Hospitalization and Survival of Medicare Patients Treated With Carboplatin Plus Paclitaxel or Pemetrexed for Metastatic, Nonsquamous, Non–Small Cell Lung Cancer | Lung Cancer | JAMA Network Open | JAMA Network
[Skip to Navigation]
Sign In
Figure 1.  Cohort Selection for Patients Receiving Chemotherapy for Stage IV, Nonsquamous, Non–Small Cell Lung Cancer
Cohort Selection for Patients Receiving Chemotherapy for Stage IV, Nonsquamous, Non–Small Cell Lung Cancer

aLung cancer diagnosis between 2008 and 2013 not from autopsy or death certificate and without missing diagnosis month for patients without end-stage renal disease.

bPrior diagnosis with localized breast, colon, or prostate cancer was permitted.

Figure 2.  Overall Survival Among Propensity-Matched Patients Receiving Carboplatin-Based Chemotherapy With Pemetrexed or Paclitaxel
Overall Survival Among Propensity-Matched Patients Receiving Carboplatin-Based Chemotherapy With Pemetrexed or Paclitaxel
Table 1.  Characteristics of Patients With Stage IV Lung Cancer Treated With Carboplatin With Paclitaxel or Carboplatin With Pemetrexed
Characteristics of Patients With Stage IV Lung Cancer Treated With Carboplatin With Paclitaxel or Carboplatin With Pemetrexed
Table 2.  Hospitalization and Survival Outcomes in Propensity-Matched Patients Receiving Carboplatin-Based Chemotherapy With Paclitaxel or Pemetrexed
Hospitalization and Survival Outcomes in Propensity-Matched Patients Receiving Carboplatin-Based Chemotherapy With Paclitaxel or Pemetrexed
1.
National Cancer Institute. Common terminology criteria for adverse events (CTCAE) version 5.0. https://ctep.cancer.gov/protocolDevelopment/electronic_applications/docs/CTCAE_v5_Quick_Reference_8.5x11.pdf. Published November 27, 2017. Accessed July 26, 2018.
2.
Manzano  JG, Luo  R, Elting  LS, George  M, Suarez-Almazor  ME.  Patterns and predictors of unplanned hospitalization in a population-based cohort of elderly patients with GI cancer.  J Clin Oncol. 2014;32(31):3527-3533. doi:10.1200/JCO.2014.55.3131PubMedGoogle ScholarCrossref
3.
Brooks  GA, Li  L, Sharma  DB,  et al.  Regional variation in spending and survival for older adults with advanced cancer.  J Natl Cancer Inst. 2013;105(9):634-642. doi:10.1093/jnci/djt025PubMedGoogle ScholarCrossref
4.
Brooks  GA, Li  L, Uno  H, Hassett  MJ, Landon  BE, Schrag  D.  Acute hospital care is the chief driver of regional spending variation in Medicare patients with advanced cancer.  Health Aff (Millwood). 2014;33(10):1793-1800. doi:10.1377/hlthaff.2014.0280PubMedGoogle ScholarCrossref
5.
Clough  JD, Strawbridge  LM, LeBlanc  TW, Hammill  BG, Kamal  AH.  Association of practice-level hospital use with end-of-life outcomes, readmission, and weekend hospitalization among Medicare beneficiaries with cancer.  J Oncol Pract. 2016;12(10):e933-e943. doi:10.1200/JOP.2016.013102PubMedGoogle ScholarCrossref
6.
Brooks  GA, Abrams  TA, Meyerhardt  JA,  et al.  Identification of potentially avoidable hospitalizations in patients with GI cancer.  J Clin Oncol. 2014;32(6):496-503. doi:10.1200/JCO.2013.52.4330PubMedGoogle ScholarCrossref
7.
Barcenas  CH, Niu  J, Zhang  N,  et al.  Risk of hospitalization according to chemotherapy regimen in early-stage breast cancer.  J Clin Oncol. 2014;32(19):2010-2017. doi:10.1200/JCO.2013.49.3676PubMedGoogle ScholarCrossref
8.
Hatfield  LA, Huskamp  HA, Lamont  EB.  Survival and toxicity after cisplatin plus etoposide versus carboplatin plus etoposide for extensive-stage small-cell lung cancer in elderly patients.  J Oncol Pract. 2016;12(7):666-673. doi:10.1200/JOP.2016.012492PubMedGoogle ScholarCrossref
9.
Warren  JL, Klabunde  CN, Schrag  D, Bach  PB, Riley  GF.  Overview of the SEER-Medicare data: content, research applications, and generalizability to the United States elderly population.  Med Care. 2002;40(8)(suppl):IV-3-IV-18.PubMedGoogle Scholar
10.
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(8):732-748. doi:10.1097/00005650-199308000-00006PubMedGoogle ScholarCrossref
11.
Berger  ML, Mamdani  M, Atkins  D, Johnson  ML.  Good research practices for comparative effectiveness research: defining, reporting and interpreting nonrandomized studies of treatment effects using secondary data sources: the ISPOR Good Research Practices for Retrospective Database Analysis Task Force Report—Part I.  Value Health. 2009;12(8):1044-1052. doi:10.1111/j.1524-4733.2009.00600.xPubMedGoogle ScholarCrossref
12.
Pak  K, Uno  H, Kim  DH,  et al.  Interpretability of cancer clinical trial results using restricted mean survival time as an alternative to the hazard ratio.  JAMA Oncol. 2017;3(12):1692-1696. doi:10.1001/jamaoncol.2017.2797PubMedGoogle ScholarCrossref
13.
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(5):373-383. doi:10.1016/0021-9681(87)90171-8PubMedGoogle ScholarCrossref
14.
Klabunde  CN, Potosky  AL, Legler  JM, Warren  JL.  Development of a comorbidity index using physician claims data.  J Clin Epidemiol. 2000;53(12):1258-1267. doi:10.1016/S0895-4356(00)00256-0PubMedGoogle ScholarCrossref
15.
Austin  PC.  An introduction to propensity score methods for reducing the effects of confounding in observational studies.  Multivariate Behav Res. 2011;46(3):399-424. doi:10.1080/00273171.2011.568786PubMedGoogle ScholarCrossref
16.
Austin  PC.  Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples.  Stat Med. 2009;28(25):3083-3107. doi:10.1002/sim.3697PubMedGoogle ScholarCrossref
17.
Austin  PC.  Optimal caliper widths for propensity-score matching when estimating differences in means and differences in proportions in observational studies.  Pharm Stat. 2011;10(2):150-161. doi:10.1002/pst.433PubMedGoogle ScholarCrossref
18.
Uno  H, Tian  L, Cronin  A, Battioui  C. survRM2: comparing restricted mean survival time. R package, version 1.0-2. http://CRAN.R-project.org/package=survRM2. Published 2015. Accessed August 28, 2018.
19.
Tian  L, Uno  H, Horiguchi  M. surv2sampleComp: inference for model-free between-group parameters for censored survival data. R package, version 1.0-5. https://CRAN.R-project.org/package=surv2sampleComp. Published 2017. Accessed August 28, 2018.
20.
Scher  KS, Hurria  A.  Under-representation of older adults in cancer registration trials: known problem, little progress.  J Clin Oncol. 2012;30(17):2036-2038. doi:10.1200/JCO.2012.41.6727PubMedGoogle ScholarCrossref
21.
Reck  M, Rodríguez-Abreu  D, Robinson  AG,  et al; KEYNOTE-024 Investigators.  Pembrolizumab versus chemotherapy for PD-L1-positive non-small-cell lung cancer.  N Engl J Med. 2016;375(19):1823-1833. doi:10.1056/NEJMoa1606774PubMedGoogle ScholarCrossref
22.
Gandhi  L, Rodríguez-Abreu  D, Gadgeel  S,  et al; KEYNOTE-189 Investigators.  Pembrolizumab plus chemotherapy in metastatic non-small-cell lung cancer.  N Engl J Med. 2018;378(22):2078-2092. doi:10.1056/NEJMoa1801005PubMedGoogle ScholarCrossref
23.
Schiller  JH, Harrington  D, Belani  CP,  et al; Eastern Cooperative Oncology Group.  Comparison of four chemotherapy regimens for advanced non-small-cell lung cancer.  N Engl J Med. 2002;346(2):92-98. doi:10.1056/NEJMoa011954PubMedGoogle ScholarCrossref
24.
Scagliotti  GV, Parikh  P, von Pawel  J,  et al.  Phase III study comparing cisplatin plus gemcitabine with cisplatin plus pemetrexed in chemotherapy-naive patients with advanced-stage non-small-cell lung cancer.  J Clin Oncol. 2008;26(21):3543-3551. doi:10.1200/JCO.2007.15.0375PubMedGoogle ScholarCrossref
25.
Gyawali  B, Prasad  V.  Pemetrexed in nonsquamous non-small-cell lung cancer: the billion dollar subgroup analysis.  JAMA Oncol. 2018;4(1):17-18. doi:10.1001/jamaoncol.2017.1944PubMedGoogle ScholarCrossref
26.
Bradley  CJ, Yabroff  KR, Mariotto  AB, Zeruto  C, Tran  Q, Warren  JL.  Antineoplastic treatment of advanced-stage non-small-cell lung cancer: treatment, survival, and spending (2000 to 2011).  J Clin Oncol. 2017;35(5):529-535. doi:10.1200/JCO.2016.69.4166PubMedGoogle ScholarCrossref
27.
Grønberg  BH, Bremnes  RM, Fløtten  O,  et al; Phase III Study by the Norwegian Lung Cancer Study Group.  Phase III study by the Norwegian lung cancer study group: pemetrexed plus carboplatin compared with gemcitabine plus carboplatin as first-line chemotherapy in advanced non-small-cell lung cancer.  J Clin Oncol. 2009;27(19):3217-3224. doi:10.1200/JCO.2008.20.9114PubMedGoogle ScholarCrossref
28.
Patel  JD, Socinski  MA, Garon  EB,  et al.  PointBreak: a randomized phase III study of pemetrexed plus carboplatin and bevacizumab followed by maintenance pemetrexed and bevacizumab versus paclitaxel plus carboplatin and bevacizumab followed by maintenance bevacizumab in patients with stage IIIB or IV nonsquamous non-small-cell lung cancer.  J Clin Oncol. 2013;31(34):4349-4357. doi:10.1200/JCO.2012.47.9626PubMedGoogle ScholarCrossref
29.
Zinner  RG, Obasaju  CK, Spigel  DR,  et al.  PRONOUNCE: randomized, open-label, phase III study of first-line pemetrexed + carboplatin followed by maintenance pemetrexed versus paclitaxel + carboplatin + bevacizumab followed by maintenance bevacizumab in patients ith advanced nonsquamous non-small-cell lung cancer.  J Thorac Oncol. 2015;10(1):134-142. doi:10.1097/JTO.0000000000000366PubMedGoogle ScholarCrossref
30.
Centers for Medicare and Medicaid Services. Medicare Part B drug average sales price: 2018 ASP drug pricing files. https://www.cms.gov/Medicare/Medicare-Fee-for-Service-Part-B-Drugs/McrPartBDrugAvgSalesPrice/2018ASPFiles.html. Accessed April 12, 2018.
31.
Sandler  A, Gray  R, Perry  MC,  et al.  Paclitaxel-carboplatin alone or with bevacizumab for non-small-cell lung cancer.  N Engl J Med. 2006;355(24):2542-2550. doi:10.1056/NEJMoa061884PubMedGoogle ScholarCrossref
32.
Ritzwoller  DP, Carroll  NM, Delate  T,  et al.  Comparative effectiveness of adjunctive bevacizumab for advanced lung cancer: the cancer research network experience.  J Thorac Oncol. 2014;9(5):692-701. doi:10.1097/JTO.0000000000000127PubMedGoogle ScholarCrossref
33.
Langer  CJ, Socinski  MA, Patel  JD,  et al.  Isolating the role of bevacizumab in elderly patients with previously untreated nonsquamous non-small cell lung cancer: secondary analyses of the ECOG 4599 and PointBreak trials.  Am J Clin Oncol. 2016;39(5):441-447. doi:10.1097/COC.0000000000000163PubMedGoogle ScholarCrossref
34.
Zhu  J, Sharma  DB, Gray  SW, Chen  AB, Weeks  JC, Schrag  D.  Carboplatin and paclitaxel with vs without bevacizumab in older patients with advanced non-small cell lung cancer.  JAMA. 2012;307(15):1593-1601. doi:10.1001/jama.2012.454PubMedGoogle ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    Original Investigation
    Oncology
    October 5, 2018

    Hospitalization and Survival of Medicare Patients Treated With Carboplatin Plus Paclitaxel or Pemetrexed for Metastatic, Nonsquamous, Non–Small Cell Lung Cancer

    Author Affiliations
    • 1The Dartmouth Institute for Health Policy and Clinical Practice, Geisel School of Medicine, Lebanon, New Hampshire
    • 2Department of Medicine, Geisel School of Medicine, Lebanon, New Hampshire
    • 3Norris Cotton Cancer Center, Lebanon, New Hampshire
    • 4Division of Population Sciences, Dana-Farber Cancer Institute, Boston, Massachusetts
    JAMA Netw Open. 2018;1(6):e183023. doi:10.1001/jamanetworkopen.2018.3023
    Key Points español 中文 (chinese)

    Question  Does the risk of hospitalization among elderly patients receiving chemotherapy for metastatic, nonsquamous, non–small cell lung cancer vary by chemotherapy regimen?

    Findings  In a cohort study of 2182 propensity-matched Medicare beneficiaries with non–small cell lung cancer, the 30-day hospitalization risk was 5% lower for patients receiving carboplatin with pemetrexed (21%) than for those receiving carboplatin with paclitaxel (26%), a statistically significant difference.

    Meaning  When multiple standard-of-care chemotherapy regimens are available in a specific treatment setting, measures of regimen-specific hospitalization risk may be relevant for treatment selection.

    Abstract

    Importance  Chemotherapy is a mainstay treatment of metastatic non–small cell lung cancer. However, little is known about the comparative risk of hospitalization associated with commonly used chemotherapy regimens.

    Objective  To evaluate the real-world association of specific lung cancer chemotherapy regimens with measures of acute hospital care (primary objective) and survival (secondary objective).

    Design, Setting, and Participants  Retrospective, propensity-matched, cohort study using the Surveillance, Epidemiology, and End Results–Medicare linked data for cancer diagnoses between 2008 and 2013, with follow-up through 2014. Patients were Medicare beneficiaries 66 years of age or older receiving initial chemotherapy for treatment of stage IV, nonsquamous, non–small cell lung cancer. Analyses were performed between September 2017 and April 2018.

    Exposures  Receipt of chemotherapy with carboplatin-pemetrexed or carboplatin-paclitaxel, with or without concurrent bevacizumab.

    Main Outcomes and Measures  The primary outcome was risk of hospitalization within 30 days of chemotherapy initiation. Secondary measures included cumulative 90-day hospitalizations, 90-day mean hospital-free survival time, and overall survival.

    Results  Of the 3310 eligible patients, 1823 received carboplatin-pemetrexed, and 1487 received carboplatin-paclitaxel. The median age at diagnosis was 73 years (interquartile range, 69-77 years), 1784 patients (53.9%) were men, and 2909 patients (87.9%) were non-Hispanic white. In total, 2182 patients were included in the propensity-matched analysis. The 30-day hospitalization risk was 20.7% (95% CI, 18.3%-23.1%) among patients receiving carboplatin-pemetrexed vs 26.0% (95% CI, 23.4%-28.6%) among patients receiving carboplatin-paclitaxel (5.3% difference; P = .003). The 90-day cumulative hospitalizations did not differ significantly between the 2 treatment groups; however, the 90-day mean hospital-free survival time was improved for patients receiving carboplatin-pemetrexed (68.4 days [95% CI, 66.5-70.4 days] vs 63.6 days [95% CI, 61.6-65.7 days]; P = .001). The median survival time was 9.0 months (95% CI, 8.4-9.5 months) with carboplatin-pemetrexed therapy vs 7.6 months (95% CI, 7.0-8.4 months) with carboplatin-paclitaxel therapy (P = .005). Stratified analyses showed no association of bevacizumab use with hospitalization risk or survival when combined with either chemotherapy regimen.

    Conclusions and Relevance  The findings from this study suggest that patients receiving carboplatin-pemetrexed compared with those receiving carboplatin-paclitaxel have a lower 30-day hospitalization risk, a greater 90-day hospital-free survival, and improved overall survival. Information about hospitalization risk may provide valuable context for evaluating real-world cancer treatment outcomes.

    Introduction

    The outcomes of palliative chemotherapy for cancer are commonly assessed in terms of overall survival. The use of survival as a summary outcome measure is necessary but not sufficient for evaluating and describing the effectiveness of palliative-intent cancer therapy because patients value not only survival but also quality of life and avoiding adverse effects. Because clinical trials are typically conducted in populations that are younger and healthier than those encountered in practice, observational data sources—including medical records, tumor registries, administrative claims, and, in particular, the Surveillance, Epidemiology, and End Results (SEER)–Medicare linked database—are an important source of information about the relative benefits and harms of alternative chemotherapy regimens.

    However, claims-based measures of drug toxicity and adverse events are underdeveloped. “Hospitalization indicated” is a general criterion for classifying chemotherapy-related adverse events as grade 3 (severe) using the Common Terminology Criteria for Adverse Events of the National Cancer Institute,1 and there is growing acknowledgment that hospitalization itself can be an adverse effect of cancer treatment. Hospitalization of patients with cancer is common,2,3 variable across regions and practices,4,5 and potentially avoidable.6 Furthermore, prior studies have shown that chemotherapy regimen selection can be associated with hospitalization risk.7,8

    The present study compares the hospitalization and survival rates among Medicare beneficiaries receiving the 2 most common cytotoxic chemotherapy regimens for metastatic, nonsquamous, non–small cell lung cancer: carboplatin with paclitaxel or carboplatin with pemetrexed. This analysis evaluates whether the choice of chemotherapy regimen was associated with hospitalization as captured by 3 alternative metrics. We secondarily evaluated survival among these patients.

    Methods
    Study Design and Data Source

    We performed a retrospective cohort study of Medicare beneficiaries receiving first-line chemotherapy for metastatic, nonsquamous, non–small cell lung cancer with either carboplatin-paclitaxel or carboplatin-pemetrexed (either regimen with or without bevacizumab). The data source for this study was the SEER-Medicare linked data for patients who received a diagnosis of lung cancer between 2008 and 2013 (representing the most current release of the SEER-Medicare linked data at the time of publication, with survival data through December 31, 2014). The SEER program of the National Cancer Institute collects uniformly reported data from population-based cancer registries, including cancer site, stage, diagnosis date, and other clinical variables.9 The SEER program data cover approximately 28% of the US population. Since 1991, the National Cancer Institute has linked SEER data with Medicare administrative data for more than 94% of SEER registry patients 65 years of age or older.10 The linked Medicare data used in this analysis included deidentified fee-for-service claims and event records from inpatient, outpatient, carrier, durable medical equipment, and Part D event files. The research question, study design, and research outcomes (primary and secondary) were all prespecified in the research protocol, as recommended in the International Society for Pharmacoeconomics and Outcomes Research (ISPOR) task force report on good research practices.11 The study was approved by the Dartmouth College institutional review board, which also waived the need for obtaining informed patient consent because the data were deidentified, and was performed under a data use agreement with the National Cancer Institute.

    Patients

    We identified all patients in the SEER-Medicare linked data who had received a new diagnosis of metastatic non–small cell lung cancer between 2008 and 2013 (stage IV as classified by the American Joint Committee on Cancer’s AJCC Cancer Staging Manual, 6th edition). We subsequently excluded patients with a squamous histologic subtype to focus our analysis on patients who were candidates to receive the chemotherapy regimens of interest. We further restricted our sample to patients who were 66 years of age or older at the time of diagnosis and who had continuous enrollment in fee-for-service Medicare (Parts A and B) in the 12 months before and 6 months after cancer diagnosis (or until death) to ensure adequate measurement of baseline comorbidities and posttreatment outcomes. Additional exclusions included prior cancer diagnosis (other than localized breast, colon, or prostate cancer) or concurrent radiation therapy (defined as receipt of radiation therapy on the same day as chemotherapy initiation). The final analytic cohort was restricted to patients receiving first-line treatment with either carboplatin-pemetrexed or carboplatin-paclitaxel (with or without concurrent bevacizumab).

    Exposure and Outcome Definitions

    The main exposure of interest was the initial chemotherapy regimen. We classified regimens by evaluating the first date of receipt of any chemotherapy agent (index date) as captured in Medicare claims from the outpatient, carrier, durable medical equipment, or Part D event files within 90 days of the first day of the diagnosis month. We then identified all chemotherapy claims from within 7 days of the index date and assigned the chemotherapy regimen based on all chemotherapy agents received within this period.

    The prespecified primary outcome measure was 30-day hospitalization risk, defined as the risk of acute inpatient hospital admission within 30 days of the first chemotherapy treatment. Additional hospitalization measures were 90-day cumulative hospitalizations (the count of acute inpatient hospitalizations) and 90-day mean hospital-free survival time (the mean number of days to the first occurrence of either hospitalization or death, restricted to a maximum value of 90 days). The restricted mean survival time has been endorsed as a nonparametric (model-free) measure that is relatively intuitive12; incorporating hospitalizations into the restricted mean survival time provides a framework for evaluating both death and the semicompeting risk of hospitalization within a single measure. We assessed overall survival at 90 days and at the median survival time.

    Statistical Analysis

    We performed descriptive analyses of patient clinical and sociodemographic characteristics, comparing the 2 patient cohorts (patients receiving carboplatin-pemetrexed vs carboplatin-paclitaxel) using 2-sided t tests for continuous variables and χ2 tests for categorical variables. Given observed imbalances in some of these characteristics, we then used propensity score matching to balance the observable baseline characteristics across the cohorts.

    The propensity model evaluated the probability that a patient would receive carboplatin-pemetrexed (vs carboplatin-paclitaxel). Demographic and clinicopathologic variables included in the propensity model were age group, sex, race/ethnicity, marital status, year of cancer diagnosis, Charlson Comorbidity Index (without points for cancer diagnosis, categorized as 0, 1, or ≥2; higher scores indicating a greater number of noncancer comorbid conditions),13,14International Classification of Diseases for Oncology, Third Revision (ICD-O-3) site and histologic codes, receipt of radiation treatment within the prior 30 days, and hospitalization within the prior 180 days. Additional variables included in the propensity model were area-level measures taken from the SEER data, including measures of census tract poverty, urban-rural residence classification, and US geographic region of residence as well as a claims-based indicator for chemotherapy treatment at hospital-based vs office-based practice sites. Further details of the propensity model are presented in Table 1. We then used a nearest-neighbor matching algorithm without replacement to obtain a 1:1 match; matches were constrained within an optimal caliper equal to 0.2 times the SD of the logit of the propensity scores.15-17 The postmatch covariate balance was verified by calculating the standardized difference of mean values for each factor in the 2 groups.

    After propensity score matching, we assessed the study outcomes of 30-day hospitalization risk (primary outcome), 90-day cumulative hospitalizations, and 90-day mean hospital-free survival time. For overall survival, we report both the 90-day overall survival probability and the median overall survival time (as calculated by the Kaplan-Meier method). Main outcome measures are reported with 95% CIs and P values. We also report descriptive characteristics of hospitalizations occurring within 90 days of chemotherapy initiation stratified by chemotherapy regimen, including information about common admission diagnoses and hospital length of stay. We assessed for differences in the mean length of stay with a 2-sample Wilcoxon test. Admission diagnoses were identified using International Classification of Diseases, Ninth Revision, Clinical Modification codes on inpatient hospital claims from the inpatient file, and diagnosis codes were grouped into descriptive categories by one of us (G.A.B).

    Finally, we performed additional analyses to evaluate the association of bevacizumab with 30-day hospitalization risk and overall survival. These stratified analyses were performed within the prematch patient cohort and are reported separately for each of the studied chemotherapy regimens. For the analysis of 30-day hospitalization risk, we fitted multivariable logistic regression models in which the main exposure of interest was receipt of bevacizumab. For the analysis of overall survival, we used Cox proportional hazards models to evaluate the association of bevacizumab use with mortality. Each of these analyses was adjusted for the same covariates included in the previously described propensity model.

    A 2-sided P ≤ .05 was considered statistically significant. Analyses were conducted between September 2017 and April 2018 using Stata, version 15 (StataCorp); R, version 3.1.1 (packages survRM218 and surv2sampleComp19); and SAS, version 9.4 (SAS Institute Inc).

    Results
    Patients

    In total, 3310 patients with stage IV, nonsquamous, non–small cell lung cancer met study inclusion criteria, including 1487 patients receiving carboplatin-paclitaxel and 1823 patients receiving carboplatin-pemetrexed. Details of the cohort selection are shown in Figure 1. The median age at diagnosis was 73 years (interquartile range, 69-77 years), 1784 patients (53.9%) were men, and 2909 patients (87.9%) were non-Hispanic white.

    Prior to propensity score matching, patients receiving carboplatin-pemetrexed and patients receiving carboplatin-paclitaxel were similar by age, sex, race/ethnicity, marital status, Charlson Comorbidity Index, and ICD-O-3 site code (Table 1). However, there were significant differences by diagnosis year (with increasing use of carboplatin-pemetrexed in more recent years) and histology (adenocarcinoma vs large cell carcinoma). Propensity score matching yielded 2182 matched patients (1091 receiving each regimen, a 73% match rate), and matching achieved satisfactory balance across variables included in the propensity model (with absolute standardized differences of less than 0.2 for all strata of variables in the propensity match) (Table 1). Bevacizumab use was more common among patients receiving carboplatin-paclitaxel than among those receiving carboplatin-pemetrexed, both before and after propensity matching (361 [33.1%] vs 244 [22.4%] among propensity-matched patients [P < .001]). Use of bevacizumab was considered to be nonindependent from the choice of chemotherapy regimen (the US Food and Drug Administration approval is specific to use with carboplatin-paclitaxel), and bevacizumab use was not included in the propensity model.

    Hospitalization and Survival by Regimen

    The 30-day hospitalization risk was 20.7% (95% CI, 18.3%-23.1%) among patients receiving carboplatin-pemetrexed and 26.0% (95% CI, 23.4%-28.6%) among patients receiving carboplatin-paclitaxel, corresponding to a 5.3% lower risk of hospitalization with carboplatin-pemetrexed (P = .003). There were also fewer cumulative hospitalizations at 90 days among patients receiving carboplatin-pemetrexed vs carboplatin-paclitaxel (585 vs 647 hospitalizations); however, the difference was not statistically significant (P = .11). The 90-day mean hospital-free survival time was significantly longer among patients receiving carboplatin-pemetrexed than among patients receiving carboplatin-paclitaxel (68.4 days [95% CI, 66.5-70.4 days] vs 63.6 days [95% CI, 61.6-65.7 days]; P = .001). Findings are summarized in Table 2.

    The mean hospital length of stay did not differ between patients receiving carboplatin-pemetrexed and patients receiving carboplatin-paclitaxel (5.7 vs 5.5 days; P = .13). The most common reasons for admission were respiratory tract complications, fever or infection, and cancer (lung cancer or metastasis); admission diagnosis codes from these categories accounted for 40% of all hospitalizations. Category distributions of admission diagnosis codes were similar for both chemotherapy regimens, including similar numbers of hospitalizations for fever or infection (100 hospitalizations of patients receiving carboplatin-paclitaxel vs 73 hospitalizations of patients receiving carboplatin-pemetrexed) and for neutropenia (15 vs 11 hospitalizations).

    Both the 90-day survival probability and the median overall survival were modestly but significantly improved for patients receiving carboplatin-pemetrexed compared with patients receiving carboplatin-paclitaxel (Table 2). The median survival was 9.0 months (95% CI, 8.4-9.5 months) with carboplatin-pemetrexed vs 7.6 months (95% CI, 7.0-8.4 months) with carboplatin-paclitaxel (P = .005); Figure 2 shows Kaplan-Meier curves of overall survival. Hospitalization and survival outcomes from the prematch (crude) cohorts were consistent with the findings from the main propensity-matched analysis (eTable in Supplement 1).

    Association of Bevacizumab With Hospitalization and Survival

    There was no evidence of an association between bevacizumab use and hospitalization among patients receiving either carboplatin-pemetrexed (adjusted odds ratio, 0.88; 95% CI, 0.66-1.18) or carboplatin-paclitaxel (adjusted odds ratio, 0.96; 95% CI, 0.76-1.28). In addition, bevacizumab use was not associated with differences in overall survival among patients receiving either carboplatin-pemetrexed (adjusted hazard ratio, 1.04; 95% CI, 0.93-1.17) or carboplatin-paclitaxel (adjusted hazard ratio, 0.92; 95% CI, 0.82-1.03).

    Discussion

    In this propensity-matched analysis of patients with stage IV, nonsquamous, non–small cell lung cancer, we found that first-line use of carboplatin-pemetrexed compared with carboplatin-paclitaxel was associated with a lower risk of hospitalization and improved survival. We reported a 5.3% reduction in the 30-day hospitalization risk among patients receiving carboplatin-pemetrexed, corresponding to a 20% reduction in hospitalizations in the 30 days following chemotherapy initiation. Treatment with carboplatin-pemetrexed was also associated with significant improvement in the median survival, by 1.4 months (P = .005). These findings are notable because carboplatin-pemetrexed and carboplatin-paclitaxel are the 2 most commonly used cytotoxic chemotherapy regimens for advanced lung cancer, and, to our knowledge, there are no large prospective studies directly comparing these regimens. Our main analysis includes 2182 patients with nonsquamous non–small cell lung cancer with a median age of 73 years, whereas existing prospective studies of each of these regimens are skewed toward younger, more pathologically heterogeneous patient populations.20

    Although the first-line treatment of advanced non–small cell lung cancer is in rapid evolution today,21,22 cytotoxic chemotherapy remains an important part of the lung cancer treatment arsenal. Pembrolizumab monotherapy is now a guideline-recommended first-line therapy for patients with programmed cell death ligand 1 (PD-L1)–positive metastatic lung cancer; however, more than half of patients in this relatively favorable subgroup experience progression within a year or less.21 Platinum-based doublets are the recommended second-line therapy after progression following first-line immunotherapy. For patients with PD-L1–negative lung cancer, combination chemoimmunotherapy with a platinum-based doublet is an emerging standard of care.22 In either setting, platinum-based doublet chemotherapy will continue to play an important role for most patients with metastatic lung cancer, highlighting the ongoing relevance of the findings reported here.

    To place our findings in context, it is helpful to review how carboplatin-pemetrexed and carboplatin-paclitaxel became the predominant cytotoxic chemotherapy regimens for treatment of metastatic, nonsquamous, non–small cell lung cancer. In 2002, a study from the Eastern Cooperative Oncology Group23 showed apparent equivalence of 4 chemotherapy doublets in the treatment of advanced non–small cell lung cancer, with a median survival of approximately 8 months for all regimens. Carboplatin-paclitaxel was the only noncisplatin combination evaluated, and subsequent widespread adoption of carboplatin-paclitaxel was based in part on its favorable toxicity profile. Pemetrexed was not approved for treatment of lung cancer until 2008, when Scagliotti and colleagues reported that the combination of cisplatin and pemetrexed appeared to offer improved survival in comparison with cisplatin and gemcitabine, although only for the nonsquamous histologic subgroup.24,25 Cisplatin remains infrequently used for Medicare patients26 owing to toxicity and patient comorbidity, and a subsequent study27 supported the use of pemetrexed with carboplatin instead of cisplatin. In this way, carboplatin-paclitaxel and carboplatin-pemetrexed became the 2 preferred chemotherapy regimens for nonsquamous non–small cell lung cancer without having been directly compared with each other in a large prospective study. The PointBreak study28 (n = 939) compared these regimens in the context of universal bevacizumab use, and the PRONOUNCE trial29 (n = 361) compared carboplatin-pemetrexed with carboplatin-paclitaxel-bevacizumab, with both studies reporting null results for their primary outcomes. In addition, the various maintenance strategies within these studies and the differences in the use of bevacizumab have left a number of unresolved questions about the comparative efficacy of these regimens.

    A key objective of the present study was to investigate new outcome measures of acute hospitalization applicable to patients receiving chemotherapy. We chose 30-day hospitalization risk as the primary outcome because the period immediately after chemotherapy initiation is known to be high risk for hospital admission and because short-term outcome measures are less likely to be affected by the competing risk of mortality. Our study confirms a startlingly high 30-day hospitalization risk among patients with lung cancer initiating chemotherapy (ie, 20.7% to 26.0% in the propensity-matched cohorts). Another measure evaluated here, the 90-day mean hospital-free survival time, provides insight into the competing risks of hospitalization and death. We contend that these measures, readily extractable from claims data, can serve as valuable proxies for adverse events during chemotherapy. A third hospitalization measure, 90-day cumulative hospitalizations, showed no difference between patients receiving either of the 2 chemotherapy regimens. This is likely because improved survival among patients receiving carboplatin-pemetrexed translated into greater time at risk of hospitalization, leading to a similar number of cumulative hospitalizations with improved hospital-free survival time.

    Overall, we found that carboplatin-pemetrexed therapy was associated with a lower risk of hospitalization and superior survival outcomes compared with carboplatin-paclitaxel therapy—a clear win for carboplatin-pemetrexed therapy from the effectiveness perspective. The relative economic value of these treatments is less clear-cut. Using the average sales price from the fourth quarter of 2017, we found that the drug cost is approximately $6000 for a 3-week cycle of carboplatin-pemetrexed and less than $50 for a 3-week cycle of carboplatin-paclitaxel. Use of bevacizumab at standard doses adds more than $8000 per treatment cycle with either regimen.30 We did not perform a formal cost-effectiveness analysis; however, any anticipated reduction in hospitalizations is very unlikely to offset the incremental cost of substituting pemetrexed for paclitaxel. Still, patients may derive substantial quality-of-life benefit from avoiding hospitalization.

    We did not find an association of bevacizumab use with either hospitalization or survival. The addition of bevacizumab to carboplatin-paclitaxel therapy led to a 2-month survival improvement in a prospective clinical trial,31 and retrospective analyses have confirmed improved survival with bevacizumab use.32,33 However, our finding of no survival advantage with the addition of bevacizumab to chemotherapy is consistent with observational studies that have challenged the utility of bevacizumab for elderly patients.33,34 As immunotherapies become first-line therapies for many patients with advanced non–small cell lung cancer, it may be appropriate to abandon costly and underwhelming vascular endothelial growth factor inhibitor therapies for most elderly patients with lung cancer.

    Strengths and Limitations

    The strengths of our analysis included the use of the linked SEER-Medicare data, which are richly annotated with cancer diagnosis information, including month of diagnosis, cancer stage, and histologic findings. Comprehensive eligibility and claims data provided high confidence that outcomes assessments were comprehensive, and the use of a propensity-matched analytic design provided strong support for the assumption of comparable patient groups. A key limitation of our analysis was that we could not fully adjust the analysis for the use of bevacizumab, an active cancer therapy that was imbalanced across the patient groups. However, our exploratory analysis provided reassurance that bevacizumab use did not appear to be an important determinant of either hospitalization or survival outcomes. Also, we could not exclude the possibility that unmeasured factors were associated with both the choice of chemotherapy regimen and the risk of hospitalization.

    Conclusions

    Treatment with carboplatin-pemetrexed for elderly patients with advanced lung cancer was associated with a 5% absolute reduction in 30-day hospitalization risk compared with that for treatment with carboplatin-paclitaxel. Carboplatin-pemetrexed therapy was also associated with more favorable survival. As immunotherapy supplements and replaces cytotoxic chemotherapy in the first-line treatment of metastatic non–small cell lung cancer, these results remain relevant to first- and second-line treatment selection and are useful for benchmarking important treatment outcomes in the Medicare population.

    Back to top
    Article Information

    Accepted for Publication: August 2, 2018.

    Published: October 5, 2018. doi:10.1001/jamanetworkopen.2018.3023

    Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2018 Brooks GA et al. JAMA Network Open.

    Corresponding Author: Gabriel A. Brooks, MD, MPH, Department of Medicine, Geisel School of Medicine, One Medical Center Drive, Lebanon, NH 03756 (gabriel.a.brooks@hitchcock.org).

    Author Contributions: Dr Brooks had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

    Concept and design: Brooks, Austin, Schrag.

    Acquisition, analysis, or interpretation of data: Brooks, Austin, Uno, Dragnev, Tosteson.

    Drafting of the manuscript: Brooks, Austin, Uno, Schrag.

    Critical revision of the manuscript for important intellectual content: Brooks, Austin, Dragnev, Tosteson, Schrag.

    Statistical analysis: Austin, Uno, Schrag.

    Obtained funding: Brooks, Schrag.

    Administrative, technical, or material support: Tosteson, Schrag.

    Supervision: Tosteson, Schrag.

    Conflict of Interest Disclosures: Dr Brooks reported receiving grants from the Conquer Cancer Foundation. Dr Tosteson reported receiving grants from the American Cancer Society during the conduct of the study. No other disclosures were reported.

    Funding/Support: This study was supported by a Career Development Award from the Cancer Conquer Foundation of the American Society of Clinical Oncology (Dr Brooks).The collection of 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 Surveillance, Epidemiology and End Results (SEER) program from the National Cancer Institute under contract HHSN261201000140C awarded to the Cancer Prevention Institute of California, contract HHSN261201000035C awarded to the University of Southern California, and contract HHSN261201000034C awarded to the Public Health Institute; and a National Program of Cancer Registries (agreement No. U58DP003862-01) from the Centers for Disease Control and Prevention awarded to the California Department of Public Health.

    Role of the Funder/Sponsor: The funders had no role in the 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 interpretation and reporting of these data are the sole responsibility of the authors. The ideas and opinions expressed herein are those of the authors and endorsement by the California Department of Public Health, National Cancer Institute, or Centers for Disease Control and Prevention or their contractors and subcontractors is not intended nor should be inferred.

    Data Sharing Statement: See Supplement 2.

    Additional Contributions: This study used the linked SEER-Medicare database. The authors acknowledge the efforts of the National Cancer Institute; the Office of Research, Development and Information, Centers for Medicare & Medicaid Services; Information Management Services, Inc; and the SEER Program tumor registries in the creation of the SEER-Medicare linked database.

    References
    1.
    National Cancer Institute. Common terminology criteria for adverse events (CTCAE) version 5.0. https://ctep.cancer.gov/protocolDevelopment/electronic_applications/docs/CTCAE_v5_Quick_Reference_8.5x11.pdf. Published November 27, 2017. Accessed July 26, 2018.
    2.
    Manzano  JG, Luo  R, Elting  LS, George  M, Suarez-Almazor  ME.  Patterns and predictors of unplanned hospitalization in a population-based cohort of elderly patients with GI cancer.  J Clin Oncol. 2014;32(31):3527-3533. doi:10.1200/JCO.2014.55.3131PubMedGoogle ScholarCrossref
    3.
    Brooks  GA, Li  L, Sharma  DB,  et al.  Regional variation in spending and survival for older adults with advanced cancer.  J Natl Cancer Inst. 2013;105(9):634-642. doi:10.1093/jnci/djt025PubMedGoogle ScholarCrossref
    4.
    Brooks  GA, Li  L, Uno  H, Hassett  MJ, Landon  BE, Schrag  D.  Acute hospital care is the chief driver of regional spending variation in Medicare patients with advanced cancer.  Health Aff (Millwood). 2014;33(10):1793-1800. doi:10.1377/hlthaff.2014.0280PubMedGoogle ScholarCrossref
    5.
    Clough  JD, Strawbridge  LM, LeBlanc  TW, Hammill  BG, Kamal  AH.  Association of practice-level hospital use with end-of-life outcomes, readmission, and weekend hospitalization among Medicare beneficiaries with cancer.  J Oncol Pract. 2016;12(10):e933-e943. doi:10.1200/JOP.2016.013102PubMedGoogle ScholarCrossref
    6.
    Brooks  GA, Abrams  TA, Meyerhardt  JA,  et al.  Identification of potentially avoidable hospitalizations in patients with GI cancer.  J Clin Oncol. 2014;32(6):496-503. doi:10.1200/JCO.2013.52.4330PubMedGoogle ScholarCrossref
    7.
    Barcenas  CH, Niu  J, Zhang  N,  et al.  Risk of hospitalization according to chemotherapy regimen in early-stage breast cancer.  J Clin Oncol. 2014;32(19):2010-2017. doi:10.1200/JCO.2013.49.3676PubMedGoogle ScholarCrossref
    8.
    Hatfield  LA, Huskamp  HA, Lamont  EB.  Survival and toxicity after cisplatin plus etoposide versus carboplatin plus etoposide for extensive-stage small-cell lung cancer in elderly patients.  J Oncol Pract. 2016;12(7):666-673. doi:10.1200/JOP.2016.012492PubMedGoogle ScholarCrossref
    9.
    Warren  JL, Klabunde  CN, Schrag  D, Bach  PB, Riley  GF.  Overview of the SEER-Medicare data: content, research applications, and generalizability to the United States elderly population.  Med Care. 2002;40(8)(suppl):IV-3-IV-18.PubMedGoogle Scholar
    10.
    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(8):732-748. doi:10.1097/00005650-199308000-00006PubMedGoogle ScholarCrossref
    11.
    Berger  ML, Mamdani  M, Atkins  D, Johnson  ML.  Good research practices for comparative effectiveness research: defining, reporting and interpreting nonrandomized studies of treatment effects using secondary data sources: the ISPOR Good Research Practices for Retrospective Database Analysis Task Force Report—Part I.  Value Health. 2009;12(8):1044-1052. doi:10.1111/j.1524-4733.2009.00600.xPubMedGoogle ScholarCrossref
    12.
    Pak  K, Uno  H, Kim  DH,  et al.  Interpretability of cancer clinical trial results using restricted mean survival time as an alternative to the hazard ratio.  JAMA Oncol. 2017;3(12):1692-1696. doi:10.1001/jamaoncol.2017.2797PubMedGoogle ScholarCrossref
    13.
    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(5):373-383. doi:10.1016/0021-9681(87)90171-8PubMedGoogle ScholarCrossref
    14.
    Klabunde  CN, Potosky  AL, Legler  JM, Warren  JL.  Development of a comorbidity index using physician claims data.  J Clin Epidemiol. 2000;53(12):1258-1267. doi:10.1016/S0895-4356(00)00256-0PubMedGoogle ScholarCrossref
    15.
    Austin  PC.  An introduction to propensity score methods for reducing the effects of confounding in observational studies.  Multivariate Behav Res. 2011;46(3):399-424. doi:10.1080/00273171.2011.568786PubMedGoogle ScholarCrossref
    16.
    Austin  PC.  Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples.  Stat Med. 2009;28(25):3083-3107. doi:10.1002/sim.3697PubMedGoogle ScholarCrossref
    17.
    Austin  PC.  Optimal caliper widths for propensity-score matching when estimating differences in means and differences in proportions in observational studies.  Pharm Stat. 2011;10(2):150-161. doi:10.1002/pst.433PubMedGoogle ScholarCrossref
    18.
    Uno  H, Tian  L, Cronin  A, Battioui  C. survRM2: comparing restricted mean survival time. R package, version 1.0-2. http://CRAN.R-project.org/package=survRM2. Published 2015. Accessed August 28, 2018.
    19.
    Tian  L, Uno  H, Horiguchi  M. surv2sampleComp: inference for model-free between-group parameters for censored survival data. R package, version 1.0-5. https://CRAN.R-project.org/package=surv2sampleComp. Published 2017. Accessed August 28, 2018.
    20.
    Scher  KS, Hurria  A.  Under-representation of older adults in cancer registration trials: known problem, little progress.  J Clin Oncol. 2012;30(17):2036-2038. doi:10.1200/JCO.2012.41.6727PubMedGoogle ScholarCrossref
    21.
    Reck  M, Rodríguez-Abreu  D, Robinson  AG,  et al; KEYNOTE-024 Investigators.  Pembrolizumab versus chemotherapy for PD-L1-positive non-small-cell lung cancer.  N Engl J Med. 2016;375(19):1823-1833. doi:10.1056/NEJMoa1606774PubMedGoogle ScholarCrossref
    22.
    Gandhi  L, Rodríguez-Abreu  D, Gadgeel  S,  et al; KEYNOTE-189 Investigators.  Pembrolizumab plus chemotherapy in metastatic non-small-cell lung cancer.  N Engl J Med. 2018;378(22):2078-2092. doi:10.1056/NEJMoa1801005PubMedGoogle ScholarCrossref
    23.
    Schiller  JH, Harrington  D, Belani  CP,  et al; Eastern Cooperative Oncology Group.  Comparison of four chemotherapy regimens for advanced non-small-cell lung cancer.  N Engl J Med. 2002;346(2):92-98. doi:10.1056/NEJMoa011954PubMedGoogle ScholarCrossref
    24.
    Scagliotti  GV, Parikh  P, von Pawel  J,  et al.  Phase III study comparing cisplatin plus gemcitabine with cisplatin plus pemetrexed in chemotherapy-naive patients with advanced-stage non-small-cell lung cancer.  J Clin Oncol. 2008;26(21):3543-3551. doi:10.1200/JCO.2007.15.0375PubMedGoogle ScholarCrossref
    25.
    Gyawali  B, Prasad  V.  Pemetrexed in nonsquamous non-small-cell lung cancer: the billion dollar subgroup analysis.  JAMA Oncol. 2018;4(1):17-18. doi:10.1001/jamaoncol.2017.1944PubMedGoogle ScholarCrossref
    26.
    Bradley  CJ, Yabroff  KR, Mariotto  AB, Zeruto  C, Tran  Q, Warren  JL.  Antineoplastic treatment of advanced-stage non-small-cell lung cancer: treatment, survival, and spending (2000 to 2011).  J Clin Oncol. 2017;35(5):529-535. doi:10.1200/JCO.2016.69.4166PubMedGoogle ScholarCrossref
    27.
    Grønberg  BH, Bremnes  RM, Fløtten  O,  et al; Phase III Study by the Norwegian Lung Cancer Study Group.  Phase III study by the Norwegian lung cancer study group: pemetrexed plus carboplatin compared with gemcitabine plus carboplatin as first-line chemotherapy in advanced non-small-cell lung cancer.  J Clin Oncol. 2009;27(19):3217-3224. doi:10.1200/JCO.2008.20.9114PubMedGoogle ScholarCrossref
    28.
    Patel  JD, Socinski  MA, Garon  EB,  et al.  PointBreak: a randomized phase III study of pemetrexed plus carboplatin and bevacizumab followed by maintenance pemetrexed and bevacizumab versus paclitaxel plus carboplatin and bevacizumab followed by maintenance bevacizumab in patients with stage IIIB or IV nonsquamous non-small-cell lung cancer.  J Clin Oncol. 2013;31(34):4349-4357. doi:10.1200/JCO.2012.47.9626PubMedGoogle ScholarCrossref
    29.
    Zinner  RG, Obasaju  CK, Spigel  DR,  et al.  PRONOUNCE: randomized, open-label, phase III study of first-line pemetrexed + carboplatin followed by maintenance pemetrexed versus paclitaxel + carboplatin + bevacizumab followed by maintenance bevacizumab in patients ith advanced nonsquamous non-small-cell lung cancer.  J Thorac Oncol. 2015;10(1):134-142. doi:10.1097/JTO.0000000000000366PubMedGoogle ScholarCrossref
    30.
    Centers for Medicare and Medicaid Services. Medicare Part B drug average sales price: 2018 ASP drug pricing files. https://www.cms.gov/Medicare/Medicare-Fee-for-Service-Part-B-Drugs/McrPartBDrugAvgSalesPrice/2018ASPFiles.html. Accessed April 12, 2018.
    31.
    Sandler  A, Gray  R, Perry  MC,  et al.  Paclitaxel-carboplatin alone or with bevacizumab for non-small-cell lung cancer.  N Engl J Med. 2006;355(24):2542-2550. doi:10.1056/NEJMoa061884PubMedGoogle ScholarCrossref
    32.
    Ritzwoller  DP, Carroll  NM, Delate  T,  et al.  Comparative effectiveness of adjunctive bevacizumab for advanced lung cancer: the cancer research network experience.  J Thorac Oncol. 2014;9(5):692-701. doi:10.1097/JTO.0000000000000127PubMedGoogle ScholarCrossref
    33.
    Langer  CJ, Socinski  MA, Patel  JD,  et al.  Isolating the role of bevacizumab in elderly patients with previously untreated nonsquamous non-small cell lung cancer: secondary analyses of the ECOG 4599 and PointBreak trials.  Am J Clin Oncol. 2016;39(5):441-447. doi:10.1097/COC.0000000000000163PubMedGoogle ScholarCrossref
    34.
    Zhu  J, Sharma  DB, Gray  SW, Chen  AB, Weeks  JC, Schrag  D.  Carboplatin and paclitaxel with vs without bevacizumab in older patients with advanced non-small cell lung cancer.  JAMA. 2012;307(15):1593-1601. doi:10.1001/jama.2012.454PubMedGoogle ScholarCrossref
    ×