Abstract
Purpose:
Previous SEER (Surveillance, Epidemiology, and End Results)-Medicare analyses have shown no definitive survival benefit for adjuvant chemotherapy (AC) with fluoropyrimidines. Impact of oxaliplatin-containing regimens for elderly stage II patients in real-world setting is unknown. We explored the utilization and outcome of AC after the Food and Drug Administration (FDA) approval of oxaliplatin.
Patients and Methods:
Patients with stage II colon cancer (2004–2011) who underwent resection were selected for this analysis. Medicare claims data were used to ascertain the administration of AC within 120 days after surgery. The primary endpoint of the analysis was overall survival. We used the Cox proportional hazards model to estimate the effect of AC while adjusting for clinical and sociodemographic variables available in SEER. To adjust for referral pattern, a source of selection bias, we conducted an instrumental variable analysis using the surgeon of record and health service area.
Results:
A total of 16,468 patients were identified and 12.1% received AC. AC recipients were significantly younger, more likely to be male, nonwhite, married, and had lower comorbidity index. Their tumors had a more advanced stage, more likely to be left sided, and were less differentiated. The hazard ratio (HR) from the Cox model showed a statistically significant survival advantage for AC (HR = 0.847, 95% confidence interval: 0.782–0.916). However, results from the instrumental variable analysis indicated that there was no definitive benefit of survival in AC recipients (HR = 1.779, 95% confidence interval: 0.927–3.415). AC use decreased over time.
Conclusions:
After controlling for referral patterns, administration of AC provided no definitive survival benefit. Future studies may elucidate the elderly population who may benefit from AC.
Keywords: adjuvant therapy, colon cancer, SEER, elderly, instrumental variable
Colon cancer is the third most common cancer in the United States. In 2019 an estimated 101,420 patients will be diagnosed with colon cancer; of those, about one quarter will have stage II disease.1 The elder subjects represent a large segment of the colon cancer population where >62% of the patients are older than 65 years at the time of diagnosis.2 Although adjuvant chemotherapy (AC) with fluoropyrimidine with or without oxaliplatin is an accepted standard for stage III colon cancer, its role for stage II disease remains controversial.3–5 Despite efforts for risk stratification of stage II patients, no subgroup is found to achieve a definitive benefit from AC.6 Consensus expert opinion recommends AC for stage II patients with high-risk features, including obstruction, perforation, and lymphovascular invasion.7
One of the largest trials that investigated the benefit of AC in patients with stage II colon cancer is the United Kingdom QUASAR (QUick And Simple And Reliable Study).8 The study reported that AC with 5-fluorouracil (5-FU) was associated with an 18% reduction in risk of death, which translated into a small, albeit meaningful 3.6% (95% confidence interval [CI]: 1.0%–6.0%) benefit in 5-year overall survival (OS). MOSAIC and NSABP C-07 established a role for addition of oxaliplatin in the adjuvant setting in patients with stage II and III disease. However, neither trial showed a definitive evidence of benefit for oxaliplatin in patients with stage II disease.4,9
In the United States, 3 SEER (Surveillance, Epidemiology, and End Results)-Medicare based observational studies did not report significant survival improvements for AC for elderly stage II disease.10–12 Moreover, one Swedish observational study also reported that AC compromised the 5-year OS of older patients with stage II disease.13
All these observational studies investigated AC treatment patterns before 2005 and in the era that 5-FU was the only available option. The treatment landscape for colorectal cancer has advanced significantly since 2004, with oxaliplatin and capecitabine added to the armamentarium of drugs for AC.3,9
The objective of this study is to explore the use of adjuvant therapy with or without oxaliplatin and its effectiveness in SEER-Medicare population (elderly patients) in more recent years. We used instrumental analysis accounting for unmeasured confounders in the care delivery patterns that can affect the outcome.
PATIENTS AND METHODS
Database (SEER-Medicare)
The SEER program covers ~34.6% of the US population.14 SEER registries collect data on patient demographics, primary tumor site, tumor morphology, stage at diagnosis, and cause of death. The SEER-Medicare data, links cancer records from SEER with Medicare’s health claims for its beneficiaries from the time of a person’s Medicare eligibility until death.15 Colon cancer incidence increases with age and the median age of diagnosis in the United States is 70 years. More than 62% of patients with colon cancer are older than 65 years at the time of diagnosis.16 Therefore, the SEER-Medicare database can serve as a platform for investigating colon cancer-related treatment patterns.
Cohort Identification
We identified colon cancer patients with stage II, diagnosed between 2004 and 2011, using SEER’s ICD-O-3 (International Classification of Diseases for Oncology, third edition)17 topographical codes C18.0-C18.9. We restricted our cohort to patients with adenocarcinoma histology with SEER’s ICD-O-3 morphologic codes: 8140, 8263, 8480, 8210, 8000, 8261, 8481, 8010, 8490, 8255, 8262, 8211, 8574, 8020, 8560, 8260, 8244, 8221, 8220, and 8144. We distinguished stage II colon cancer by incorporating information about tumor extension, tumor lymph nodes involvement, and tumor metastasis status according to American Joint Committee Cancer (AJCC) staging guideline including pT3, pT4a, pT4b; N0; M0.
We selected patients who were over 65 years at their diagnosis and limited our cohort to patients who had surgery within 120 days of their initial diagnosis using inpatient and outpatient ICD-9 procedure codes (45.7× and 45.8×). We excluded patients who died within 90 days after the surgery as these patients were potentially too ill to receive AC. Only patients covered by Medicare parts A and B during the year before their cancer diagnosis were included in this analysis. Patients with enrollment in Health Maintenance Organization plans do not have claims data and were therefore excluded from the analysis. We also searched for chemotherapy claims before surgery and excluded patients who received chemotherapy within 180 days before surgery (Supplement Fig. 1, Supplemental Digital Content 1, http://links.lww.com/AJCO/A326).
Demographic and tumor-related variables were extracted from SEER data. These include, age at diagnosis, race, sex, marital status, tumor grade, tumor location, number of lymph nodes examined, differentiation grade, poverty level (census tract level data), and registry region. Some of the characteristics are composite and calculated using a set of existing variables. National Cancer Institute (NCI) Comorbidity Index score was calculated using the NCI comorbidity macros.18 We divided stage II patients into 2 subgroups: IIA and IIB/C, following AJCC seventh edition guidelines and stratified our cohort by their cancer location—right side versus left side.
Identification of AC
We defined AC as chemotherapy with 5-FU, capecitabine, with or without oxaliplatin within 120 days from the date of surgery using ICD-9 diagnosis/procedural codes, Healthcare Common Procedure Coding System (HCPCS) codes, and Revenue Center codes in Medicare claims. Patient with claims for chemotherapy drugs, chemotherapy administration, or medical evaluation of chemotherapy in either hospital outpatient facility (Medpar), or physician supplier (carrier) claims within the 120 days after surgery, were labeled as an AC recipient (Supplemental Table 1, Supplemental Digital Content 1, http://links.lww.com/AJCO/A326).
Statistical Analysis
For the descriptive analysis, we used χ2 to compare proportions between recipients of AC versus non-recipients. Among recipients of AC, we identified those who received oxaliplatin-containing regimen and reported their clinical characteristics. Furthermore, we assessed the trends in AC and oxaliplatin use from 2004 to 2011.
The primary outcome of this study was OS, which was defined as the time from 90 days after surgery until death or the end of the observation period (December 31, 2013). We used the Cox proportional hazards model to estimate the OS between patients who received versus those who did not receive AC. We included age at diagnosis, NCI Comorbidity Index score, race, sex, marital status, tumor grade, tumor location, number of lymph nodes examined, differentiation grade, poverty level, and region as covariates.
To determine whether an unmeasured confounding factor was likely to negate the observed result, we calculated the E-value.19,20 The E-value measures whether an unmeasured confounder affects the treatment and outcome, while simultaneously considering the measured covariates.
We used instrumental variable regression analysis, which controls for potential unmeasured confounders in treatment assignment by estimating treatment effects using only the variation in treatment choices determined by variation related to the instrument, analogous to variations that result from randomization.21 The instrumental variable analysis adjusts for the effects of both observable and unobservable characteristics. The key step in conducting instrumental variable analysis is identifying instruments, which significantly affect treatment choice (AC or not) but are not directly related to the health outcome (see Supplemental File for Details of Methodology, Supplemental Digital Content 1, http://links.lww.com/AJCO/A326).
Our instrumental variables were constructed based on the treatment pattern in 2 levels: surgeons and health service areas. We identified the surgeon of record from claims data. We also grouped patients into health service areas that were defined by SEER. Then we calculated the monthly lagged and accumulated proportions of patients who received AC from patients who were treated by the same surgeon on record or in the same health service area. These instrumental variables generate a measure of treatment effects for “marginal patients,” like the patients who were indifferent to choosing between AC or observation, but their actual treatment choices were driven by the treatment preferences from the health service area or the surgeon that treated the patient.22–24
The development of these instruments was based on 2 criteria: First, whether they had a statistically significant impact on the acceptance of AC. Second, whether they balanced the observable characteristics. For the first criteria, we measured the joint significance of instrumental variables from the first-stage regression. For the second criteria, we split our cohort by the median value of each instrumental variable into 2 parts, respectively: above-median group and below-median group. Associations between the below/above group and demographic and clinical characteristics were analyzed. Tests of association were performed with χ2 tests. For details, see Supplement Table 2 (Supplemental Digital Content 1, http://links.lww.com/AJCO/A326).
We plotted the nonparametric survival curves between patients who received AC and those who did not with the methods of Kaplan-Meier. We also generated the predicted survival curves based on a spline-based Royston-Parmar model. We used a covariate-adjusted flexible survival model25 to estimate the predicted average survival probability under 2 scenarios. One scenario is everyone in the population has AC. Another scenario is everyone does not have AC. For each patient in the population, the model assumes every covariate variable (demographic/clinical factor, like age, sex, NCI Index, etc.) took the value of their observed values. We then plotted the difference in survival probability based on the fitted model26 (as shown in Fig. 2B).
FIGURE 2.
A, Kaplan-Meier survival curves in patients with adjuvant chemotherapy (AC) and those without AC without adjustment for any known variables included in Table 2 (age, comorbidity index, sex, tumor location, T stage, tumor differentiation, number of lymph nodes, and year of diagnosis) or referral patterns on the levels of health service areas and surgeons. B, Standardized survival curves (with 95% confidence interval), with adjustment based on a spline-based model with instrumental variable regression analysis (observable variables and referral patterns). The survival curves display the differences in survival outcomes under two counterfactual scenarios: everyone in the population has AC and everyone does not have AC.
All statistical tests were 2 sided and assessed for significance at the 5% level. We used STATA (Stata Statistical Software: Release 15, 2017; StataCorp LLC, College Station, TX) for the Survival analysis and SAS (SAS Institute Inc., Cary, NC) for all other analyses.
RESULTS
We identified 16,468 patients with stage II colon cancer between 2004 and 2011 who met eligibility criteria, 12.1% of them received AC (n=1996). The median age of the cohort was 80 years and the median age of the patients who received AC was 73 years. Compared with patients who did not receive AC, the recipients were statistically significantly younger, more likely to be male, nonwhite, and married. Moreover, they were more likely to have lower comorbidity scores, have left-sided tumors, advanced stage, poorly differentiated or undifferentiated tumors, and fewer lymph nodes examined (Table 1).
TABLE 1.
Demographic and Clinical Characteristics for Patients With Stage II Colon Cancer Diagnosed Between 2004 and 2011 (Deleted P-Value)
| n (%) |
|||
|---|---|---|---|
| Total (N=16,468) | Without Adjuvant Chemotherapy (N=14,472) | With Adjuvant Chemotherapy (N=1996) | |
|
| |||
| Age at diagnosis (y) | |||
| 65–74 | 4890 (29.70) | 3735 (25.80) | 1155 (57.90) |
| 75–85 | 7514 (45.60) | 6755 (46.70) | 759 (38.00) |
| 86+ | 4064 (24.70) | 3982 (27.50) | 82 (4.10) |
| National Cancer Institute Comorbidity Index | |||
| 0 | 7597 (46.10) | 6511 (45.00) | 1086 (54.40) |
| 1–3 | 5993 (36.40) | 5306 (36.70) | 687 (34.40) |
| >3 | 2878 (17.50) | 2655 (18.30) | 223 (11.20) |
| Sex | |||
| Male | 7079 (43.00) | 6100 (42.20) | 979 (49.00) |
| Female | 9389 (57.00) | 8372 (57.80) | 1017 (51.00) |
| Marital status | |||
| Married | 7823 (47.50) | 6624 (45.80) | 1199 (60.10) |
| Single, separated, divorced | 2426 (14.70) | 2094 (14.50) | 332 (16.60) |
| Windowed | 5549 (33.70) | 5145 (35.60) | 404 (20.20) |
| Unknown | 668 (4.10) | 607 (4.20) | 61 (3.10) |
| Race | |||
| White | 14,453 (87.80) | 12,734 (88.00) | 1719 (86.10) |
| Black | 1295 (7.90) | 1132 (7.80) | 163 (8.20) |
| Other and unknown | 720 (4.30) | 606 (4.20) | 114 (5.80) |
| Poverty level | |||
| 0% to <5% poverty | 4105 (24.90) | 3616 (25.00) | 489 (24.50) |
| 5% to <10% poverty | 4432 (26.90) | 3899 (26.90) | 533 (26.70) |
| 10% to <20% poverty | 4710 (28.60) | 4106 (28.40) | 604 (30.30) |
| 20% to 100% poverty | 2927 (17.80) | 2575 (17.80) | 352 (17.60) |
| Unknown | 294 (1.80) | 276 (1.90) | 18 (0.90) |
| Urban/rural | |||
| Big metro | 8689 (52.80) | 7580 (52.40) | 1109 (55.60) |
| Metro | 4752 (28.90) | 4207 (29.10) | 545 (27.30) |
| Urban | 1048 (6.40) | 954 (6.60) | 94 (4.70) |
| Less urban | 1600 (9.70) | 1400 (9.70) | 200 (10.00) |
| Rural | 378 (2.30) | 330 (2.30) | 48 (2.40) |
| Region | |||
| West | 6259 (38.00) | 5576 (38.50) | 683 (34.20) |
| South | 4029 (24.50) | 3507 (24.20) | 522 (26.20) |
| Midwest | 2531 (15.40) | 2187 (15.10) | 344 (17.20) |
| Northeast | 3649 (22.20) | 3202 (22.10) | 447 (22.40) |
| Tumor location | |||
| Left-sided colon cancer | 4949 (30.10) | 4209 (29.10) | 740 (37.10) |
| Right-sided colon cancer | 11,519 (69.90) | 10,263 (70.90) | 1256 (62.90) |
| Tumor stage | |||
| IIA | 14,472 (87.90) | 12,911 (89.20) | 1561 (78.20) |
| IIB | 1114 (6.80) | 910 (6.30) | 204 (10.20) |
| IIC | 882 (5.40) | 651 (4.50) | 231 (11.60) |
| Tumor grade | |||
| Well differentiated | 1208 (7.30) | 1067 (7.40) | 141 (7.10) |
| Moderately differentiated | 11,673 (70.90) | 10,322 (71.30) | 1351 (67.70) |
| Poorly differentiated | 2977 (18.10) | 2559 (17.70) | 418 (20.90) |
| Undifferentiated | 296 (1.80) | 248 (1.70) | 48 (2.40) |
| Not determined | 314 (1.90) | 276 (1.90) | 38 (1.90) |
| Lymph node examined | |||
| < 12 lymph node examined | 5008 (30.40) | 4297 (29.70) | 711 (35.60) |
| ≥ 12 lymph node examined | 11,460 (69.60) | 10,175 (70.30) | 1285 (64.40) |
| Diagnosis year | |||
| 2004 | 2750 (16.70) | 2300 (15.90) | 450 (22.50) |
| 2005 | 2693 (16.40) | 2348 (16.20) | 345 (17.30) |
| 2006 | 2671 (16.20) | 2300 (15.90) | 371 (18.60) |
| 2007 | 2353 (14.30) | 2092 (14.50) | 261 (13.10) |
| 2008 | 2114 (12.80) | 1899 (13.10) | 215 (10.80) |
| 2009 | 1419 (8.60) | 1274 (8.80) | 145 (7.30) |
| 2010 | 1285 (7.80) | 1178 (8.10) | 107 (5.40) |
| 2011 | 1183 (7.20) | 1081 (7.50) | 102 (5.10) |
Among recipients of AC, 43.39% (n=866) received oxaliplatin-containing regimen. Patient characteristics are reported in Table 2.
TABLE 2.
Characteristics of Recipients of Oxaliplatin (Deleted P Values and Some Rows)
| n (%) |
|||
|---|---|---|---|
| Adjuvant Chemotherapy Recipients (N=1996) | Without Oxaliplatin (N=1130) | With Oxaliplatin (N=866) | |
|
| |||
| Age at diagnosis (y) | |||
| 65–74 | 1155 (57.90) | 567 (50.20) | 588 (67.90) |
| 75–85 | 759 (38.00) | 494 (43.70) | 265 (30.60) |
| 86+ | 82 (4.10) | 69 (6.10) | 13 (1.50) |
| National Cancer Institute Comorbidity Index | |||
| 0 | 1086 (54.40) | 589 (52.10) | 497 (57.40) |
| 1–3 | 687 (34.40) | 400 (35.40) | 287 (33.10) |
| >3 | 223 (11.20) | 141 (12.50) | 82 (9.50) |
| Sex | |||
| Male | 979 (49.00) | 563 (49.80) | 416 (48.00) |
| Female | 1017 (51.00) | 567 (50.20) | 450 (52.00) |
| Tumor location | |||
| Left-sided colon cancer | 740 (37.10) | 394 (34.90) | 346 (40.00) |
| Right-sided colon cancer | 1256 (62.90) | 736 (65.10) | 520 (60.00) |
| Tumor stage | |||
| IIA | 1561 (78.20) | 944 (83.50) | 617 (71.20) |
| IIB | 204 (10.20) | 95 (8.40) | 109 (12.60) |
| IIC | 231 (11.60) | 91 (8.10) | 140 (16.20) |
| Tumor grade | |||
| Well differentiated | 141 (7.10) | 79 (7.00) | 62 (7.20) |
| Moderately differentiated | 1351 (67.70) | 773 (68.40) | 578 (66.70) |
| Poorly differentiated | 418 (20.90) | 230 (20.40) | 188 (21.70) |
| Undifferentiated | 48 (2.40) | 27 (2.40) | 21 (2.40) |
| Not determined | 38 (1.90) | 21 (1.90) | 17 (2.00) |
| Lymph node examined | |||
| < 12 lymph node examined | 711 (35.60) | 426 (37.70) | 285 (32.90) |
| ≥ 12 lymph node examined | 1285 (64.40) | 704 (62.30) | 581 (67.10) |
| Diagnosis year | |||
| 2004 | 450 (22.50) | 360 (31.90) | 90 (10.40) |
| 2005 | 345 (17.30) | 224 (19.80) | 121 (14.00) |
| 2006 | 371 (18.60) | 194 (17.20) | 177 (20.40) |
| 2007 | 261 (13.10) | 113 (10.00) | 148 (17.10) |
| 2008 | 215 (10.80) | 99 (8.80) | 116 (13.40) |
| 2009 | 145 (7.30) | 62 (5.50) | 83 (9.60) |
| 2010 | 107 (5.40) | 39 (3.50) | 68 (7.90) |
| 2011 | 102 (5.10) | 39 (3.50) | 63 (7.30) |
Utilization of AC declined over time: from 16.36% in 2004 to 8.62% in 2011 (Fig. 1). However, among patients who received AC, the percentages of receipt of oxaliplatin—increased from 20.00% in 2004 to 61.76% in 2011.
FIGURE 1.
Adjuvant chemotherapy (AC) use among stage II colon cancer patients from 2004 to 2011.
The 3-year OS rate was 80.96% in the AC group versus 73.10% in the untreated group (Fig. 2A). The Cox model indicated that the hazard ratio (HR) of AC while adjusting for SEER variables was 0.846 (95% CI: 0.782–0.916). The point estimator of E-value for AC was 1.49 (upper limit of the CI estimator: 1.32). This means that an unmeasured covariate beyond SEER covariates could bias the AC treatment benefit as measured in COX model with a relative risk association of at least 1.49 with both AC and survival outcomes. For more details on E-value, see Supplement Table 3 (Supplemental Digital Content 1, http://links.lww.com/AJCO/A326).
Using instrumental variables analysis, the estimated risk of mortality trended higher in those who received AC (HR = 1.779, 95% CI: 0.927–3.415). The predicted standard survival curves (Fig. 2B) indicated that the survival improvements due to AC was uncertain. The 95% CI of predicted survival curves for patients who used AC was wide and the upper parts overlapped with the survival curves for patients who did not receive AC. For instrumental variable analysis our sample was not the same as the unadjusted Cox model since the patients who were treated in January 2004 did not have data about previous treatment patterns for either surgeon or health service area (152 patients were removed from the analysis).
The strength of the instrumental variables was examined by their joint significance in the first-stage equation. Our instrumental variables were highly statistically significant in the first-stage equation: χ2 score is 26.73 (P <0.001). Supplemental Table 2 (Supplemental Digital Content 1, http://links.lww.com/AJCO/A326) displayed how our instrumental variables balanced the observed characteristics. Although the χ2 tests indicated that there were still some differences, grouping patients by the value of the instruments, indeed, narrowed the differences in the observed characteristics. Moreover, the residual generated from the first stage was significant (P =0.025) which is equivalent to Hausman tests of endogeneity in the linear setting.
Similar to previously published data8–10,27 we show that female sex, younger age, higher differentiation grade, as well as adequate lymph node evaluation number of lymph nodes confer better prognosis in both Cox and IV models (Table 3).
TABLE 3.
Survival Analysis With Cox Proportional Hazard Model and Instrumental Variable Regression Analysis
| Cox Model (N=16,468) |
Instrumental Variable Analysis+Cox Model (N=16,316) |
|||||
|---|---|---|---|---|---|---|
| HR | P | HR (95% CI) | HR | P | HR (95% CI) | |
|
| ||||||
| Adjuvant chemotherapy | 0.847 | 0.000 | 0.782–0.916 | 1.779 | 0.083 | 0.927–3.415 |
| Age at diagnosis | 1.056 | 0.000 | 1.053–1.060 | 1.064 | 0.000 | 1.056–1.072 |
| NCI Comorbidity Index | 1.207 | 0.000 | 1.194–1.220 | 1.216 | 0.000 | 1.200–1.233 |
| Sex (reference: male) | ||||||
| Female | 0.731 | 0.000 | 0.696–0.767 | 0.729 | 0.000 | 0.692–0.768 |
| Marital (reference: not married) | ||||||
| Married | 0.816 | 0.000 | 0.776–0.857 | 0.799 | 0.000 | 0.757–0.844 |
| Race (reference: nonwhite) | ||||||
| White | 1.042 | 0.249 | 0.972–1.117 | 1.041 | 0.328 | 0.961–1.127 |
| Tumor stage (reference: stage IIB/C) | ||||||
| Stage IIA | 0.640 | 0.000 | 0.601–0.682 | 0.686 | 0.000 | 0.624–0.754 |
| Tumor site (reference: left) | ||||||
| Right | 0.944 | 0.022 | 0.899–0.992 | 0.961 | 0.147 | 0.910–1.014 |
| No. lymph nodes examined: (reference: <12) | ||||||
| ≥ 12 | 0.808 | 0.000 | 0.771–0.846 | 0.825 | 0.000 | 0.781–0.871 |
| Tumor grade (reference: well/moderately) | ||||||
| Poorly/not | 1.084 | 0.004 | 1.026–1.146 | 1.065 | 0.045 | 1.002–1.133 |
| Poverty level (reference: 5%–100% Poverty) | ||||||
| 0%–5% poverty | 0.926 | 0.005 | 0.877–0.978 | 0.925 | 0.006 | 0.876–0.977 |
| Region (reference: West) | ||||||
| South | 1.155 | 0.000 | 1.090–1.223 | 1.152 | 0.000 | 1.086–1.223 |
| Midwest | 1.020 | 0.561 | 0.954–1.091 | 0.995 | 0.882 | 0.927–1.068 |
| Northeast | 1.047 | 0.135 | 0.986–1.112 | 1.031 | 0.340 | 0.968–1.098 |
| Residual term | NA | NA | NA | 0.470 | 0.025 | 0.243–0.910 |
CI indicates confidence interval; HR, hazard ratio; NA, not available; NCI, National Cancer Institute.
DISCUSSION
Our results of the instrumental variable regression analysis suggest that it is unlikely that AC, including oxaliplatin based regimens, provides a survival benefit in the elderly Medicare population with stage II colon cancer. This is despite the fact that recipients of AC were younger and healthier and had higher risk disease. The findings are in line with other published studies, including 3 SEER-Medicare analyses10–12 and a systematic review,28 in the era of 5-FU–based chemotherapy. We further show that despite a decline in the use of AC, the aggressiveness of treatment (defined as the use of oxaliplatin-containing regimen) is increasing.
To our knowledge this is the first instrumental variable analysis based on a large population-based cohort for addressing observed and unobserved confounders in the treatment of stage II colon cancer. We took advantage of the SEER-Medicare database variation in lagged treatment patterns across 187 different health service areas in developing our instrumental variables. Several studies have produced similar instrumental variables that were based on the variation across health service areas.29–31 We also developed one instrumental variable from the surgeon of record. Although surgeons are not the decision maker for AC, their referral pattern to a medical oncologist may influence the treatment delivery. The strong associations between our instrumental variables and treatment choices, indicated the good reliability of our instrumental variables. While in the Cox model the HR of AC was 0.846 (95% CI: 0.782–0.916), The HR in the instrumental variable model was 1.779 (95% CI: 0.927–3.415). This discrepancy is attributable to differences in the area and surgeon level treatment preference controlled by the instrumental analysis.
Our results are in agreement with previous SEER-Medicare studies and provide the advantage of instrumental variable analysis. Schrag et al10 reported that 27% of stage II patients, diagnosed between 1991–1996, received AC with no survival benefit; HR 0.91 (95% CI: 0.77–1.09). Similarly, O’Connor et al11 examined the AC use pattern for those diagnosed between 1992 and 2005. The percentage of AC use among patients without/with poor prognostic features was 19% and 21%, respectively. They also reported no survival benefit for AC in patients with stage II colon cancer without/ with poor prognostic features (HR=1.02; 95% CI: 0.84–1.25 and HR=1.03, 95% CI: 0.94–1.13), respectively. Finally, Weiss et al12 reported that 18% of stage II patients (n=2941) with right-sided cancer and 22% (n=1693) with left-sided cancer, diagnosed 19921995, received AC. However, no survival benefit was observed for those with right-sided (HR=0.97, 95% CI: 0.87–1.09; P =0.64) or left-sided cancer (HR=0.97, 95% CI: 0.84–1.12; P =0.68).
Intriguingly, the usage of adjuvant therapy for stage II over time is declining from 27% in patients diagnosed 1991–1996, to 20% for those diagnosed 1992–2005, and now to 12% for those diagnosed 2004–2011 in our report. This perhaps reflects the better acquaintance with effectiveness and side-effect of AC in elderly patients by oncology community and is similar to the declining use of AC for breast cancer in recent years as reported by Kurian et al.32
The results of our analysis contrast with the QUASAR trial. Several patients and disease-related factors may explain this difference. First, the median age in QUASAR enrollees was 63 (interquartile range [IQR]: 56 to 68), and in our cohort was 79 (IQR: 73 to 84)—the subgroup analysis from the QUASAR trial indicated that patients younger than 70 years were more likely to benefit from AC. Second, is the adequacy of staging, 64% of patients enrolled on the QUASAR had <12 lymph nodes examined, in contrast, 70% of our patients had >12 lymph nodes examined. Inadequately staged patients may be stage III and thus more likely to benefit from chemotherapy.33 Last, the QUASAR trial included patients with rectal cancer, while our cohort was only composed of patients with colon cancer.
Our study is in conflict with a retrospective analysis from the National Cancer Database (NCDB). In this study, 153,110 stage II colon cancer patients that were diagnosed from 1998 to 2006 were identified and adjuvant treatment was associated with improved survival (HR=0.76; P <0.001) adjusted for age.34 The authors did not limit the timing of chemotherapy, so it is possible that some patients with metastatic disease were included in their cohort. In contrast, we only included patients who started adjuvant therapy within 120 days from their date of surgery and the median duration of chemotherapy, based on claims, was 5.1 months (IQR: 2.8 to 5.7). In addition, NCDB data only has the intent for chemotherapy and the number of cycles is not one of their data variables.
SEER-Medicare and NCDB studies were observational and the potential for bias limits the causal relationship between AC and survival outcome. Even though they all employed the propensity score matching to decrease the impact of selection bias, the method can only create matched pairs based on observed characteristics. If unobservable factors were a major source of bias, then bias would remain an issue and propensity score methods would not eliminate the bias.35 In principle, the instrumental variable analysis is more robust than the propensity score matching since it adjusts for both observable and unobservable potential sources of bias.29
Our study has several limitations. First, our sample only included patients who were older than 65 years, which limits the generalizability of our findings to the younger population. Second, we did not have information about patients’ microsatellite instability (MSI) status. MSI is an important factor in treatment decisions for AC in patients with stage II colon cancer, since patients with MSI-high tumors may not benefit from 5-FU AC.36 Third, the instrumental variable analysis does not guarantee that all potential biases has been eliminated.
In conclusion, our study is the first to incorporate the instrumental variable analysis in investigating the effect of AC among patients with stage II colon cancer using SEER-Medicare dataset. We show no definitive survival benefit for AC with fluoropyrimidine with or without oxaliplatin in the elderly patients with stage II disease. Future studies need to focus on the identification of the elderly population with stage II who may benefit from AC.
Supplementary Material
Footnotes
The authors declare no conflicts of interest.
Supplemental Digital Content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website, www.amjclinicaloncology.com.
REFERENCES
- 1.Siegel RL, Miller KD, Jemal A. Cancer statistics, 2019. CA Cancer J Clin. 2019;69:7–34. [DOI] [PubMed] [Google Scholar]
- 2.Howlader N, Noone AM, Krapcho M, et al. SEER Cancer Statistics Review, 1975–2013. Bethesda, MD: National Cancer Institute; 2016. [Google Scholar]
- 3.André T, Boni C, Mounedji-Boudiaf L, et al. Oxaliplatin, fluorouracil, and leucovorin as adjuvant treatment for colon cancer. N Engl J Med. 2004;350:2343–2351. [DOI] [PubMed] [Google Scholar]
- 4.André T, Boni C, Navarro M, et al. Improved overall survival with oxaliplatin, fluorouracil, and leucovorin as adjuvant treatment in stage II or III colon cancer in the MOSAIC trial. J Clin Oncol. 2009;27:3109–3116. [DOI] [PubMed] [Google Scholar]
- 5.Schmoll H-J, Tabernero J, Maroun J, et al. Capecitabine plus oxaliplatin compared with fluorouracil/folinic acid as adjuvant therapy for stage III colon cancer: final results of the NO16968 randomized controlled phase III trial. J Clin Oncol. 2015;33: 3733–3740. [DOI] [PubMed] [Google Scholar]
- 6.Tournigand C, André T, Bonnetain F, et al. Adjuvant therapy with fluorouracil and oxaliplatin in stage II and elderly patients (between ages 70 and 75 years) with colon cancer: subgroup analyses of the Multicenter International Study of Oxaliplatin, Fluorouracil, and Leucovorin in the Adjuvant Treatment of Colon Cancer trial. J Clin Oncol. 2012;30:3353–3360. [DOI] [PubMed] [Google Scholar]
- 7.Benson Iii AB, Schrag D, Somerfield MR, et al. American Society of Clinical Oncology recommendations on adjuvant chemotherapy for stage II colon cancer. J Clin Oncol. 2004;22:3408–3419. [DOI] [PubMed] [Google Scholar]
- 8.Quasar Collaborative Group. Adjuvant chemotherapy versus observation in patients with colorectal cancer: a randomised study. Lancet. 2007;370:2020–2029. [DOI] [PubMed] [Google Scholar]
- 9.Yothers G, O’Connell MJ, Allegra CJ, et al. Oxaliplatin as adjuvant therapy for colon cancer: updated results of NSABP C-07 trial, including survival and subset analyses. J Clin Oncol. 2011;29:3768. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Schrag D, Rifas-Shiman S, Saltz L, et al. Adjuvant chemotherapy use for Medicare beneficiaries with stage II colon cancer. J Clin Oncol. 2002;20:3999–4005. [DOI] [PubMed] [Google Scholar]
- 11.O’Connor ES, Greenblatt DY, LoConte NK, et al. Adjuvant chemotherapy for stage II colon cancer with poor prognostic features. J Clin Oncol. 2011;29:3381. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Weiss JM, Schumacher J, Allen GO, et al. Adjuvant chemotherapy for stage II right-sided and left-sided colon cancer: analysis of SEER-medicare data. Ann Surg Oncol. 2014;21:1781–1791. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Yang L, Ma Q, Yu Y-Y, et al. Efficacy of surgery and adjuvant therapy in older patients with colorectal cancer: a STROBE-compliant article. Medicine. 2014;93:e266. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.National Cancer Institute. Surveillance, Epidemiology, and End Results (SEER) Program. Overview of the SEER Program. 2019. Available at: https://seer.cancer.gov/about/overview.html. [Google Scholar]
- 15.Warren JL, Klabunde CN, Schrag D, et al. Overview of the SEER-Medicare data: content, research applications, and generalizability to the United States elderly population. Med Care. 2002;40:IV-3–18. [DOI] [PubMed] [Google Scholar]
- 16.Howlader NNA, Krapcho M, Miller D, et al. SEER Cancer Statistics Review, 1975–2016, based on November 2018 SEER Data Submission. Bethesda, MD: National Cancer Institute; 2019. [Google Scholar]
- 17.Jack A, Percy C, Shanmugarathan S, et al. International Classification of Diseases for Oncology (3rd edition). Geneva, Switzerland: World Health Organization; 2000. [Google Scholar]
- 18.Klabunde CN, Potosky AL, Legler JM, et al. Development of a comorbidity index using physician claims data. J Clin Epidemiol. 2000;53:1258–1267. [DOI] [PubMed] [Google Scholar]
- 19.Haneuse S, VanderWeele TJ, Arterburn D. Using the E-value to assess the potential effect of unmeasured confounding in observational studies. JAMA. 2019;321:602–603. [DOI] [PubMed] [Google Scholar]
- 20.VanderWeele TJ, Ding P. Sensitivity analysis in observational research: introducing the E-value. Ann Intern Med. 2017;167:268–274. [DOI] [PubMed] [Google Scholar]
- 21.Cameron AC, Trivedi PK. Microeconometrics: Methods and Applications. UK: Cambridge University Press; 2005. [Google Scholar]
- 22.Angrist JD, Imbens GW, Rubin DB. Identification of causal effects using instrumental variables. J Am Stat Assoc. 1996;91:444–455. [Google Scholar]
- 23.Heckman J.Instrumental variables: a study of implicit behavioral assumptions used in making program evaluations. J Hum Resour. 1997;32:441–462. [Google Scholar]
- 24.Harris KM, Remler DK. Who is the marginal patient? understanding instrumental variables estimates of treatment effects. Health Serv Res. 1998;33:1337. [PMC free article] [PubMed] [Google Scholar]
- 25.Royston P, Lambert PC. Flexible Parametric Survival Analysis Using Stata: Beyond the Cox Model. College Station, TX: StataCorp; 2011. [Google Scholar]
- 26.Lambert P: STPM2_STANDSURV: Stata module to obtain standardized survival curves after fitting an stpm2 survival model; 2018. [Google Scholar]
- 27.André T, De Gramont A, Vernerey D, et al. Adjuvant fluorouracil, leucovorin, and oxaliplatin in stage II to III colon cancer: updated 10-year survival and outcomes according to BRAF mutation and mismatch repair status of the MOSAIC study. J Clin Oncol. 2015; 33:4176–4187. [DOI] [PubMed] [Google Scholar]
- 28.Figueredo A, Charette ML, Maroun J, et al. Adjuvant therapy for stage II colon cancer: a systematic review from the Cancer Care Ontario Program in evidence-based care’s gastrointestinal cancer disease site group. J Clin Oncol. 2004;22:3395–3407. [DOI] [PubMed] [Google Scholar]
- 29.Hadley J, Yabroff KR, Barrett MJ, et al. Comparative effectiveness of prostate cancer treatments: evaluating statistical adjustments for confounding in observational data. J Natl Cancer Inst. 2010;102: 1780–1793. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.McDowell BD, Chapman CG, Smith BJ, et al. Pancreatectomy predicts improved survival for pancreatic adenocarcinoma: results of an instrumental variable analysis. Ann Surg. 2015;261:740. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Xu H, Xia Z, Jia X, et al. Primary tumor resection is associated with improved survival in stage IV colorectal cancer: an instrumental variable analysis. Sci Rep. 2015;5:16516. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Kurian AW, Bondarenko I, Jagsi R, et al. Recent trends in chemotherapy use and oncologists’ treatment recommendations for early-stage breast cancer. J Natl Cancer Inst. 2018;110:493–500. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Hutchins G, Southward K, Handley K, et al. Value of mismatch repair, KRAS, and BRAF mutations in predicting recurrence and benefits from chemotherapy in colorectal cancer. J Clin Oncol. 2011;29:1261–1270. [DOI] [PubMed] [Google Scholar]
- 34.Casadaban L, Rauscher G, Aklilu M, et al. Adjuvant chemotherapy is associated with improved survival in patients with stage II colon cancer. Cancer. 2016;122:3277–3287. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.King G, Nielsen R. Why propensity scores should not be used for matching. Political Analysis. 2019;27:435–454. [Google Scholar]
- 36.Sargent DJ, Marsoni S, Monges G, et al. Defective mismatch repair as a predictive marker for lack of efficacy of fluorouracil-based adjuvant therapy in colon cancer. J Clin Oncol. 2010;28:3219. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.