Fax: (212) 305-3412
Patterns of care and treatment outcomes for elderly women with cervical cancer
Article first published online: 28 OCT 2011
Copyright © 2011 American Cancer Society
Volume 118, Issue 14, pages 3618–3626, 15 July 2012
How to Cite
Sharma, C., Deutsch, I., Horowitz, D. P., Hershman, D. L., Lewin, S. N., Lu, Y.-S., Neugut, A. I., Herzog, T. J., Chao, C. K. and Wright, J. D. (2012), Patterns of care and treatment outcomes for elderly women with cervical cancer. Cancer, 118: 3618–3626. doi: 10.1002/cncr.26589
- Issue published online: 2 JUL 2012
- Article first published online: 28 OCT 2011
- Manuscript Accepted: 24 AUG 2011
- Manuscript Revised: 16 AUG 2011
- Manuscript Received: 13 JUL 2011
- cervical cancer;
- cervical carcinoma;
- radical hysterectomy
Cervical cancer is common in the elderly. The authors examined the patterns of care, treatment, and outcomes of elderly women with cervical cancer.
Women with cervical cancer diagnosed between 1988 and 2005 and registered in the Surveillance, Epidemiology, and End Results database were analyzed. Patients were stratified by age: <50, 50 to 59, 60 to 69, 70 to 79, and ≥80 years. Multivariate logistic regression models were constructed to examine treatment; cancer-specific survival was examined using Cox proportional hazards models.
A total of 28,902 women were identified, including 2543 women 70 to 79 years old and 1364 ≥80 years. For women with early stage (IB1-IIA) tumors, primary surgery was performed in 82.0% of women <50 years old compared with 54.5% of those 70 to 79 years old and 33.2% of those ≥80 years old (P < .0001). For women treated surgically, lymphadenectomy was performed in 66.8% of women <50 years old versus 9.1% of patients ≥80 years old (P < .0001). Compared with patients <50 years old, those >80 years old were less likely to undergo radical hysterectomy (odds ratio [OR], 0.10; 95% confidence interval [CI], 0.07-0.14) and lymphadenectomy (OR, 0.11; 95% CI, 0.08-0.16) and to receive adjuvant radiation therapy (OR, 0.06; 95% CI, 0.01-0.35). Among women with stage IIB-IVA disease, use of brachytherapy declined with age (P < .0001). For women with stage IB1-IIA tumors, the hazard ratio for death from cancer was 1.35 (95% CI, 1.16-1.58) for women 70 to 79 years old and 2.08 (95% CI, 1.72-2.48) for those ≥80 years old compared with younger women.
Elderly women with cervical cancer are less likely to undergo surgery, receive adjuvant radiation, and receive brachytherapy. After adjusting for treatment disparities, cancer-specific mortality is higher in older women. Cancer 2012;3618–3626. © 2011 American Cancer Society.
Worldwide cervical cancer remains a major cause of cancer-associated morbidity and mortality in women.1 Despite screening efforts, >3600 women die each year from cervical cancer in the United States.2 Although several effective treatments exist for cervical cancer, therapy can be intensive, often involves multiple therapeutic modalities, and can be associated with substantial treatment-associated morbidity.
Cervical cancer demonstrates a bimodal distribution, with peaks between 30 and 39 years of age and between 60 and 69 years of age. The older age group is becoming an increasingly prevalent portion of the US population. Demographic estimates suggest that the number of people 65 years of age or older will reach approximately 72 million, or 19% of the population, by 2030.3 Cancer incidence is greatest in the elderly population. Population-based data estimates that between 2003 and 2007, 54% of all new cancer cases and 70% of cancer-related deaths occurred in patients 65 years of age or older.4 As the population continues to grow, the allocation of cancer treatment to the elderly will become an even greater concern.
Despite the frequency of cancer in the elderly, treatment disparities are common. Older patients receive less aggressive cancer therapy than their younger counterparts or receive cancer care that deviates from the standards of care for several cancer sites.5-8 In a study of elderly breast cancer patients, 26% of women aged ≥80 years without comorbidity received breast-conserving surgery alone or no surgery compared with 6% of women aged 67 to 79 years.7 In addition to treatment delivered, age often has an independent effect on decreasing survival.7, 8
For cervical cancer, several institutional studies have demonstrated treatment-related disparities despite evidence that suggests elderly women with cervical cancer often tolerate treatment well.8-15 The objective of our study was to perform a population-based analysis of the patterns of treatment and survival for elderly women with cervical cancer. We examined access to standard cervical cancer treatments stratified by stage and compared utilization for younger and older women.
MATERIALS AND METHODS
Data from the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) database were used. SEER is a population-based tumor registry that records data on approximately 26% of the US population.16 SEER is composed of several geographically distinct tumor registries.17 Data from SEER 17 registries were used. All data were publicly available, deidentified, and exempt from institutional review board review.
Women with invasive cervical cancer diagnosed between 1988 and 2005 were analyzed. Only patients with squamous cell carcinomas, adenocarcinomas, or adenosquamous carcinomas were included in the analysis. Subjects were stratified by age into the following groups for analysis: <50 years, 50 to 59 years, 60 to 69 years, 70 to 79 years, and ≥80 years. Clinical and pathologic data, including race (white, black, or other), marital status, and tumor grade, were collected. Year of diagnosis was classified as 1988 to 1993, 1994 to 1999, or 2000 to 2005. Patients were categorized based on the geographic area of residence at the time of diagnosis: central (Detroit, Iowa, Kentucky, Louisiana, Utah), eastern (Connecticut, New Jersey, Atlanta, rural Georgia), or western (Alaska, California, Hawaii, Los Angeles, New Mexico, San Francisco, San Jose, Seattle).
Primary treatment was classified as either surgery, radiation, or no cancer-directed therapy. Surgery was classified as 1 of the following procedure types: excision/destruction, simple hysterectomy, radical hysterectomy, hysterectomy unspecified, or other/unknown surgery type. Performance of lymphadenectomy was noted for each subject. Receipt of radiation was recorded as external beam with or without brachytherapy (intracavitary radiation). SEER lacks data on chemotherapy, and as such this was not included. Staging information was derived from the American Joint Cancer Committee staging information and recorded extent of disease codes. Each patient's stage was determined using the revised 2009 staging criteria of the International Federation of Gynecologists and Obstetricians.18
Each patient's vital status was recorded. Cancer-specific survival was calculated as the number of months from cancer diagnosis to date of death from cancer. Patients who were alive at last follow-up were censored. Frequency distributions between categorical variables were compared using chi-square tests. Logistic regression models were developed to describe predictors of radical hysterectomy, lymphadenectomy, adjuvant radiation, and vaginal brachytherapy. In the logistic regression analysis, we modeled all the categories of each variable controlling for all other variables in the model. Both the univariate and multivariate analyses were stratified by stage as either early stage (IB1-IIA) or advanced stage (IIB-IVA). Cox proportional hazards models were developed to examine cancer-specific survival. In Cox proportional hazards analyses, we modeled the cancer-specific mortality hazard ratios (HRs) comparing women by age and accounting for other prognostic variables and treatment. Kaplan-Meier curves were generated to examine cancer-specific survival. All analyses were performed with SAS version 9.2 (SAS Institute, Cary, NC).
A total of 28,902 women with cervical cancer were identified. The cohort included 15,790 (54.6%) women <50 years of age, 5326 (18.4%) aged 50 to 59, 3879 (13.4%) aged 60 to 69, 2543 (8.8%) who were 70 to 79 years of age, and 1364 (4.7%) ≥80 years of age (Table 1). The older women were more commonly nonwhite, more often lived in the eastern or central United States, and were more often unmarried (P < .05 for all). Older women more commonly had squamous cell tumors (P < .0001) and more frequently had poorly differentiated neoplasms (P < .0001). Older women were also more likely to have advanced stage disease at presentation. At diagnosis, 1058 (6.7%) women <50 years of age had stage IIIB tumors compared with 230 (16.9%) of the women in the oldest age strata (P < .0001).
|Characteristic||<50 Years||50-59 Years||60-69 Years||70-70 Years||≥80 Years||P|
|Year of diagnosis||<.0001|
Table 2 displays treatment trends stratified by age and stage. Among women with early stage (IB1-IIA) tumors, primary surgery was performed in 82.0% of women <50 years of age, compared with 54.5% of women aged 70 to 79 and 33.2% of those ≥80 years of age (P < .0001). For those women who underwent primary surgery, use of radical hysterectomy, the operative procedure of choice, decreased with age from 58.9% of patients <50 years old to 20.4% in those ≥80 years of age (P < .0001). Among patients treated with surgery, lymphadenectomy was undertaken in 66.8% of women <50 years of age versus 29.0% of those aged 70 to 79 years and 9.1% of patients ≥80 years old (P < .0001).
|Characteristic||<50 Years||50-59 Years||60-69 Years||70-70 Years||≥80 Years||P|
|Type of surgerya||<.0001|
|Use of external beam and brachytherapyb||<.0001|
In multivariate analysis, age remained a strong predictor of treatment for women with early stage tumors (Table 3). Compared with patients <50 years of age, women aged 70 to 79 years (odds ratio [OR], 0.32; 95% confidence interval [CI], 0.27-0.37) and those >80 years old (OR, 0.10; 95% CI, 0.07-0.14) were less likely to undergo radical hysterectomy. Among those women with early stage tumors who were treated surgically, women 70 to 79 years old (OR, 0.28; 95% CI, 0.23-0.34) and those aged >80 years (OR, 0.11; 95% CI, 0.08-0.16) were less likely to undergo lymphadenectomy. For patients with stage IB1-IIA neoplasms who underwent primary surgery and had nodal disease, women >80 years of age were less likely (OR, 0.06; 95% CI, 0.01-0.35) to receive adjuvant radiotherapy. Other predictors of primary treatment included race, year of diagnosis, area of residence, marital status, histology, grade, and stage.
|Characteristic||Radical Hysterectomy for All Stage IB1-IIA Patients||Lymphadenectomy for Stage IB1-IIA Patients Who Underwent Surgery||Radiation for Node-Positive Stage IB1-IIA Patients Who Underwent Surgery||Brachytherapy for Stage IIA-IVA Patients Treated With Radiation|
|50-59 years||0.80 (0.72-0.88)a||0.65 (0.57-0.73)a||0.87 (0.60-1.26)||0.88 (0.76-1.01)|
|60-69 years||0.57 (0.51-0.64)a||0.50 (0.43-0.58)a||1.21 (0.68-2.15)||0.77 (0.66-0.91)a|
|70-79 years||0.32 (0.27-0.37)a||0.28 (0.23-0.34)a||1.37 (0.57-3.32)||0.73 (0.61-0.87)a|
|≥80 years||0.10 (0.07-0.14)a||0.11 (0.08-0.16)a||0.06 (0.01-0.35)a||0.46 (0.37-0.57)a|
|Black||0.68 (0.60-0.76)a||0.62 (0.53-0.72)a||0.64 (0.42-0.97)a||0.77 (0.66-0.89)a|
|Other||1.07 (0.95-1.20)||1.22 (1.03-1.45)a||1.23 (0.75-2.01)||1.34 (1.11-1.61)|
|Unknown||0.80 (0.51-1.27)||0.34 (0.20-0.58)a||—||0.69 (0.18-2.63)|
|Year of diagnosis|
|1994-1999||0.84 (0.76-0.92)a||1.01 (0.89-1.15)||1.48 (0.99-2.22)||0.92 (0.78-1.10)|
|2000-2005||0.48 (0.44-0.53)a||1.11 (0.98-1.25)||1.63 (1.14-2.34)a||0.44 (0.38-0.50)a|
|Central||0.84 (0.76-0.92)a||0.66 (0.59-0.74)a||0.82 (0.56-1.19)||1.53 (1.34-1.76)a|
|East||0.91 (0.82-1.00)||0.78 (0.69-0.89)a||0.67 (0.46-0.97)a||1.14 (0.98-1.32)|
|Single||0.80 (0.74-0.86)a||0.84 (0.76-0.93)a||1.16 (0.87-1.55)||0.90 (0.80-1.02)|
|Unknown||0.55 (0.45-0.67)a||0.56 (0.44-0.70)a||4.27 (1.00-18.21)||0.89 (0.67-1.20)|
|Adenocarcinoma||1.17 (1.07-1.29)a||1.30 (1.15-1.47)a||0.92 (0.62-1.37)||0.94 (0.77-1.15)|
|Adenosquamous||1.48 (1.28-1.70)a||1.49 (1.21-1.82)a||1.76 (0.99-3.12)||0.95 (0.72-1.26)|
|2||1.35 (1.18-1.55)a||1.85 (1.56-2.19)a||1.26 (0.64-2.48)||1.01 (0.77-1.32)|
|3||1.36 (1.18-1.56)a||2.38 (1.99-2.84)a||1.45 (0.73-2.88)||0.92 (0.71-1.20)|
|Unknown||0.69 (0.59-0.80)a||0.88 (0.74-1.05)||1.73 (0.80-3.77)||0.90 (0.69-1.18)|
|IB2||0.43 (0.38-0.47)a||1.09 (0.91-1.30)||0.86 (0.61-1.22)||—|
|IB NOS||0.43 (0.39-0.46)a||0.33 (0.30-0.37)a||1.04 (0.71-1.52)||—|
|IIA||0.22 (0.19-0.25)a||0.50 (0.42-0.61)a||1.44 (0.87-2.37)||—|
Similar disparities were noted for patients with advanced stage (IIB-IVA) disease (Table 2). No cancer-directed therapy was allocated to 138 (3.9%) women <50 years of age, compared with 80 (7.3%) of those aged 70 to 79 years and 74 (12.1%) of the women ≥80 years old (P < .0001). Among women treated with radiation, use of brachytherapy, the standard of care, declined with age from 66.7% in those <50 years old to 58.9% in patients aged 70 to 79 years and to 46.3% in women ≥80 years old (P < .0001). Accounting for other prognostic factors, women with advanced stage tumors who were treated with radiation and were 70 to 79 years old (OR, 0.73; 95% CI, 0.61-0.87) or ≥80 years old (OR, 0.46; 95% CI, 0.37-0.57) were less likely to receive brachytherapy (Table 3). Other predictors of use of brachytherapy were race, year of diagnosis, area of residence, and stage.
Cancer-specific survival was then examined. Accounting for treatment disparities and other prognostic factors, elderly women were more likely to die from their tumors. For women with stage IB1-IIA tumors, the HR for death from cervical cancer was 1.35 (95% CI, 1.16-1.58) for women 70 to 79 years old and 2.08 (95% CI, 1.72-2.48) for those ≥80 years old compared with younger women. Similarly, women 70 to 79 years old (HR, 1.12; 95% CI, 1.01-1.24) and ≥80 years old (HR, 1.47; 95% CI, 1.29-1.68) with stage IIB-IVA disease were more likely to die from their cancers (Table 4). In Kaplan-Meier analyses, age negatively impacted cancer-specific survival for stage IB1 (P = .02) and stage IIIB (P = .0002; Fig. 1) disease.
|Characteristic||Stage IB1-IIA||Stage IIB-IVA|
|50-59 years||1.00 (0.88-1.14)||0.92 (0.84-1.00)|
|60-69 years||1.01 (0.88-1.16)||0.94 (0.86-1.04)|
|70-79 years||1.35 (1.16-1.58)a||1.12 (1.01-1.24)a|
|≥80 years||2.08 (1.74-2.48)a||1.47 (1.29-1.68)a|
|Black||1.35 (1.20-1.53)a||1.16 (1.06-1.27)a|
|Other||1.00 (0.86-1.15)||0.76 (0.68-0.86)a|
|Unknown||0.21 (0.05-0.83)a||1.64 (0.82-3.30)|
|Year of diagnosis|
|1994-1999||0.85 (0.76-0.95)a||0.96 (0.88-1.05)|
|2000-2005||0.89 (0.79-1.00)a||0.82 (0.75-0.89)a|
|Central||1.07 (0.96-1.19)||1.04 (0.96-1.12)|
|East||1.00 (0.88-1.13)||0.96 (0.87-1.05)|
|Single||1.05 (0.96-1.16)||1.05 (0.98-1.13)|
|Unknown||0.81 (0.63-1.05)||1.12 (0.94-1.33)|
|Adenocarcinoma||1.38 (1.22-1.55)a||1.49 (1.34-1.66)a|
|Adenosquamous||1.53 (1.30-1.80)a||1.09 (0.93-1.29)|
|2||1.81 (1.45-2.25)a||1.14 (0.96-1.34)|
|3||2.46 (1.98-3.07)a||1.42 (1.21-1.67)a|
|Unknown||1.35 (1.07-1.70)a||1.12 (0.95-1.32)|
|Radiation||2.21 (1.99-2.45)a||1.34 (1.23-1.46)a|
|No treatment||3.49 (2.84-4.30)a||3.38 (2.93-3.90)a|
|IB NOS||1.26 (1.10-1.43)a||—|
Our findings add to a growing body of literature suggesting that elderly patients with cancer are treated differently than their younger counterparts.5-7, 19 A study of nearly 23,000 patients with a variety of tumors noted that use of curative therapy declined with age and that elderly patients were more likely to forego treatment.6 We noted several treatment disparities for elderly women with cervical cancer. Compared with younger patients, women older than 70 years with early stage tumors were less likely to undergo primary surgery and less likely to receive adjuvant therapy for high-risk features, and older women with advanced stage tumors less often received high-quality radiotherapy. Even after adjusting for treatment disparities, elderly women with cervical cancer were more likely to die from their tumors than younger women.
Early stage cervical cancer (stage IB-IIA) is treated either with radical hysterectomy or primary radiotherapy; survival is similar for the 2 modalities.20 In the United States, there is a strong bias toward surgical management; >80% of women are treated surgically.11 Despite the overall patterns of care, we noted substantial biases for the treatment of elderly women with early stage tumors. Only 55% of women aged 70 to 79 years and 33% of those aged >80 years were treated with primary surgery, compared with 82% of patients <50 years old. Even if elderly women underwent surgery, they were often not treated with radical hysterectomy, the standard of care. Despite the biases we noted in access to surgery, institutional studies have suggested that elderly women tolerate radical hysterectomy and regional lymphadenectomy well.9, 10, 21, 22 Further population-based studies are clearly warranted to determine the morbidity of radical hysterectomy in the elderly.
The use of adjuvant treatment in early stage patients with negative prognostic features after surgery decreases the risk of locoregional recurrence and improves survival.23, 24 The Southwest Oncology Group demonstrated the superiority of chemoradiation over radiation alone for patients with high-risk features including nodal metastasis, parametrial spread, or positive margins.24 For women at intermediate risk of recurrence with various combinations of large tumor size, deep cervical invasion, or lymphovascular space invasion, radiotherapy decreases the risk of local recurrence.23 We noted that elderly women were less likely to receive adjuvant radiotherapy even when the indication was relatively clear. Among lymph node-positive women with early stage tumors who underwent primary surgery, patients ≥80 years of age were >90% less likely to receive radiation than younger women.
Radiation in combination with chemotherapy is the treatment of choice for women with advanced stage cervical cancer.24-27 Patterns of care studies have reported that uptake of chemotherapy has been relatively rapid, increasing from 34% in 1997 to 85% in 2001.11 Despite the addition of chemotherapy to treatment plans, adequate delivery of radiation remains the most important factor in disease control for advanced stage disease. The dose of radiation delivered with external beam radiation (both whole pelvic and parametrial boost) is inadequate to treat macroscopic disease and must be combined with intracavitary brachytherapy to attain adequate local control.15 In our series, fewer than half of the women ≥80 years of age and only 59% of those aged 70 to 79 years received brachytherapy. Prior studies have also noted disparities in access to brachytherapy; Mitchell and coworkers noted that 11% of patients 60 to 69 years old, 20% aged 70 to 79 years, and 40% of those ≥80 years of age did not receive intracavitary radiation.28
Despite the disparities in treatment with radiation, recent studies have reported that elderly women tolerate pelvic radiotherapy and brachytherapy well.13-15, 29-31 The European Organization for Research and Treatment of Cancer found that there were no age-related differences in either early or late toxicities in >1600 patients treated with pelvic radiation.30 Even among patients >80 years of age it appears that radiotherapy is well tolerated.31 Care should be taken to optimize radiation therapy in elderly patients who are treated. Grant and colleagues noted that 32% of women aged >75 years treated with pelvic radiotherapy did not complete therapy, and treatment interruptions occurred in 42% of patients.32 Cisplatin-based chemotherapy is now frequently used in combination with radiation. It appears that chemoradiation is well tolerated, but further studies examining the long-term outcomes of chemoradiation in the elderly are needed.13
Despite the inclusion of a large number of patients, we recognize several important limitations. Perhaps most importantly, SEER lacks medical comorbidities; therefore, we were unable to risk adjust our population. It is well known that performance status and comorbidity not only impact the allocation of treatment, but also exert a direct effect on survival.5, 6, 33 SEER lacks data on chemotherapy, and registry data may not capture all women treated with radiotherapy.34 Although undercoding of radiation would have been balanced among the age strata, further work is required to explore the use of chemotherapy in elderly women with cervical cancer. As central pathology review is not provided, we limited our analysis to only patients with the most common histologic subtypes of cervical cancer. Many of the effects we noted regarding treatment disparities became gradually more pronounced with age, and there was not 1 specific age cutpoint that we could define. Lastly, SEER does not capture recurrence or subsequent treatment. Given this limitation, we analyzed cancer-specific survival as an endpoint.
In addition to the finding that elderly women with cervical cancer are treated less aggressively than younger patients, we noted that even with standard of care therapy, older women are more likely to die from their tumors. When adjusted for stage and treatment, women >70 years of age were more likely to die from cervical cancer than their younger counterparts, suggesting that age >70 years is a negative prognostic factor in cervical cancer. Given the guarded prognosis of elderly patients with cervical cancer, our study raises the question of whether these women should be treated aggressively. Outcomes data suggest that survival for cervical cancer is in general favorable; 5-year survival for early stage tumors is 80% or greater, and even for stage IIIB neoplasms 5-year survival approaches 35% in some subsets of patients.20, 35 Given these data, treatment in the elderly should be individualized and take into account not only tumor characteristics but also comorbidities, functional status, and patient preferences.36, 37 For other tumor types, comprehensive geriatric assessment, a multidisciplinary treatment approach that includes tumor, treatment, and patient characteristics to guide treatment, have shown success and may be of value in elderly women with cervical cancer.38 Further studies to examine the factors that underlie treatment disparities for elderly women with cervical cancer and comparative effectiveness research to individualize the management of older women with cervical cancer are needed.
No specific funding was disclosed.
CONFLICT OF INTEREST DISCLOSURES
The authors made no disclosures.
- 3U.S. Population Projections. 2011. Available at: http://www.census.gov/population/www/projections/summarytables.html Accessed on February 1, 2011.
- 4SEER Cancer Statistics. 2011. Available at: http://www.seer.cancer.gov Accessed on…
- 16Surveillance, Epidemiology, and End Results, SEER*Stat Database: Incidence-SEER 9 Regs Limited-Use, Nov 2006 Sub (1973-2004), National Cancer Institute, DCCPS, Surveillance Research Program, Cancer Statistics Branch, released April 2007, based on the November 2006 submission. Available at: http://.seer.cancer.gov Accessed on October 1, 2009.
- 17Surveillance, Epidemiology, and End Results (SEER) Program, Data Quality. National Cancer Institute, 2007. Available at: http://seer.cancer.gov/about/ Accessed on October 1, 2009.
- 25Randomized comparison of fluorouracil plus cisplatin versus hydroxyurea as an adjunct to radiation therapy in stage IIB-IVA carcinoma of the cervix with negative para-aortic lymph nodes: a Gynecologic Oncology Group and Southwest Oncology Group study. J Clin Oncol. 1999; 17: 1339-1348., , , et al.