Data are from the US Military Health System.30
Customize your JAMA Network experience by selecting one or more topics from the list below.
Manjelievskaia J, Brown D, McGlynn KA, Anderson W, Shriver CD, Zhu K. Chemotherapy Use and Survival Among Young and Middle-Aged Patients With Colon Cancer. JAMA Surg. 2017;152(5):452–459. doi:10.1001/jamasurg.2016.5050
Are young and middle-aged persons with colon cancer more likely to receive postoperative systemic chemotherapy and, as a result, have better survival than older persons?
This cohort study found that although young and middle-aged patients were 2 to 8 times more likely to receive postoperative systemic chemotherapy for colon cancer than older patients, there was no significant survival difference by age group.
The additional use of postoperative systemic chemotherapy among young and middle-aged persons without matched survival improvement suggests possible overuse of chemotherapy among younger than older adults with colon cancer.
Treatment options for patients with young-onset colon cancer remain to be defined and their effects on prognosis are unclear.
To investigate receipt of adjuvant chemotherapy by age category (18-49, 50-64, and 65-75 years) and assess whether age differences in chemotherapy matched survival gains among patients diagnosed as having colon cancer in an equal-access health care system.
Design, Setting, and Participants
This cohort study was based on linked and consolidated data from the US Department of Defense’s Central Cancer Registry and Military Heath System medical claims databases. There were 3143 patients aged 18 to 75 years with histologically confirmed primary colon adenocarcinoma diagnosed between 1998 and 2007. This study was conducted from December 2015 to August 2016.
Patients who underwent surgery and postoperative systemic chemotherapy.
Main Outcomes and Measures
The primary outcome measure of the study was overall survival of patients who only received surgery and those who received both surgery and postoperative systemic chemotherapy.
Of the 3143 patients, 1841 were men (58.6%). Young (18-49 years) and middle-aged (50-64 years) patients were 2 to 8 times more likely to receive postoperative systemic chemotherapy compared with older patients (65-75 years) across all tumor stages. Middle-aged patients with stage I (odds ratio, 5.04; 95% CI, 2.30-11.05) and stage II (odds ratio, 2.42; 95% CI, 1.58-3.72) disease were more likely to receive postoperative chemotherapy compared with older patients. Both groups were more likely to receive multiagent chemotherapy than were older patients (patients aged 18-49 years: odds ratio, 2.48; 95% CI, 1.42-4.32 and patients aged 50-64 years: odds ratio, 2.66; 95% CI, 1.70-4.18). Among patients who received surgery and postoperative systemic chemotherapy, no significant differences were observed in survival among age groups (the 95% CIs of hazard ratios included 1 for young and middle-aged patients compared with older patients for all tumor stages).
Conclusions and Relevance
In an equal-access health care system, we found potential overuse of chemotherapy among young and middle-aged adults with colon cancer. The addition of postoperative systemic chemotherapy did not result in matched survival improvement.
Colorectal cancer is the third leading cause of cancer mortality in the United States, with an expected 134 490 new cases and 49 190 deaths in 2016.1 While incidence and mortality rates of colorectal cancer among adults aged 50 years and older have decreased in the United States in recent years,2 the same trend has not been observed for patients aged 20 to 49 years. The decreased mortality among people aged 50 years or older may be related to the use of colorectal cancer screening recommended for individuals in this age group.3-5 However, survival studies have shown inconsistent results. While some studies showed poorer survival for young patients as compared with their older counterparts,6,7 others found better or comparable survival among younger patients.8-13
Effective treatment increases survival from colon cancer. While surgery is the primary curative option for patients with stage II or III colon cancer, adjuvant chemotherapy is recommended for stage III and IV tumors.14-18 For some patients with stage IV colon cancer, chemotherapy may be used as primary treatment.19 While among patients with stage III colon cancer, adjuvant chemotherapy regimens significantly increase disease-free survival,14-16 less is known about the benefits of adjuvant chemotherapy use in patients with earlier-stage cancers13,20 because some studies have shown no significant improvement in survival.21,22 Thus, routine adjuvant chemotherapy is currently not the standard of care,17,18 and its use remains controversial for early-stage colon cancer.9 Nevertheless, a considerable proportion of patients are still receiving this treatment.20,23 It is possible that chemotherapy is overused in clinical practice without substantial gain in survival.
Because prognosis of colon cancer among young patients is not well known, it is difficult to advise about adjuvant chemotherapy,13 and treatment options for these patients remain unclear.24 Thus, overuse of chemotherapy may be more likely for younger patients. A study by Kneuertz et al13 found that young patients are more likely to receive aggressive therapies, such as multiagent chemotherapy regimens, regardless of their tumor stage at diagnosis. This treatment approach has not resulted in matched survival gains and the study findings suggest overtreatment of younger patients with colon cancer.13
Receipt of treatment is related to a patient's insurance status and access to medical care. In regard to adjuvant treatment of cancer, patients with private insurance may receive adjuvant treatment more frequently than those with no insurance or government-sponsored insurance,20,25 and patients with Medicare may be more likely to receive chemotherapy than those with Medicaid and private insurance plans.26 Insurance status may vary by age and thus may confound comparisons of different age groups in adjuvant treatment. As shown in the study by Kneuertz et al,13 young patients with colon cancer were less likely to be insured and more likely to have Medicaid or other government insurance.13 Their study adjusted for insurance status to reduce its potential confounding effects on age comparisons in adjuvant chemotherapy and survival. Nevertheless, because the study was based on a nationwide database, there were extensive differences in health insurance. Thus, it remains unclear whether there was potential residual confounding by insurance status and types, which vary in many aspects such as coverage, out-of-pocket costs, and accessible medical facilities.
Therefore, a study that compares receipt of adjuvant chemotherapy among different age groups in an equal-access health care system, in which members are more similar and have comparable access to medical care, is warranted. An examination in an equal-access health care system can minimize the potential effects of insurance status and types, thus providing additional evidence concerning the effect of age on the use of adjuvant chemotherapy and its benefits.
The US Department of Defense (DoD) Military Health System (MHS) provides universal health care to its beneficiaries including active-duty members, retirees, and their dependents, making it an ideal setting to assess whether differences in cancer treatment and survival exist among patients diagnosed as having colon cancer. Many MHS beneficiaries receive free care. Although some plans have out-of-pocket costs (enrollment fees, deductibles, copayments, and insurance premiums), these costs are much lower than those for civilian health plans. In 2013, annual out-of-pocket costs for families of active-duty members and retirees younger than age 65 years who were enrolled in TRICARE Prime were $93 and $976, respectively, while the corresponding costs were $5463 and $5905 for their civilian health maintenance organization counterparts.27 Therefore, the DoD health care system provides a unique opportunity to assess the relationship between age and adjuvant chemotherapy in an equal-access setting.
The objective of this study was to use data from the medical claims database and cancer registry of the DoD to investigate receipt of adjuvant chemotherapy by age category and assess whether age differences in treatment matched survival gains among patients diagnosed as having colon cancer.
Conducted from December 2015 to August 2016, this study was based on linked and consolidated data from the DoD’s Central Cancer Registry and MHS Data Repository medical claims database. The Central Cancer Registry data contain demographic factors, cancer diagnosis, treatment, and vital status from patients with cancer diagnosed and/or treated at military treatment facilities. The Central Cancer Registry follows the North American Association of Central Cancer Registries guidelines and uses the International Classification of Diseases for Oncology for cancer diagnosis. The MHS Data Repository database contains administrative and medical claims data for inpatient and outpatient services provided at both military treatment facilities and nonmilitary treatment facilities paid for by the DoD. It includes clinical diagnoses of all medical conditions, which are coded using the International Classification of Diseases, Ninth Revision, and diagnostic and treatment procedures, which are coded using International Classification of Diseases, Ninth Revision; Current Procedural Terminology; or Healthcare Common Procedure Coding System codes. The data linkage project was reviewed and approved by the institutional review boards of the Walter Reed National Military Medical Center, Defense Health Agency, and the National Institutes of Health Office of Human Subjects Research. Patient consent was waived because the study was based on large existing data and patients could not be identified in the linked and consolidated data.
Linked data were queried for patients aged between 18 and 75 years with histologically confirmed primary colon adenocarcinoma (International Classification of Diseases for Oncology, Third Revision, topography site codes [C180-189]) diagnosed between January 1, 1998, and December 31, 2007. Patients with prior primary tumors or multiple primary colon tumors were excluded. Also excluded were patients with unknown age. Patients were categorized into 3 groups based on age at onset of colon cancer: 18 to 49 (young), 50 to 64 (middle), and 65 to 75 (older) years. There were 3143 patients, of whom, 671 were young, 1599 were middle-aged, and 873 were older.
Treatment modalities were the variables of interest, which included surgery and postoperative systemic chemotherapy. Surgical procedures included segmental colectomy, total colectomy or proctocolectomy, or other/multivisceral resection (Current Procedural Terminology codes: 44204, 44205, 44140, 44160, 44207, 45.7x, and 45.8x). Chemotherapy regimens were classified as single-agent or multiagent and had to have been reported after the patient’s surgery date. Single-agent chemotherapy was 5-fluorouracil (Healthcare Common Procedure Coding System code J9190) (with or without leucovorin [J0640]) or capecitabine based (Healthcare Common Procedure Coding System codes J8520 and J8521). Multiagent regimens were defined as the addition of oxaliplatin (J9263), irinotecan (J9206), or floxuridine (J9200).
Other variables used in this study included demographic variables, tumor characteristics, and comorbidities. Demographic variables included sex, race/ethnicity (non-Hispanic white, non-Hispanic black, and other/unknown), marital status (married, single, other, and unknown), military sponsor branch (Army, Navy, Marines, Air Force, and other), active-duty status (yes, no, and unknown), and benefit type (TRICARE Prime, TRICARE Non-Prime, and other/unknown) defined as the patient’s insurance status at any point in the 3 months prior to diagnosis or 3 months following diagnosis. Tumor characteristics included tumor location, stage, and grade. Tumor location was characterized as right (cecum, ascending colon, and hepatic flexure, respectively), transverse, left (splenic flexure, descending colon, and sigmoid colon, respectively), and overlapping/unknown. Tumor stage was determined according to the American Joint Committee on Cancer28 staging recommendations and categorized as stage I, stage II, stage III, stage IV, and unknown. Tumor grade was classified as I (well differentiated), II (moderately differentiated), III and IV (poorly or undifferentiated), and unknown. Comorbidities were categorized according to the Charlson Comorbidity Index.29
The primary outcome measure of the study was overall survival of patients who received both surgery and postoperative systemic chemotherapy compared with those who only received surgery. Cancer-specific survival data were not available.
We first described demographic, tumor, treatment, and survival characteristics by age category. To compare different age groups in receipt of chemotherapy, we used multivariable logistic regression models adjusted for potential confounders. A potential confounder was defined as a variable that was associated with both age group and chemotherapy, including sex, race/ethnicity, tumor location, tumor grade, comorbidities, military branch, marital status, and active-duty status. We then stratified multivariable logistic models by tumor stage at diagnosis. Odds ratios (ORs) and 95% CIs were calculated.
Overall and stage-specific survival was assessed using Kaplan-Meier plots with log-rank test statistics. Multivariable Cox models were used to estimate hazard ratios (HRs) and 95% CIs to assess the effect of treatment on overall survival of patients with colon cancer. The models were adjusted for the same potential confounding factors as were the logistic regression models. Proportional hazards assumptions were assessed using parallel log-log plots and nonsignificant Schoenfeld residuals. Variables failing proportional hazard assumptions were included in stratified survival models. Follow-up began at the date of cancer diagnosis and ended with date of death or date of last contact. Study follow-up was censored at December 31, 2007 (the last date of available linked data). Median length of follow-up was 3.3 years.
Statistical analyses were performed using SAS version 9.3 (SAS Institute Inc). All tests of significance were 2-tailed with α levels set at .05.
Table 1 shows the distributions of demographic variables and tumor characteristics by age category. Patients aged 18 to 49 years were less likely to be male, non-Hispanic white, and have earlier stages at diagnosis as compared with those aged 50 to 64 and 65 to 75 years. Young patients were also more likely to be married and to be active-duty members. Young and middle-aged patients were more likely to have TRICARE Prime and overlapping and unknown tumor locations and less likely to have right-sided tumors than patients aged 65 to 75 years. Young patients had fewer comorbidities than persons aged 50 years and older.
Young patients received more postoperative chemotherapy (n = 465; 69.3%) than middle-aged (n = 849; 53.1%) and older (n = 370; 42.4%) patients. Of those patients who received chemotherapy, young and middle-aged patients had a higher percentage of multiagent regimen use than older patients. However, the percentages of unknown regimen type were high, especially for the older group. Receipt of surgery was similar across all 3 age groups (92.0%, 91.3%, and 92.1%, for young, middle-aged, and older patients, respectively). Survival status was similar for young and middle-aged patients (72.0% and 71.6% alive, respectively) and lower for older patients (61.4% alive).
After adjusting for the potential confounders, young patients were more likely to receive postoperative systemic chemotherapy as compared with older patients across all tumor stages (Table 2). The ORs were 7.98 (95% CI, 2.88-22.11), 4.22 (95% CI, 2.23-7.98), 2.30 (95% CI, 1.01-5.22), and 2.43 (95% CI, 1.26-4.70) for young patients compared with older patients with tumors of stages I, II, III, and IV, respectively. Middle-aged patients with stage I (OR, 5.04; 95% CI, 2.30-11.05) and stage II (OR, 2.42; 95% CI, 1.58-3.72) disease were more likely to receive postoperative chemotherapy compared with older patients.
Table 3 shows the use of multiagent regimens relative to single-agent regimens. After adjusting for potential confounders, younger and middle-aged patients were also more likely to receive multiagent regimens (OR, 2.48; 95% CI, 1.42-4.32 and OR, 2.66; 95% CI, 1.70-4.18, respectively) as compared with patients aged 65 to 75 years. The Figure displays survival by tumor stage. Survival was greatest for young patients and least for older patients. This trend was consistent across all stages.
Table 4 shows survival by tumor stage among patients with surgery alone and those with both surgery and postoperative chemotherapy. While there were too few younger patients to calculate an HR for surgery alone, all HR point estimates were lower than 1 after adjusting for the potential confounders. Among patients treated with surgery alone, a significantly better survival was observed for middle-aged patients with stage I (HR, 0.29; 95% CI, 0.13-0.62) and stage IV (HR, 0.47; 95% CI, 0.22-0.98) disease than their older counterparts. Younger patients with stage III disease who received surgery alone had better survival than older patients (HR, 0.01; 95% CI, 0.01-0.89). Among patients who received surgery and postoperative systemic chemotherapy, no significant differences were observed in survival between age groups. In addition, the HR was not lower for surgery and chemotherapy than the HR for surgery alone, given age group and tumor stage.
Our study examined whether younger and middle-aged patients with colon cancer are overtreated in the MHS, an equal-access health care system. We found that young and middle-aged patients are nearly 2 to 8 times more likely to receive postoperative systemic chemotherapy after surgery compared with patients aged 65 to 75 years, regardless of tumor stage at diagnosis. Furthermore, young and middle-aged adults were 2.5 times more likely to receive multiagent chemotherapy regimens and most patients with information on chemotherapy regimens underwent multiagent regimens, suggesting a tendency toward more intense treatments. While young and middle-aged adults who only underwent surgery had better survival compared with older patients, no significant differences in survival were seen between young/middle-aged and older patients who received surgery plus postoperative systemic chemotherapy after adjustment for the potential confounders. The study suggests that more use of chemotherapy in younger patients did not result in additional survival benefits.
Using the data from an equal-access system, our findings are consistent with the study by Kneuertz et al13 suggesting that the association between age and chemotherapy/survival is similar when access to care and the opportunity to receive treatment are comparable among age groups. Both studies imply the overuse of postoperative chemotherapy in younger patients. According to national guidelines,17,18 select patients with stage II disease, such as those with inadequately sampled nodes, T4 lesions, or poorly differentiated histology, can be considered for adjuvant chemotherapy. However, there has been no solid evidence on the effectiveness of chemotherapy among this subset of patients.17 Our findings, as well as those of Kneuertz et al,13 show higher chemotherapy use as a whole and higher use of multiagent regimens specifically among younger patients compared with older patients for stages I and II as well as stages III and IV. Decision making regarding chemotherapy involves physician recommendation and patient acceptance. Physicians are often reluctant to recommend chemotherapy or more intensive regimens to older patients owing to higher prevalence of comorbid diseases, adverse effects of the treatment, and chronological age itself.31,32 These considerations are less likely to be concerns for younger patients and, thus, physicians may be more likely to recommend chemotherapy. The patient’s decision to receive chemotherapy may also be affected by financial and geographic barriers to care and involvement in the decision-making process,31,33 which may vary by age. A study on colorectal cancer chemotherapy, which had a small number of patients and no adjustment for potential confounders, showed that younger patients preferred more information about the disease and chemotherapy and wanted to be more involved in treatment decisions.33
A higher use of adjuvant chemotherapy among younger patients was demonstrated in our study, as well as in the study by Kneuertz et al.13 Unfortunately, while younger patients were more likely to receive chemotherapy, their survival was not correspondingly improved with this addition, suggesting overtreatment for younger patients. These findings have significant clinical and economic implications. Patients with cancer who receive chemotherapy are vulnerable to its toxicity and adverse effects and may have reduced quality of life. As a result, patients may undergo decreased physical,34,35 functional,34,36,37 emotional,34,35 and social35,37 well-being, although these changes might be mitigated over time.34,35,38 In addition to the effects on patients’ quality of life, overuse of chemotherapy increases economic burdens on society because the cost of chemotherapy for colon cancer is high.39 Therefore, appropriate use of chemotherapy in colon cancer treatment should be investigated and evaluated in further research.
Our study had some limitations. First, some DoD beneficiaries might have non-DoD health insurance, and treatment information might not be complete for these patients because these health care services are not paid for by the DoD and thus not included in our data. Patients aged 65 to 75 years are likely also covered by Medicare. As we did not have access to Medicare data, it is possible that there are missing treatment data for Medicare recipients. However, as shown in Table 1, the older group had a similar percentage of surgery compared with the younger groups, suggesting that the effect of missing data might be limited. Second, there was a higher proportion of unknown regimens for the older group than the younger groups. We do not exclude the possible effect of such missing information, which was differential among age groups, on the receipt of single-agent vs multiagent regimens. Third, sample sizes were small for some stratified analyses. Owing to the limited number of deaths and small sample of younger patients with surgery alone, an HR was not able to be calculated for stage I. Finally, because all-cause death rather than disease-specific death was the outcome of the study, it is possible that some deaths are not attributable to colon cancer. Nevertheless, we adjusted for comorbidities in our analysis.
In an equal-access health care system, we found potential overtreatment of young and middle-aged adults with colon cancer. While younger patients who underwent surgery tended to have better survival than older patients, the addition of postoperative systemic chemotherapy did not result in matched survival improvement. Most of the young patients received postoperative systemic chemotherapy, including multiagent regimens, which are currently not recommended for most patients with early-stage colon cancer. Our findings suggest overtreatment of young and middle-aged adults with colon cancer.
Corresponding Author: Kangmin Zhu, PhD, MD, Division of Military Epidemiology and Population Sciences, John P. Murtha Cancer Center, Walter Reed National Military Medical Center–Bethesda, 11300 Rockville Pike, Ste 1120, Rockville, MD 20852 (email@example.com).
Accepted for Publication: October 16, 2016.
Published Online: January 25, 2017. doi:10.1001/jamasurg.2016.5050
Author Contributions: Dr Zhu and Ms Manjelievskaia had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Manjelievskaia, Brown, Shriver, Zhu.
Acquisition, analysis, or interpretation of data: Manjelievskaia, Brown, McGlynn, Anderson, Zhu.
Drafting of the manuscript: Manjelievskaia, Brown, Zhu.
Critical revision of the manuscript for important intellectual content: Manjelievskaia, McGlynn, Anderson, Shriver, Zhu.
Statistical analysis: Manjelievskaia, Brown, Zhu.
Obtained funding: Shriver.
Administrative, technical, or material support: Shriver.
Conflict of Interest Disclosures: None reported.
Funding/Support: This project was supported by the John P. Murtha Cancer Center, Walter Reed National Military Medical Center via the Uniformed Services University of the Health Sciences under the auspices of the Henry M. Jackson Foundation for the Advancement of Military Medicine and by the intramural research program of the National Cancer Institute. The original data linkage was supported by the former US Military Cancer Institute and Division of Cancer Epidemiology and Genetics, National Cancer Institute.
Role of the Funder/Sponsor: The funding organizations supported procurement/maintenance of the data and the time and efforts spent by the authors on the study. The funding organizations had no direct role in design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Disclaimer: The views expressed in this article are those of the authors and do not necessarily reflect the official views of the Departments of the Navy and Army, the Uniformed Services University of the Health Sciences, the Department of Defense, National Cancer Institute, or the US government. Nothing in the presentation implies any federal/Department of Defense endorsement.
Additional Contributions: We thank those from ICF Macro, Kennel and Associates Inc, the former TRICARE Management Activity, the former Armed Forces Institute of Pathology, the National Cancer Institute, and the former US Military Cancer Institute for their contributions to or support for the original data linkage project. We also thank Stephanie Shao, MS, from John P. Murtha Cancer Center, Walter Reed National Military Medical Center, for her analysis on the number of patients at risk for the Figure. Ms Shao did not receive compensation for her contribution.
Create a personal account or sign in to: