| No Radiation | 31 | 0.5 | 208 | 3.4 | 373 | 6.1 | |
| p | 0.0626 | 0.2331 | <0.0001 | ||||
| Ulcers/Perforations | |||||||
| Radiation | <11 | <1.0* | <11 | <1.0* | 55 | 2 | |
| No Radiation | <11 | <1.0* | 22 | 0.4 | 71 | 1.2 | |
| p | 0.1521 | 0.8198 | 0.0018 | ||||
| Fistula | |||||||
| Radiation | <11 | <1.0* | 14 | 0.5 | 67 | 2.5 | |
| No Radiation | 17 | 0.3 | 38 | 0.6 | 87 | 1.4 | |
| p | 0.4909 | 0.5317 | 0.0007 | ||||
| GI Hospitalization (Include all primary Dx codes only for those hospitalized) | |||||||
| Radiation | 14 | 0.5 | 54 | 2.0 | 274 | 10.1 | |
| No Radiation | 29 | 0.5 | 56 | 0.9 | 315 | 5.2 | |
| p | 0.8157 | <0.0001 | <0.0001 | ||||
| GI Procedures | |||||||
| Endoscopies | |||||||
| Radiation | 203 | 7.5 | 219 | 8.1 | 1148 | 42.2 | |
| No Radiation | 562 | 9.2 | 409 | 6.7 | 2143 | 35.3 | |
| p | 0.0060 | 0.0261 | <0.0001 | ||||
| Surgical Procedures | |||||||
| Radiation | <11 | <1.0* | 13 | 0.5 | 85 | 3.1 | |
| No Radiation | 11 | 0.2 | 49 | 0.8 | 116 | 2.1 | |
| p | 04640 | 0.0888 | 0.0029 | ||||
| Bladder Diagnosis | Before Dx toxicity (12 months before) | Early Toxicity (1-6 months) | Later toxicity (6-60 months) | ||||
|---|---|---|---|---|---|---|---|
| Count | % | Count | % | Count | % | ||
| Any Bladder toxicity | |||||||
| Radiation | 366 | 13.5 | 388 | 14.3 | 974 | 35.8 | |
| No Radiation | 813 | 13.4 | 941 | 15.5 | 1941 | 31.9 | |
| p | 0.9214 | 0.1398 | 0.0004 | ||||
| Cystitis | |||||||
| Radiation | 86 | 3.2 | 67 | 2.5 | 271 | 10 | |
| No Radiation | 178 | 2.9 | 142 | 2.3 | 448 | 7.4 | |
| p | 0.5544 | 0.7187 | <0.0001 | ||||
| Incontinence | |||||||
| Radiation | 65 | 2.4 | 125 | 4.6 | 470 | 17.3 | |
| No Radiation | 200 | 3.3 | 388 | 6.4 | 953 | 15.7 | |
| p | 0.0223 | 0.0009 | 0.0601 | ||||
| Hemorrhage | |||||||
| Radiation | 248 | 9.1 | 128 | 4.7 | 471 | 17.3 | |
| No Radiation | 455 | 7.5 | 261 | 4.3 | 836 | 13.8 | |
| p | 0.0092 | 0.3862 | <0.0001 | ||||
| Obstruction | |||||||
| Radiation | <11 | <1.0* | 12 | 0.4 | 24 | 0.9 | |
| No Radiation | <11 | <1.0* | 11 | 0.2 | 40 | 0.7 | |
| p | 0.2850 | 0.0272 | 0.2529 | ||||
| Fistula | |||||||
| Radiation | <11 | <1.0* | <11 | <1.0* | 18 | 0.7 | |
| No Radiation | <11 | <1.0* | 21 | 0.3 | 24 | 0.4 | |
| p | 0.5342 | 0.3274 | 0.0934 | ||||
| Bladder Hospitalization (Include all primary Dx codes only for those hospitalized) | |||||||
| Radiation | <11 | <1.0* | <11 | <1.0* | 15 | 0.6 | |
| No Radiation | <11 | <1.0* | <11 | <1.0* | 18 | 0.3 | |
| p | 0.4089 | 0.4089 | 0.0703 | ||||
| Bladder Procedures | |||||||
| Cistoscopy | |||||||
| Radiation | 112 | 4.1 | 108 | 4.0 | 277 | 10.2 | |
| No Radiation | 194 | 3.2 | 252 | 4.1 | 430 | 7.1 | |
| p | 0.0286 | 0.6999 | <0.0001 | ||||
| Surgical Procedures | |||||||
| Radiation | 41 | 1.5 | 67 | 2.5 | 272 | 10.0 | |
| No Radiation | 93 | 1.5 | 208 | 3.4 | 438 | 7.2 | |
| p | 0.9351 | 0.0168 | <0.0001 | ||||
Due to confidentiality issues, counts with less than 11 women are presented as “<11” and the corresponding percentages are reported as “<1%”.
GI = Gastrointestinal; Dx = Diagnosis
The SEER-Medicare database has Medicare claims starting in 1991. For this reason staging for cancer is based on the 1988 AJCC staging system
The most common late GI toxicity was inflammation such as colitis and diarrhea; rates were 7.6% higher among irradiated women (42.2% of irradiated women vs. 34.6% of non-irradiated women) (Table 3). The most common bladder toxicities were hemorrhage and incontinence; hematuria rates were 1.6% higher among irradiated women compared to those non-irradiated and incontinence rates were 3.5% higher. A total of 10.1% of irradiated women experienced lower GI toxicities serious enough to require hospitalization over the 5-year period, compared to 5.2% of women who did not receive radiation. For bladder toxicities the percentage was considerably lower; less than 1% of women who received radiation required hospitalization for late bladder toxicities.
When medical procedures were examined, the rate of diagnostic endoscopies was 7% higher in irradiated women and the difference was statistically significant (p<.0001) (Table 3). Also, the percentage of patients who required lower GI surgical procedures was significantly higher in women who received radiation compared to women who did not receive radiation (3.1% vs. 2.1%, p=0.003). Two fifths of irradiated women had a lower GI endoscopy over the five years after cancer diagnosis and one tenth of irradiated women had a cystoscopy. A 2.8% increase in risk of having bladder surgical procedures was also observed and the difference was significant (p<.0001).
Hazard ratios (HR) of late GI and bladder toxicities were obtained for the entire sample (n=23,591). Two models were created for each toxicity type. For late GI toxicities the HR was 1.38 with a 95% confidence interval (95% CI) between 1.33 and 1.43. After adjusting for all sociodemographic covariates, comorbidities and concomitant treatments (surgery and chemotherapy), the magnitude of the HR decreased but remained statistically significant (HR 1.31, 95% CI 1.26-1.36). Similarly, for late bladder toxicities, the unadjusted ratio was HR 1.22, 95% CI 1.16-1.27 and after controlling for all covariates the magnitude decreased but remained statistically significant (HR 1.13, 95% CI 1.08-1.19). We additionally analyzed the interaction between radiation and all covariates for GI and bladder toxicity in the complete sample and the interaction between grade 4 vs. 1 and radiation (parameter estimate 0.28; p=.005) was the only significant one at the 0.05 level. This results can be observed in Figure 2.
Figure 2.
Adjusted1 time-to-event curves for (A) time until any late gastrointestinal (GI) toxicity and (B) time until any late bladder toxicity, stratified by radiation group (n=23,591)
Adjusted time to event curves using the total population of women diagnosed with uterine cancer (n=23,591) are depicted in Figure 2a for GI toxicities and in Figure 2b for bladder. Figure 2a depicts the adjusted curve for women with GI toxicities. Women receiving radiation had higher toxicity rates compared to women who did not receive radiation. Five years after initial treatment, irradiated women had 8.1% higher risk of late GI toxicities compared to non irradiated women (Absolute Risk [AR] 0.081; 95% Confidence Interval [95% CI] 0.065-0.098). Figure 2b depicts the adjusted curve for women with bladder toxicities. Similarly, irradiated women had higher toxicity rates. The absolute risk for bladder toxicities was 3.5% for irradiated women compared to non irradiated women (AR 0.035; 95% CI 0.019-0.051).
Table 4 shows the odds associated with developing any GI and bladder diagnosis among women receiving radiation. Multivariable analyses found that women with advancing age, those who had chemotherapy, those with higher comorbidities, those with GI or bladder diagnoses 12 months prior to cancer diagnosis and women with early GI or bladder toxicities had statistically significant with higher risk of both GI and bladder late toxicities. Higher tumor grade only increased the risk for GI late toxicities while being Medicaid eligible only increased the risk of late bladder toxicities. Having surgery decreased the risk of late bladder toxicities.
Table 4.
Multivariate analysis of factors associated with any late GI and bladder diagnosis or hospitalization with a late GI and bladder diagnosis in 8,006 women treated with primary radiation therapy.
| Any late GI diagnosis adjusted HR (95% CI) | Any late bladder diagnosis adjusted HR (95% CI) | |
|---|---|---|
| Age (every 10 year increase) | 1.04 (1.00-1.09) | 1.08 (1.02-1.14) |
| Ethnicity | ||
| White | 1.00 | 1.00 |
| Black | 0.93 (0.81-1.06) | 0.93 (0.78-1.11) |
| Hispanic | 0.93 (0.79-1.09) | 1.03 (0.83-1.27) |
| Other | 0.87 (0.70-1.07) | 0.88 (0.66-1.15) |
| AJCC Stage | ||
| Stage 0 | 1.00 | 1.00 |
| Stage I | 0.91 (0.29-2.84) | 0.22 (0.08-0.61) |
| Stage II | 0.99 (0.32-3.09) | 0.26 (0.10-0.71) |
| Stage III | 1.03 (0.33-3.22) | 0.26 (0.09-0.70) |
| Stage IV | 1.25 (0.40-3.92) | 0.34 (0.13-0.94) |
| Tumor Grade | ||
| Grade1 | 1.00 | 1.00 |
| Grade2 | 0.94 (0.86-1.02) | 0.88 (0.78-0.98) |
| Grade3 | 1.04 (0.95-1.14) | 0.95 (0.85-1.08) |
| Grade4 | 1.24 (1.07-1.43) | 1.06 (0.88-1.29) |
| Medicaid eligible (yes) | 0.97 (0.89-1.06) | 1.27 (1.14-1.41) |
| Chemotherapy (yes) | 1.35 (1.26-1.44) | 1.45 (1.32-1.58) |
| Surgery group (yes) | 0.92 (0.82-1.04) | 0.74 (0.64-0.86) |
| Charlson Comorbidity Index | ||
| 0 | 1.00 | 1.00 |
| 1 | 1.14 (1.05-1.23) | 1.19 (1.07-1.31) |
| 2 | 1.21 (1.07-1.36) | 1.17 (0.99-1.37) |
| 3 | 1.36 (1.17-1.57) | 1.32 (1.09-1.60) |
| Had previous GI diagnosis (12 months prior) | 1.41 (1.31-1.53) | |
| Had early GI diagnosis (0-6 months after) | 1.45 (1.35-1.55) | |
| Had previous bladder diagnosis (12 months prior) | 1.51 (1.36-1.67) | |
| Had early bladder diagnosis (0-6 months after) | 1.75 (1.59-1.93) |
Models also adjusted for education, marital status, SEER region, unknown stage and grade and year of diagnosis.
GI= Gastrointestinal; HR = Hazard Ratios; AJCC = American Joint Committee on Cancer; SEER = Surveillance, Epidemiology and End Results (SEER) Program; The SEER-Medicare database has Medicare claims starting in 1991. For this reason staging for cancer is based on the 1988 staging system
Discussion
In this study we evaluated lower GI and bladder toxicity after radiation therapy for uterine cancer. Among the women who received radiation, an absolute increase in late GI toxicities and bladder toxicities was observed. Inflammation was the most common GI toxicity. Incontinence and hematuria were the most common bladder toxicities. Differences in late GI toxicities were significantly higher for women in the radiation group and this trend remained over time. Conversely, differences in late bladder toxicities were significantly higher for irradiated women but the trend did not hold over time. Diagnostic procedures, particularly endoscopies and cystoscopies, were also elevated among women treated with radiation. Despite these rates, most toxicities appear to have been mild to moderate since only 10% of irradiated women required hospital admission for a lower GI diagnosis and less than 1% of irradiated women required hospital admission for a bladder diagnosis. In the multivariable analysis, advancing age, having chemotherapy, higher comorbidities and prior or early GI and bladder diagnoses were predictive of late toxicities among irradiated women.
The rates of GI and bladder late toxicities reported in our study are comparable to previous studies [12, 20, 21]. For example, in a multi-center clinical trial performed in the Netherlands, 691 women with stage I endometrial cancer were evaluated and 5-year rates of late GI, urinary tract, vagina and bone complications were reported in 26% of women who received radiation compared to 4% of women who did not receive radiation [12], a 22% absolute increase in late toxicities. When the rates were estimated by complication site (i.e. GI and bladder), the GI rates were 20% and the bladder rates were 8% over the same time period [12]. Few other studies have evaluated toxicity at 5-years of follow-up [11, 12]. Despite these rates, identification of late toxicities was based on the French-Italian glossary, a very broad and general classification for complications related to radiation therapy[22]. Thus, rates of complications reported in our study are probably an overestimation and are not specific.
Direct comparisons of results obtained from patient surveys and toxicities assessed by medical claims pose limitations because they measure different aspects of GI and bladder toxicities, in addition to comparing patient-reported symptoms with physician reported diagnoses. Our approach using claims-based data may ascertain more mild toxicities, or toxicities that are relevant for the physician but not necessarily bothersome for the patients, especially if physicians have incentives to register all diagnoses, regardless of how minor they are. Despite these limitations, the use of claims data provides advantages over other data sets. Toxicities are registered precisely over time and it is likely that additional transient symptoms are captured. In addition, claims data allow us to compare baseline rates of GI and bladder diagnoses between the irradiated and non-irradiated women, allowing calculation of absolute risk.
Previous literature has discussed radiation treatment parameters for uterine cancer including optimal doses and volume [5, 10, 13, 21]. Less attention has been directed towards individual characteristics that predispose women to treatment toxicity. In our study, individual characteristics such as older age and higher comorbidities increase the risk of both GI and bladder toxicities. Other individual factors, such as GI and bladder diagnosis 12 months before cancer, and early toxicities were also associated with higher risk of late toxicities. Conversely, education decreased the risk of both GI and bladder diagnoses. Thus, despite the increased risk of toxicities due to advancing age, age alone cannot be considered a contraindication to radiation therapy or other types of cancer treatment[10, 23].
Previous studies have also reported an increased survival and reduction in recurrence of cancer with increased acute toxicity in women treated for some types of uterine cancer with concomitant radiation therapy and chemotherapy [24]. They also report that the prevalence rates of late toxicities are unclear. Our data adds to this literature by finding that chemotherapy increases the prevalence of late GI and bladder toxicities in our cohort of women. Future analyses need to identify other factors that can influence toxicities related to chemotherapy, determine prevalence rates and establish potential mechanisms to explain late toxicity rates in older women with dual therapy.
Our study has some limitations. First, we determined diagnoses consistent with toxicity using Medicare claims. Our rates of diagnoses are likely to underestimate all toxicities and some women may have toxicities not included through billing codes. Second, few details regarding radiation treatment were available. Radiation fields and applied doses are important predictors of late toxicities that were not available. Third, this study probably covered a period during which some improvements to treatment planning were implemented. Despite this, no significant decrease in toxicity over time was observed in the unadjusted or adjusted analyses. Fourth, there could be a bias when including some diagnostic procedures to identify late toxicities. Nevertheless, identification of differences in number of procedures performed by treatment group is an important indicator of healthcare provision. Finally, grouping beam and brachytherapy might influence the results obtained. This approach is likely to underestimate the effect of radiation on late toxicities.
In conclusion, our data show that there are many women who develop GI and bladder late toxicities following radiation therapy for uterine cancer. Older women, women who receive chemotherapy and those with a higher comorbidity burden were at higher risk of developing both GI and bladder toxicities after radiation. The severity of both types of toxicities, however, appeared to be low. Given that radiotherapy is currently one of the cornerstones of treatment for uterine cancer, it is important to have better information on the long-term effects of radiation therapy. Even if most toxicities described are not serious enough to result in hospitalization, they significantly reduce quality of life in women who survive cancer. The heterogeneity of the aging process highlights the importance of considering a variety of individual factors when determining outcomes from radiation therapy in women with uterine cancer.
Conflict of Interest Statement
The authors disclose no actual or potential conflict of interest including any financial, personal or other relationships with other people or organizations within 24 months of beginning the submitted work that could inappropriately influence, or be perceived to influence, their work.
Acknowledgements
This study used the linked SEER-Medicare database. The interpretation and reporting of these data are the sole responsibility of the authors. The authors acknowledge the efforts of the Applied Research Program, NCI; the Office of Research, Development and Information, CMS; Information Management Services (IMS), Inc.; and the Surveillance, Epidemiology, and End Results (SEER) Program tumor registries in the creation of the SEER-Medicare database.
The collection of the California cancer incidence data used in this study was supported by the California Department of Public Health as part of the statewide cancer reporting program mandated by California Health and Safety Code Section 103885; the National Cancer Institute's Surveillance, Epidemiology and End Results Program under contract N01-PC-35136 awarded to the Northern California Cancer Center, contract N01-PC-35139 awarded to the University of Southern California, and contract N02-PC-15105 awarded to the Public Health Institute; and the Centers for Disease Control and Prevention's National Program of Cancer Registries, under agreement #U55/CCR921930-02 awarded to the Public Health Institute. The ideas and opinions expressed herein are those of the author(s) and endorsement by the State of California, Department of Public Health the National Cancer Institute, and the Centers for Disease Control and Prevention or their Contractors and Subcontractors is not intended nor should be inferred. The authors acknowledge the efforts of the Applied Research Program, NCI; the Office of Research, Development and Information, CMS; Information Management Services (IMS), Inc.; and the Surveillance, Epidemiology, and End Results (SEER) Program tumor registries in the creation of the SEER-Medicare database.
This study was supported by a grant (5R01CA133069-03) from the National Cancer Institute (PI J. Freeman). Infrastructure support was provided by the Sealy Center on Aging at the University of Texas Medical Branch.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
The SEER-Medicare database has Medicare claims starting in 1991. For this reason staging for cancer is based on the AJCC 1988 staging system.
Curves were adjusted by age, AJJC stage, grade, ethnicity, marital status, Medicaid eligible, education, chemotherapy treatment, surgery, comorbidities, SEER region, diagnoses 12 months before initial radiation and early toxicities.
References
- 1.Jemal A, Siegel R, Ward E, Hao Y, Xu J, Thun MJ. Cancer statistics, 2009. CA Cancer J Clin. 2009;59:225–49. doi: 10.3322/caac.20006. [DOI] [PubMed] [Google Scholar]
- 2.SEER Cancer Statistics Review Horner MJ, Ries LAG, Krapcho M, Neyman N, Aminou R, Howlader N, et al., editors. National Cancer Institute . 2009. 1-17-2010.
- 3.Jemal A, Clegg LX, Ward E, Ries LA, Wu X, Jamison PM, et al. Annual report to the nation on the status of cancer, 1975-2001, with a special feature regarding survival. Cancer. 2004;101:3–27. doi: 10.1002/cncr.20288. [DOI] [PubMed] [Google Scholar]
- 4.Practice guidelines in Oncology - Uterine Neoplasms. National Comprehensive Cancer Network . 2008. 12-5-2009.
- 5.Nag S, Erickson B, Parikh S, Gupta N, Varia M, Glasgow G. The American Brachytherapy Society recommendations for high-dose-rate brachytherapy for carcinoma of the endometrium. Int J Radiat Oncol Biol Phys. 2000;48:779–90. doi: 10.1016/s0360-3016(00)00689-1. [DOI] [PubMed] [Google Scholar]
- 6.Rauthe G. Management of Reactions and Complications Following Radiation Therapy. In: Vahrson HW, editor. Radiation Oncology of Gynecological Cancers. Berlin: Springer. 1997:433–54. [Google Scholar]
- 7.Giordano SH, Lee A, Kuo YF, Freeman J, Goodwin JS. Late gastrointestinal toxicity after radiation for prostate cancer. Cancer. 2006;107:423–32. doi: 10.1002/cncr.21999. [DOI] [PubMed] [Google Scholar]
- 8.Saibishkumar EP, Patel FD, Sharma SC. Evaluation of late toxicities of patients with carcinoma of the cervix treated with radical radiotherapy: an audit from India. Clin Oncol (R Coll Radiol ) 2006;18:30–7. doi: 10.1016/j.clon.2005.06.013. [DOI] [PubMed] [Google Scholar]
- 9.Jereczek-Fossa B, Jassem J, Nowak R, Badzio A. Late complications after postoperative radiotherapy in endometrial cancer: analysis of 317 consecutive cases with application of linear-quadratic model. Int J Radiat Oncol Biol Phys. 1998;41:329–38. doi: 10.1016/s0360-3016(98)00050-9. [DOI] [PubMed] [Google Scholar]
- 10.Mundt AJ, Mell LK, Roeske JC. Preliminary analysis of chronic gastrointestinal toxicity in gynecology patients treated with intensity-modulated whole pelvic radiation therapy. Int J Radiat Oncol Biol Phys. 2003;56:1354–60. doi: 10.1016/s0360-3016(03)00325-0. [DOI] [PubMed] [Google Scholar]
- 11.Creutzberg CL, van Putten WL, Koper PC, Lybeert ML, Jobsen JJ, Warlam-Rodenhuis CC, 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;355:1404–11. doi: 10.1016/s0140-6736(00)02139-5. [DOI] [PubMed] [Google Scholar]
- 12.Creutzberg CL, van Putten WL, Koper PC, Lybeert ML, Jobsen JJ, Warlam-Rodenhuis CC, et al. The morbidity of treatment for patients with Stage I endometrial cancer: results from a randomized trial. Int J Radiat Oncol Biol Phys. 2001;51:1246–55. doi: 10.1016/s0360-3016(01)01765-5. [DOI] [PubMed] [Google Scholar]
- 13.Yalman D, Arican A, Ozsaran Z, Celik OK, Yurut V, Esassolak M, et al. Evaluation of morbidity after external radiotherapy and intracavitary brachytherapy in 771 patients with carcinoma of the uterine cervix or endometrium. Eur J Gynaecol Oncol. 2002;23:58–62. [PubMed] [Google Scholar]
- 14.Athanassiou H, Antonadou D, Coliarakis N, Kouveli A, Synodinou M, Paraskevaidis M, et al. Protective effect of amifostine during fractionated radiotherapy in patients with pelvic carcinomas: results of a randomized trial. Int J Radiat Oncol Biol Phys. 2003;56:1154–60. doi: 10.1016/s0360-3016(03)00187-1. [DOI] [PubMed] [Google Scholar]
- 15.Tewari KS, DiSaia PJ. Radiation therapy for gynecologic cancer. J Obstet Gynaecol Res. 2002;28:123–40. doi: 10.1046/j.1341-8076.2002.00036.x. [DOI] [PubMed] [Google Scholar]
- 16.Potosky AL, Riley GF, Lubitz JD, Mentnech RM, Kessler LG. Potential for cancer related health services research using a linked Medicare-tumor registry database. Med Care. 1993;31:732–48. [PubMed] [Google Scholar]
- 17.Schrag D, Virnig BA, Warren JL. Linking tumor registry and medicaid claims to evaluate cancer care delivery. Health Care Financ Rev. 2009;30:61–73. [PMC free article] [PubMed] [Google Scholar]
- 18.Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40:373–83. doi: 10.1016/0021-9681(87)90171-8. [DOI] [PubMed] [Google Scholar]
- 19.Klabunde CN, Potosky AL, Legler JM, Warren JL. Development of a comorbidity index using physician claims data. J Clin Epidemiol. 2000;53:1258–67. doi: 10.1016/s0895-4356(00)00256-0. [DOI] [PubMed] [Google Scholar]
- 20.Anacak Y, Yalman D, Ozsaran Z, Haydaroglu A. Late radiation effects to the rectum and bladder in gynecologic cancer patients: the comparison of LENT/SOMA and RTOG/EORTC late-effects scoring systems. Int J Radiat Oncol Biol Phys. 2001;50:1107–12. doi: 10.1016/s0360-3016(01)01527-9. [DOI] [PubMed] [Google Scholar]
- 21.Fayed A, Mutch DG, Rader JS, Gibb RK, Powell MA, Wright JD, et al. Comparison of high-dose-rate and low-dose-rate brachytherapy in the treatment of endometrial carcinoma. Int J Radiat Oncol Biol Phys. 2007;67:480–4. doi: 10.1016/j.ijrobp.2006.09.004. [DOI] [PubMed] [Google Scholar]
- 22.Chassagne D, Sismondi P, Horiot JC, Sinistrero G, Bey P, Zola P, et al. A glossary for reporting complications of treatment in gynecological cancers. Radiother Oncol. 1993;26:195–202. doi: 10.1016/0167-8140(93)90260-f. [DOI] [PubMed] [Google Scholar]
- 23.Balducci L, Ershler WB. Cancer and ageing: a nexus at several levels. Nat Rev Cancer. 2005;5:655–62. doi: 10.1038/nrc1675. [DOI] [PubMed] [Google Scholar]
- 24.Green J, Kirwan J, Tierney J, Vale C, Symonds P, Fresco L, et al. Concomitant chemotherapy and radiation therapy for cancer of the uterine cervix. Cochrane Database Syst Rev. 2005:CD002225. doi: 10.1002/14651858.CD002225.pub2. [DOI] [PMC free article] [PubMed] [Google Scholar]